Response to Comments: MolDX: Molecular Testing for Solid Organ Allograft Rejection

The following comment was submitted to Palmetto GBA and Noridian:

Introduction

On behalf of the International Society for Heart and Lung Transplantation (ISHLT), the leading global multidisciplinary society dedicated to the care of patients with advanced heart and lung disease, we appreciate the opportunity to comment on the proposed Local Coverage Determination (LCD) for Molecular Testing for Solid Organ Allograft Rejection.

ISHLT strongly supports Medicare coverage of validated molecular tests for rejection surveillance in solid organ transplantation. These tools are increasingly embedded in evidence-based guidelines, including ISHLT’s own, and are vital to patient-centered, equitable, and cost-effective transplant care.

Surveillance Frequency Limitations The draft LCD proposes limiting surveillance molecular testing after year one to only two timepoints per year across all organ types. We are concerned that this restriction does not reflect the clinical realities of transplant care and may create barriers to individualized patient management.

• Heart Transplantation: The 2023 ISHLT Guidelines for the Care of Heart Transplant Recipients recommend lifelong monitoring at a transplant center for the possibility of acute and/or chronic rejection. For clinically stable patients, quarterly visits (including labs and biomarker-based assessments) are recommended in years two and three post-transplant, with adjustments for higher-risk patients. Thus, quarterly molecular surveillance (4x/year) aligns with prevailing practice and guidelines.

• Lung Transplantation: ISHLT has not established lung-specific guidelines for molecular surveillance. However, continuity of care is essential, and it often is clinically inappropriate to shift abruptly from intensive monitoring in the first year to only two surveillance assessments thereafter. In practice, patients undergoing molecular surveillance often receive more than two assessments after the first year, reflecting the clinical need for ongoing individualized monitoring. Physicians caring for lung transplant recipients require access to validated tools, including molecular diagnostics, and the discretion to determine frequency of use based on individual patient needs. Restrictive caps risk undermining consistency of care and could compromise patient outcomes.

Recommendation:

Given the statement in the document that “Should national consensus guidelines be generated to demonstrate that a cfDNA (or other molecular biomarker) could improve outcomes at a different schedule or with a different number of testing timepoints comprising the service, these altered schedules would be compliant with this policy without need for amendment as it would fulfill existing coverage requirements listed above.”, we urge MolDX to increase the allowable surveillance frequency to at least four timepoints per year beyond year one for heart transplantation, consistent with ISHLT guidelines¹, and to permit physician discretion for lung transplantation to ensure continuity of care and individualized patient management.

Clinical Utility and Complement to Biopsy

In lung transplantation, a suite of diagnostic tools is utilized by clinicians to monitor the lung allograft giving consideration to cost effectiveness, safety and institutional practices. It is noted that lung biopsy is an invasive procedure with inherent risks, especially noted in some patient cohorts including pneumothorax, bleeding, anesthetic administration and respiratory failure. Consequently, retaining access to frequent molecular testing provides a more nuanced and flexible approach to graft assessment, potentially reducing unnecessary biopsies and risk to some patient groups. Quarterly testing after 12 months aligns with key assessment milestones and clinical review of lung transplant recipients.

Importantly, molecular diagnostics are not used in isolation but as part of a comprehensive care model including imaging, serologies, and clinical evaluation. Limiting access to these tools undermines clinicians’ ability to optimize immunosuppression—balancing rejection risk with avoidance of infection, malignancy, and other drug toxicities. Therefore, we recommend modifying the language related to coverage of testing beyond year one to reflect this, e.g., “After the first year, surveillance timepoints will be limited to a decreased frequency of 4 per year unless other evaluations including but not limited to imaging, serologies and clinical evaluation indicate the presence of increased risk for allograft dysfunction.”

Accessibility and Equity Considerations

Medicare beneficiaries represent a substantial proportion of the U.S. transplant population. Restrictive limits on molecular surveillance risk creating inequities in access to care, particularly for patients in rural or resource-limited settings where biopsy access may be limited or higher risk. Ensuring appropriate test availability is essential to reducing health disparities and aligning with ISHLT’s mission of equitable care worldwide.

Conclusion

ISHLT commends MolDX for recognizing the clinical value of donor-derived cell-free DNA (dd-cfDNA) and gene expression profiling (GEP) in transplant surveillance. To better align policy with current guidelines and patient needs, we respectfully recommend that the final LCD:

1. Increase the post–year one frequency limitation to quarterly testing (≥4 per year) for heart transplantation, consistent with ISHLT guidelines.
2. Permit clinical discretion for lung transplantation to maintain continuity of care and equip physicians with the tools needed to manage individual patient needs.
3. Ensure that molecular testing will be covered beyond the first year when other factors indicative of increased risk of allograft dysfunction are present.
4. Affirm the role of molecular diagnostics as complementary to, and in some cases preferable to, invasive biopsy for long-term patient management.

Reference provided for review.

Thank you for your comments.

Please note that comments from multiple stakeholders that involve the same topics are addressed together here as a single response. Additionally, we have added a qualifying statement to the section on the 2022 CAC meetings, acknowledging the additional evidence that has been developed since then, and have included the relevant evidence in the Summary of Evidence and Analysis sections of the LCD.

1. Surveillance Testing Frequencies

First, in some of the comments received, we observed apparent confusion about the distinction between the for-cause and surveillance indications. This seems to be a significant driver of the requests for an increased frequency of surveillance tests, particularly regarding kidney transplants. We address it here.

First, testing patients with a clinical suspicion of rejection (as defined in the LCD) who require a change in immunosuppression constitutes the for-cause indication. Examples of such scenarios received in the comments were inappropriately provided as a rationale for requesting an increased surveillance testing frequency. Similarly, follow-up testing “in the event that complications arise” (as noted in the comments as another reason for requesting an increased surveillance testing frequency) may also constitute a for-cause indication IF the complications are relevant to the for-cause indication and testing is performed in accordance with the LCD.

For additional clarity on this matter, we have added specificity to the definitions regarding the intended uses of the test and have more clearly delineated the distinction between the for-cause and surveillance indications.

Note also that test frequency limits are not applicable to the for-cause indications. However, it is always expected that the medical record also documents the necessary clinical evidence justifying for-cause testing in compliance with the requirements of the LCD.

Second, regarding physician discretion and physician-directed protocols, the LCD clearly states that “Physicians know best how to manage the care of their patients. However, as a Medicare Administrative Contractor (MAC), we are required by statute to provide coverage for and reimburse services determined to be reasonable and necessary (R&N) according to the published medical evidence.” MACs are not required to pay for anything a physician wants without evidentiary support regarding its clinical validity and utility. Evidence for what is reasonable and necessary (R&N) comes from the peer-reviewed published medical literature. Further, specifying the R&N use cases for a covered service regarding the number of tests, testing type, frequency, and timing is included in the holistic evaluation and determination by the MAC contractors, based on the available evidence. Here, we note that the field of transplantation medicine is rapidly evolving and still working on establishing the optimal frequencies for surveillance testing, the most appropriate tests or combination of tests, and the optimal quantitative threshold(s) for various clinical scenarios. As a result, institutions are using their own variable surveillance protocols, as can be seen in the published literature as well as in the comments received. Finally, the LCD affirms that molecular and proteomic testing for allograft rejection (AR) is covered beyond the first year post-transplantation in accordance with the LCD criteria. However, the testing frequencies for the surveillance indication do have frequency limits and these differ between the first year post-transplantation and later, in accordance with the published evidence. Further, while organ transplant recipients may remain at risk for rejection for the life of their allografts, the CV and CU of the relevant analytes do change over time, and this is an ongoing area of research in allograft rejection monitoring. Further, as the level of risk changes over time, existing surveillance protocols differ greatly between years 1 and subsequent years, with some stopping surveillance altogether by year 3 or 5, depending on organ transplanted. On the basis of the evidence that is available, we have assimilated the current data and have determined the surveillance testing frequencies that are R&N per policy. Below, we provide additional responses to comments received specifically regarding surveillance testing.

1. Heart: For post-operative follow-up visits, the current ISHLT guidelines include biopsy or “other non-invasive methods as appropriate” separately from clinic visits and other routine lab tests in Table 13, Sample: Follow-Up Visits and Test Schedule). Accordingly, there is a discrete schedule outlined for biopsy (or “other non-invasive methods as appropriate”) which does not include any timepoints between months 12 and 24 and includes only 1 per year between years 2 and 5. Further, under Topic 1: Frequency of Routine Tests and Clinic Visits in Heart Transplant Recipients, “blood work” is again listed separately from tests used for surveillance tests for acute rejection and here, those specific surveillance tests are outlined in detail: “Surveillance EMB, and noninvasive rejection monitoring [Gene Expression Profiling (Allomap), DSA, BNP and high sensitivity troponins, donor-derived cell-free DNA].” Specifically, the guidelines state that “post-transplant monitoring for DSA should be minimally performed at 1-, 3-, 6-, and 12 months post-transplant. Thereafter, patients should be monitored annually, except for high-risk patients, who require more frequent surveillance.” Therefore, coupling the schedules for surveillance allograft rejection monitoring and routine visits does not seem consistent with the ISHLT guidelines, though this was referenced in the comments as a rationale for increased surveillance testing.

Further, some programs stop routine rejection surveillance at different times post-transplant. However, given the available evidence, we have expanded the post-year 1 surveillance testing frequency for Heart in the LCD.

2. Lung and Kidney:

We have added evidence and/or modified the testing schedules according to the available literature,

2. Combination testing using dd-cfDNA and GEP for allograft rejection in heart transplantation:

We have added more information regarding the expanded role of GEP in the 2023 ISHLT guidelines, the impact of molecular testing on the number of biopsies performed in the relevant studies, and the more recent analysis published by Moayedi and Teuteberg. We have also more clearly defined the roles of dd-cfDNA and GEP as indicators of graft injury and rejection.

Further, in light of the additional evidence and comments received, we have modified the final LCD regarding the use of combination testing with GEP and dd-cfDNA for heart allograft surveillance.

3. The use of GEP tests in Kidney Transplantation

There was some confusion in the comments received regarding whether the discussion in the LCD around the use of GEP tests was only relevant to heart transplantation.

First, we note that the following statement in the LCD is clearly relevant only to heart transplantation- “Further, while GEP will continue to be covered in heart transplantation as it has become a standard of practice, providers should be aware that they may be performing an inferior service if performed alone.”

We have also included publications that have evaluated GEPs in kidney transplantation.

Finally, the Billing Article includes a Table of tests that have met the criteria of the LCD. Any such GEP tests for use in Kidney transplantation will be included there.

4. The use of multiple molecular tests per patient encounter –

The LCD states that “For a given patient encounter, only one molecular test for assessing allograft status may be performed. A test may include more than one assay; however, multianalyte and combination tests must demonstrate superiority and additive benefit when compared to respective single analytes or components.”

First, if physicians are testing only for one rejection type, it is expected that they use the most appropriate test for that purpose. If they are testing for more than one rejection type, it is expected that they use the most appropriate test for that indication. A multi-analyte or combination test may be the most appropriate test, depending on the specific clinical indication. In this case, the combination test constitutes the test, which is a singular service composed of more than one assay/analyte.

As such, we maintain the LCD criterion regarding the use of one molecular test per patient encounter, as that test service may be comprised of more than one assay.

For added clarity, we also modified the following statement in the Analysis section of the LCD as follows - “Given the limited but evolving evidence, molecular surveillance testing for allograft rejection is considered by this contractor to be a service that may be comprised of multiple assays and timepoints with evidence to support it as R&N.”

5. Testing in Multiple Transplants –

Evidence regarding Multiorgan transplantation (MOT) has been included in the LCD. We additionally note that the Billing Article includes a Table with listed tests that have been demonstrated compliance with policy. Importantly, these have been approved only for the specific indications listed therein.

To avoid confusion regarding MOTs vs prior transplants of the same organ type, we have removed the statement in the Billing Article regarding tests not having been approved in patients who have undergone multiple transplants.

6. Simultaneous Molecular/Proteomic Test and Biopsy –

A non-invasive allograft rejection (AR) test may precede and/or inform the need for biopsy in certain situations, whereas it may obviate the need for biopsy in others. However, a molecular test and biopsy cannot be performed simultaneously, barring the rare exceptional case, as outlined in the LCD.

7. Issues relevant to the accompanying Billing Article –
1. Bundled services – We have revised the Billing Article for clarity and provide instructions for services billed as a series of assays or as a single assay/timepoint, in accordance with the LCD.
2. Table – The table identifying tests that have met LCD criteria is included in the Billing Article.

The following comment was submitted to Palmetto GBA:

I am writing to express strong support for the inclusion of donor-derived cell-free DNA (dd-cfDNA) testing in the proposed MolDX Local Coverage Determination (LCD) for patients who have undergone Simultaneous Pancreas-Kidney (SPK) transplantation.

Since 2022, dd-cfDNA has been clinically validated and increasingly adopted as a non-invasive biomarker for monitoring allograft health and detecting acute rejection in SPK recipients. A growing body of peer-reviewed evidence demonstrates that dd-cfDNA performs with high sensitivity and specificity, comparable to its established utility in kidney-only transplantation.

Notably:

• Ventura-Aguiar et al. (2022) showed that dd-cfDNA levels correlate strongly with pancreas graft rejection, offering a sensitive and reliable diagnostic tool for SPK patients.
• Williams et al. (2022) reported early clinical experience using dd-cfDNA for surveillance post-SPK, highlighting its ability to distinguish between rejection, injury, and quiescence.
• Yoo et al. (2023) provided baseline dd-cfDNA data from high-volume U.S. centers, reinforcing its clinical applicability and reproducibility.
• Liu et al. (2024) conducted a prospective study demonstrating that dd-cfDNA levels significantly rise in cases of biopsy-proven acute rejection, with an optimal cutoff of 1.81% yielding an AUC of 0.86, sensitivity of 95.7%, and specificity of 57.1%.

These findings underscore dd-cfDNA’s value as a dynamic, donor-specific biomarker that enables early detection of rejection, reduces reliance on invasive biopsies, and supports timely clinical intervention. Given the complexity and risk profile of SPK transplantation, the availability of a validated, non-invasive surveillance tool is not only clinically advantageous but essential for improving patient outcomes.

In light of this robust evidence base, I urge MolDX to include dd-cfDNA testing in its LCD for SPK transplant recipients. Doing so will align coverage policy with current clinical standards and ensure equitable access to a proven tool for transplant monitoring.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA, CGS, and Noridian:

The American College of Cardiology (ACC) appreciates the opportunity to comment on the proposed update to the Local Coverage Determination (LCD) regarding molecular testing for solid organ allograft rejection. The ACC is a global leader dedicated to transforming cardiovascular care and improving heart health for all. For more than 75 years, the ACC has empowered a community of over 60,000 cardiovascular professionals across more than 140 countries with cutting-edge education and advocacy, rigorous professional credentials, and trusted clinical guidance. From its world-class JACC Journals and NCDR registries to its Accreditation Services, global network of Chapters and Sections, and CardioSmart patient initiatives, the College is committed to creating a world where science, knowledge and innovation optimize patient care and outcomes. Learn more at www.ACC.org or connect on social media at @ACCinTouch.

We respectfully offer the following comments and suggested clarifications:

Surveillance Frequency Beyond Year One

The proposed LCD states that “after the first year, surveillance timepoints may continue at a decreased frequency of 2 per year.” We recommend revising this to reflect the variability in clinical practice and evidence from registries such as the Surveillance HeartCare^® Outcomes Registry (SHORE), which used quarterly surveillance in years 2 and 3, followed by biannual testing. Many centers continue surveillance every 3 months during the second year, recognizing it as a particularly fragile period. We suggest the language be revised to:

“After the first year, surveillance timepoints may continue at a frequency of 2–4 per year, or more frequently if abnormalities develop.”

Simultaneous Testing and Biopsy

The current language states that “a test and biopsy cannot be performed simultaneously.” We recommend changing “cannot” to “should not” to allow for rare but clinically justified exceptions. Specifically, we propose:

“A test and biopsy should not be performed simultaneously for surveillance. However, in very high-risk patients with overt signs and symptoms of rejection, a molecular test and biopsy may be performed together in urgent situations ONLY when the information derived from the test is complementary to the biopsy, established to meet CV requirements outlined above for that information, and demonstrated to improve outcomes in patients when performed along with biopsies. We expect these situations to be extremely rare.”

These clarifications would better reflect real-world clinical practice and ensure that the policy supports appropriate, patient-centered care without introducing unnecessary restrictions.

The ACC appreciates the opportunity to contribute comments and values the Medicare Administrative Contractor’s effort to support evidence-based use of molecular diagnostics in transplant care.

Reference was provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

I have reviewed the draft policy and would like to offer comment.

The International Society for Heart and Lung Transplantation (ISHLT) guidelines for the care of heart transplant recipients were established based on a consensus of published clinicians in the transplant space, both nationally and internationally.

As it relates to the ISHLT guidelines and GEP, the acknowledgement in the draft policy only focuses on one sentence “that the ISHLT guidelines affirm that the impact of GEP-guided surveillance on long-term clinical outcomes in heart transplantation still needs further study.” However, the policy completely omits the two sections where GEP/AlloMap is acknowledged in its explicit recommendations:¹

Gene Expression Profiling (GEP) (i.e., Allomap) of peripheral blood can be used in low-risk patients between 2 months and 5 years after HT to identify adult recipients who have low risk of current ACR to reduce the frequency of EMB. Data in children does not allow a general recommendation of GEP as a routine tool at present (Class IIa, Level of Evidence: B)

The purpose of the follow-up visits is to monitor for rejection and screen for adverse events and may include the following: (1) a complete physical examination; (2) review of medications and changes to medications based on the results of the examinations; (3) blood work; (4) echocardiogram; (5) coronary angiography. Adjunct Intravascular imaging can be considered if expertise available, as Maximal Intimal Thickening (MIT) > 0.3 mm in the first year has been shown to have prognostic value; (6) Surveillance EMB, and noninvasive rejection monitoring [Gene Expression Profiling (Allomap), DSA, BNP and high sensitivity troponins, donor-derived cell-free DNA] (7) additional education and/or interaction with members of the multi-disciplinary team. An example of a follow-up visits and test schedule is presented in Table 13 (below). (Class I, Level of Evidence B)

Importantly, both recommendations expand the role of GEP, compared to the prior 2010 guidelines in which GEP was recommended for patients beginning at 6 months and was not acknowledged as part of the routine follow-up of patients. Even more critically, the draft policy omits the two publications on which the recommended expanded role of GEP was based, including a randomized controlled clinical utility study demonstrating that GEP could safely replace surveillance biopsies as early as 2 months post-transplant and could be used to adjust immunosuppression.^2,3 To date, GEP is the only molecular test that currently has such data available. This should be acknowledged in the draft policy.

With respect to combination testing, it is important to note that the first SHORE manuscript was not available at the time the guidelines were published. The description of the significantly improved specificity for combination testing in the SHORE manuscript is logical, given that the tests are not redundant, and appears congruent with clinical practice.⁴ A negative GEP can be leveraged to avoid a biopsy in patients with an elevated dd-cfDNA. An elevated GEP can indicate a biopsy in patients with a normal dd-cfDNA but a higher pre-test probability for rejection, a strategy that is supported by GEP’s higher sensitivity of ACR.^4,5 Both dd-cfDNA and GEP are supported by evidence for use in the complex care of heart transplant recipients, and clinicians should determine how best to integrate these tests into clinical practice.

References were provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

Thank you for the opportunity to offer my comments on the draft policy. I am a transplant cardiologist and the medical director of the heart transplant program. I believe the draft policy does not accurately reflect the current complementary roles of gene expression profiling (GEP) and donor-derived cell-free DNA (dd-cfDNA) and the literature that supports this.

GEP was developed at a time when antibody medicated rejection (AMR) was not recognized as a clinical entity within cardiac transplantation and as such, is a tool used to assess for the risk of acute cellular rejection (ACR) by measuring the recipient’s immune system status. Dd-cfDNA is a marker of graft injury and thus is often elevated in the context of ACR or AMR, though I note that current data supports it has better sensitivity for AMR then ACR.¹ Moreover, both tests have imperfect specificity for biopsy evidence of rejection. However, given their orthogonality, the specificity for ACR is much improved when both tests are abnormal, as elegantly showed in the first SHORE publication.²

The policy seems to want to simplify cardiac surveillance by allowing only one molecular at a time. And while this would potentially simplify test interpretation, it would be a step away from personalized medicine and would almost certainly trigger more unnecessary biopsies, as a single test has both poorer specificity and a higher positivity rate.²

Here, we have leaned into the complexity, but also the richness of the data to create an evidenced based detailed approach to dual test interpretation (protocol attached) that leverages the complementary nature of these tests to make sure that only patients who truly need an invasive test, receive one.

I hope that you will reconsider your position on combination testing and encourage clinicians to take full advantage of the innovative research that has occurred in this space over the last 20 years which has finally gained traction in moving away from the archaic practice of routine surveillance biopsies.

References and image were provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA, CGS, Noridian, and WPS:

Thank you for the opportunity to submit these comments in response to the draft Local Coverage Determination (dLCD) entitled MolDX: Molecular Testing for Solid Organ Allograft Rejection. The Association for Molecular Pathology (AMP) is an international medical and professional association representing approximately 3,100 physicians, doctoral scientists, and medical technologists involved with laboratory testing based on knowledge derived from molecular biology, microbiology, genetics, and genomics. Our membership includes experts in molecular testing from the academic medicine, private and hospital-based clinical laboratories, the government, and the in vitro diagnostics industry that work to ensure coverage policies meet the highest standard of care and allow patients access to needed molecular testing.

AMP thanks MolDX for the inclusion of coverage for molecular tests for solid organ allograft rejection. We believe providing a noninvasive method for assessing organ rejection is a critical component of transplant care and are pleased that this dLCD provides coverage for this essential service. To improve compliance with the dLCD, we encourage MolDX to provide clarification on two aspects in the final LCD to ensure that the document provides a clear, comprehensive, and accurate representation of the coverage limitations.

Coverage Indications, Limitations and/or Medical Necessity

The acronym for Donor-Derived Cell-Free DNA within the definitions provided in this policy is inconsistent with scientific and clinical terminology. The acronym cfDNA is widely recognized as the abbreviation for cell-free DNA in general and is not specific to the donor-derived version of cell-free DNA. Using cfDNA to represent Donor-Derived Cell-Free DNA introduces ambiguity and may cause confusion regarding the scope of coverage for relevant testing. Instead, AMP recommends MolDX use the widely accepted abbreviation dd-cfDNA for Donor-Derived Cell-Free DNA, as used by organizations such as the American Society of Transplant Surgeons and the National Institutes of Health¹.

While AMP has interpreted the dLCD to have positive coverage of molecular testing for Solid Organ Allograft Rejection, ambiguity in the coverage or non-coverage of dd-cfDNA and Gene Expression Profiling (GEP) testing for transplant rejection remains. For instance, the Summary of Evidence includes the following contrasting statements:

“In heart transplantation, molecular biomarker testing has been used for over a decade and, unlike the outdated Banff guidelines for kidney transplantation (at the time of this work), the ISHLT guidelines do recommend the use of molecular biomarker testing in the evaluation of heart transplant recipients.”

“Though GEP will continue to be covered in heart transplantation only as it has become a standard of practice, providers should be aware that they may be performing an inferior service.”

AMP is concerned that this ambiguity could inadvertently restrict patient access to these potentially life-saving diagnostics in situations where coverage might be applicable. To prevent such confusion, AMP requests that the document be revised to specify the coverage scenarios for dd-cfDNA and GEP testing in the Coverage Indications, Limitations, and/or Medical Necessity section. This would help ensure that both patients and providers have a clear understanding of when these tests are covered, thereby improving access and supporting better clinical decision-making.

The Association for Molecular Pathology thanks MolDX for the opportunity to comment on MolDX: Molecular Testing for Solid Organ Allograft Rejection.

Reference was provided for review.

Thank you for your comment. Please see Response #1.

Additionally, we have revised “cfDNA” to “dd-cfDNA” in the LCD.

The following comment was submitted to Palmetto GBA:

DdcfDNA testing for surveillance has become a useful tool in our toolbox for taking care of transplant patients and to a large extent has become standard of practice.

Given current evidence that dd-cfDNA significantly outperforms serum creatinine in identifying subclinical rejection and allograft dysfunction (Graver AS, Lee D, Power DA, Whitlam JB. Understanding Donor-derived Cell-free DNA in Kidney Transplantation: An Overview and Case-based Guide for Clinicians. Transplantation. 2023 Aug 1;107(8):1675-1686. doi: 10.1097/TP.0000000000004482. Epub 2022 Dec 29. PMID: 36579675.), it is clinically appropriate to assess dd-cfDNA levels at more frequent timepoints that align more closely with serum creatinine utilization, at least 4 times a year even after the first year. Therefore we would recommend at least 4X/year surveillance testing in first year and beyond.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA:

Eurofins U.S. Clinical Diagnostics, including Eurofins Transplant Genomics, is committed to improving patient outcomes through advanced diagnostic technologies. Our tests inform clinical decision-making and enhance patient care for transplant recipients, as well as for those with infectious diseases, cardiovascular disease, immune disorders and allergies. Our parent company, Eurofins Scientific (EUR.PA) is the global leader in food, environment, pharmaceutical and cosmetic product testing, discovery pharmacology, forensics, advanced material sciences, and agroscience contract research services.

For transplant patients, Eurofins Transplant Genomics provides innovative cfDNA (TRAC^®) and gene expression profile (TruGraf^®) tests. The use of Eurofins Transplant Genomics’ tests, especially when used in combination, has been demonstrated to detect, with higher positive and negative predictive values than either test alone, the likelihood of allograft rejection. These advancements improve the ability of clinicians to identify patients with early, pre-clinical rejection and thus hold the potential to save countless kidney allografts. The benefits of these innovations, with a lengthy history of Medicare (MolDX) coverage, could significantly improve the current standard of care, reduce unnecessary and countless rejections, and improve the quality of life for transplant patients.

We appreciate the opportunity to comment on MolDX’s recently proposed local coverage determination (LCD) for Molecular Testing for Solid Organ Allograft Rejection and its associated Billing and Coding Article. Our comments aim to support long-term graft success, enhance patient monitoring, and ensure that physicians can apply medical judgment in delivering care. We respectfully submit the following feedback for your consideration.

Recommendation #1: Surveillance Testing Frequency Post-Transplant

Laboratory testing is an essential component of every nephrology visit, serving as a critical tool for evaluating and managing transplant recipients. In the absence of a clear clinical guideline for monitoring beyond two years post-transplant, it is concerning that the proposed policy would arbitrarily limit coverage to just two molecular tests in year two and none thereafter. This limitation does not reflect the individualized nature of post-transplant care, where certain patients – particularly those at increased risk for rejection or infection – may require more frequent testing to ensure the safe and effective management of immunosuppression. For example, 30% and 12% of kidney transplant patients develop BK viruria and viremia, respectively, often requiring alterations of their immunosuppression (Viruses 2022, 14(8), 1616; https://doi.org/10.3390/v14081616). While many patients may not need additional testing, those who do could face compromised care if access to molecular biomarkers is unnecessarily restricted.

As noted in the proposed LCD, “The importance of graft rejection and immunosuppression was discovered early on following the development of transplantation … though with the current standard-of-care for managing solid organ transplant patients, rejection remains a common problem with a high frequency of graft failure at 5 and 10 years.” If rejection continues to be a significant concern even under standard-of-care protocols, it is reasonable to question why non-invasive molecular assays would not be covered under physician-directed protocols based on medical necessity.

In contrast to invasive biopsies – which decrease in frequency over time due to safety concerns – non-invasive molecular diagnostics offer a non-invasive means for routine surveillance, are designed to inform the need for biopsy, and may result in fewer biopsies over time. These innovations enable earlier detection and more proactive management of complications without subjecting patients to higher-risk procedures. Restricting access to such tools could inadvertently undermine the very goals of improving long-term graft survival and patient outcomes.

Due to the absence of guidelines for protocol biopsy use in kidney transplant patients past the first year post-transplant, we recommend revising the proposed surveillance testing limits to be based solely on medical necessity as determined by clinical need, or reflecting the only guideline that currently exists which is in first year post transplant and is currently 4 times per year, quarterly. This approach better supports risk-based, physician-directed care. Patients with heightened rejection risk – due to prior rejection, infection, or immunologic complexity – may require more frequent testing to ensure safe and effective immunosuppressive management.

Specifically, we recommend revising the LCD language as follows:

As Proposed
“After the first year, surveillance timepoints may continue at a decreased frequency of 2 per year.”

Suggested Revision
“A maximum of 4 surveillance timepoints within any year, per year, post transplantation. Additional molecular tests may be ordered when clinically indicated and deemed medically necessary by the treating clinician.”

Modifying the proposed coverage limitations to create a clinician-directed exception to the proposed limits on the number of covered tests would be fully consistent with chapter 13, section 13.5.4 of the Program Integrity Manual, which requires MACs to consider whether a service has been “ordered… by a qualified professional” and “meets, but does not exceed, the patient’s medical need” when determining whether a service is “reasonable and necessary.”

Recommendation #2: Use of Multiple Molecular Tests Per Encounter

We respectfully request clarification and revision of the restriction limiting physicians to one molecular test per patient encounter. This limitation may unintentionally prevent physicians from using complementary assays that provide distinct, clinically valuable information. For example, cfDNA assays and gene expression profile tests are often used together to detect different types of rejection – antibody-mediated and T-cell mediated, respectively. Limiting their combined use could delay diagnosis or reduce the accuracy of patient assessments, especially in cases of mixed rejection. It has been shown that cfDNA would miss 80% of T-cell mediated rejection and, conversely, when using gene expression tests, it has been shown that testing could miss some cases of antibody mediated rejection. (Clin J Am Soc Nephrol. 2021 Oct;16(10):1539-1551. doi: 10.2215/CJN.05530421 https://pubmed.ncbi.nlm.nih.gov/34620649/)

The current draft language states that multianalyte tests must “demonstrate superiority” to single analytes but does not define how such superiority is determined. We recommend clarifying whether this means demonstration through peer-reviewed literature (e.g., improved NPV, sensitivity, or specificity), technical assessments by MolDX, or another metric.

To this end, we urge MolDX to permit the use of multiple assays when each is directed at detecting different forms of rejection and when supported by clinical judgment.

As Proposed

“… multianalyte and combination tests must demonstrate superiority and additive benefit when compared to respective single analytes or components.”

Suggested Revision

“… multianalyte and combination tests must demonstrate superiority and additive benefit when compared to respective single analytes or components, except when surveilling for different rejection types or mixed rejection.”

Recommendation #3: Bundling and Billing Clarity

We seek additional clarification regarding the application of bundling in the proposed LCD. Specifically, it is unclear whether all bundled tests must be completed before billing and whether the pricing of bundles aligns with the CLFS and CPT code standards. This ambiguity could result in billing delays or discourage the ordering of medically necessary tests.

To improve clarity, we recommend at a minimum stating: “The price of each bundle is equal to the product of the number of timepoints determined to be reasonable and necessary according to the LCD criteria and the CLFS rate for the applicable CPT code.”

That said, the Social Security Act requires the Centers for Medicare and Medicaid Services (CMS) to establish a fee schedule for clinical diagnostic laboratory tests. Social Security Act § 1833(h)(1). CMS sets initial CLFS rates for tests described by new or substantially revised codes via crosswalk or gapfill. 42 C.F.R. § 414.508(b). Then, once private payer rate information is reported to CMS (as required under the Protecting Access to Medicare Act (PAMA)), CMS updates the rates for those codes to reflect the volume-weighted median of reported data. 42 C.F.R. § 414.507. The Social Security Act prohibits CMS from adjusting nationally-established CLFS rates for any reason, except as outlined above.

We thus suggest eliminating the bundling construct entirely and allowing laboratories to bill for each covered test using established CPT codes and paid at established CLFS rates. To the extent that MolDX chooses to retain the bundling construct, we request that (i) the language describing the application of it to be broadened for clarity and (ii) specific examples be used to illustrate how the construct would work in practice.

Recommendation #4: Clarifying Coverage of GEP in Kidney Transplant Patients

We recommend clarifying that Gene Expression Profiling (GEP) remain covered for solid organ transplants. For example, TruGraf is a GEP test approved for stable kidney transplant patients and is used to assess immune quiescence. As proposed, the LCD statement that “GEP will continue to be covered in heart transplantation only” suggests a potential change in coverage for GEP and could cause confusion for clinicians and patients on whether Medicare covers these tests.

This underscores a broader concern: the draft LCD differentiates molecular assays by purpose rather than the type of assay. In practice, the clinical use cases for cfDNA surveillance differ substantially from those for GEP surveillance. MolDX’s previous table with the context of use specified for each approved test provided valuable, assay-specific guidance to clinicians, helping them match the appropriate molecular test to patient need. In contrast, MolDX’s updated table indicates a test description and broad categories of use, signaling coverability but not differential clinical utility. This shift has caused confusion for physicians and may inadvertently limit optimal test selection for patient care.

Conclusion

In summary, Eurofins encourages MolDX to revise the proposed LCD to:

• Restore the context of use to MolDX’s table for approved tests;
• Allow surveillance testing beyond year one for kidney transplant patients;
• Permit the use of multiple complementary molecular assays when clinically indicated;
• Eliminate the bundling concept or clarify bundling and billing guidance to align with CPT and CLFS practices;
• Clarify that GEP surveillance tests continue to be covered where already approved in solid organ transplant, including for kidney;
• Ensure that medical necessity, as determined by the treating physician, guides test utilization.

These recommendations are intended to protect transplant recipients and ensure that innovation in diagnostic testing continues to support better patient outcomes and long-term graft survival.

Thank you for your comment. Please see Response #1.

Additionally –

The LCD states that “Clinical validity (CV) of any analytes (or expression profiles) measured must be established through a study published in the peer-reviewed literature for the intended use of the test in the intended population. The degree of validity must be similar or superior to established and covered tests under this policy (see associated coverage Articles).”

Further, the LCD states that “The test successfully completes a MolDX Technical Assessment that will ensure that AV, CV, and clinical utility criteria set in this policy are met to establish the test as reasonable and necessary.”

Thus, equivalence or superiority is demonstrated through these means.

Further, the LCD differentiates molecular tests by their intended use, and these can notably be performed using different assays and methodologies. The scope of the LCD is the assessment of a transplanted allograft for rejection status, inclusive of the different types of rejection. While tests may use different methodologies to interrogate different aspects of acute rejection, they are expected to perform with equivalent or superior performance to established and covered tests under the policy. Multianalyte and combination tests must demonstrate superiority and additive benefit when compared to respective single analytes or components in the assessment of a transplanted allograft for the assessment of rejection status.

Finally, while we understand the confusion regarding the following “GEP will continue to be covered in heart transplantation only,” we note that the remainder of that statement provides its actual context and meaning. Nevertheless, we have removed ”only” for clarity.

The following comment was submitted to Palmetto GBA:

In the DEX registry, labs like CareDx have two entries per test, one for surveillance and one for cause testing.

In the new proposed billing article, clear distinction is made between surveillance and for-cause testing.

1. MolDX should keep the primary term, "SURVEILLANCE" testing rather than per protocol. First, this is in keeping with other precedents, like well established DEX text. Second, "surveillance" is used in other similar contexts like post surgical MRD surveillance for cancer relapse. "Surveillance" guides the reader to a familiar medical concept. In contrast, "per protocol" can mean anything (McDonald's makes Big Macs "per protocol.") The reader should not have to crosswalk or guess between one term in DEX and one in the LCD.
2. The LCD coverage section at top should clearly distinguish between the surveillance and for-cause testing. This is clearly the pivotal issue in the domain, as reflected in the CAC, the body of the LCD, and guidelines. For-cause testing triggered by signs and symptoms should have one set of rules that are easily found. In contrast, surveillance testing (if covered) should have its own different and clear set of rules (absent signs and symptoms.) One approach would be to have a header, "For Cause testing" followed by several bullet points. Then a different header, "Surveillance Testing" and its own specific several bullet points. I believe this reflects how clinicians as well as CMS medical auditors would approach any particular claims under pre or post pay review.

Hyperlink provided for review.

Thank you for your comment. Please see Response #1.

Additionally –

We have chosen to keep “protocol” included in the “Surveillance” testing definition, to further denote that this testing is used per protocol (i.e. is not for-cause).

The following comment was submitted to Palmetto GBA:

On behalf of the American Society of Transplantation (AST), an organization representing more than 5,000 transplant professionals dedicated to advancing the field of transplantation and improving patient care, I am writing to express our concerns regarding the newly proposed MolDX local coverage determination which would limit reimbursement coverage for molecular diagnostic testing in solid organ transplantation.

Transplant professionals use molecular diagnostic testing broadly, but most commonly for heart, lung, and kidney recipients, as a non-invasive method to monitor for allograft injury and assess for rejection. Transplant diagnostics is a rapidly evolving area in transplant medicine. Neither the optimal frequency of testing nor the specific test or combination of tests to be used has been established. Therefore, we believe it would be premature to specify a limit on the number of tests, testing type, frequency, or timing with respect to solid organ transplantation. The proposed reimbursement changes will directly impact the way these tests are used for patient care. We do not support restricting a physician’s professional judgment in considering the best testing option for their patients and their patients’ individual circumstances.

We recognize that the MolDX Program was developed in 2011 to identify, establish coverage for, and determine appropriate reimbursement for molecular diagnostic tests. We strongly support research into the clinical utility of molecular diagnostics in transplantation. However, in the absence of definitive studies, we do not support reimbursement changes that would limit physician choice should they wish to incorporate molecular diagnostic testing into the care of their patients.

We thank you in advance for your consideration of this request and look forward to future dialogue.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA:

On behalf of the Coalition for 21st Century Medicine (C21),¹ thank you for the opportunity to submit comments regarding the above-captioned proposed local coverage determination (LCD). Multiple C21 members furnish advanced molecular diagnostic tests that fall within the scope of this proposed LCD. As a result, C21 has a material interest in the proposed LCD, as well as the predictability and operations of the MolDX program more generally.

For nearly 15 years, C21 member laboratories have participated in and appreciated collaboration in the development, implementation, and refinement of the MolDX program. The MolDX program offers test developers a transparent, data-driven pathway to Medicare coverage for advanced molecular diagnostic tests based on the program’s assessment of an assay’s analytical validation, clinical validation, and clinical utility data as summarized in a Technical Assessment (TA) dossier. MolDX coverage decisions are further informed by the program medical directors’ understanding of the appropriate use of molecular diagnostics in clinical practice, as informed by published, peer-reviewed literature, clinical standards (e.g., professional society guidelines) and feedback from practicing clinicians who use the results of such tests in patient management. By considering this information in a holistic manner with a defined process, the MolDX program ensures that Medicare beneficiaries have timely, appropriate access to advanced diagnostic testing options that reflect – or in many cases, advance – the applicable standard of care.

Surveillance Testing Coverage and Limitations

C21 strongly supports the Draft LCD’s explicit coverage for molecular testing used in patient surveillance following organ transplantation (i.e., for the detection of subclinical rejection). As MolDX acknowledges in the draft policy, the clinical utility of molecular testing in this critical patient population is well-established in the published clinical literature and clinical guideline statements.² The innovation of detecting donor derived cell free DNA within a blood sample from a patient with a newly transplanted organ has advanced the care of these patients and reduced the frequency of invasive biopsies. We further appreciate the inclusion of substantive language confirming such coverage within a Draft LCD and not within a Billing and Coding Article, as such inclusion allows for the greatest transparency and stakeholder input.

We have concerns, however, that MolDX surprisingly proposed new limits on the number of surveillance tests that it will cover during a particular time period. Instead of establishing limits on the frequency of such services, C21 urges MolDX to continue current policy under which treating clinicians have the flexibility to order surveillance tests on a cadence based on their individualized assessment of what is medically necessary for that specific patient.

Chapter 13, section 13.5.4 of the Program Integrity Manual allows contractors to establish LCDs based on their assessment of whether the service is safe and effective, not experimental and investigational, and appropriate (including the duration and frequency that is considered appropriate for the service). By confirming coverage for surveillance tests, MolDX acknowledges that the first two prongs – safe/effective and not experimental/investigational – are met. With respect to the third prong – appropriateness – the Manual requires contractors to consider whether the service is furnished consistent with accepted standards of medical practice, ordered and furnished by qualified personnel, and meets (but does not exceed) the patient’s medical need, among other criteria. Allowing treating clinicians the flexibility to order surveillance tests at a different or more frequent cadence would be fully consistent with this instruction.

Alternatively, if MolDX insists on retaining the proposed limits on the number of surveillance tests offered each year, C21 requests that MolDX revise the LCD to explicitly allow testing on a more frequent basis when supported by the treating clinician’s attestation of medical necessity. As MolDX notes in the policy, clinicians know best how to treat their patients and the policy should be revised to better embrace that view. Creating such an exception would confirm that MolDX intends only to curb potentially inappropriate utilization with the updated policy, and is not attempting to limit the amount of payment made for medically necessary services.

Restriction to One Molecular Test per Patient Encounter

In addition to setting new annual limits on surveillance testing, MolDX proposes to restrict the number of molecular tests that may be ordered and covered during a single patient encounter with a new higher standard of evidence (e.g. superiority and additive benefit when compared to respective single analytes or components). We request this policy not be finalized.

In all clinical situations – not just in transplant – clinicians use their training and judgement to identify and order the best test or combination of tests to inform patient management. For example, clinicians routinely order multiple chemistry tests (e.g., basic metabolic panel, complete blood count) during a single patient encounter, and Medicare routinely pays for such services. Medicare contractors may reasonably determine that coverage for multiple laboratory testing services is inappropriate if those services provide duplicative information. However, all such tests should be covered if the analytes measured are ordered by the patient’s treating clinician, are not duplicative, and provide differentiated information that inform patient management. It is unclear why MolDX proposes to treat assays intended for the detection of transplant rejection differently than Medicare treats laboratory testing services for many other conditions.

Our concerns about MolDX restrictions on multiple molecular tests are not unique to molecular tests assessing allograft status and is reflected more broadly in how MolDX considers multiple molecular tests to constitute a panel vs. individual services that may be separately billable. As an example, Article (“Defining panel services in MolDX”) states that a service is a panel “if multiple analytes (more than 1) are measured in parallel, regardless of how they are performed or reported.” This criterion is very broad, and has resulted in a MolDX policy where, for genomic profiling tests, RNA analysis cannot be separately billed when a separate DNA analysis is ordered. As a result, labs cannot use the specific CPT codes assigned for each of these services, even when the services are not ordered together and the procedures are run entirely separately and provide non-duplicative information. In these instances, MolDX directs labs to use the unspecified CPT code 81479, which runs counter to chapter 10 of the NCCI manual that states “Providers/suppliers shall report the HCPCS/CPT code that describes the procedure performed to the greatest specificity possible.” Consistent with NCCI instructions, other MACs have policies that allow distinct molecular tests, including DNA and RNA analyses, to be billed separately in the same patient encounter.

Bundled Payments

In the Draft Billing and Coding Article at the very end of the Draft LCD, MolDX proposes to create surveillance testing “bundles” under which it would pay for “multiple tests comprising the service.” These bundles correlate with the new frequency limits on surveillance testing proposed in the Draft LCD but include no reference to actual payment rates, nor does the article describe a transparent method for calculating the same.

The Social Security Act requires CMS to establish a fee schedule for clinical diagnostic laboratory tests.³ CMS sets initial CLFS rates for tests described by new or substantially revised codes via crosswalk or gapfill.⁴ Then, once private payer rate information is reported to CMS (as required under PAMA), CMS updates the rates for those codes to reflect the volume-weighted median of reported data.⁵ The Social Security Act prohibits CMS from adjusting nationally-established CLFS rates for any reason, except as outlined above.⁶

CMS has established (or is in the process of establishing) national rates for nearly every test service affected by the proposed LCD. Given these established rates, if MolDX decides to proceed with “bundled” payments to facilitate claims processing, we encourage MolDX to confirm that it will pay for each bundle at a rate equivalent to the current CLFS rate TIMES the number of tests included in the bundle. We are unsure, however, as to why these bundles would be proposed unless the reason is to establish payments that differ from the nationally established CLFS rates. Insofar as the tests included within any such bundles – either under this LCD or otherwise – have individual CPT codes with rates established on the CLFS, the creation of such “bundles” is not necessary because individual test claims can – and must – be paid at the existing CLFS rate.

While bundled payment mechanisms can be accommodated when billing Medicare Part B, they add a significant logistical burden for labs to track samples that should or should not be billed in each instance. Additionally, most Medicare Advantage (MA) plans require each service be individually billed and do not have the ability to process bundled claims. These payors typically refuse to manage bundled billing. With more than half of Medicare beneficiaries now enrolled in MA, the disconnect between FFS and MA creates a significant operational challenge. This challenge is further compounded by the fact that MA plans are not required to follow language in coverage articles and must only follow the language in the LCD itself. Therefore, MA plans may not effectively cover bundled services as MolDX intends, leading to excessive claim denials. As a result of this additional complexity, bundled billing should only be employed as a last resort when no alternative payment mechanism is available to process payment for the services.

As always, C21 appreciates the opportunity to comment on the Draft LCD.

References were provided for review.

Thank you for your comment. Please see Response #1.

Additionally, this comment exemplifies a severe misunderstanding of current policy which does not include an ambiguously flexible and open-ended cadence for surveillance testing. We also note that LCDs interrogate the evidence, which may change with time, and provide criteria for coverage accordingly. Finally, we limit our comments here to the relevant LCD only.

The following comment was submitted to Palmetto GBA and Noridian:

I am writing as a practicing transplant nephrologist with extensive experience in managing kidney transplant recipients to provide comments on the Draft LCD for Molecular Testing for Solid Organ Allograft Rejection, issued on July 17, 2025. I appreciate the effort to update coverage policies based on evolving evidence and commend the inclusion of surveillance testing for subclinical rejection using molecular diagnostics like donor-derived cell-free DNA (dd-cfDNA). However, I strongly urge reconsideration of the proposed frequency limits for surveillance testing-four tests in the first year post-kidney transplant and two annually thereafter-as these restrictions may hinder optimal patient management, early rejection detection, and immunosuppression individualization. Instead, coverage should either be at the clinician's discretion or at a minimum align with clinically validated protocols, such as the Assessing Rejection in Transplants Schedule (ARTS)¹ from the DART2 and KOAR³ studies, allowing at least seven tests in the first year (monthly for months 1-4, then quarterly) and quarterly thereafter.

Surveillance with dd-cfDNA serves critical purposes: detecting subclinical rejection before irreversible allograft injury and enabling precise immunosuppression titration, which varies by patient. Medium- to long-term graft survival has stagnated, partly due to immunosuppression-related complications, and ongoing surveillance is essential beyond the first year. The draft's limits, tied to outdated biopsy frequencies (historically ~4 in year one due to procedural risks), do not account for dd-cfDNA's non-invasive nature and favorable risk-benefit profile. Several studies demonstrate dd-cfDNA elevations precede biopsy-proven rejection by months, allowing early intervention and improved outcomes."" The KDIGO guidelines suggest using the lowest planned doses of immunosuppression by month 4 post-transplant, suggesting that at minimum patients would be surveilled monthly during this period with ongoing surveillance quarterly thereafter to ensure immunosuppression minimization does not subject the patient to graft injury or rejection. Additionally, the KOAR study showed rejection yield was 39% with elevated dd-cfDNA in surveillance versus 7% without, enhancing pre-test probability for biopsy-proven acute rejection (BPAR). Similarly, the ProActive study found dd-cfDNA elevations before rejection, supporting proactive management.'

The American Society of Transplant Surgeons (ASTS)" statement explicitly endorses dd-cfDNA for surveillance, noting clear clinical utility in longitudinal testing despite lacking prospective trials on optimal frequency. It highlights the DART study's ARTS protocol-monthly testing months 1-4, then quarterly-as empirically sound, aligning with KDIGO guidelines for other non-invasive markers like creatinine (daily initially, then 2-3 times weekly, weekly, biweekly, monthly, and bimonthly). This regimen (7 tests in year one) focuses on high-risk early periods (planned reductions of immunosuppression) while enabling ongoing monitoring. Quarterly testing post-year one (4 annually) better supports immunosuppression optimization, especially for high-risk patients like those with marginal organs under the Increasing Organ Transplant Access (IOTA)⁹ model, who face elevated delayed graft function and rejection risks.

Additionally, evidence from large cohorts, including ADMIRAL (1,092 recipients)¹⁰ and a Nature Medicine study (2,882 recipients)¹¹, confirms dd-cfDNA's performance in diverse populations, including repeat transplants (8-15.5% of cases), with similar discrimination between quiescence and rejection. Longitudinal monitoring yields high positive (77%) and negative (85%) predictive values for antibody-mediated rejection (AMR), outperforming single-point testing. Serial dd-cfDNA also tracks treatment response and de nova donor-specific antibody (DSA) development, with elevations predicting risks. No studies link higher frequencies to harm; conversely, restrictions may delay detection, increasing the need for biopsy and costs for returning to dialysis.

Furthermore, dd-cfDNA is emerging as a valuable marker not only for detecting injury but also for confirming resolution of injury, as presented in multiple abstracts at the recently concluded World Transplant Congress, San Francisco, USA 2025. This evolving role supports its use in immunomodulation strategies, where clinicians can tailor immunosuppression based on dynamic dd-cfDNA trends. Single-center studies and real-world practices are actively exploring these applications, laying the groundwork for larger, multicenter trials. Unlike creatinine, a late and functional marker, dd-cfDNA offers a real-time, injury-specific signal that can guide proactive care. Curtailing its use through rigid frequency caps risks stifling innovation and delaying the broader understanding and integration of this novel biomarker into transplant medicine.

In summary, while the draft advances coverage, frequency caps undermine physician discretion and evidence-based care. I recommend removing limits or adopting ARTS­ aligned frequencies (7 in year one, 4 annually thereafter) to reflect dd-cfDNA's utility in improving graft outcomes in transplant and repeat transplant recipients.

References were provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

The American Association of Kidney Patients (AAKP) appreciates the opportunity to comment on the above-referenced draft Local Coverage Determination (LCD). Established in 1969, AAKP is the nation’s oldest and largest fully independent kidney‐patient organization. We advocate for policies that improve transplant outcomes, protect patient choice, and use taxpayer dollars responsibly. AAKP believes kidney disease is fundamentally and inextricably both a healthcare and workforce issue. Our view is informed by our history and policy expertise. AAKP helped secure the bipartisan passage and authorization of the modern End Stage Renal Disease program in 1972, legislation signed into law by President Richard Nixon on the advice of his senior advisors, including Office of Management and Budget Director Caspar Weinberger. From the outset, the ESRD program recognized dialysis as a temporary bridge and rehabilitative therapy that could enable Americans with kidney failure to remain in the workforce while awaiting the preferred therapy of organ transplant. AAKP has never forgotten that dialysis was intended to be a bridge to transplantation, not a substitute.

1. Background

In light of this history, we believe every safe medical advancement designed to improve long-term transplant survival and the capacity of patients to maintain their lives and livelihoods is a matter of national, not simply medical, interest. American taxpayers spend over $100 billion a year in direct kidney care, including dialysis, and transplant remains the platinum standard and most cost-effective treatment for kidney failure. Any proposed policy decision that negatively impacts transplant care or transplant outcomes is, therefore, and in our view, contrary to both patient and national interests.

Regrettably, since 2023, flawed policy processes and decisions pursued by the MolDX program and embraced by Medicare Advisory Committees (MACs), have restricted and limited transplant patient access to molecular blood testing for organ transplant health surveillance. The actions of these MACs transformed what was once heralded as an optimistic innovation for patient-centered kidney care into an unnecessarily controversial and highly burdensome issue among transplant recipients and organ donors. The MAC policy effort to limit patient access over the past several years also created tremendous confusion and frustration among transplant experts and across transplant centers patients entrusted to co-manage their life-saving transplants. Despite the fact these policy changes originally occurred under the auspices and approval of the Biden Administration and appointed and career officials at the Department of Health and Human Services and the Centers for Medicare and Medicaid Services (CMS), AAKP believes MolDX officials bear ongoing responsibility and accountability for all known, and as yet unknown, negative consequences borne by transplant recipients.

The proposed LCD, at best, makes a marginal effort to repair the serious damage and confusion MolDX created around molecular blood test access over the past three years. We were pleased to read that MolDX has evolved in its policy position and now accepts the advice originally offered by a majority of their transplant expert consultants that these tests are a critical, non-invasive surveillance alternative for determining early organ rejection when compared to status quo, and often life-threatening, invasive biopsies. Somewhat ironically, these same positive transplant expert views of molecular blood tests, along with strongly supportive patient insights, were known over a decade ago by HHS, CMS and CMS contractors. In fact, these expert and patient views shaped the basis for the 2017 CMS decision to approve molecular blood test coverage. AAKP played an integral role in the 2017 decision as a national stakeholder working to advance transplant innovations, improve patient survival and reduce avoidable taxpayer costs driven by organ failure, dialysis, disability and dependency.

Yet, at worst, the draft reflects the ongoing inability of MolDX to fully respect and take seriously the ongoing and legitimate concerns raised publicly by transplant recipients, organ donors and the broader kidney community during the past three years regarding the challenges associated with long-term transplant health and outcomes. Current generation immunosuppressive drugs approved for use by the Food and Drug Administration for kidney transplants were approved over twenty years ago. The clinical endpoint upon which they were approved measured one-year graft survival. Today, over ninety-three percent of kidney transplants meet that endpoint. Unfortunately these same FDA approved transplant drugs pose serious threats to patients, including nephrotoxicity to the transplanted organ, severe gastrointestinal issues and cancer risks. As of yet, there are no long-term transplant clinical outcome measures for use by kidney patients or transplant professionals. Prior to the 2017 CMS coverage of molecular blood tests, the only effective way for transplant professionals to test for organ failure or rejection were needle biopsies, which pose a threat to the transplanted kidney and too frequently confirm rejection after the window for an effective intervention has closed. This is why molecular blood tests were viewed by transplant recipients as a game changing victory in the battle for improved long-term transplant outcomes and avoidance of a second or third transplant. Even to the most uninformed outside observer, the arc of innovation in the transplant space over the past decade has highlighted the unmistakable fact that transplant patient care is a highly specialized and individualized endeavor.

Every transplant patient in America understands this reality – because they live it every day. So do organ donors, because every organ donor, living and deceased, has invested in their gift the hope that the life of the transplant recipient will be extended and made more resilient to the unique challenges that may lie ahead.

As AAKP reviewed the proposed LCD, we were stunned that, once again, MolDX put their hand on the scale to limit patient and transplant patient access to these tests and, once again inserted themselves between transplant patients and the professionals they trust. The draft has particularly negative and onerous access limits for transplant recipients who have worked diligently to maintain their organs beyond one year. After the past three years of controversy caused by MolDX and despite hours of public testimony, Congressional inquiries, media stories and the heartfelt concerns of tens of thousands of organ transplant recipients questioning policy changes since 2023 – the MolDX program still presumes it knows better than America’s transplant experts, patients and organ donors. We find the proposed draft antithetical to patient-centered transplant care and a fundamental rejection of the principle of patient care choice. The proposed policy is culturally contrary to the kidney community’s commitment to making certain patients’ lives remain fully protected and the shared belief that kidney experts are the best arbiters for determining the right life-saving innovation, at the right time, for the right patient.

Over the past month, AAKP has been engaged on Capitol Hill and in local Congressional offices discussing Government Determinants of Health (GDoH) and their impacts on transplant patients and organ donors. GDoH are instances when government officials and or government contractors, either knowingly or unwittingly, advance policies that are contrary to known patient and national interests. As we have discussed with elected leaders, we believe the ongoing actions of MolDX in regard to long-term transplant health and related life-saving innovations are examples of GDoH. Further, we believe the previous policy changes and proposed changes stand in sharp contrast the bipartisan national goals for transplantation outlined in President Donald Trump’s historic 2019 Executive Order on Advancing American Kidney Health. AAKP is very proud of the central role we played in developing and advancing this policy and believe MolDX owe the American people, and the Trump Administration, a transparent explanation for defending and proposing policies that undermine this policy as well as national, patient and taxpayer interests.

In summary, I reiterate and emphatically urge you to consider the following:

1. Importance of Molecular Blood Tests

Molecular blood tests are a critical, non-invasive tool for early detection of organ rejection. Before the Centers for Medicare & Medicaid Services (CMS) approved coverage for these tests in 2017, clinicians relied primarily on needle biopsies—an invasive procedure that can damage the transplanted kidney and often confirms rejection only after intervention is less effective. Today, more than ninety-three percent of kidney transplants survive one year, but long-term graft survival remains a challenge. Molecular testing offers the best currently available method to monitor graft health beyond that first year and guide timely clinical action.

1. Concerns with the Draft LCD
2. Restrictions on Access. Since 2023, policy changes advanced by the MolDX program and adopted by MACs—have curtailed patient access to molecular testing. The draft LCD makes only marginal progress toward remedying this problem and still imposes onerous limits, particularly on recipients more than one year post-transplant.
3. Process Deficiencies. Despite extensive public testimony, Congressional inquiries, and input from transplant experts and patients, MolDX continues to override the judgment of clinicians who manage highly individualized transplant care.
4. Conflict with National Policy Goals. The draft LCD undercuts bipartisan transplant objectives, including those embodied in the 2019 Executive Order on Advancing American Kidney Health, which prioritizes innovation to extend graft longevity and reduce reliance on dialysis.
5. Economic and Societal Impact

Taxpayers spend more than $100 billion annually on kidney care, with dialysis costs far exceeding those of successful transplantation. Policies that jeopardize transplant longevity therefore impose substantial fiscal and social burdens. Restricting molecular testing risks avoidable graft failures, re-listing for transplantation, return to dialysis, and loss of workforce participation.

1. Requested Revisions

AAKP urges MolDX to:

• Restore broad coverage for molecular blood tests consistent with the 2017 CMS decision.
• Eliminate arbitrary time limits on testing after the first post-transplant year.
• Defer to treating transplant specialists, in consultation with their patients, to determine when molecular testing is clinically warranted.

Conclusion

Transplant recipients, organ donors, and the broader kidney community rely on policymakers to support innovations that safeguard long-term graft health. The proposed LCD, as drafted, falls short of that obligation. We respectfully request that MolDX revise the policy to ensure timely, equitable access to molecular blood testing.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

I have had an opportunity to review this draft policy and feel the need to outline some misconstruing of the reviewed data, especially as it relates to the SHORE publication, for which I was an author.

I am a transplant cardiologist and previously as Medical Director of the Heart Transplant Program.

The purpose of the SHORE publication was to evaluate the performance characteristics of combination testing with GEP and dd-cfDNA in the primary context it is used: surveillance.¹ Given that in this context, the pre-test probability is low, we intentionally chose to optimize the combined use to maximize its specificity and positive likelihood ratio. Thus, without changing longstanding thresholds for either test, we appropriately defined a positive test as one in which both GEP and dd-cfDNA were elevated. With that definition and, unsurprisingly, given that the tests are measuring different aspects of rejection, the specificity and positive likelihood ratio were clearly superior with combination testing than with either GEP or dd-cfDNA alone. The notion that the increase in specificity was not notable is incorrect. It was both a statistically significant difference in specificity and contributed to both a statistically significant and clinically meaningful increase in the positive likelihood ratio. That this higher specificity/+LR was clinically meaningful is evident from the fact that clinicians within SHORE more frequently biopsied patients with a dual positive test compared to patients in whom dd- cfDNA was positive but GEP was negative.¹

It is true that if all you had available was dd-cfDNA, an ACR positivity rate of 7% might be high enough to justify a biopsy- but given that GEP is available, it identifies the patients with a positive dd-cfDNA into those at higher risk, at approximately 9% and lower risk at approximately 4%. Thus enables us as clinicians to achieve the goal of personalized care. By stripping the clinician of the ability to have the complimentary GEP, you are guaranteeing a higher reflex biopsy rate. I have personally safely deferred biopsies when the dd-cfDNA was elevated and the GEP was reassuringly negative.

Finally, I was troubled by the contrasting of SHORE with the paper by Rodgers et al.² Despite acknowledging the low pre-test probability of rejection during surveillance, the authors chose to define a positive combined test as either a positive GEP or a positive dd-cfDNA. This optimizes for overall sensitivity but cripples specificity. Equally critically, the study was a single center study (n=112) that only contained 10 acute cellular rejections and was severely underpowered to compare sensitivities of the test. I think this needs to be acknowledged in the policy.

Ultimately, there is an ongoing role for combination molecular testing for heart transplant with GEP and dd-cfDNA. I recognize that there are resource considerations but disallowing combination testing with GEP and dd-cfNDA is not good medicine and not good for patients. It will subject more patients to costly invasive procedures with well- known complication rates.

References provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

I am currently the Associate Director of Advanced Heart Failure, Cardiac Transplantation, and Mechanical Circulatory Support.

I appreciated the opportunity to review the draft LCD on molecular testing in cardiac transplantation. I have some concerns about the interpretation of the presented data and restrictions on testing that would follow should the LCD be implemented as described.

First off, I would like to emphasize that the literature is definitive- combination testing with GEP and dd-cfDNA brings value to clinicians and patients. In the first SHORE manuscript, patients who had both a positive dd-cfDNA and GEP were more likely to have acute cellular rejection than patients with an elevated dd-cfDNA and a negative GEP. This made a lot of sense to us as transplant clinicians and was congruent with what we were seeing in our patients.¹ While dd-cfDNA represents a major innovation in the surveillance of heart transplant patients, it is not a perfect predictor of whether a biopsy is likely to reveal rejection. As demonstrated in the SHORE manuscript, the risk of ACR is highly modified by whether the associated GEP is positive or negative. This was not just statistically significant but is also clinically meaningful.

As described in the reflex biopsy section of the SHORE manuscript (which was not mentioned in the draft LCD), my colleagues and I have personally deferred biopsies for mildly elevated dd-cfDNA if the GEP was low. Furthermore, given the limited sensitivity of dd-cfDNA for ACR, I have at times detected acute cellular rejection with a normal dd-cfDNA but an elevated GEP.

The data from SHORE was so compelling that we sought to confirm and expand its findings in our own heart transplant patient population, most of whom were not enrolled in SHORE. In a manuscript that we have just submitted with revisions, we also demonstrated that not only do patients with positive GEP and dd-cfDNA have a higher risk of current rejection but are also at high risk of future adverse events. Dd-cfDNA alone is not sufficient to precisely risk stratify patients. Importantly, in certain cases, we have begun modifying immunosuppression in some of these patients, so precisely characterizing risk with combined molecular testing is critical to minimize those exposed to long term adverse effects of these drugs.

Finally, I have significant concerns about the limitation on surveillance with molecular testing proposed by the local coverage determination. We have found that patients in the second- and third-year post transplant still benefit from close surveillance to allow further individualization of their immunosuppression. While immunosuppression has been shown to help prevent organ rejection, it does not come without risks. Overexposure to immunosuppression can lead to serious infection and malignancies as well as kidney injury. We often see patients quarterly during this time frame and we continue to reduce immunosuppression in those that have reassuring combined molecular testing results. Limiting our ability to fully assess these patients with their molecular profile will limit our ability to fully take advantage of these clinic visits and target the right drug combination for each patient.

Reference was provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

As the Associate Medical Director overseeing lung transplant recipients and their complex care, I am writing to provide clinical feedback on the draft Local Coverage Determination (LCD), which outlines coverage parameters for molecular testing in solid organ transplant rejection, including donor-derived cell-free DNA (dd-cfDNA) in lung transplant recipients.

The inclusion of dd-cfDNA testing in the draft policy is a welcomed acknowledgment of its growing role in transplant medicine, and once-monthly testing in year one is appropriate based on existing literature and our own personal experiences. It is our current protocol to utilize dd-cfDNA in posttransplant recipients quarterly after year one. This is especially critical given that these patients are no longer undergoing surveillance bronchoscopies to monitor for acute rejection and infections but remain at high risk for both entities. This relates to the highly immunogenic nature of lung allografts and the need for greater immunosuppression relative to other solid organ transplants. However, the proposed limitations past year one raise significant concerns from a clinical management perspective:

1. Rigid Surveillance Cadence

The draft policy proposes a fixed schedule of 12 dd-cfDNA tests in the first-year post-transplant and 2 annually thereafter. This approach does not reflect the clinical variability in lung transplant patients, particularly those with heightened immunologic risk or complex post-operative courses. Surveillance strategies must remain adaptable to individual patient needs, guided by clinical judgment rather than arbitrary limits. While 12 tests in year one is most likely adequate, 2 in the subsequent years is not.

2. Clinical Utility Standards

While we support the need for robust validation, the requirement that dd-cfDNA demonstrate superiority over existing modalities may inadvertently exclude valuable adjunctive tools. Dd-cfDNA has shown strong correlation with acute rejection and infection risk in peer-reviewed studies and real-world practice. Its utility lies not only in replacing biopsy in certain patients, but in enhancing diagnostic precision and reducing reliance on invasive procedures. While the draft LCD references the Keller et al clinical utility study as its basis for 12 surveillance episodes in year one post-transplant, the draft policy diverges from Keller et al in subsequent years. The draft policy states that 2 tests a year post-transplant are reasonable and necessary in subsequent years while Keller et al used quarterly testing (4 test per year) in subsequent years. From Keller et al: “All four centers developed consensus guidance regarding the frequency of asymptomatic dd-cfDNA monitoring and interpretation with monthly surveillance plasma dd-cfDNA testing for patients less than 1-year post transplant and every 3 months for patients > 1-year post-transplant.” Similar to Keller et al., we surveil our clinically stable patients monthly in the first-year post-transplant and quarterly thereafter. Prior to dd-cfDNA we used transbronchial biopsies (TBBx) during these routine visits and now we are able to also rely on dd-cfDNA. The final policy’s surveillance frequencies should be consistent with the published data from Keller et al.: testing monthly in the first-year post-transplant and at least quarterly in the subsequent years post-transplant.

3. Equity and Access Considerations

The proposed restrictions may disproportionately affect patients in rural or underserved areas, where access to biopsy and transplant centers is limited. Dd-cfDNA offers a non-invasive, scalable solution that aligns with CMS’s goals of improving access and reducing disparities in care.

I would recommend allowing for individualized surveillance protocols based on clinical risk and transplant center and physician discretion or at a minimum, 12 in the first year-post-transplant followed by quarterly testing there after (4 per year) for clinically stable post-transplant recipients. dd-cfDNA is a welcomed personalized medicine tool for lung transplant recipients. We appreciate CMS’s commitment to evidence-based policy and urge reconsideration of these provisions to ensure optimal care for lung transplant recipients.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA:

Devyser Genomic Laboratories appreciates the opportunity to submit comments in support of coverage for molecular testing for allograft rejection. As discussed below, Devyser is developing a molecular test for allograft rejection and operates a clinical laboratory in the MolDX jurisdiction. Our comments on the proposed local coverage determination (LCD) and associated draft Billing and Coding Article are intended to ensure a transparent and sustainable coverage framework that promotes innovation while ensuring Medicare beneficiaries have access to the most advanced diagnostic technologies available. Devyser plans to commercially launch our TransplantTrace cfDNA Kidney test in early 2026 and we look forward to meeting with you in advance to continue the technical assessment process.

Support for Coverage of Molecular Testing for Allograft Rejection

Devyser strongly supports LCD's recognition of the clinical utility of molecular testing for solid organ allograft rejection detection. The draft LCD appropriately acknowledges the well­ established evidence base supporting the use of dd-cfDNA testing in transplant medicine, as reflected in peer-reviewed literature and clinical guidelines. OurTransplantTrace cfDNA Kidney test will contribute to this evidence base and provide clinicians with additional options for patient management.

The innovation of detecting donor-derived cell-free DNA has indeed transformed transplant care by providing a less invasive alternative to tissue biopsies while delivering clinically actionable information. We appreciate MolDX's commitment to ensuring Medicare beneficiaries have access to these advances in precision medicine.

While our initial market entry will focus on "for-cause" indications rather than surveillance testing, we note that the proposed LCD appropriately recognizes both use cases. We support the flexible approach to for-cause testing that allows clinicians to order molecular tests based on their clinical judgment and patient-specific factors.

Comments Regarding Bundled Coding Proposals

While we support coverage for molecular testing, we want to raise questions regarding the surveillance testing "bundles" coding approach referenced in the draft Billing and Coding Article. The proposed creation of bundled codes raises several important issues.

The draft Billing Article references bundled payments but does not specify the actual payment rates. This lack of transparency makes it difficult for stakeholders to assess the potential impact of the proposed payment methodology. We request that MolDX provide clear information about proposed payment rates and the methodology used to establish these rates.

If MolDX proceeds with bundled coding approach, we strongly urge that each bundle be paid at a rate equivalent to the current CLFS rate multiplied by the number of individual tests included in the bundle. The Social Security Act requires CMS to establish a fee schedule for clinical diagnostic laboratory tests, and Medicare contractors must pay claims for covered services at the national Clinical Laboratory Fee Schedule (CLFS) rate when it has been established.¹ CMS sets CLFS rates through crosswalk or gapfill processes for new codes, and then updates these rates based on private payer data reported under PAMA.² The Social Security Act specifically prohibits CMS from adjusting nationally-established CLFS rates except through these defined processes.³

Given that individual CPT codes for the tests covered under this LCD have established CLFS rates, the creation of "bundled" payments appears unnecessary and potentially problematic.

Bundled payment structures that do not reflect established CLFS rates could create barriers to market entry for innovative diagnostic companies like Devyser. Our TransplantTrace cfDNA Kidney test represents significant investment in research and development, and appropriate reimbursement is essential to support continued innovation in this field.

Artificially constrained payment rates through bundling could discourage the development of next-generation diagnostic technologies that could benefit patients.

Alternative Recommendation

Rather than implementing bundled payments, we recommend that MolDX continue to process claims for individual tests using their respective CPT codes and established CLFS rates. This approach would:

• Ensure compliance with federal payment requirements
• Maintain transparency in the reimbursement process Support continued innovation in molecular diagnostics
• Provide predictable reimbursement for covered services

Conclusion

Devyser appreciates MolDX's continued leadership in establishing evidence-based coverage policies for advanced molecular diagnostics. We respectfully request that MolDX reconsider the bundled payment approach in favor of maintaining individual test billing at established CLFS rates. This approach would better serve the interests of Medicare beneficiaries, healthcare providers, and diagnostic innovation companies while ensuring compliance with federal payment requirements.

References were provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

I have had an opportunity to review the policy and would like to highlight some important considerations, as I have concerns about the way the data in the draft policy was portrayed as it relates to GEP and combination testing with dd-cfDNA.

I am a past president of the International Society of Heart and Lung Transplantation, a transplant cardiologist, and serve as the Medical Director of the heart transplant program. I also have extensive experience with both GEP and dd-cfDNA in the management of heart transplant patients and have integrated both into our standard protocols.

1) The first was regarding Henricksen et al., a study for which I was senior author and used data from Stanford’s heart transplant program.¹ The draft policy states that this study “had very few episodes of AR and was underpowered to draw meaningful conclusions regarding an outcomes-based benefit from combination testing”

This characterization that the study was underpowered regarding an outcomes-based benefit ignores the impact of combination testing on biopsies. The study was adequately powered to assess the large and statistically significant reduction in biopsies in those undergoing combination testing falling from a median of 10 to a median of 4 biopsies per patient (p<0.01). From a patient perspective, having fewer invasive procedures, while still having the same survival and rejection free survival, is definitively an important outcome. This data should be added to the draft LCD.

Moreover, the draft policy incorrectly states that the paper “failed to consider that testing by cfDNA alone could provide a superior alternative to GEP, though their data was consistent with this interpretation.” While the study used a historical control and could not change the comparator, the data provided in table 3 of the manuscript proves that both GEP and dd-cfDNA contributed to the important reduction in biopsies. Of the 37 patients who had a positive dd-cfDNA only 10 had a positive GEP, of whom 90% had a biopsy. But of the remaining 27 positive dd-cfDNA with negative GEP we only biopsied 22%, clearly demonstrating the utility of GEP in contributing to the dramatic reduction in biopsies. To be clear, without GEP, it is likely that we would have performed approximately 75% more biopsies in the contemporary cohort.

• 2) The SHORE manuscript cited confirmed two main points:²

1. It reproduced the clinical utility of combination testing that we had demonstrated in our single center publication. A negative GEP led to significantly fewer unnecessary biopsies in patients with a positive dd-cfDNA. This critical data was completely omitted from the draft policy.
2. This approach clearly demonstrated that patients with a positive dd-cfDNA, but a negative GEP were at less than half the risk of having ACR than patients with a positive dd-cfDNA and a positive GEP. The statement in the draft policy that as a group, patients with a dd-cfDNA are at moderately high risk of ACR is true but elides the point that GEP distinguishes groups at higher risk vs those at lower risk for ACR. This should be acknowledged.

3) The cited Rodger’s paper defines a positive combined test as a either a positive GEP or a positive dd-cfDNA.³ This is not that way the test should be used for surveillance as it maximizes false positives and reduces its specificity. We published an analysis on their publicly available data set and showed that when a positive combined test is defined as both a positive GEP and dd-cDNA, the results were nearly identical to the results from the SHORE manuscript.⁴ The likelihood ratio for the combined test was highest for combination testing. This peer-reviewed analysis and publication should be included in the draft policy.

4) That same publication also demonstrates that in higher risk monitoring scenarios only GEP can achieve an adequate sensitivity to exclude ACR and that in this scenario combination testing with dd-cfDNA should be used when both the risk of ACR and AMR is elevated.⁴

5) My colleagues and I have submitted an additional analysis that is currently undergoing revisions demonstrating that only patients with elevated dd-cfDNA and elevated GEP are at risk for future graft dysfunction and cardiovascular death potentially requiring intensification of monitoring and therapy. Thus, curtailing clinicians’ ability to combine GEP and dd-cfDNA testing will not only result in more unnecessary biopsies, but potentially other unnecessary testing and therapies.

Thank you in advance for integrating all the above into the finalized LCD.

References were provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

I have reviewed the above draft policy on molecular testing in solid organ transplantation and would like to offer a few comments.

I would first like to note that the policy’s description of the data surrounding gene expression profiling does not include two important studies. CARGO-II was conducted at multiple centers across the United States, Canada and Europe that re-demonstrated clinical validity of GEP.¹ eIMAGE was a randomized clinical utility study that was conducted at my institution that demonstrated that GEP could safely be used instead of endomyocardial biopsies to surveil patients as early as 2 months post-transplant and to adjust immunosuppression.² Both studies contributed to the 2023 ISHLT guidelines which expanded their recommendation for GEP to be used as early as 2 months post cardiac transplant.³ This should be added to the draft policy.

With respect to combined testing with GEP and dd-cfDNA, while I agree the impact of dd-cfDNA on biopsy yield may be greater than GEP, I do think that they are indeed complementary tests. The data from the SHORE publication, a very large study in the context of cardiac transplant research, demonstrates that GEP appears to significantly moderate the risk of acute cellular rejection detection in those with an abnormal dd-cfDNA.⁴ Moreover, clinicians appear to consider both tests when deciding on whom to biopsy. Ultimately, I believe both dd-cfDNA and GEP have extensive data to support their use, and it should be clinicians that decide how to use these tests in their clinical practice.

References were provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

I am writing as a practicing transplant surgeon with extensive experience in managing kidney transplant recipients to provide comments on the Draft LCD for Molecular Testing for Solid Organ Allograft Rejection, issued on July 17, 2025. I appreciate updated coverage policies based on evolving evidence and commend the inclusion of surveillance testing for subclinical rejection using molecular diagnostics like donor-derived cell-free DNA (dd-cfDNA). I do however have concerns with establishing surveillance caps, this was a surprise and not expected. As the draft policy noted, “Physicians know best how to manage care for their patients.” It is well known that there is high heterogeneity between populations served by transplant centers and between individual patients and their risk for rejection. Instead of establishing limits on using dd-cfDNA for surveillance, I urge MolDX to adopt an approach whereby the treating clinicians have the flexibility to order surveillance tests on a different or more frequently based on their individualized assessment of what is medically necessary for that specific patient or at a minimum align with clinically validated protocols such as the ARTS protocol from DART and KOAR, allowing at least seven tests in the first year (monthly for months 1-4, then quarterly) and quarterly thereafter.

While the Draft LCD recognizes that prospective trials designed to optimize the frequency of dd-cfDNA testing in stable, asymptomatic adult renal transplant recipients are lacking, multiple experts agree that developing such data is impractical. However, the evidence for the clinical utility of dd-cfDNA surveillance testing is clear, as is the utility of testing longitudinally.

For maintenance immunosuppression, the KDIGO guidelines recommend using the lowest planned doses of immunosuppressive medications by month 4 providing there has been no rejection or delayed graft function. The current “ARTS” surveillance testing schedule of monthly tests from months 1 to 4 and quarterly tests thereafter aligns with this post-transplant period of increased risk of rejection while maintenance immunosuppression is being optimized. This was the surveillance schedule which was developed and used in the DART and KOAR studies which demonstrated that dd-cfDNA levels were elevated up to 4 months before biopsy confirmation of ABMR, thus allowing clinicians to identify these patients several months in advance of other clinical signs or symptoms of rejection and before irreversible damage to the allograft.

Surveillance with dd-cfDNA serves many critical purposes: detecting subclinical rejection before irreversible allograft injury and enabling precise immunosuppression titration, which varies by patient. Thus, testing to establish individual patient baseline and trajectory of initial dd-cfDNA is required.

Medium- to long-term graft survival has stagnated for over 30 years, partly due to immunosuppression-related complications and antiquated surveillance modalities, so that newer and better surveillance is essential beyond the first year. The draft's limits, tied to outdated biopsy frequencies (historically ~4 in year one due to procedural risks), do not account for dd-cfDNA's non-invasive nature and favorable risk-benefit profile. KDIGO guidelines recommend increased frequencies for non-invasive testing (serum creatinine, urine protein excretion) for allograft function/rejection than they do for biopsy in large part due to the more favorable risk-benefit profile of non-invasive testing. Dd-cfDNA fits a similar risk-benefit profile of serum creatine and proteinuria, however it is a far more sensitive test for early rejection. Thus, it makes more sense to align dd-cfDNA testing for rejection to other non-invasive testing frequencies rather than invasive testing via biopsy.

Several studies demonstrate conclusively that dd-cfDNA elevations precede biopsy-proven rejection by many months, allowing early intervention and improved outcomes. These same studies show that other clinical tests (e.g., creatinine, urine protein) are not sensitive, specific or reliable for early detection of treatable rejection.

The KDIGO guidelines suggest using the lowest planned doses of immunosuppression by month 4 post-transplant suggesting that at minimum patients should be surveilled monthly during this period with ongoing surveillance quarterly thereafter to ensure immunosuppression minimization does not subject the patient to graft injury or rejection.

The KOAR study showed rejection yield was 39% with elevated dd-cfDNA in surveillance versus 7% without, enhancing pre-test probability for biopsy-proven acute rejection (BPAR), demonstrating that this test focuses resources of biopsy on those patients that need them most.

The American Society of Transplant Surgeons (ASTS) statement explicitly endorses dd-cfDNA for surveillance, noting clear clinical utility in longitudinal testing despite lacking prospective trials on optimal frequency. It highlights the DART study's ARTS protocol—monthly testing months 1-4, then quarterly—as empirically sound. This regimen (7 tests in year one) focuses on high-risk, early periods (planned reductions of immunosuppression) while enabling ongoing monitoring. Quarterly testing post-year one (4 annually) better supports immunosuppression optimization, especially for high-risk patients like those with marginal organs under the Increasing Organ Transplant Access (IOTA) model, who face elevated delayed graft function and rejection risks.

Evidence from large cohorts, including ADMIRAL (1,092 recipients) and a Nature Medicine study (2,882 recipients), confirms dd-cfDNA's performance in diverse populations, including repeat transplants (8-15.5% of cases), with similar discrimination between quiescence and rejection. Longitudinal monitoring yields high positive (77%) and negative (85%) predictive values for antibody-mediated rejection (AMR), outperforming single-point testing. Serial dd-cfDNA also tracks treatment response and de novo donor-specific antibody (DSA) development, with elevations predicting risks. No studies link higher frequencies to harm; conversely, restrictions may delay detection, increasing the need for biopsy and costs for returning to dialysis.

In summary, while the draft advances coverage, frequency caps undermine physician discretion and evidence-based care. I recommend removing limits or adopting ARTS-aligned frequencies (7 in year one, 4 annually thereafter) to reflect dd-cfDNA's utility in improving graft outcomes in transplant and repeat transplant recipients.

References were provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

I have had an opportunity to review the draft policy, and I would like to express my reservations about its content.

I am transplant cardiologist. I have spent more than a decade researching donor-derived cell-free DNA (dd-cfDNA) as a marker for rejection in organ transplantation.

The draft policy incorrectly positions dd-cfDNA and gene expression profiling (GEP) as simply indicators of rejection: dd-cfDNA an indicator of antibody mediated rejection (AMR) and acute cellular rejection, and GEP as an indicator of acute cellular rejection (ACR). In fact, the two tests, while both molecular, are completely different tests with dd-cfDNA being a marker of graft injury and GEP being a marker of immune system activation.

When I was asked in 2022 whether there was any data supporting the use of GEP and dd-cfDNA together, I answered that while there was a rationale, there was no data. However, this has definitively changed over the past three years. I was the first author of the SHORE publication that sought to answer this question directly and the conclusive answers it provided were:¹

1) Combination testing with GEP and dd-cfDNA more accurately identifies cardiac transplant patients undergoing surveillance who are at risk of acute cellular rejection.

2) Clinicians use both tests when making decisions about whom to biopsy.

I disagree with the implication and interpretation of the study that dd-cfDNA alone is good enough in assessing risk, as the patients with a positive dd-cfDNA had a 7% risk of ACR compared to a 9.2% risk of ACR in those with a dual positive result. Patients with a positive dd-cfDNA but a negative GEP are at much lower risk and could potentially avoid a biopsy if complete risk stratification were applied. By unnecessarily grouping all patients with positive dd-cfDNA together, we would be committing some patients to unnecessary invasive procedures.

In the same manuscript, we demonstrated that clinicians understood this and managed patients accordingly. In patients who had a positive dd-cfDNA, physicians performed 33% fewer biopsies in those with a negative GEP compared to those with a positive GEP, a trend that became more pronounced over time. Overall, this led to fewer biopsies over time and overall fewer biopsies than is typically performed in patients not undergoing molecular surveillance- all while maintaining excellent outcomes. This data should be included in the policy.

Of note, I am also the corresponding author on a forthcoming publication that now shows that not only are patients with a dual positive GEP and dd-cfDNA at risk of ACR, they are also at risk of future graft dysfunction and cardiovascular death. Patients with an elevated dd-cfDNA but a negative GEP were not at risk in this study.

The only data presented in the draft policy that refuted the complementarity of GEP and dd-cfDNA was contained in the manuscript by Rodgers et al.² I had previously reviewed this paper in detail. It was a much smaller single center study that aimed to evaluate whether GEP and dd-cfDNA had superior performance characteristics to dd-cfDNA alone. The authors elected to define positive combined tests as either a positive GEP or a positive dd-cfDNA which led to two issues:

1) It optimized the test’s sensitivity, which is much less relevant in the context of surveillance

2) It was not powered to evaluate differences in sensitivity of tests since it only included 10 episodes of acute cellular rejection.

While I understand a goal of policy development is to have a comprehensive literature review, these limitations need to be included in the descriptions of the studies.

Finally, I would like to express my concerns about capping surveillance at 2 episodes per year after the first-year post-transplant. Surveillance, particularly frequent surveillance, after the first-year post-transplant is controversial in the cardiac transplant community- but this controversy stems directly from the increased difficulty in performing surveillance biopsies later post-transplant with both a higher chance of complications and a higher likelihood of obtaining non-diagnostic material.³ In the era of non-invasive molecular surveillance, this limitation is no longer relevant. We typically see our patients and perform molecular surveillance quarterly in the second year and continue to adjust immunosuppression based on these results. We know that medium to long term survival of heart transplant patients has remained stagnant for the last two decades and generally our patients succumb to complication from over-immunosuppression.⁴ I genuinely believe that extended surveillance with ongoing optimization of immunosuppression is the only way we will extend the life of our patients and prevent long-term complications. As a national payer, Medicare should be encouraging close monitoring. If centers are seeing their patients quarterly, they should have the freedom to completely assess them with all tools available.

Thank you for considering these important points in the finalized LCD policy.

References were provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

As organizations dedicated to serving the transplant community, we appreciate the opportunity to comment on the proposed Local Coverage Determination (LCD). We thank the Centers for Medicare and Medicaid (CMS) and the Medicare contractors (MACs) for issuing a Draft LCD and soliciting feedback through a comment period versus issuing coverage restrictions through Billing and Coding Articles.

Our organizations provide comprehensive support, education, advocacy, and connection for transplant recipients, donors, families, caregivers, and professionals. Together we are committed to advocating for policies that support transplant patient access to the latest innovations in care, including non-invasive, post-transplant diagnostic testing.

Every transplant represents both a second chance at life and the start of a complex, lifelong medical journey. Patients undergo intense medication regimens, financial and emotional pressures, and uncertainty about their long-term health. Timely and reliable diagnostic tools can mean the difference between catching early signs of rejection and facing devastating setbacks. That is why access to innovative, non-invasive blood tests is so critical to the communities we serve.

We are encouraged that CMS and its contractors recognize the importance of maintaining coverage for innovative blood tests that have transformed post-transplant care in recent years. These tests have reduced the reliance on invasive and often painful or even harmful biopsies. They allow clinicians to detect signs of rejection earlier and with greater confidence. We strongly support CMS’s affirmation of coverage, acknowledging the real-world benefits that patients have seen from these tools.

We are concerned, however, with the proposed limitations on surveillance testing frequency. Continued access to donor-derived cell-free DNA (dd-cfDNA) testing is critical for detecting early signs of rejection and guiding clinical decision-making based on the personal needs of the patient.

Leading clinician organizations, including the American Society of Transplant Surgeons (ASTS), have emphasized that surveillance testing should be flexible and determined by physician judgment, with higher frequency testing available when medically necessary, and consistent with national consensus guidelines.

A one-size-fits-all approach does not reflect this evidence or the real-world needs of diverse transplant populations. Factors such as donor type, immunologic risk, comorbidities, and prior rejection episodes vary widely. Limiting testing to a handful of time points may miss early signs of immune activation. By the time clinical symptoms appear, irreversible damage may have already occurred.

The field of transplant is undergoing its first reform in over 40 years and is an actively evolving field of medicine. There are over 90,000 Americans currently waiting for a kidney transplant. For those who receive a donor kidney, effective post-transplant care requires access to tools that allow their care teams to monitor them, ensuring the best long-term outcomes, and preventing the need to be retransplanted. The Increasing Organ Transplant Access (IOTA) initiative calls for both transplant centers and organ procurement organizations (OPOs) to potentially increase the use of previously discarded organs to facilitate more transplants. There has also been an increase in the use of organs from donation after circulatory death (DCD) donors. The use of these organs will require care teams to monitor patients even more closely post-transplant.

A surveillance strategy that prioritizes minimum testing over personalized testing may inadvertently delay earlier detection of rejection, compromising a transplant’s longevity. We ask that CMS revise the proposed policy to align with expert recommendations and allow physicians the discretion to order additional testing when in the best interest of their patients.

By ensuring access to non-invasive molecular testing, CMS can help protect the health of transplant recipients, strengthen their quality of life, and safeguard the extraordinary gift of organ donation.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

I am a transplant cardiologist and the associate medical director of the heart transplant program. I was the first author of the referenced article: The End of Endomyocardial biopsy?: A practical guide for noninvasive heart transplant rejection surveillance publication.¹

I would like to clarify the following:

1) When describing the complementarity of GEP and dd-cfDNA our intention was to describe the two tests as mechanistically quite different which goes beyond that one test may operate better in ACR and the other better in AMR. In fact, they are measuring two completely different aspect of the pathophysiology of rejection- GEP is assessing the status of the recipients immune system and dd-cfDNA is assessing the status of the graft.

2) Given the above, our intention in the manuscript was not to advance the idea that “only dd-cfDNA alone has the potential to replace or significantly reduce the need for EMB.” In fact, clinicians continue to use and will continue to use multimodal non-invasive tests to surveil their patients including GEP, dd-cfDNA but also clinical examination, echocardiography, donor specific antibodies and drug trough levels. In the paper, we suggested that we needed data to understand how best to use multimodal molecular testing and to quantify the effect of each result, not that multimodal testing was not necessary when felt to be clinically helpful. At the time of submission, such data was not available.

Since our published manuscript, I was pleased to see the publication of the first SHORE manuscript which precisely answered our proposed question.² As we had postulated, a negative dd-cfDNA does moderate the risk of an elevated GEP but also, a negative GEP moderates the risk of ACR with a positive dd-cfDNA. This data is helpful in determining the need for EMB. For example amongst those with modestly elevated dd-cfDNA with a negative GEP and without risk factors for antibody mediated rejection an EMB might not be necessary.

I think this sequence of events is exactly how scientific discourse should occur. Clinicians and scientists highlight gaps in data that is then answered by robust data analysis in large studies. I hope that going forward, and as the scientific field evolves, medical policy will continue to support evidence based use of multimodal molecular testing with the use of GEP and dd-cfDNA allowing clinicians to deliver the best possible care for their patients.

References provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

I have reviewed the draft Local Coverage Determination policy. While I appreciate the detailed review of the literature, there were important errors and omissions that I would like to highlight. I am a transplant cardiologist and have been involved in clinical practice and research for more than 30 years and led the initial development of the AlloMap test for cardiac transplant rejection. The draft LCD should add important considerations regarding the clinical validation of the test and its utility in post-transplant care.

AlloMap has been extensively studied: 1) AlloMap was validated in two separate rigorous clinical validation studies, CARGO and CARGO II.^1,2 These studies did not rely on a single clinical pathologist determination of quiescence or rejection. Rather, biopsies were independently reviewed by three separate pathologists for the diagnostic determination of the reference standard to which AlloMap was then compared. This is important, as we and others have documented tremendous variability in pathological interpretation of endomyocardial biopsies with a false positive rate as high as 40%.² No other molecular tests in transplantation have undergone this level of study design, i.e. against a centrally reviewed biopsy. 2) To date, only AlloMap has been formally studied in randomized controlled clinical utility studies. IMAGE³ and subsequently eIMAGE⁴ definitively proved that AlloMap could safely replace surveillance biopsies as early as 2 months post-transplant. This is the reason that the recent ISHLT guidelines expanded their recommendation for AlloMap and acknowledged the AlloMap could be used for acute cellular rejection surveillance beginning at two months post-transplant.⁵ These important studies, as well as their acknowledgement by the 2023 guidelines should be included in the policy.

While dd-cfDNA has proven invaluable in the management of my transplant patients, I definitively view it as an additive test to AlloMap, not a replacement test. Importantly, while dd-cfDNA can detect rejection through evaluation of donor allograft injury, AlloMap detects early evidence of acute cellular rejection through the profiling of allograft recipient’s gene expression of peripheral blood mononuclear cells. As neither test has perfect performance characteristics for detection of rejection and the tests are measuring different aspects of the pathophysiology of rejection, the tests would be accurately described as complementary and adjunctive to one another. I have personally used the tests together to: A) Enhance rejection detection, including when AlloMap was elevated and dd-cfDNA was normal. B) Safely avoid biopsies when dd-cfDNA was mildly elevated but AlloMap was completely normal.

For these reasons, I would caution policy makers accusing clinicians who choose to use GEP testing as using an "inferior service." In my experience, as noted above, and in the recent SHORE publication,⁶ combination testing with AlloMap and dd-cfDNA demonstrates additive benefit and superiority over using either testing alone, allowing us to safely spare patients from unnecessary, invasive and costly biopsies while maintaining excellent survival, and, thus, helping to achieve high value healthcare.

References were provided for review.

Thank you for your comment. Please see Response #1.

The following was submitted to Palmetto GBA:

Introduction

Natera commends MolDX for drafting a thoughtful and comprehensive proposed LCD for molecular testing in solid organ allograft rejection. We appreciate the inclusion of recent data and publications that reflect a deep understanding of the evolving role of donor-derived cell-free DNA (dd-cfDNA) in transplant care.

We are particularly encouraged to see the proposed expansion of coverage for surveillance testing in kidney, heart, and lung transplantation. To further optimize patient care and align coverage with current evidence, we respectfully submit the following comments and suggested revisions.

1. Maximum Number of Surveillance Timepoints

The proposed LCD limits surveillance testing to:

• Kidney: 4 timepoints in the first year
• Heart: 12 timepoints in the first year
• Lung: 12 timepoints in the first year
• Two timepoints per year after year one

Recommended Revision: Based on published evidence and clinical consensus, we recommend:

• Kidney: at least 7 timepoints in the first year
• Heart: at least 12 timepoints in the first year
• Lung: at least 12 timepoints in the first year
• All organs: at least 4 timepoints per year thereafter

Rationale:

• Kidney:

○ KDIGO guidelines recommend frequent monitoring with multiple biomarkers (serum creatinine, BKV, EBV), typically at least 7 times in the first year. American Journal of Transplantation 2009;9(Suppl3):Sii-Sii

○ The American Society of Transplant Surgeons (ASTS) issued a position statement (ASTS-Statement-on-donor-derived-cell-free-DNA-(dd-cfDNA)—-Updated-Oct.-2024.pdf) explicitly endorsing 7 surveillance tests in year one and 4 tests annually thereafter for kidney transplant recipients.

○ dd-cfDNA significantly outperforms creatinine in detecting rejection, and recent prospective studies (e.g., Bromberg et al., Am J Transplant, 2025) confirm its predictive value months prior to biopsy-proven rejection.

• Heart:

○ ISHLT guidelines recommend quarterly surveillance after the first year, with biomarker, imaging, and molecular diagnostics.

○ dd-cfDNA has been validated as an early and noninvasive marker of rejection (Kim et al., Am J Transplant, 2025; Agbor-Enoh et al., Circulation, 2021).

• Lung:

○ Standard of care involves quarterly surveillance post–year one with spirometry, imaging, and laboratory tests.

○ Multiple multicenter studies, including those by the GRAfT consortium, support dd-cfDNA monitoring on a monthly basis during year one and quarterly thereafter.

2. Clarification on Bundled Payments

The accompanying Billing & Coding Article defines one unit of service as a bundle of surveillance timepoints. We respectfully note that Natera’s Prospera^® cfDNA test is a PLA-coded assay priced by CMS at $2,753 per assay.

Recommendation: Any bundled payment methodology should be calculated using the established CMS CLFS rate as the baseline, reflecting the number of medically necessary timepoints approved in the final LCD.

3. Coverage for Multi-organ Transplantation

The proposed LCD states: “To-date, NO test has been approved for use in patients who have undergone multiple transplants.”

Recommended Revision: While this may apply broadly, sufficient evidence exists for simultaneous pancreas-kidney (SPK) transplantation. Multiple studies (Williams et al., Transplant Direct, 2022; Ventura-Aguiar et al., Transplantation, 2022; Liu et al., Research Square, 2024) demonstrate dd-cfDNA as a validated and reliable biomarker for rejection surveillance in SPK. Accordingly, coverage should extend to SPK patients.

Conclusion

Natera strongly supports MolDX’s proposed LCD expansion for dd-cfDNA in transplant surveillance. To ensure alignment with current clinical evidence and best practices, we respectfully recommend:

1. Increasing surveillance frequency to 7 timepoints in the first year for kidney, 12 timepoints in the first year for heart and lung, and at least 4 timepoints annually thereafter.
2. Adopting bundled payment rates derived from the established CLFS rate for Prospera.
3. Including coverage for SPK transplantation, where evidence demonstrates clinical validity and utility.

We appreciate the opportunity to provide input and remain committed to collaborating with MolDX to ensure optimal patient outcomes.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

As organizations representing the heart transplant patient community, we appreciate the opportunity to submit a public comment in response to the proposed local coverage determination (LCD) regarding molecular testing for solid organ allograft rejection. Our organizations are committed to supporting heart transplant patients and their families, whether through medical system navigation, financial relief, housing stability, mental and emotional support, and much more.

We understand that a successful transplant is only just the beginning of a new care journey for patients, and monitoring for potential signs of rejection or failure is critical for preserving their transplanted organ. That is why together we are deeply committed to advocating for evidence-based coverage policies that support patient access to innovative, non-invasive, post-transplant diagnostic testing.

We emphasize our strong support for the affirmation of surveillance testing coverage for heart transplant recipients. We are concerned, though, with two unexpected, proposed changes in coverage that we feel are not supported by current professional guidelines or standards of care, and will negatively impact post-transplant care for the heart recipients we collectively represent:

1. Loss of coverage for multimodality testing, which combines the use of donor-derived cell-free DNA (dd-cfDNA) and gene expression profiling (GEP) together.
2. Adding coverage restrictions or caps on surveillance testing frequency.

Our community has relied upon multimodality testing with GEP and dd-cfDNA together to monitor heart transplants for early signs of rejection so clinicians can intervene before dysfunction occurs. Multimodality testing has also been a critical tool to help avoid unnecessary biopsies.

Patient testimonials and studies have demonstrated that heart transplant patients experience less anxiety and pain with non-invasive testing compared to invasive surveillance biopsies. Without access to multimodality testing, more patients will undergo biopsies and experience the associated complications. Further, in cases of heart transplant failure, there is no alternative lifeline for patients — re-transplantation is the only option. With exceedingly high emotional and financial costs associated with a heart transplant, coverage for innovative monitoring tools such as multimodality testing is crucial for transplant preservation and patient health.

Finally, the International Society of Heart and Lung Transplantation (ISHLT) guidelines recommend GEP and dd-cfDNA as part of the follow-up visit and testing schedule. The recommendation was noted by the guideline to be stronger for non-invasive testing (GEP and dd-cfDNA) than endomyocardial biopsy.

We ask CMS to carefully consider and defer to the clinician community’s expert recommendations and feedback on surveillance testing and remove any proposed caps. Through our joint voices, we hope to underscore the importance of surveillance testing and physician choice when determining what is in the best interest of the patient’s health, as no two patients, or their needs, are the same.

On behalf of the many heart transplant recipients who have benefitted from access to multimodality, non-invasive, post-transplant diagnostic tests, we thank you for the affirmation of surveillance coverage, which recognizes the importance of early detection of potential rejection. We urge your careful consideration of professional guidelines and clinician recommendations for surveillance testing frequencies as well as restoring coverage for multimodality testing in the final proposed LCD.

Thank you for the opportunity to lend our perspective on these important issues.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA:

I am writing to advocate for the continued and broader use of donor-derived cell-free DNA (dd-cfDNA) testing as a routine screening tool in the management of kidney transplant recipients. It has been supported by KDIGO guidelines to perform frequent monitoring of kidney transplant allografts, several times a year by many biomarkers and test to ensure the health of the transplanted kidney, for example serum creatinine is recommended to be checked at least 7 times in the first year.

As you know, timely detection of allograft injury is critical to improving long-term graft outcomes. While serum creatinine and other conventional markers remain standard, they often lack sensitivity and may indicate damage only after significant injury has occurred. In contrast, dd-cfDNA testing offers a non-invasive, highly sensitive method to detect early signs of rejection—sometimes before functional decline is evident.

Numerous studies have demonstrated that elevated levels of dd-cfDNA correlate strongly with active rejection, and that regular surveillance can allow for earlier intervention, better patient stratification, and potentially reduced need for invasive biopsies. In clinical practice, this tool could be particularly valuable in identifying subclinical rejection and in tailoring immunosuppression regimens to individual patients.

Given the increasing body of evidence supporting dd-cfDNA testing, I believe it is time to integrate this approach more fully into routine post-transplant monitoring protocols. Doing so could lead to improved graft survival, reduced morbidity, and ultimately, better quality of life for our transplant patients.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA:

Thank you for including surveillance with donor-derived cell-free DNA (dd-cfDNA) in the latest proposed LCD. I am writing to advocate for an expansion of the LCD to cover additional testing.

Kidney transplantation is a life-saving intervention, yet long-term graft survival remains challenged by both acute and chronic allograft injury. Traditional monitoring strategies—such as serum creatinine, proteinuria, and protocol biopsies—lack sensitivity and specificity, often identifying injury only after irreversible damage has occurred.

Donor-derived cfDNA offers a validated, noninvasive biomarker that provides earlier and more accurate detection of allograft injury. Multiple studies have demonstrated that dd-cfDNA levels correlate with acute rejection, subclinical injury, and long-term graft prognosis. Routine surveillance with dd-cfDNA enables timely intervention, reduces unnecessary biopsies, and supports precision in immunosuppressive management.

Based on current evidence and clinical practice patterns, I recommend more flexibility of testing surveillance schedule. In discussion with my colleagues, it has been proposed that up to 7 dd-cfDNA tests in the first post-transplant year and up to 4 in the following years. This approach balances early intensive monitoring—when the risk of rejection and immune-mediated injury is highest—with sustained long-term surveillance to safeguard graft health. This is particularly important given the 5-month lead-time for the detection of ABMR and 2-month lead time for TCMR. (Bromberg JS, Bunnapradist S, Samaniego-Picota M, Anand S, Stites E, Gauthier P, Demko Z, Prewett A, Armer-Cabral M, Marshall K, Kaur N, Bloom MS, Tabriziani H, Bhorade S, Cooper M; ProActive Investigators. Elevation of Donor-derived Cell-free DNA Before Biopsy-proven Rejection in Kidney Transplant. Transplantation. 2024 Sep 1;108(9):1994-2004. doi: 10.1097/TP.0000000000005007. Epub 2024 Aug 20. PMID: 38595232; PMCID: PMC11335081). It also aligns with the normal frequency of laboratory assessment post-transplant.

In my clinical experience, routine incorporation of dd-cfDNA surveillance has meaningfully altered patient management. I have seen cases where rising dd-cfDNA levels prompted early biopsy and treatment, preventing irreversible rejection and preserving graft function. Conversely, stable dd-cfDNA trends have allowed us to safely reduce unnecessary biopsies and adjust immunosuppression with greater confidence. This has not only improved patient outcomes but also reduced anxiety for recipients and streamlined care delivery.

Given the significant implications for patient outcomes, health system efficiency, and stewardship of scarce donor organs, the integration of dd-cfDNA surveillance using this schedule represents a clinically sound and evidence-based advancement in post-transplant care.

Thank you for your consideration of this important measure to optimize care for kidney transplant recipients.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

I have reviewed the LCD on molecular testing in cardiac transplantation specified above and would like to offer my perspective on the document and the review of the data.

I am currently the Director of Advanced Heart Failure and Cardiac Transplantation, a professor of medicine, and practicing transplant cardiologist.

I must first explain that while dd-cfDNA and GEP are both molecular tests that can evaluate rejection risk- they are very different tests, one examining injury to the heart and the other assessing the reaction of the patient’s immune system. Because of that, the idea that us, as clinicians, should be restricted to using only one while surveilling our transplant patients for rejection is illogical. There may be situations where only one or the other is warranted, but most often surveillance with combined testing is the best choice for our patients. The two tests measure different aspects of rejection and neither performs perfectly for predicting what a surveillance biopsy might show.

The value of combined testing is precisely what we set out to assess and characterize in SHORE, for which I am a member of the steering committee, and an author of the first publication.¹ The manuscript shows how both tests together accurately risk stratify patients. I take exception with the characterization that there was no “notable” improvement in the specificity of the combined testing. In fact, the increase in specificity with a combined positive test, yielded a full point increase in the positive likelihood ratio that was statistically significant. This should be acknowledged by the draft policy.

Furthermore, the notion that the risk of ACR in those patients with a positive dd-cfDNA at 7% was similar to those with a dual positive result at 9% was misses the value GEP brings at determining which of the patients with a positive dd-cfDNA have 4% risk of ACR and which have a 9 percent risk. The value of GEP in this context is to help avoid biopsies in patients who do not need them. We definitively demonstrated that in the section that described how clinicians responded to different molecular test results. Compared to patients with a dual positive test, clinicians appropriately biopsied those with a positive dd-cfDNA but negative GEP 33% less. This data should be added to the policy. By preventing clinicians from leveraging both tests together, invasive biopsies will undoubtedly be performed more frequently.

Furthermore, the SHORE authors submitted a manuscript based on data I presented at ISHLT this year and are currently revising for publication in response to a positive review. This impending publication demonstrates that these patients with dual positive results are not only at the highest risk for ACR on a current biopsy but are at risk of future cardiovascular death and graft dysfunction, even when there is no evidence of rejection on the biopsy. This risk was not seen in patients with a positive dd-cfDNA and negative GEP, again demonstrating that GEP defines a group of patients with elevated dd-cfDNA that are at highest risk and which ones are not. Having the GEP available as part of the molecular assessment profile can spare the ones with a negative GEP more intense monitoring and potentially unnecessary increases in toxic immunosuppression.

We have confirmed all of the above in an analysis that we expect will soon be published after also receiving a positive review. In this manuscript evaluating cardiac transplant patients followed with combined molecular testing, most of whom were not enrolled in the SHORE registry, we demonstrate that GEP, independent of dd-cfDNA, confers risk of current rejection and future adverse events with those with combined positive results having the highest risk.

Finally, despite being one of the largest heart transplant centers in the country, we have committed to continuing frequent surveillance beyond the first-year post-transplant and usually see our patients quarterly in year 2 and 3. We have found that this has allowed us to continue to optimize immunosuppression and allow patients to be on exactly the right dose for the patient. If we are prepared to see these patients four times yearly, we should be enabled to obtain as much useful information as possible in the Medicare population.

I think it would be a tremendous disservice to the transplant community and cardiac transplant patients to prevent clinicians from combining these two complementary molecular tests during surveillance and limiting surveillance episodes to only twice yearly in year 2 and 3.

Reference provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

As the Medical Director for lung transplantation, I am writing to provide clinical feedback on Draft LCD, which outlines coverage for donor-derived cell-free DNA (dd-cfDNA) testing in solid organ transplant recipients, including lung transplant patients.

The inclusion of dd-cfDNA testing in the draft policy is a positive step forward, recognizing its growing role in transplant surveillance. Monthly testing during the first year post-transplant is consistent with current literature and our clinical experience. In my practice, we also routinely perform quarterly dd-cfDNA testing beyond the first year, as lung transplant recipients remain at elevated risk due to their unique immunologic and anatomic vulnerabilities.

However, the proposed fixed testing schedule raises concerns. Lung transplant patients are not a homogeneous group, and their post-transplant trajectories vary widely. A rigid testing cadence does not reflect the clinical realities of transplant medicine. Physicians must retain the ability to tailor surveillance strategies to individual patient needs, based on clinical judgment, risk factors, and evolving circumstances.

Limiting dd-cfDNA testing to two per year after the first year may compromise care for patients with complex courses or heightened immunologic risk. The Keller et al. study, which the draft policy references, supports quarterly testing beyond year one. All four centers in that study agreed on monthly testing in year one and quarterly testing thereafter. Our practice aligns with this approach, and we believe the final policy should reflect this consensus.

We strongly recommend that the final policy allow for physician-directed testing protocols, enabling transplant centers to order dd-cfDNA testing based on individual patient risk and clinical discretion. At a minimum, coverage should include 12 tests in the first year and quarterly testing thereafter for stable patients. This approach supports personalized medicine and ensures that transplant recipients receive the most appropriate and timely care.

Thank you for your commitment to evidence-based policy.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA:

On behalf of the College of American Pathologists (CAP), we thank you for the opportunity to comment on Palmetto GBA Proposed LCD - MolDX: Molecular Testing for Solid Organ Allograft Rejection.

The CAP is the world’s largest organization of board-certified pathologists and the leading provider of laboratory accreditation and proficiency testing programs. The CAP serves patients, pathologists, and the public by fostering and advocating for excellence in the practice of pathology and laboratory medicine worldwide.

The CAP commends MolDX for recognizing the role of plasma-based molecular diagnostic methods that help monitor solid organ transplant and provide information to help optimize immunosuppressive therapy in post-transplant Medicare beneficiaries. Molecular experts from the CAP have reviewed the proposed policy and agree that the coverage criteria, as outlined, are reasonable based upon currently available evidence.

Thank you again for the opportunity to review and comment on this proposed policy.

Thank you for your comment and support of this LCD.

The following comment was submitted to Palmetto GBA:

On behalf of Verici Dx, I am writing to submit public comments on Proposed LCD (“MolDX: Molecular Testing for Solid Organ Allograft Rejection”) in support of establishing coverage criteria for allograft rejection testing. Verici Dx appreciates MolDX’s efforts to make updates to the LCD to clarify the coverage criteria. Verici Dx offers and performs the Tutivia test described by CPT 0320U and Z-code identifier. Tutivia has been covered by Medicare via a MolDX TA under the current MolDX LCD for solid organ allograft rejection testing.

We respectfully make the following recommendations and clarifications in order to ensure that the finalized LCD enables clinicians to order medically necessary testing for allograft rejection.

1. Ensure that the list of covered for cause testing indications reflects the use of molecular testing to guide appropriate medication management.
2. Clarify the policy governing coverage of multiple molecular tests per encounter.
3. Provide examples of favorable benefit to risk profile for coverage of molecular tests.
4. Clarify or remove bundled payment methodology for surveillance indication.
5. Clarify or remove language suggesting superiority of cfDNA testing over GEP testing.
6. Clarify or remove language that no tests have been approved for patients with multiple transplants.

Verici greatly appreciates your consideration of these important comments.

1. Addition of Medication Management to For Cause Testing

First, Verici Dx recommends clarification of the application of the definition of “for cause” testing to include use of a molecular test to inform medication management. The LCD states that for cause testing includes testing “[t]o assist in the evaluation of adequacy of immunosuppression or response to treatment, wherein a non-invasive or minimally invasive test can be used in lieu of a tissue biopsy”, as well as testing “[f]or further evaluation of allograft status for the probability of allograft rejection after a physician-assessed pretest review of clinical and biological factors concerning for risk of rejection.”

We want to highlight the importance of for cause testing for the purpose of evaluation of medication changes. In some clinical cases, a pre-test may not indicate injury or rejection signs; however, medication modulation is purposeful and different from regular surveillance protocols in stable patients. We believe that it is important to clarify that assessments related to medication changes constitute for cause testing even if injury or rejection signs are not present. An example may be a patient with previous history of malignancy for whom the treating physician believes that they are able to minimize immunosuppression protocols to reduce risk of recurrence of the malignancy but also wants more purposeful monitoring when reducing the immunosuppression.

We encourage MolDX to clarify that this clinical use of biomarker testing falls under coverage of testing “to assist in the evaluation of adequacy of immunosuppression or response to treatment” in the Final LCD. This clarification is important to differentiate the medication management indication, which might not be ordered in lieu of a biopsy, and for which there may not yet be signs of rejection from a pre-test. However, the clinician is aware that he/she is making changes that increase the risk of rejection, and there is a clinical benefit to the patient if this can be identified early should a rejection occur.

2. Clarification of Rules Governing Coverage of Multiple Tests Per Encounter

Under the Proposed LCD, “[f]or a given patient encounter, only one molecular test for assessing allograft status may be performed.” The Draft Billing and Coding Article states that “[a]ny additional molecular services billed after the first will be denied.”

We encourage MolDX to clarify how this requirement is to be implemented. Our understanding is that this is to be applied on a test-by-test basis – in other words, if a treating physician orders multiple molecular tests for allograft rejection status from a given laboratory for a single patient encounter, claims will be denied for all tests after the first one.

We do not understand this coverage criterion to mean that if a treating physician determines that it is medically necessary to order a cfDNA-based allograft rejection test from one laboratory and a GEP-based test from another laboratory, only the first of these tests to be billed would be payable. This is a determination best made by the treating physician.

Moreover, it is not clear how a laboratory would reliably be able to ensure that the treating physician had not ordered a test from a different laboratory. This proposed coverage criterion, if finalized, is likely further to cause transplant center clinical staff and patient concerns regarding their clinician’s ability to order tests that they deem medically necessary and that have historically been covered under the LCD, given the risk that these claims may go on to be denied based on information unavailable to the performing laboratory at the time of test order.

3. Provision of Examples of Favorable Benefit to Risk Profile

Also requires that “[f]or minimally or non-invasive tests, the benefit to risk profile of the molecular test is considered by the ordering clinician to be more favorable than the benefit to risk profile of a tissue biopsy, or a tissue biopsy cannot be obtained, when the test and biopsy provide similar information. For example, this may be the case if a biopsy is considered medically contraindicated in a patient.”

Verici Dx agrees that the assessment of a molecular test’s benefit to risk profile relative to tissue biopsy is appropriately within the clinical determination of the treating physician. However, we do recommend that the final LCD offer guidance relative to indications where the literature indicates that the benefit to risk profile for biomarker testing may be favorable. Such examples will reduce administrative burdens and streamline beneficiary access for those indications for which MolDX agrees that a favorable benefit to risk profile may exist.

For instance, in obese patients, biopsy is more challenging, presenting higher risks. In obese patients who underwent robotic surgery for transplant, biopsy procedures typically require general anesthesia, again posing higher risk. Biomarker testing may thus have a more favorable benefit to risk profile relative to biopsy in obese patients. Because obese patients experience a chronic state of inflammation, it is important to assess biomarker performance in this population. Tutivia testing assessed in clinical validation and in real-world use produced an accuracy of 84.21% in obese patients (1).

Most notably, Verici Dx believes that the clinical data for Tutivia is supportive of a favorable benefit to risk profile for molecular test for solid organ allograft rejection in patients with previous failed transplant with test performance accuracy matched to first-time kidney transplant recipients (2). Often patients with previous failed transplant(s), currently representing about 15% of the transplant waiting list, may have multiple previous failed allografts present and some will have an intraperitoneal kidney. Biomarker testing may thus have a more favorable benefit to risk profile relative to biopsy in patients with multiple kidney transplants.

The clinical data analysis for Tutivia supports the benefit of molecular testing to assess allograft rejection in this population. As an RNA signature, Tutivia uses next-generation sequencing relying on gene expression in the peripheral blood of kidney transplant recipients to inform the algorithm and does not rely on methods requiring differentiating donor(s) from recipient. In the Tutivia clinical validation study (3), 21 of 151 (14%) patients had at least one previous failed kidney transplant. To assess impact of previous failed transplants on Tutivia results, the performance characteristics in the overall cohort and relative to previous failed kidney transplant recipients are provided below.

The clinical characteristics (sensitivity, specificity, NPV and PPV) of the multi-kidney subset (n=21) of the Tutivia validation set are nearly identical to the performance characteristics from the overall cohort where 130 (86%) of patients are first time sole kidney transplant recipients. This evidence demonstrates that Tutivia can support clinicians in decision making in this important sector of the transplant population.

4. Clarification or Removal of Bundling Proposal for Surveillance Indication

Verici Dx strongly supports the LCD’s proposed coverage for molecular testing for surveillance of allograft rejection. We note that the Proposed LCD states that “[a]t this time, the current evidence supports a maximum number of surveillance timepoints for evaluation in the first-year post-transplantation as follows: Kidney (4), Heart (12), and Lung (12). After the first year, surveillance timepoints may continue at a decreased frequency of 2 per year.” We request additional clarification on the proposal in the Billing and Coding Article for a requirement for a unique Z-code for each of the following indications for every covered test: for cause testing, surveillance in Year 1, and surveillance in subsequent years.

We support frequency limitations that are grounded in the clinical literature and provide adequate discretion for treating physicians; however, we request additional detail and opportunity for further comment on the proposal to establish multi-test bundles identified by new Z-codes.

First, any proposal for bundling should reflect the pricing for covered tests that are priced nationally on the Clinical Laboratory Fee Schedule (CLFS) when furnished for a surveillance indication. To the extent MolDX utilizes bundling to make payment for nationally-priced tests covered under a finalized LCD, MolDX is required to price those bundles at the number of tests per bundle TIMES the CLFS rate for the test.

Additionally, we request clarification of how bundled payment for solid organ allograft rejection surveillance testing would work from an implementation and operational standpoint. As an example, there likely will be scenarios where a treating physician orders one molecular test for the initial surveillance testing after transplant and then switches to a different test. In our experience, some transplant centers order Tutivia testing in the first months after transplant because cfDNA can give them highly variable results early post-transplant. However, the treating physician may switch to cfDNA for later timepoints. Under a bundled approach, it is not clear how testing would be billed in this circumstance. In contrast, without the bundled approach, the performing laboratory could simply bill for the test it performs.

Given these operational complexities, we believe that the proposed bundling of molecular tests covered under draft LCD for payment for the surveillance indication is not the best approach. To the extent MolDX finalizes this bundling methodology, it is required to price the bundle at the number of tests times the CLFS rate for CLFS-priced tests, and we urge MolDX to provide clarification of how it will address the scenarios described above.

5. Revision of Analysis of Evidence Language on GEP Testing

We note that the Analysis of Evidence section of the Proposed LCD, in its discussion of the use of combined cfDNA and GEP testing in heart transplant patients, states that “the data support that cfDNA testing may be superior to GEP” and that studies supporting combined cfDNA and GEP testing “ failed to consider that testing by cfDNA alone could provide a superior alternative to GEP, though their data was consistent with this interpretation.” The LCD goes on to state that “[t]hough GEP will continue to be covered in heart transplantation only as it has become a standard of practice, providers should be aware that they may be performing an inferior service. Should evolving literature clearly support cfDNA as superior to GEP for non-invasive testing in heart transplantation and demonstrate no clear value for GEP testing, this coverage decision will be modified to reflect the new evidence.”

While we cannot speak to the clinical evidence for GEP testing versus cfDNA testing in heart transplant patients, the evidence supporting GEP testing in kidney transplant patients using Tutivia is robust, as demonstrated by the extensive data reviewed by MolDX in the recent TA process for Tutivia.

In addition to the analytical and clinical validity that was included in the TA, real world evidence data has demonstrated that Tutivia has utility in areas of clinical need where cfDNA does not perform well, including patients with previous failed transplants as shown in table 1 and (2), or with multi-organ transplants, table 2 and (4), patients experiencing other forms of injury, such as BK viremia / BK nephropathy (4), patients who are early post-transplant with delayed graft function (DGF) (5), and patients who are severely obese (1).

Given this evidence recently reviewed by MolDX, we respectfully request that MolDX modify the Analysis of Evidence section to avoid provider misconceptions about the relative clinical performance of cfDNA and GEP testing. At minimum, MolDX should make clear that the statements suggesting superiority of cfDNA testing to GEP testing pertain only to heart transplant testing.

Clarification of Tutivia Coverage for Patients with Multiple Transplants

Draft Article currently states that “[t]o-date, NO test has been approved for use in patients who have undergone multiple transplants.” Based on the clinical evidence discussed above and shown in Table 2 below, Tutivia’s TA submission was approved in April 2025 (retroactive to November 2024), with no limitation for recipients of multiple kidney transplants. To accurately reflect this fact, we respectfully request that MolDX modify the above statement upon finalization to reflect that Tutivia is approved for use in patients who have undergone multiple transplants.

In conclusion, Verici appreciates and supports MolDX’s efforts to ensure the appropriate evolution of coverage criteria for solid organ allograft rejection testing, and, in particular, to clarify coverage of surveillance testing. As noted, we support finalization of the Proposed LCD and Draft Billing and Coding Article with the above modifications.

Tables and references were provided for review.

Thank you for your comment. Please see Response #1.

Additionally –

• Per the LCD, the evidence considered includes the peer-reviewed published literature. Abstracts do not fulfill this requirement.

The following comment was submitted to Palmetto GBA:

I am writing to strongly advocate for more frequent use of donor-derived cell-free DNA (dd-cfDNA) surveillance testing in the care of kidney transplant recipients. As a transplant nephrologist, I have witnessed firsthand the limitations of conventional monitoring strategies and the significant benefits of incorporating cfDNA into routine surveillance. It is my understanding that the current goal to only allow coverage for 4 tests in the first year after transplant and 2 in the following year which is inadequate.

Traditional monitoring approaches—such as serum creatinine trends, proteinuria, or donor-specific antibody testing—are often late markers of graft injury. By the time abnormalities are detected, irreversible damage may already have occurred. In contrast, dd-cfDNA provides an earlier and more sensitive indication of allograft injury, whether due to rejection, infection, or other insults, allowing for timely diagnostic evaluation and intervention.

Studies have consistently demonstrated that rising cfDNA levels precede creatinine changes and correlate with biopsy-proven rejection. More frequent surveillance increases the likelihood of detecting subclinical injury, enabling interventions that preserve graft function and improve long-term outcomes. For patients with high immunologic risk, recent rejection, or fluctuating kidney function, closer cfDNA monitoring is particularly valuable.

Early detection translates into earlier treatment, reducing the need for invasive biopsies, minimizing the risk of chronic rejection, and ultimately prolonging graft survival. We often use these tests as well to follow improvement after rejection treatment which would reduce need for biopsy in this context as well. Given the tremendous costs—both financial and human—of graft loss and return to dialysis, proactive monitoring through regular cfDNA testing represents both a clinically sound and economically responsible strategy.

In summary, routine and more frequent cfDNA surveillance has the potential to transform kidney transplant monitoring by providing earlier, more actionable insights. I strongly urge consideration of this evidence-based practice for the benefit of our patients and the long-term success of their transplants.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

CareDx is the leading precision medicine company focused on the discovery, development, and commercialization of clinically differentiated, high-value healthcare solutions for transplant patients, including non-invasive molecular testing. We are submitting this letter in response to the Draft Local Coverage Determination (LCD) - “MolDX: Molecular Testing for Solid Organ Allograft Rejection.”

We appreciate that the Draft LCD appropriately reflects the many validated uses of non-invasive molecular testing, including surveillance. However, we have serious concerns with the removal of coverage for certain testing, the establishment of new limits on surveillance testing, and the application of bundled payments. As you work to finalize an LCD with evidence-based decisions, we want to highlight conclusions in the Draft LCD that do not reflect the full extent of the publicly available clinical evidence. We agree that this area of medicine is still “actively evolving” and as such we provide additional information that was not referenced in the Draft LCD for your review and inclusion in a final policy.

We also note that the transplant care infrastructure in the United States is experiencing major changes aimed at increasing donations and transplantations. The organ donation system is experiencing the first reforms in over forty years, shifting to a national allocation system with new oversight to encourage more transplantation. These reforms coupled with an increase in organs procured via donation after circulatory death (DCD) have led to a marked increase in the numbers of transplanted organs. In fact, last year, a record number of 48,149 solid organs were transplanted and the percentage of organs procured via DCD increased by 23.5% over the previous year. While clinical practices have improved the ability of DCD organs to function in some recipients, documentation by the United Network for Organ Sharing (UNOS) shows a significant increase in hospitalization for rejection for patients who receive these organs, increasing the need to closely monitor these patients post-transplant. There are also federal payment policy changes underway in Medicare aimed at increasing kidney transplants and decreasing the large numbers of donated kidneys that are discarded each year. These organ donation and transplant payment reforms must be taken into account when considering changes to policies that impact the ability of clinicians to monitor transplant patients for rejection.

Further, transplantation incurs lower annual expenditures than dialysis, making it not only a more cost-effective option but also a preferred treatment due to its potential for better long-term outcomes and quality of life. The overall Medicare expenditures for non-invasive testing are small compared to the treatment costs of transplanting or retransplanting organs and the clinical benefits are critical to help protect both the patient and the transplanted organ.

Given transplant clinical practice is actively evolving to meet the clinical needs of recipients, and the importance of rejection testing is ever increasing given the changes in procurement, we request the following changes in the Final LCD:

1. Maintain coverage for HeartCare (AlloMap (GEP) with AlloSure (dd-cfDNA)).
2. Update the published evidence for AlloMap to reflect its level 1 clinical utility evidence.
3. Do not finalize limits for “surveillance” testing – or at least, finalize twelve for heart and lung in the first year, increase to seven for kidney in the first year and increase to quarterly for all organs in subsequent years.
4. Include coverage of testing for patients who have undergone “multiple transplants.”
5. Do not finalize bundled payments; maintain the already established fee-for-service model with the national payment rates according to CPT codes.

We believe these changes will ensure the best care for transplant recipients and support the Administration’s transplant reforms, without jeopardizing program integrity. We provide justification and details for each request below:

1. Maintain coverage of HeartCare (AlloMap (GEP) ordered with AlloSure (dd-cfDNA)).

The Draft LCD proposes to restrict clinicians to a single molecular test per patient encounter unless the combination of tests demonstrates a new higher hurdle of superiority and additive benefit when compared to the single test components. In applying this new standard, the Draft LCD wrongly proposes to withdraw coverage for HeartCare - AlloMap (GEP) and AlloSure (dd-cfDNA) ordered together. This testing should continue to be covered, as it has since 2020, because the two tests are complementary, are superior when used together, and provide additive benefit.

1. GEP and dd-cfDNA are complementary molecular tests that provide non-overlapping analysis of analytes:

While both tests are molecular, they can be considered orthogonal. GEP tests measure host immune system activity by analyzing the expression level of 20 genes while dd-cfDNA assays measure evidence of graft injury by quantifying the percentage of cell-free DNA in the bloodstream from the transplanted organ.¹ Their orthogonality make them akin to other combinations of tests routinely used in cardiovascular care, including B-type natriuretic peptide (BNP) and echocardiography in patients followed for heart failure, and electrocardiograms and troponins in patients suspected of having coronary artery disease.

With respect to the type of rejection they can help detect, the GEP test (AlloMap) was validated to detect acute cellular rejection (ACR) (as antibody-mediated rejection (AMR) was not clinically recognized during AlloMap’s development and subsequent FDA review) and dd-cfDNA tests (e.g. AlloSure) have since been shown to be able to detect both. However, the performance characteristics of dd-cfDNA differs depending on the type of rejection being assessed and the utility of GEP and dd-cfDNA depend on the context in which they are used. For example, GEP can achieve a higher sensitivity than dd-cfDNA, which can be important in higher-risk monitoring situations.² On the other hand, dd-cfDNA is highly sensitive to AMR.² Individually each test, however, has somewhat limited specificity for ACR which can be enhanced significantly with combination testing. This is particularly relevant in the surveillance context when subclinical ACR is the primary concern, as it is universally treated, though the pretest probability for detecting it on any individual biopsy is low.¹

Thus, GEP and dd-cfDNA assays are distinct diagnostic tools and combining the testing results facilitates more precise and accurate diagnoses because of the meaningful differences in how the two different types of tests operate and yield results.

As noted in the Draft LCD, when one or more tests, each analyzing different analytes and measuring different pathophysiology, are used during a patient encounter and have been shown to be superior and provide additional benefit beyond that of each test, then both tests should be deemed reasonable and necessary, even in the same encounter. This is in fact the case for HeartCare (combination testing with GEP and dd-cfDNA) which has demonstrated clinical superiority to either GEP or dd-cfDNA testing alone. Unfortunately, the Draft LCD did not consider definitive evidence of the clinical validity and utility of combination testing for rejection surveillance which is the main clinical context in which molecular testing in heart transplant is used.¹

1. The literature demonstrates that GEP and dd-cfDNA together are superior to using either alone:

The prospective, multi-center SHORE registry study and other literature definitively support combination testing with AlloMap and AlloSure as a valid, clinically effective, and medically necessary testing strategy, meeting the definition of reasonable and necessary.^A

First, SHORE was one of the largest studies in cardiac transplantation and published in the most respected heart transplant journal. The initial publication included 2,077 patients and 161 episodes of acute cellular rejection that were paired with molecular testing. The Draft LCD provides no description of the significant size of the SHORE study, especially relative to the small Rodgers study relied upon by the Draft LCD in non-covering HeartCare (Table 1).

Second, as illustrated in the table below, the SHORE study showed significant improvement with combination testing compared to dd-cfDNA alone in both specificity and positive likelihood ratio (+LR), which are the most relevant performance characteristics for surveillance, when the pre-test probability of ACR is low (Table 2). The +LR represents how many fold the odds of finding rejection increase with a positive test.

The Draft LCD’s statement that combination testing in the SHORE registry did not have any “clear benefit in specificity” is simply incorrect. The specificity improved from 84.6% [83.5%-85.5%] with dd-cfDNA testing alone to 91.7% [91.0%-92.4%] with combination testing.¹ With non-overlapping confidence intervals, this improvement was statistically significant. Importantly, the analysis of evidence in the Draft LCD does not acknowledge that the improvement in specificity yielded a clinically meaningful and statistically significant improvement in +LR 3.9 [3.08-4.96] vs. 2.9 [2.4-3.5] p<0.01) which integrates both sensitivity and specificity and is the most important performance characteristic in the surveillance context.¹ The Draft LCD lacks consideration for this context when critiquing the sensitivity of combination testing. In surveillance, the pre-test probability of detecting ACR on any single biopsy is already very low at approximately 2-3%.^1,3-5 This makes the test’s negative likelihood ratio (- LR) (driven by its sensitivity) clinically redundant as a negative test cannot meaningfully impact the post-test probability. The relevant performance characteristic to highlight is the test’s +LR (driven by specificity) because positive tests can meaningfully impact the post-test probability in the surveillance context.

Third, in the context of high-risk monitoring, where the pre-test probability is higher than routine surveillance, the Draft LCD analysis fails to appropriately recognize the utility of combination testing for ACR and AMR detection and optimized –LR. AlloMap, but not dd-cfDNA, can achieve sufficient –LR to rule out ACR, while dd-cfDNA, but not AlloMap, can achieve sufficient –LR to rule out AMR. Combination testing, therefore, would be critical in scenarios where the pre-test probability for ACR and AMR are both higher than routine surveillance.²

Fourth, as illustrated in the table below, SHORE also showed the value of combination testing in stratifying risk as dual positives had a significantly greater biopsy yield (Table 3).

The Draft LCD’s statement that “the prevalence of AR was highest in the first 30 days, making predictive values very different early versus later in the [SHORE] study” is true but irrelevant. None of the data described in this table relied on any rejection that happened before 55 days.

The Draft LCD also seeks to diminish the complementary value of AlloMap by arguing that the biopsy yield for all positive dd-cfDNA of 7% was “similar” to the dual positive tests at 9.2%. However, this does not consider the significant clinical value of AlloMap GEP in stratifying risk. In the population of positive dd-cfDNA and negative GEP, the biopsy yield was 4.3%, less than half those with positive dd- cfDNA and positive GEP, where the biopsy yield was 9.2%. This difference illustrates the enhanced ability to stratify risk from combination testing.

1. GEP and dd-cfDNA when used together significantly improves patient care:

The improved biopsy yield with combination testing has critical implications for patient care because the ability to stratify risk enables a significant reduction in biopsies. The SHORE study showed that the addition of AlloMap testing to dd-cfDNA testing resulted in a clinically significant decrease in biopsies. While the Draft LCD’s summary of evidence highlights the clinical utility data of combined testing from two single center studies and asserts that the associated reduction in biopsies was primarily driven by negative dd-cfDNA, it leaves out the extensive clinical utility data from the SHORE registry study publication from 52 centers.¹ These data clearly demonstrate that a negative AlloMap in the context of a positive dd-cfDNA is in fact associated with a significant avoidance of biopsies.¹

Specifically, in the first 2 years post-transplant, the reflex biopsy rate for a positive dd-cfDNA was 36% lower when accompanied by a negative GEP compared to when accompanied by a positive GEP (22.8% [19.9-25.9] vs. 35.4% [32.0%, 38.9%]).¹ Importantly, the summary of evidence also overlooks the important data that the contribution of a negative AlloMap on biopsy avoidance increased as the study progressed. As clinicians learned the performance characteristics of combination testing this translated into an overall significant reduction in both first- and second-year biopsies over time from 5.9 [5.6-5.3] biopsies down to 5.3 [5.1-5.4] biopsies and 1.5 [1.3-.7] biopsies down to 0.9 [0.8-1.0] biopsies respectively.¹

On the other hand, the Draft LCD, by suggesting that dd-cfDNA without AlloMap sufficiently risk stratified patients by grouping all to a 7% risk, would commit patients to significantly more biopsies.

Using the first year reflex biopsy rate provided in SHORE for the most recent year of enrollment, and assuming patients with a positive dd-cfDNA (but without the availability of an AlloMap result) would be biopsied at the same frequency as the patients with AlloMap positive and dd-cfDNA positive results, the approach of only using dd-cfDNA as a marker of ACR risk, would lead to approximately 34% more biopsies for dd-cfDNA positive results (Table 4).

With 4,500 patients (report data from OPTN)⁶ undergoing 54,000 surveillance episodes (4,500 multiplied by 12 surveillance episodes) in the first-year post-transplant and at least 35,000 transplant patients undergoing 70,000 surveillance episodes (35,000 multiplied by 2 surveillance episodes) beyond the first-year post-transplant, that 34% increase represents an extra 1,840 biopsies annually (Table 5). If transplant patients received surveillance testing quarterly beyond year 1, which is the surveillance protocol followed by many transplant centers today, omitting GEP would lead to an extra 2,979 biopsies annually.

In fact, if the separation of reflex biopsy rates between dd-cfDNA positive/GEP positive and dd- cfDNA positive/GEP negative continues to grow, as seen in SHORE, the annual excess biopsies would be substantially higher. For example, Henricksen et al. documented a reflex biopsy rate for dd-cfDNA positive/GEP positive of 90% but only 22% for cfDNA positive/GEP negative.⁷ In this scenario, patients with a positive dd-cfDNA (without any GEP result) would receive 69% more biopsies and there would be an overall annual excess biopsy rate of 5,900 biopsies/year (Table 5).

1. The data is consistent with recently presented data as well as clinical practice:

A manuscript currently under review, based on oral abstracts presented at the International Society of Heart and Lung Transplantation annual congresses in 2024 and 2025, and expected to be published soon, further demonstrates that combination testing allows for a precise identification of the population of cardiac transplant patients not only at risk of current rejection but at risk of future adverse events and that this risk cannot be as precisely defined with dd-cfDNA alone (Figure 1).^8,9

As shown in the figure below, patients who had a simultaneously positive GEP and dd-cfDNA result between 2-6 months were at the highest risk of graft dysfunction or death from rejection in the following 12 months compared to patients that did not have a simultaneously positive result (p=0.009). Importantly, patients with a positive dd-cfDNA but a negative GEP, were not at increased risk of adverse events compared to patients with dual negative results (adjusted HR 0.98; 95% CI 0.46-2.07; p=0.95), again emphasizing the validity of and additive benefit of combination testing. Also, there was no difference in dd-cfDNA levels in combination testing results that were GEP positive or GEP negative, so the GEP result was the only way to definitively ascertain risk. Therefore, the implications of not allowing clinicians the benefit of an AlloMap result at the same encounter to refine the risk of future events are even more profound than just receiving excessive invasive biopsies. Patients with positive dd-cfDNA but not at risk of future events could receive a plethora of unnecessary testing and follow-up. More critically, they may receive unnecessary intensification of immunosuppression along with all of the well- documented associated complications, including infections, malignancies, and renal dysfunction.

The data is also consistent with the clinical community’s widespread adoption of combination testing, with it being routinely used by dozens of leading academic medical centers across the country including Stanford, Columbia, Tampa General, University of Chicago, UCLA, Mayo Clinic, and Cleveland Clinic among others.

1. The 2022 CAC meeting feedback is outdated:

To the extent the Draft LCD relies on the November 2022 contractor advisory committee (CAC) meetings, a nearly three-year-old discussion with experts, that meeting was dependent on data that is now outdated. Specifically, the statement included in the Draft LCD that there is a lack of data to support the routine use of molecular combination testing is no longer true. In fact, subject matter experts from the 2022 CAC were subsequently authors on the SHORE publication (published in September 2024) demonstrating clinical utility and validity of combination molecular testing already noted above.¹

1. The Draft LCD unduly relies on Rodgers et al, and the view that other studies are consistent with Rodgers, to claim that dd-cfDNA may alone be sufficient and that combination testing (AlloMap and dd-cfDNA) is no longer reasonable and necessary:

Neither Rodgers,⁴ nor the other studies referenced in the Draft LCD, support the policy’s withdrawal of coverage for combination testing with AlloMap and AlloSure (HeartCare).

First, Rodgers incorrectly defined positive tests. In the context of a low pre-test probability of rejection, which is the case in a surveillance setting, the main value of molecular testing is in its ability to identify high-risk patients; this is generally accomplished through the test’s specificity and +LR, a fact explicitly acknowledged by Rodgers et al. However, when they evaluated combination testing, the authors made the critical error of defining a positive combined test as either a positive GEP or a positive dd- cfDNA, rather than both a positive GEP and a positive dd-cfDNA, thus crippling the combined test’s specificity and leading to their erroneous conclusion. We note that the correct approach was taken in an independent analysis of the same data used by Rodgers et al. This corrected analysis was performed and published by Moayedi and Teuteberg, yet that publication was not included in the summary of evidence.² When compared appropriately in Moayedi et al. (dual-positive vs. dd-cfDNA only positive) specificity for dual positives was 93% compared to 82% for dd-cfDNA positive only and the positive likelihood ratio for dual positive testing was 3x greater than dd-cfDNA alone (4.97 vs. 1.66).² Given the confidence intervals for the +LR of each of the tests, in the Rodgers et al data, the positive likelihood ratio for combination testing was the only test that definitively increased the post-test probability.⁴

Second, Rodgers was significantly underpowered. Importantly, even in the context of evaluating which test has the superior sensitivity and then defining a positive combined test as either a positive GEP or positive dd-cfDNA, Rodgers et al. could not conclude from their data that there was no improvement in sensitivity with combination testing, since the study was significantly underpowered to detect differences in sensitivity with only 10 episodes of acute cellular rejection (ACR). This is clear from the wide confidence intervals displayed of both sensitivities’ point estimates (Figure 3 in the publication; up to 100% for combined testing); the confidence intervals were notably missing from the Draft LCD’s Summary of Evidence but should be noted and included as important information.

Third, Rodgers is the only support used for the Draft LCD’s conclusion about combination testing in heart transplant recipients. The assertion within the Draft CLD that Rodger’s conclusion “agreed with other reports that dd-cfDNA alone has the potential to replace or significantly reduce the need for endomyocardial biopsy (EMB)” is incorrect. None of the other cited reports make the claim that dd- cfDNA alone could replace surveillance biopsies. In fact, Holzhauser et al explicitly acknowledged the complementary nature of AlloMap and dd-cfDNA testing.¹⁰ Moreover, none of the cited publications considered data from SHORE, as SHORE post-dated these publications but is now available for evaluation.

In summary, this combined testing has been affirmatively covered by MolDX since 2020 and should continue to be covered. The SHORE study definitively demonstrated that combination testing with AlloMap and AlloSure provided additive benefit and superiority over either test alone. Further, the best available evidence clearly demonstrates that eliminating coverage of combination testing will lead to more unnecessary biopsies and will also likely lead to more adverse events from immunosuppression. The only study (Rodgers et al) that directly disputes the validity of combined molecular testing is methodologically flawed and underpowered. This is not an appropriate basis to disturb longstanding coverage, particularly given the strength of the SHORE evidence.

1. Update the published evidence for AlloMap to reflect its level 1 clinical utility evidence

The discussion of AlloMap’s evidence within the Draft LCD does not reflect the extensive and unique data supporting its validation and demonstrated clinical utility.

First, from an evidence-based medicine perspective, the AlloMap test has level 1 evidence (IMAGE, eIMAGE) – prospective, randomized trials.^11,12 As illustrated in figure 2, level 1 evidence is in fact the highest level of evidence.

Second, the strength of this evidence is reflected in guidelines. Based on the most recent randomized controlled clinical utility study and an updated clinical validity study that again used a centrally reviewed biopsy as the gold standard, the International Society for Heart and Lung Transplantation (ISHLT) updated its 2023 guidelines recommending an even earlier use of GEP expression profiling (AlloMap) beginning at 2 months post-transplant (versus 6 months in the older 2010 guideline).¹³ Further, the recommendation for using AlloMap was noted by the guidelines to be supported by stronger evidence than endomyocardial biopsy. The expanded ISHLT recommendation for AlloMap and the two supporting studies should be included in the final policy.

The Draft LCD’s negative commentary on the AlloMap test is directly at odds with these guidelines, which the policy otherwise recognizes as an important source of evidence and which MACs must consider in developing policy.^[D]

Third, from a clinical perspective, AlloMap has significant and unique validation and data supporting its utility. Illustrated in Table 6 below, compared to dd-cfDNA, AlloMap can achieve a higher sensitivity and negative likelihood ratio (92.3% [80.9%, 97.8%] and 0.2[0.07-0.59], respectively) to exclude ACR in higher-risk monitoring situations. This is important because ACR is the most common form of subclinical rejection, and it is uniformly treated.^1,2

Fourth, the policy wrongly relies on an Area Under the Curve (AUC) comparison between AlloMap and dd-cfDNA from two different studies - De Vlaminck et al.¹⁴ and Deng et al.¹⁵ - to suggest that dd-cfDNA has superior test performance to AlloMap-testing and is sufficient alone. While the policy correctly recognizes that the AUCs were from different studies, it fails to acknowledge that the studies were of fundamentally different design, thus making AUC comparisons impossible. Deng et al only evaluated the performance of AlloMap in the detection of acute cellular rejection (ACR) and used centrally adjudicated biopsies as the reference standard, while De Vlaminck included cases of antibody- mediated rejection (AMR) and biopsies that were not centrally reviewed. Moreover, the discussion in the Draft LCD failed to mention that De Vlaminck relied on a non-commercially available assay that had not been analytically validated and required genotyping of both the donor and recipient, making it an inappropriate basis to draw conclusions to commercial dd-cfDNA assays currently available for cardiac transplant monitoring.

Finally, the Final LCD must also remove the subjective and inappropriate language related to using the AlloMap service… that “providers should be aware that they may be performing an inferior service.” A Medicare Administrator Contractor’s (MAC’s) role in the LCD process is to make coverage policy determinations about whether a particular item or service is “reasonable and necessary,”^[A] not to comment on, direct, or otherwise influence medical practice for covered technologies or services. The Social Security Act expressly prohibits such interference with the practice of medicine in recognition that clinical determinations about the most appropriate test or therapy are best rendered by the treating clinician based on the medical needs of their patient.^B Moreover, as discussed earlier, the language is not an accurate reflection of the published evidence nor the value of the AlloMap test and is therefore inconsistent with LCD evidentiary standards.^C

1. Do not finalize limits for “surveillance” testing – or at least, finalize twelve for heart and lung in the first year, increase to seven for kidney in the first year and increase to quarterly for all organs in subsequent years.

As detailed below, testing frequency should be left to the judgment of the treating clinician or at least increased to reflect current clinical practice and guidelines.

1. Surveillance testing should be left to the clinician’s discretion given the wide range of risk in patients; limits will jeopardize outcomes
   • Transplant patients require personalized treatment plans; some may need more testing timepoints and others may need less

This newly introduced testing restriction is inconsistent with providing personalized medicine and would be a step backward. Molecular testing is not only clinically valuable but also increasingly accessible, cost-effective, and patient-friendly. Rather than curtailing its use, the focus should be on expanding access, refining interpretation, and integrating it into standard care protocols. Molecular testing represents the future of transplant medicine; clinicians should not face sudden arbitrary limits.

The need to provide personalized medicine is particularly acute in transplant patients. This is because allograft rejection surveillance must be highly personalized to: (1) detect subclinical rejection so it can be treated before rejection and potential irreparable injury to the allograft; and (2) facilitate individualization of immunosuppression to assure patients are on the optimal amount of immunosuppression necessary to prevent organ rejection, no more and no less.

The risk of rejection and necessary drugs and drug levels needed to prevent rejection varies from patient to patient (and by transplant type) and changes significantly over time. It can widely vary even later post-transplant when the risk of rejection and need for immunosuppression is generally lower. Frequency caps will significantly impair clinicians’ ability to individually manage patients’ immunosuppression, which requires constant reassessment to ensure enough is prescribed to prevent rejection while minimizing the toxicities associated with these medications which include serious infections, post-transplant malignancies and renal toxicity.

The reliance on personalized medicine in transplantation is likely to be heightened in the coming years. New Medicare payment models are working to increase kidney transplants and decrease the large numbers of donated organs that are discarded each year.¹⁶ Moreover, there has been a dramatic increase in organs procured via donation after cardiac/circulatory death (DCD) and in at least heart transplant these organs are at elevated risk of rejection into at least the second year post transplant.¹⁷ The increased use of organs at elevated risk will require more frequent surveillance and closer monitoring.

Though the natural history data of untreated subclinical acute rejection is limited (as the identified entity is usually treated), the consensus is that it can progress to graft dysfunction over weeks to months. This is supported in the literature by the ISHLT recommendation that consideration should be given to treating moderate to severe subclinical rejection with intravenous steroids.¹³ Moreover, recent renal literature suggests that transplant glomerulopathy may develop within weeks to months after isolated microvascular inflammation is detected on biopsy.¹⁸ This relatively short indolent period of at most 6 months makes infrequent (i.e. biannual) surveillance potentially dangerous. Twice yearly monitoring, even with a test that had 80% sensitivity would miss 1 in 5 rejections per year.¹⁹

• To the extent the frequency limits are premised on the frequency of surveillance biopsies, that limitation is inconsistent with the benefit to risk profile of non-invasive testing.

Prior to the introduction of molecular testing, rejection surveillance monitoring was accomplished with surveillance biopsies. Surveillance biopsies across all organs have significant limitations - they are invasive and come with a well-documented complication rate and increased expenditures. Moreover, in cardiac transplantation, biopsies can also become more technically challenging to perform over time, as scar tissue develops both at the site of the insertion of the bioptome as well as the site from which endomyocardial tissue is sampled. This can lead to receiving non-diagnostic results that cannot be used for patient management.²⁰

While the risk of rejection diminishes over time, historic practice around the frequency of rejection surveillance, which generally decreased over time, was driven as much by the invasiveness of the procedure, known associated complications, and increasing difficulty in obtaining diagnostic material, as it did to the lower risk of rejection over time. The lower number of surveillance biopsies in subsequent years may have made sense when the alternative required multiple invasive procedures where the risks outweighed the benefits; however, in the era of reliable noninvasive molecular surveillance, it no longer does.

In fact, local coverage policies in the past and even in the Draft LCD have recognized the difference in risk/benefit between invasive and non-invasive surveillance methods, since AlloSure was first covered by MolDX. The response to comments from the Final LCD for AlloSure expressly delinked non-invasive dd-cfDNA testing frequencies from established biopsy frequencies: “Since a biopsy is not frequently performed for surveillance due to its invasive nature, the use of AlloSure may be performed at a frequency established for other non-invasive tests.”²¹ And, the current Draft LCD rightly acknowledges a more favorable benefit to risk profile for non-invasive tests in one of the coverage criteria. We are concerned that adopting limits based on minimum timepoints without taking into account the more favorable benefit to risk profile of non-invasive tests and the heterogeneity of patient risk profiles may do more harm than good given the intensity of caring for a patient post-transplant.

• Program integrity interests and prior claims utilization data should not be the basis for testing limits.

While we appreciate CMS’s responsibility to maintain program integrity, this important interest should not result in sudden and arbitrary coverage restrictions in this area. The testing services identified as eligible for coverage under the Draft LCD each have unique and specific CPT codes as well as Z-codes based on their intended use increasing transparency and accountability to billing practices and strengthening program integrity.

Likewise, to the extent the proposed limitations are based on Medicare claims utilization data, such data is not an appropriate basis for making coverage decisions. Coverage should be based on what is reasonable and medically necessary. This type of utilization data is not a recognized form of evidence of clinical utility and is a particularly poor proxy for assessing medical necessity given the widespread variation in patient needs discussed above.

1. To the extent that any surveillance frequency limits are finalized they should be supported by either guidelines for other non-invasive testing for rejection, societal consensus or published studies representing current clinical practice:
   • Heart surveillance testing limits if finalized should be 12 times in the first year and increased to quarterly in subsequent years.

In heart transplantation, the Draft LCD purports to base its proposed surveillance frequencies (12 in first year and 2 in subsequent years) on International Society for Heart and Lung Transplantation (ISHLT) guidelines as well as a study from the Genomic Research for Transplantation.^13,22 While the Draft Policy’s establishment of 12 surveillance tests in the first year post-transplant aligns with ISHLT guidelines, the frequency in the subsequent years should be higher based on the same guidelines, an increasingly higher risk patient population, and clinical practice all support surveillance frequencies of 12 in the first year and at least quarterly thereafter.

First, with respect to ISHLT:

• The guidelines recognize that lifelong follow-up at a cardiac transplant center is necessary to monitor for the possibility of acute and/or chronic rejection.
• The guidelines recommend, for an uneventful clinical course, follow-up with clinic visits and lab tests 12 times in the first-year post-transplant and every 3-6 months thereafter.
• The guidelines further recognize that this frequency should be increased for patients with complex medical or psychosocial conditions.

Second, as discussed above, the percentage of higher risk patients is increasing. In 2024, donation after cardiac death represented 18% of all donations for cardiac transplantation and is on a dramatic upward trajectory. A large recent UNOS study documented a significant increase in hospitalization for rejection in patients who receive these donations with follow-up that extends into the second year.²³ While the precise longer-term impact remains unknown, a relatively intense non-invasive surveillance schedule would be critical for precisely characterizing this impact, identifying trends early and treating the rejection while it is still in the subclinical phase.

Finally, published US transplant center surveillance protocols mirror the ISHLT recommendations of 12 surveillance episodes in year one and quarterly thereafter.

Therefore, if testing limits are finalized for heart transplants, they should be increased to quarterly (4 per year) in subsequent years and finalized at 12 times in the first-year post-transplant.

• Lung surveillance testing limits if finalized should be 12 times in the first year and increased to quarterly in subsequent years.

Lung surveillance monitoring is conducted for the life of the patient, and its utility extends far beyond monitoring for rejection. Chronic Lung Allograft Dysfunction (CLAD) is the leading cause of death in patients one year post transplant and the median survival of bilateral lung transplant recipient is just over 6 years, the lowest among major organ transplants.¹³ There are a multitude of factors that can increase a patient's risk of CLAD and death post-transplant that include immune activation, ACR, AMR, infection, and gastrointestinal complications. External exposures, such as environmental factors, which do not go away over time, put patients at greater risk of injury given the lung allograft’s contact with the outside environment that the other solid organ counterparts do not experience. For all these reasons, lung transplant patients are at increased risk of rejection, and therefore monitored closely for the first year and continue routinely and more frequently for the life of the organ.

In the existing literature utilizing dd-cfDNA to monitor lung transplant recipients, the study protocols utilize quarterly testing in post operative years two and three. While the Draft LCD references the Keller et al clinical utility study²⁴ as its basis for establishing 12 surveillance episodes in year one post-lung-transplant, the Draft LCD diverges from Keller et al in subsequent years by proposing 2 tests a year post-transplant are reasonable and necessary instead of quarterly testing (4 test per year). From Keller et al: “All four centers developed consensus guidance regarding the frequency of asymptomatic dd- cfDNA monitoring and interpretation with monthly surveillance plasma dd-cfDNA testing for patients less than 1-year post transplant and every 3 months for patients > 1-year post-transplant.”²⁴

Therefore, if testing limits remain, they should be consistent with the published data from Keller et al. in both time points: 12 in the first-year post-transplant and increased to quarterly (4 per year) in the subsequent years post-transplant.

• Kidney surveillance testing limits if finalized should be increased to 7 times in the first year and increased to quarterly in subsequent years.

The American Society of Transplant Surgeons (ASTS) recommendation for the frequency of surveillance testing in kidney transplant recipients recognizes that the most “robust clinical data” from the AlloSure Routine Testing Schedule (ARTS) supports this testing cadence and its linkage to time points of concern: “Prospective trials designed to optimize the frequency of dd-cfDNA testing in stable, asymptomatic adult transplant recipients are lacking. However, they also recognize that obtaining such data may prove impractical and the evidence for the clinical utility of dd-cfDNA surveillance testing is clear, as is the utility of testing longitudinally. While the optimal surveillance testing frequency is unknown, ASTS suggests the most robust clinical data in surveillance testing comes from the DART study, which utilized the “ARTS” surveillance testing schedule.” The ARTS protocol recommended testing at monthly intervals from month 1-4, and then quarterly thereafter beginning at month 6. ASTS goes on to further state, “This schedule was developed empirically based on the KDIGO recommended testing frequencies for other noninvasive post-transplant tests and focused the most frequent testing at the time of highest immunologic risk and the most kinetic portion of the post-transplant period and then decreased frequencies later in the post-transplant course.”^25,26

If implemented, the Draft LCD’s proposed testing limits would likely lead to missed rejection detection and the potential for early intervention, particularly in subsequent years when only two tests are proposed to be allowed. As previously mentioned, the relatively short indolent period of subclinical rejection of at most 6 months makes infrequent (i.e. biannual) surveillance potentially dangerous. Testing must align to timepoints that allow clinicians to identify these patients several months in advance of other clinical signs or symptoms of rejection and before irreversible damage to the allograft. Early interventions have been shown to improve outcomes;^27,28 however, they can only occur if the surveillance intervals are frequent enough.

Accordingly, if surveillance testing limits for kidney are implemented in the Final LCD, they should be increased to reflect the guidance from ASTS and at least match the ARTS schedule of 7 in the first year and quarterly in subsequent years. Overall, we believe surveillance testing limits should not be finalized and instead the current policy should continue which allows for personalized clinician judgment. While differing protocols, recommendations, and guidance exist for surveillance testing frequencies, these frequencies are merely suggestive and should not be the basis of hard caps on testing.

Should surveillance testing limits be finalized, at a minimum, the limits must be increased to reflect the increasing risk profile of transplant patients, published studies and current clinical practice, and the more favorable risk/benefit profile of non-invasive testing over biopsies.

1. Include coverage of dd-cfDNA in patients who have undergone multiple transplants.

The Draft Billing Article states “To date: NO test has been approved for use in the patients who have undergone multiple transplants.” The final policy should reflect the published evidence which supports dd-cfDNA's clinical validation in recipients with simultaneous kidney-pancreas transplantation and also in recipients with repeat kidney transplants.

1. AlloSure Kidney is validated for use in patients who have undergone simultaneous pancreas- kidney transplantation:

Simultaneous kidney-pancreas transplants (SPK) are typically performed in patients with type 1 diabetes (and selected patients with type 2 diabetes) and advanced chronic kidney disease. This procedure involves transplantation of kidney and pancreas allografts from the same deceased donor. SPK has been shown to significantly improve quality of life and increase life expectancy of patients with diabetes and chronic kidney disease. One- and 5-year pancreas graft survival rates are now comparable to those of kidney, liver, and heart transplants. Studies in these patients highlight the importance of diagnosis and treatment of pancreas rejection. Niederhaus, et al.²⁹ reported on the incidence, risk factors, and outcomes of pancreas rejection. The incidence of rejection within 1-year post-transplant in patients undergoing for- cause biopsies was 21%, with relatively even distribution between ABMR, TCMR, and mixed rejection.

Although the majority of pancreas rejection episodes were successfully treated, 20% of pancreas grafts failed within a year of diagnosis. In another study, Dong et al.³⁰ also demonstrated the association of pancreas rejection with subsequent graft loss. An observational, prospective single center study conducted by Williams et al.³¹ demonstrated dd-cfDNA (AlloSure) differed between patients with rejection (median=2.25%) versus those with injury (0.36%) or quiescence (0.18%) (p=0.0006). 97% of patients without rejection had dd-cfDNA <0.5%, consistent with published data in kidney transplantation. Biopsy- confirmed rejection was associated with elevations in dd-cfDNA levels with a median level of 2.4%.

A multi-center study conducted by Yoo et al.³² also demonstrated dd-cfDNA (AlloSure) values differed between SPK patients with rejection or infection (median=0.56%) versus those who were stable (0.28%). Of the SPK recipients with dd-cfDNA > 1.0%, 36.8% of patients had evidence of infection or rejection. 100% of recipients without infection or rejection had dd-cfDNA <1.0%, consistent with the previously published rejection threshold of 1.0% for kidney transplant recipients.

1. AlloSure Kidney is also validated for use in patients who have undergone repeat transplants:

In 2024, 27,759 kidneys were transplanted in the US; 2,614 kidneys were transplanted into patients who had one or more prior kidney transplants. Over the past decade, approximately 10% of patients who received kidney transplants have had a prior kidney transplant that failed.

The ADMIRAL Study³³ evaluated dd-cfDNA in 1,092 kidney transplant recipients from 7 US transplant centers. 8% of these patients had a prior kidney transplant. In the Nature Medicine publication,³⁴ which evaluated 2,882 kidney transplant recipients from the US and Europe, 447 patients (15.5%) had a prior kidney transplant. Both of these large studies demonstrated that the performance of dd-cfDNA (AlloSure) in repeat transplant recipients was similar as in de novo kidney transplant recipients and clearly differentiated patients who were immunologically quiescent versus those that had rejection.

1. Do not finalize bundled payments; maintain the already established fee-for-service model with the national payment rates according to CPT codes.

We provide feedback below on the Draft Billing Article that was briefly referenced at the conclusion of the Draft LCD via a hyperlink. The Draft Article sets forth a requirement that laboratories performing services must obtain nine new Z-codes. These codes represent different intended uses (for-cause and surveillance) as well as different payment “bundles” for surveillance testing. In discussing the billing requirements with these new codes, the Draft Article introduces a “bundle of testing timepoints,” and notes the “price of each bundle reflects the number of timepoints demonstrated to be R&N according to the criteria in the accompanying LCD.” We request that this bundled payment language not be finalized.

First, we are concerned that the MACs have not specifically solicited comment on the Draft Article, and the Draft Article is not substantively discussed in the Draft LCD, except for a brief hyperlink reference.

Second, the Draft Article appears to be altering coding and/or payment standards in a manner that is inconsistent with applicable statutory and CMS requirements.

LCDs are statutorily limited to determinations of whether or not a particular item or service is “reasonable and necessary” and therefore coverable on a contractor-wide basis.^A CMS has also issued long-standing Program Integrity Manual Guidance making clear that contractors are only permitted to issue billing and coding articles to “help implement” the coverage policies in the LCDs.^B

On its face, the Draft Article goes beyond permissibly “help[ing] implement” the LCD coverage policies. The Draft LCD contains no discussion of the Draft Article’s new coding and payment bundling framework. MACs cannot use a Draft Article to create an entirely new coding framework that does not appear in the associated Draft LCD and is inconsistent with current, nationally established rates and coding for the applicable tests.^C

Third, the bundled payment structure would represent a change in pricing for the tests and MACS cannot depart from the national Clinical Laboratory Fee Schedule (CLFS) rates for these tests via LCD or billing article.

Fourth, to the extent bundled payments nonetheless remain, we expect that the reimbursement for the bundles is equal to the test rates on the fee schedule for each test included in the bundle, as multiplied by the timepoints referenced in the Draft LCD to reflect the number of tests performed. This is because there is no authority for a MAC to set pricing where, as here, each of the tests listed by CPT code in the Draft Article have proprietary and unique codes with national rates that have been (or are being) determined by CMS through the Annual Public Meeting process.^D

A change in payment rates would also be inconsistent with the fact that MolDX and the participating MolDX jurisdictions themselves provided the foundation for the current CLFS rates. Although the Crosswalk methodology has been the payment determination methodology employed by CMS for the more recent codes in this space, the first test code was priced through Gapfill and the rate was established by MolDX to determine the median amount which became the national fee schedule amount.

Finally, if the goal of these bundles, however, is to manage quantity limits on the testing, we understand edits are available to MACs to prevent paying claims above a certain amount without requiring bundling of payments. Given that specific codes with CLFS rates already exist for all these tests, bundled payments with additional new Z-codes are both unnecessary and improper, as MACs cannot supplant CMS-established national codes and rates through either an LCD or a billing article.

In conclusion, CareDx respectfully urges the reconsideration of key provisions in the Draft LCD to ensure that transplant patients continue to receive the highest standard of care. The proposals—if finalized without modification—risk undermining the progress made in transplant medicine and the broader goals of federal transplant reform. By maintaining coverage for combination molecular testing, updating the clinical evidence record, preserving personalized surveillance protocols, and rejecting bundled payment models, CMS can better support clinicians in delivering personalized, cost-effective, and life-saving care. We appreciate your attention to these concerns and look forward to continued collaboration in shaping policies that reflect the evolving needs of the transplant community.

Tables, figures, and references were provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

I am writing as a transplant nephrologist with over a decade of experience managing kidney transplant recipients.. I would like to provide comments on the Draft LCD for Molecular Testing for Solid Organ Allograft Rejection, issued on July 17, 2025.

I have been using dd-cfDNA in the management of my patients for over 5 years. Dd-cfDNA identifies which of my patients have developed rejection before clinical signs of rejection are present, i.e. subclinical rejection. I authored a peer-reviewed manuscript that was published in 2019 which studied kidney transplant recipients who were found to have subclinical antibody mediated rejection (AMR) via surveillance allograft biopsies. These patients underwent treatment for their subclinical rejection and were noted to have significantly improved graft survival when compared to patients who developed clinical AMR. Indeed, the graft survival of these patients with subclinical AMR treated was similar to the patients who were found to have no rejection on their surveillance biopsies. This would suggest early identification and treatment of subclinical rejection offers an opportunity to intervene before irreversible damage is done to the renal allograft.

dd-cfDNA has clearly been demonstrated to identify patients with subclinical rejection significantly better than other standard of care studies such as serum creatinine or proteinuria and is currently the best biomarker available for identifying subclinical rejection. Limiting the number of dd-cfDNA surveillance testing to 4 tests during the first year and 2 tests in subsequent years will delay the diagnosis of subclinical rejection which I believe will reduce the longevity of these allografts.

Our transplant team recently published our center’s experience when we followed a surveillance protocol of 7 tests during the first year. We demonstrated that dd-cfDNA levels would increase several months prior to the other clinical signs of rejection would be present. In addition, we showed that dd-cfDNA levels would decrease in patients after they underwent treatment of their rejection. In contrast, the patients who had biopsies with no rejection had stable low levels of dd-cfDNA. This study also included over 25 patients who had prior kidney transplants.

In summary, dd-cfDNA is currently the most powerful biomarker for identifying patients with subclinical acute rejection and can also be used to monitor patients undergoing treatment of their rejection. Any frequency caps will undermine the physician’s ability to use this valuable test when medically necessary. I thus strongly advocate removal of any caps for dd-cfDNA testing in kidney transplant recipients.

Figure and references were provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA:

I am writing to request reconsideration of the proposed Local Coverage Determination (LCD) statement: “To-date, NO test has been approved for use in patients who have undergone multiple transplants,” specifically as it pertains to the use of donor-derived cell-free DNA (dd-cfDNA) monitoring in simultaneous pancreas-kidney (SPK) transplant recipients. Summary of Clinical Need and Evidence SPK transplantation is a well-established therapy for patients with type 1 diabetes and end-stage renal disease. Monitoring for pancreas graft rejection in this population is challenging, as traditional biochemical markers such as serum amylase and lipase lack specificity and sensitivity for early or subclinical rejection. Protocol biopsies, while informative, are invasive and carry procedural risks.[1] At our center, we do not perform pancreas biopsies and thus rely on empiric treatments that may expose patients to unnecessary immunosuppression and increased risk of infection. Based on our experience, biomarkers of pancreas rejection, especially in the setting of kidney transplant, are desperately needed.

Recent advances have identified dd-cfDNA as a promising noninvasive biomarker for detecting allograft injury and rejection. The Banff 2022 Pancreas Transplantation Multidisciplinary Report, developed by the American Society of Transplant Surgeons and the American Society of Transplantation, highlights dd-cfDNA as a sensitive tool for the diagnosis of antibody-mediated rejection (AMR) in pancreas transplantation. Notably, dd-cfDNA elevation precedes changes in standard clinical markers and outperforms serum lipase in the diagnosis of acute rejection, with an area under the curve (AUC) of 0.84 compared to 0.74 for lipase.[1] The guideline further notes that dd-cfDNA is particularly useful for longitudinal monitoring after the early post-transplant period, with levels normalizing within 45–80 days post-transplant and subsequent increases correlating with rejection or infection.[1]

A prospective study of 36 SPK recipients demonstrated that dd-cfDNA, measured using the Prospera assay, detected biopsy-proven acute rejection with a sensitivity of 85.7% and specificity of 93.7% when using a quantity cutoff of 70copies/mL. This performance was superior to both serum lipase (AUC 0.74) and amylase (AUC 0.46), supporting the clinical utility of dd-cfDNA as an early and accurate marker of pancreas graft rejection in SPK patients.[2] Importantly, dd-cfDNA elevation was shown to precede clinical rejection episodes, providing a window for earlier intervention.[1] Based on this data, we are in the process of incorporating this into our routine surveillance protocol for simultaneous pancreas-kidney and pancreas after kidney transplant recipients.

Clinical Utility in Multi-Organ Transplantation

While the literature on dd-cfDNA in all multi-organ transplant scenarios is still evolving, the evidence for SPK transplantation is robust and growing. The Banff 2022 guidelines explicitly recognize the utility of dd-cfDNA in SPK recipients, and recent studies confirm its diagnostic accuracy and potential to reduce unnecessary biopsies and improve graft outcomes.[1][2]Furthermore, dd-cfDNA offers advantages in situations where biopsy is contraindicated or refused, and may facilitate more personalized immunosuppression strategies.[3][4]

Recommendation

Given the above, there is sufficient evidence to support the clinical validity and utility of dd-cfDNA monitoring in SPK transplant recipients. The current LCD statement, which categorically excludes all multi-organ transplant patients from coverage, does not reflect the nuanced and organ-specific evidence base. Restricting access to dd-cfDNA testing in SPK patients may result in delayed diagnosis of rejection, increased reliance on invasive biopsies, and potentially worse graft outcomes.

Request

I respectfully request that the LCD be revised to allow coverage of dd-cfDNA testing in SPK transplant recipients, consistent with the current evidence and expert consensus. Such a revision would align coverage policy with the best available clinical data and the recommendations of leading transplant societies. Transplant centers that perform pancreas transplantation need additional tools to improve pancreas allograft survival and overall patientcare and dd-cfDNA offers that is easy to perform and for clinicians to understand.

References were provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

I am writing as a [ENTER SPECIALTY HERE] with [INSERT #] years experience in managing kidney transplant recipients. I currently practice at [ENTER TRANSPLANT CENTER HERE]. I would like to provide comments on the Draft LCD for Molecular Testing for Solid Organ Allograft Rejection, issued on July 17, 2025. I appreciate the process of including specialist feedback to update coverage policies.

It was nice to see the inclusion of dd-cfDNA for surveillance testing to identify subclinical rejection. I do however have concerns with establishing surveillance caps, this was a surprise and not expected. As the draft policy noted, “Physicians know best how to manage care for their patients.” It’s also well understood that there is heterogeneity between the populations served by transplant centers and between individual patients and their risk for rejection. Instead of establishing limits on using dd-cfDNA for surveillance, I urge MolDX to adopt an approach whereby the treating clinicians have the flexibility to order surveillance tests on a different or more frequent cadence based on their individualized assessment of what is medically necessary for that specific patient or at a minimum align with clinically validated protocols such as the ARTS1 protocol from DART2 and KOAR,3 allowing at least seven tests in the first year (monthly for months 1-4, then quarterly) and quarterly thereafter.

While the Draft LCD recognizes that prospective trials designed to optimize the frequency of dd-cfDNA testing in stable, asymptomatic adult renal transplant recipients is lacking, multiple experts agree that developing such data is impractical. However, the evidence for the clinical utility of dd-cfDNA surveillance testing is clear, as is the utility of testing longitudinally.

For maintenance immunosuppression, the KDIGO4 guidelines recommend using the lowest planned doses of immunosuppressive medications by month 4 providing there has been no rejection or delayed graft function. The current “ARTS” surveillance testing schedule of monthly tests from months 1 to 4 and quarterly tests thereafter aligns with this post-transplant period of increased risk of rejection while maintenance immunosuppression is being optimized. This was the surveillance schedule which was developed and used in the DART and KOAR studies which demonstrated that dd-cfDNA levels were elevated up to 4 months before biopsy confirmation of ABMR, thus allowing clinicians to identify these patients several months in advance of other clinical signs or symptoms of rejection and before irreversible damage to the allograft.

In addition, studies have demonstrated that kidney transplant patients will have variable baseline dd-cfDNA levels based upon their level of sensitization (cPRA%), prior kidney transplant status and other clinical variables; thus obtaining serial surveillance levels to determine a patient’s individual baseline will be beneficial in the management of these patients.

Surveillance with dd-cfDNA serves critical purposes: detecting subclinical rejection before irreversible allograft injury and enabling precise immunosuppression titration, which varies by patient. This can only be accomplished through routine surveillance with the ARTS protocol of 7 surveillance episodes in the first year followed by quarterly testing thereafter.

Finally, it’s important to note that the KDIGO guidelines recommend increased frequencies for non-invasive testing (serum creatinine, urine protein excretion) for allograft function/rejection than they do for biopsy in large part due to the more favorable risk-benefit profile of non-invasive testing. It should be noted that dd-cfDNA fits a similar risk-benefit profile of serum creatine and proteinuria however it is a far more sensitive test in picking up signs of early rejection. That said, it would make more sense to align dd-cfDNA testing for rejection to other non-invasive testing frequencies rather than invasive testing via biopsy.

In summary, while the draft advances coverage, frequency caps undermine physician discretion and evidence-based care. I recommend removing limits or adopting ARTS-aligned frequencies (7 in year one, 4 annually thereafter) to reflect dd-cfDNA's utility in improving graft outcomes in transplant and repeat transplant recipients.

References were provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA, CGS, Noridian, and WPS:

Thank you very much for joining the American Clinical Laboratory Association (ACLA) for our recent discussion with payors about documentation to support the medical necessity of laboratory tests. ACLA members benefitted greatly from hearing your perspective and that of other payors and from sharing ideas about how to make that aspect of the claim submission process more seamless. We appreciate the time you spent with us and your active engagement in the conversation.

This letter includes ACLA’s comments on the draft Local Coverage Determination MolDX: Molecular Testing for Solid Organ Allograft Rejection¹(“dLCD”). As you know, ACLA is the national trade association representing leading laboratories that deliver essential diagnostic health information to patients and providers by advocating for policies that expand access to the highest quality clinical laboratory services, improve patient outcomes, and advance the next generation of personalized care. Many ACLA member laboratories submit claims to Medicare Administrative Contractors (MACs) that follow MolDX policies and have an interest in the program’s coverage determinations and operations.

We believe strongly that MolDX should not proceed with the proposal to create surveillance testing “bundles” under which it would pay for “multiple tests comprising the service.” This would add a layer of complexity that is unnecessary in the transplant setting. The “bundles” correlate to new proposed frequency limits on surveillance testing but do not include information about payment rates, nor does the article that is associated with the dLCD describe a transparent method for calculating the rates. ACLA is concerned that this “bundling” process would result in payment for laboratory tests covered under the policy that differ from the rates established by the Centers for Medicare & Medicaid Services (CMS). A departure from the rates established by CMS is prohibited under law: Any rate established by CMS through the process set forth in 42 U.S.C. § 1395m-1(b)(1)(A) “shall continue to apply until the year following the next data collection period” and “shall not be subject to any adjustment (including any geographic adjustment, budget neutrality adjustment, annual update, or other adjustment)”.² For a test that was priced via crosswalking or gapfilling, that rate is to stay in place “until applicable information is available to establish a payment amount under the methodology described in 42 C.F.R. § 414.507(b)” (which implements 42 U.S.C. § 1395m-1(b)(1)(A)).³ Neither Congress nor CMS has authorized a different pricing mechanism, such as “bundling” a series of the same test, for a test code that has been priced under the processes set forth in 42 U.S.C. § 1395m-1. Therefore, bundled billing should be employed only when no alternative payment mechanism exists for a particular service, e,g., if a national coverage determination explicitly non-covers serial testing, which is not the case in transplant testing and monitoring.

Finally, ACLA members are concerned that the “bundling” requirement would create confusion regarding coverage requirements and lead to high rates of claim denials by Medicare Advantage (MA) plans that require laboratories to use Z-codes for claims submissions. Laboratories note that some MA plans that require Z-codes also currently require individual submission of test claims in all cases and do not have the ability to process bundled claims.

ACLA urges MolDX not to proceed with “bundling” the tests as proposed and instead instruct laboratories to file claims for the individual services as they are provided. If MolDX decides to proceed with the “bundled” payments proposal, MolDX should clarify that it will pay for each bundle at a rate equivalent to the current CLFS rate times the number of tests included in the bundle.

References were provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA:

I am writing in support of the use of more frequent donor-derived cell-free DNA (dd-cfDNA) testing in the surveillance of kidney transplant recipients, as outlined in Proposed LCD.

I am a transplant clinician and I try to follow well-established KDIGO guidelines, as such recommending frequent monitoring for kidney allograft dysfunction in the first year post-transplant. Standard-of-care includes testing multiple markers — such as serum creatinine, BKV, and EBV — at least seven times in the first year, and serum creatinine at least four times per year thereafter.

In line with above, current evidence clearly shows that dd-cfDNA significantly outperforms serum creatinine in identifying subclinical rejection and allograft dysfunction (Graver et al., Transplantation, 2023). This test is able to detect injury earlier, before overt biochemical or clinical signs appear, and thus offers an opportunity for earlier intervention and improved graft survival.

Recent prospective studies (Bromberg et al., Transplantation, 2024; Bromberg et al., Am J Transplant, 2025) demonstrate that dd-cfDNA elevation can precede biopsy-proven antibody-mediated rejection by up to five months and T-cell mediated rejection by up to two months. In my own clinical practice, I have observed too many cases where earlier detection via dd-cfDNA allowed us to initiate treatment prior to irreversible injury, avoiding the functional decline and fibrosis often seen when relying solely on serum creatinine trends.

Given this solid evidence, it is in my opinion, clinically appropriate and necessary to assess dd-cfDNA at a frequency that at least mirrors serum creatinine monitoring — especially in the first year post-transplant when the risk of rejection is highest. Doing so aligns testing with current surveillance standards, improves rejection detection yield, and ultimately supports better patient and graft outcomes. Instead of limiting the number of cfDNA tests per year by virtue of cutting reimbursements, in my humble opinion MolDX should rather focus of lowering the amounts of reimbursements issued to the diagnostic companies preforming commercial measurement of cfDNA.

I strongly urge MolDX to recognize the medical necessity of more frequent dd-cfDNA testing as part of routine surveillance for kidney transplant recipients.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA:

I appreciate the opportunity to comment on the proposed LCD regarding molecular testing for solid organ allograft rejection. While the recognition of donor-derived cell-free DNA (dd-cfDNA) as a surveillance tool is an important step, the proposed limitation of four tests in the first year and two annually thereafter is not sufficient to meet clinical need.

As a transplant nephrologist, I have seen firsthand that rejection often occurs outside of these restricted time windows. In multiple cases, dd-cfDNA identified subclinical rejection before serum creatinine changes or overt clinical symptoms developed. Early detection allowed us to intervene and reverse rejection, preserving graft function that might otherwise have been lost. These outcomes would not have been possible under the proposed testing frequency.

Published evidence also supports that dd-cfDNA can detect antibody-mediated rejection months before biopsy confirmation and T-cell–mediated rejection weeks in advance. Given this, more frequent surveillance is clinically appropriate and essential to optimizing patient and graft survival. I strongly urge MolDx to revise the proposed LCD to allow for greater flexibility in testing frequency, consistent with clinical judgment and patient needs.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA:

I am writing in response to the Proposed LCD regarding coverage for molecular testing in solid organ allograft rejection. While I commend MolDx for recognizing the value of donor-derived cell-free DNA testing, I am deeply concerned that the proposed coverage of only four tests in the first year and two tests per year thereafter is insufficient to meet the clinical needs of kidney transplant recipients.
It is routine standard-of-care and recommended by well-established KDIGO guidelines (Am J Transplant. 2009;9 Suppl 3:S1–155), to perform frequent monitoring for kidney allograft dysfunction using multiple molecular markers. These guidelines recommend serum creatinine and other markers of kidney function at least seven times in the first year, and serum creatinine at least four times per year thereafter.

Given the current evidence that dd-cfDNA significantly outperforms serum creatinine in identifying subclinical rejection and allograft dysfunction (Graver AS et al., Transplantation. 2023;107(8):1675–1686), it is clinically appropriate to assess dd-cfDNA levels at a frequency similar to serum creatinine monitoring, rather than the highly limited testing schedule proposed.
Recent published evidence further demonstrates the clinical importance of serial dd-cfDNA surveillance:

• Bromberg JS et al., Transplantation. 2024;108(9):1994–2004: dd-cfDNA elevations were observed up to 5 months prior to biopsy-proven ABMR and up to 2 months prior to biopsy-proven TCMR.
• Bromberg JS et al., Am J Transplant. 2025; Epub ahead of print: dd-cfDNA significantly improved the diagnostic yield of transplant biopsies, confirming its role as an essential surveillance tool.

These findings underscore that rejection often develops silently over months, and infrequent monitoring risks missing early, treatable rejection events.

In my practice, we have cared for several patients who underwent kidney transplantation with positive serial dd-cfDNA and stable creatinine. One notable case involved a patient in their 40s with stable serum creatinine levels. However, a series of dd-cfDNA tests indicated a high-risk level, despite a negative result just a month earlier. When we repeated the test a few weeks later, the elevated levels persisted while the creatinine remained normal and stable. A biopsy was performed, which confirmed rejection. Fortunately, this rejection was successfully treated, and her dd-cfDNA levels decreased back to the negative risk baseline after the treatment. Thanks to these findings, we were able to diagnose the rejection early. One year later, her creatinine levels are still normal. Before the introduction of dd-cfDNA testing, this patient would have likely presented in a few years with worsening proteinuria or elevated creatinine levels, necessitating a biopsy that would have shown advanced, irreversible scarring in the kidney, which did not happen here, as we were able to detect it early on and treat it. Early and preemptive identification and treatment are essential for managing patients after a transplant. Limiting access to this innovative approach in the care of post-kidney transplant patients may lead to late recognition of rejection, which could have been detected early and treated effectively.

I understand the expenses tied to this test, but the costs involved in ensuring the longevity of a kidney transplant are much higher than the test's expense. It’s essential that we enhance our post-transplant care through innovative methods and focus on our patients' needs. Even years after their transplant, kidney transplant patients should have access to advanced care, including early detection markers for rejection—beyond just serum creatinine, which is not very sensitive.

For dd-cfDNA to be used in a way that truly aligns with established transplant standards and evidence, Medicare coverage should support more frequent serial monitoring that parallels conventional surveillance schedules—at least 7 tests in the first year and 4 tests per year thereafter. Anything less risks undermining the full clinical value of this technology and compromising patient outcomes.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA:

In regards to this proposal, I am transplant nephrologist for more than 10 years and have experienced the introduction of dd cf DNA in our practice since its inception.

As a transplant nephrologist and interventional nephrologist (personally performing > 250 renal allograft biopsies /year), and in view of recent robust data of dd cfDNA use for surveillance and monitoring rejection treatment, I believe that the policy needs to be reviewed to allow more testing beyond the first year post transplant.

Thank you for your comment. Please see Response #1.

The following comment was submitted to CGS, Noridian, and WPS:

I write to express ASTS’ concerns about the proposed Local Coverage Determination: Molecular Testing for Solid Organ Allograft Rejection (the “proposed LCD”). ASTS is a medical specialty society representing approximately 2,400 professionals dedicated to excellence in transplant surgery and to the patients that we serve. Our mission is the advancement of the art and science of transplant surgery through patient care, research, education, and advocacy.

We appreciate that data for the rapidly emerging field of molecular diagnostic testing are still maturing and that the global costs associated with such testing, while not insignificant, are less costly than biopsy and its associated complications or graft failure and return to dialysis (in the case of kidney) or repeat transplantation. However, molecular diagnostic testing has indisputable clinical utility and may provide massive clinical and economic benefits in the early detection and management of solid organ allograft rejection. Utilizing molecular testing for detection of allograft injury has emerged as a standard of care that can directly aid clinical decision making and may improve patient and allograft survival. We support the access of transplant patients to these diagnostic technologies and believe that continued Medicare coverage of these tools is critical. We recently issued an updated ASTS White Paper on molecular diagnostic testing (Appendix A), which was utilized as source material by MolDX in formulating the proposed LCD (Appendix B)

We note the significant changes made by MolDX to the proposed LCD relative to the prior LCD and congratulate them for the wise removal of the linkage of molecular testing to tissue biopsy. We note that the proposed LCD recognizes the clinical utility of both surveillance and for-cause testing. We are heartened by the progress made and confident that those changes will benefit patients. However, we have scientific, ethical, and process-related concerns about the proposed LCD which are enumerated below.

We are puzzled that the coverage limitations in the proposed LCD have been put forward at a time when CMS has clearly acknowledged that transplantation is the best, and most cost-effective, treatment option for those with ESRD, and without apparent consideration of the potential impact on innovation in the field or on patient care.

The proposed local coverage determination (LCD):

1) Specifically constrains clinicians’ ability to surveil their kidney transplant recipients for rejection utilizing molecular testing by limiting surveillance testing in kidney transplant recipients to four tests in the first year after transplant and two tests per year in subsequent years.

2) Specifically constrains clinicians’ ability to surveil their heart transplant recipients for rejection utilizing molecular testing by limiting surveillance testing to twelve tests in the first year after transplant and two tests per year in subsequent years.

3) Specifically constrains clinicians’ ability to surveil their lung transplant recipients for rejection utilizing molecular testing by limiting surveillance testing to twelve tests in the first year after transplant and two tests per year in subsequent years.

4) Does not provide support for the utilization of molecular testing in recipients of simultaneous kidney and pancreas transplants or recipients of repeat kidney transplants despite evidence of analytical validity (AV) and clinical validity (CV) in this patient population.

5) Does not support multimodality or combination testing or concomitant use of multiple tests in heart transplant recipients.

We feel that these features of the proposed LCD may substantively compromise patient care as detailed below. Our concerns regarding the surveillance testing limits in the proposed LCD extend to abdominal and thoracic transplant organs.

1. The proposed LCD’s limits on the frequency of surveillance testing are too low.

The primary purpose of surveillance (or protocol) biopsies in kidney transplant recipients is the detection of subclinical acute rejection (subAR), i.e., rejection occurring in the absence of overt clinical symptoms or laboratory abnormalities. The role of routine surveillance biopsies in the era of modern immunosuppression has come into question, and the utilization of such biopsy protocols has decreased over the years.^{1, 2} In a survey of US transplant programs in 2017, only a minority (38%) reported performing any surveillance biopsies, with very few (17%) doing so in a universal, non-risk stratified fashion.³ The most recent KDIGO guidelines on the management of kidney transplant recipients, published in 2009, do not recommend surveillance biopsies, instead concluding that “RCTs are needed to determine when the benefits of protocol biopsies outweigh harm.”⁴ However, the same guidelines recommend treating subAR recognizing its association with eGFR decline, chronic graft injury, and graft loss, along with evidence that treatment may ameliorate the risk of these adverse outcomes.^4-9 This is inherently conflicting guidance, as subAR treatment is advised, but none of the modalities endorsed in the KDIGO guidelines for monitoring graft function (urine volume, urine protein excretion, serum creatinine, or ultrasound) are useful for its detection and the surveillance biopsies that historically have revealed the presence of subAR are not recommended.⁴

Molecular testing with donor-derived cell-free DNA (dd-cfDNA) provides an ideal solution to this conundrum. Several studies have demonstrated a strong correlation between dd-cfDNA and the finding of histologic rejection identified on surveillance biopsies, establishing a non-invasive between dd-cfDNA and molecular histology offers the possibility of detecting subAR that may not yet be apparent on traditional histologic tissue examination.¹¹

The proposed LCD tethers the ability to perform reimbursable surveillance testing with dd-cfDNA to published surveillance biopsy protocols, limiting patient access and essentially conflating the risk/benefit calculations for a non-invasive blood test with a resource-intensive interventional procedure (tissue biopsy) that is neither guideline-endorsed nor broadly utilized for kidney allograft surveillance (as opposed to for-cause use) in clinical practice.^3,4 We maintain that transplant professionals, in partnership with patients, require the exercise of clinical discretion in determining the frequency of molecular surveillance testing in these challenging patients. We are also cognizant of the need for all transplant stakeholders to be responsible stewards of scarce resources. Guardrails on surveillance testing frequencies are logical, and we have openly advocated for transplant professionals utilizing these testing modalities to enshrine practice patterns or protocols that prevent inadvertent overutilization. By the same token, the cost of inadequate surveillance can be measured in organs and lives lost, and those losses are both tragic and inordinately expensive.

Testing frequency should therefore be based on relevant patient-centric factors, including the immunologic risk of a particular patient, the risks of biopsy in that patient, and their clinical judgement regarding the time points post-transplant at which such surveillance are warranted. Molecular testing is non-invasive and carries a diametrically different risk-benefit profile than invasive tissue biopsy. The robust clinical validity of molecular testing and its utility in clinical decision-making suggest that surveillance testing frequencies should parallel those of other non-invasive surveillance tests. While the optimal surveillance testing frequency is unknown, the most robust clinical validation study for kidney allograft surveillance testing comes from the DART study, which utilized the so-called “ARTS” surveillance testing schedule. This protocol recommended testing at monthly intervals from month one through four, and then quarterly thereafter. This would suggest a reasonable kidney allograft surveillance frequency of seven tests in the first-year post-transplant and four tests per year thereafter.

This schedule was developed empirically based on the KDIGO recommended testing frequencies for other noninvasive post-transplant tests and temporally correlates the highest testing frequency with the period of highest immunologic risk and recommending decreased frequencies later in the post-transplant course. Importantly, longitudinal surveillance testing at this frequency provides an early baseline for the allograft and then captures data at critical post-transplant immunologic timepoints.¹² This correlates with existing molecular testing surveillance protocols at many transplant centers and would alleviate patient and transplant provider concerns about the proposed LCDs stringent limits on surveillance test frequency.

Surveillance of renal transplant recipients does not only serve the clinical imperative of detecting subAR. The other critical importance of allograft surveillance is to facilitate individualization of the immunosuppressive burden faced by patients. One of the reasons for the paucity of improvement in long-term outcomes is that the burden of over immunosuppression carries significant risk of morbidity and mortality for patients. Surveillance of clinically stable patients allows for a data-driven approach to the careful decrease of immunosuppressive burden in these patients. Again, the surveillance testing frequency stipulated in the proposed LCD is inadequate for that long-term follow up goal.

The types of kidneys utilized, and the types of recipients transplanted continue to change in ways that increase the need for robust surveillance frequencies and the need for clinically validated tools for that surveillance. The Centers for Medicare and Medicaid Innovation (CMMI) IOTA model started in July of this year and is specifically designed to increase the utilization of kidneys at elevated risk of discard, including many biologically marginal, high-KDPI organs. IOTA is a mandatory model into which approximately half of the nation’s kidney transplant programs have been enrolled by CMS. The model will predictably, and by intent, result in higher incidences of delayed graft function (DGF) and higher prevalence of suboptimal allograft function. While many aspects of the IOTA model are laudatory, achieving IOTA objectives while protecting patients will likely require more frequent surveillance and closer monitoring.

Surveillance of renal transplant recipients is bound by the basic clinical tenant that rejection risk persists for the life of the allograft. While the risk of acute rejection, and therefore subAR, is highest in the first year after transplant, a sudden, steplike decrease in risk to the allograft, and thus the patient, does not occur at the transition from post-transplant month-12 to post-transplant month-13. As we have detailed above, kidney recipient surveillance testing frequencies should accommodate a minimum of seven tests in year one and four annual tests thereafter.

The proposed LCD limits surveillance testing frequencies for heart and lung recipients to 12 tests in year one, and 2 tests in all subsequent years. All organ transplant recipients remain at risk for rejection for the life of their allografts.^13,14 The prognosis of cardiac allograft rejection worsens when it is recognized at a more advanced stage, and more timely treatment of mild forms of rejection yields a more favorable prognosis.^15,16 Thus, efficacious and timely surveillance for rejection is a critical component in the management of cardiac transplant patients and a significant determinant of long-term survival rates.¹³ Importantly, rejection surveillance utilizing molecular testing may also allow safe, monitored, individualization of immunosuppression for transplant patients.¹³ Molecular testing shows promise in its ability to allow clinicians to safely lower immunosuppression in stable cardiac transplant recipients. Molecular testing surveillance techniques provide adequate sensitivity and specificity to provide early detection of allograft injury and allow timely treatment while avoiding the dangers of myocardial biopsy.¹⁷

Historically, while only biopsies provided sufficient sensitivity to detect early rejection, it is now clear that biopsies, in fact, may cause harm. Unavoidable risks of biopsies include bleeding, infection, pain, and anxiety. Biopsy-related complications can be devastating to patients as well as enormously expensive for payers.^13,18-21 In an actively surveilled population receiving modern immunosuppression the yield of biopsies is quite low.¹³ Finally, it has also increasingly been recognized that biopsies are severely limited in their real-world sensitivity and specificity, due to both significant interobserver variability and sampling error.^16,22,23 Because of these limitations, over the years, transplant centers have reduced the utilization of surveillance (protocol) biopsies while not reducing the frequency or duration of other forms of non-invasive surveillance, including molecular diagnostic testing.¹³ Importantly, the risk of heart allograft rejection, and the intensity of surveillance needed to monitor for it, do not decrease as precipitously over time as the surveillance testing frequencies allowed in the proposed LCD would imply or safely allow.

Surveillance molecular testing is not associated with pain, anxiety or life-threatening complications and can even be done at the patient’s home, sparing exposure of patients to hospital pathogens and limiting hospital resource consumption.^13,18,21 Furthermore, molecular testing is not subject to interobserver variability or sampling errors and therefore, may in fact have superior sensitivity and specificity than biopsies.²⁴ Therefore, molecular testing, performed at an adequate frequency, suitably replaces most cardiac surveillance biopsies. Specifically, extended surveillance with molecular testing is now reasonable both to detect late rejection and allow clinicians the ability to provide individualization of immunosuppression regimens throughout the transplant recipient’s lifetime.

2. The proposed LCD makes no provision for the use of molecular testing in simultaneous pancreas and kidney (SPK) recipients nor in kidney transplant recipients who have received one or more prior kidney transplants.

While the number of simultaneous pancreas and kidney (SPK) transplants performed nationally remains small, the risks to allograft health faced by these patients are more significant than in kidney alone recipients, and the risks of allograft pancreas biopsy are much higher than those associated with kidney biopsy. Importantly, acute rejection can occur in the pancreatic allograft of an SPK recipient discordantly from rejection in that recipient’s renal allograft. That makes the utilization of molecular testing for allograft surveillance critically important in these patients. Fortunately, data supporting the clinical utility of molecular testing utilizing dd-cfDNA continues to accrue. An observational, prospective single center study of SPK recipients conducted by Williams et al demonstrated that dd-cfDNA levels were significantly different (p=0.0006) between patients with rejection (median 2.25%) versus those with allograft injury (median 0.36%) or quiescence (median 0.18%). Among patients without rejection, 97% had dd-cfDNA <0.5%. This finding was consistent with published data in kidney transplantation. Biopsy-confirmed rejection was associated with elevations in dd-cfDNA levels with a median level of 2.4%. a median dd-cfDNA of 2.4%.²⁵

Yoo et al. demonstrated that dd-cfDNA values differed between SPK patients with rejection or infection (median 0.56%) versus those who were stable (median 0.28%). Of the SPK recipients with dd-cfDNA > 1.0%, 36.8% had evidence of infection or rejection. 100% of recipients without infection or rejection had dd-cfDNA levels <1.0%, consistent with the previously published rejection threshold of 1.0% for kidney transplant recipients.²⁶ These and other studies support the clinical utility of dd-cfDNA in SPK recipients, and we urge MolDX to incorporate that emerging standard of care in the proposed LCD.

Kidney recipients facing the need for retransplantation already represent a sizeable proportion of all kidney transplant candidates, and the prevailing trend of more aggressive utilization of biologically challenged and high-KDPI kidneys will likely increase that proportion. A total of 27,759 kidneys were transplanted in the US in 2024. Of those, 2,614 kidneys were transplanted into prior kidney recipients. Over the past decade, approximately 10% of patients who received kidney transplants had a prior kidney transplant that had failed.²⁷ The ADMIRAL Study evaluated dd-cfDNA in 1,092 kidney transplant recipients from seven US transplant centers, 8% of these recipients had a prior kidney transplant.²⁸ Aubert evaluated 2,882 kidney transplant recipients from the US and Europe, 447 (15.5%) of which had had a prior kidney transplant.²⁹ Both of these large studies demonstrated that the performance of dd-cfDNA was similar in de novo and repeat kidney transplant recipients and clearly differentiated immunologic quiescent and rejection in both groups. The proposed LCD should contain language, and reference data, supporting surveillance and for-cause testing kidney recipients who have had prior kidney transplants.

3. The proposed LCD inappropriately restricts multimodality testing and concomitant use of multiple tests.

Multi-modality molecular surveillance testing using gene expression profiling (GEP) and dd-cfDNA provides complementary, not redundant, information about a transplant recipient. Gene expression profiling measures mRNA levels of peripheral blood mononuclear cells while dd-cfDNA measures levels of circulating DNA released by an injured allograft.²³ Multimodal assessment utilizing both dd-cfDNA and GEP in heart transplant recipients provides additional clinical utility over their use in isolation. Combined dd-cfDNA and GEP testing provides information on distinct biologic processes, with dd-cfDNA providing insight about graft injury, and GEP providing insight about recipient immune system activation. In the context of rejection surveillance, where the prevalence of disease is low, it is generally accepted that the most important characteristic of a surveillance test is its ability to predict which of the surveilled patients are most likely to have rejection.^30,31 The characteristics of the test that closely aligns with this objective is the positive likelihood ratio. For patients undergoing surveillance for acute cellular rejection, the magnitude of the positive likelihood ratio of any one commercially available molecular test using recommended thresholds is modest and at most 2.9.^31,32,35,36 When results in a transplant recipient exceed normal thresholds for both GEP and dd-cfDNA the likelihood ratio is approximately 4, which provides a robust ability to accurately identify patients with underlying acute cellular rejection.,³⁵ The clinical utility achieved by using dd-cfDNA and GEP concomitantly allows clinicians to dramatically reduce their use of biopsies, with attendant benefits for patients.^33,34,35 Importantly, the SHORE study, published by Kush et al., was not available at the time we published the most recent version of our molecular diagnostics position statement. It is an exceptionally large multi-center study in the cardiac transplant space with convincing clinical validity and utility data supporting the clinical utility of combination GEP and dd-cfDNA testing in heart transplant recipients.

We recognize that results obtained utilizing legacy surveillance techniques are suboptimal. Further, the persistent failure to achieve meaningfully better long-term renal allograft survival despite massive improvements in short-term patient and allograft survival remains one of the cardinal failures of the transplant endeavor. Molecular testing is a standard of care that holds the promise of changing kidney allograft surveillance and for-cause testing paradigms. Molecular diagnostic testing may allow us to unlock long sought and badly needed gains in long-term patient and allograft survival in kidney transplant recipients. This technology is already a standard of care in the management of our patients, and the limitation of patient access to these platforms stipulated in the proposed LCD will be detrimental to advancement of the field in general and the care of these vulnerable patients specifically. We respectfully and strongly urge MolDX to increase or eliminate the surveillance testing frequency limits in the proposed LCD and to stipulate coverage for SPK and retransplant patients as well as multimodal testing of GEP/dd-cfDNA in heart transplants.

Appendices and references were provided for review.

Thank you for your comment. Please see Response #1.

The following comment was submitted to Palmetto GBA and Noridian:

The American Kidney Fund (AKF) appreciates the opportunity to provide comments on the proposed LCD: MolDX: Molecular Testing for Solid Organ Allograft Rejection.

AKF fights kidney disease on all fronts as the nation’s leading kidney nonprofit. AKF works on behalf of the 37 million Americans living with kidney disease, and the millions more at risk, with an unmatched scope of programs that support people wherever they are in their fight against kidney disease—from prevention through transplant. Through programs of prevention, early detection, financial support, disease management, clinical research, innovation and advocacy, no kidney organization impacts more lives than AKF. AKF is one of the nation’s top-rated nonprofits, investing 96 cents of every donated dollar in programs, and holds the highest 4-Star rating from Charity Navigator and the Platinum Seal of Transparency from GuideStar.

AKF Comments on the Proposed LCD:

Molecular diagnostic tests, including donor-derived cell-free DNA (dd-cfDNA) testing, plays a critical role as an accurate, non-invasive tool to monitor kidney function and detect graft injury and rejection in transplant patients. Because of this, AKF supports the proposed LCD as it relates to the for-cause coverage of molecular diagnostic tests, as outlined by the coverage criteria in the proposed LCD. However, we urge the Medicare contractor to remove the proposed surveillance testing limits for kidney transplant recipients and allow physicians to use their discretion in using non-invasive molecular diagnostic tests to prevent organ rejection, which is highly specific to each patient. If the Medicare contractor finalizes a surveillance testing limit for kidney transplant recipients, we recommend the limit be set at 7 in the first-year post-transplant, and quarterly in subsequent years (i.e. 4 per year).

As the Medicare contractor stated in the LCD, “Physicians know best how to manage the care of their patients.” We strongly agree with this statement, particularly in the context of post-transplant care and preventing organ rejection. There are several clinical factors that can influence the best course of treatment for a kidney recipient, and there may be unforeseen changes or circumstances that occur during a year that may prompt the need for additional surveillance testing to monitor kidney function or detect graft injury. Removing the proposed surveillance testing limits or raising the limit to 7 in the first year-year post transplant and 4 times a year in subsequent years will ensure physicians are not hindered by lower limits that could impact their ability to detect subclinical rejection and manage the care of their patients.

The Medicare contractor based its proposed surveillance frequency limits on the frequency of surveillance biopsies performed at transplant centers. However, non-invasive tests such as dd-cfDNA have a far more favorable risk to benefit profile than biopsies and should not be bound by the same surveillance frequency limits that are more applicable to outdated surveillance biopsy protocols. Studies of non-invasive tests such as dd-cfDNA have demonstrated their ability to detect elevated dd-cfDNA and allow for early intervention to address graft injury or rejection, months before ABMR (antibody-mediated rejection) and TCMR (T-cell-mediated rejection) were detected on biopsy.^1,2,3,4 Given the effectiveness of non-invasive tests and their more favorable risk to benefit profile compared to biopsies, the lower surveillance frequency limits should not be applied.

The Importance of Molecular Diagnostic Testing to Kidney Transplant Recipients

To ensure the best health outcomes for a person with a kidney transplant, providers need to consistently monitor how well the new kidney is functioning. A blood draw to measure serum creatinine is a routine and inexpensive test, but creatinine is not a very sensitive marker for identifying early kidney injury. There may be significant injury to a transplanted kidney before creatinine is elevated, and treatments may not be as effective.

A kidney biopsy is considered the most informative test to diagnose issues with a transplanted kidney, but it is an invasive procedure with a less desirable risk to benefit profile. Because of this, the biopsy is not ideal for ongoing kidney surveillance.

Donor-derived cell-free DNA tests have been a trusted test for over 15 years for heart transplant recipients and over 5 years in kidney transplant recipients. The testing of dd-cfDNA has resulted in improved early detection of graft injury and rejection, and can also reassure the provider and the patient that there is no subtle, unrecognized damage or rejection going on with the transplant if the level of dd-cfDNA is normal. Prior to the adoption of dd-cfDNA tests, if a patient’s creatinine was elevated for unclear reasons, they may have been asked to have a kidney biopsy to exclude the possibility of rejection, infection, or other acute damage to their transplant. Now with the use of dd-cfDNA testing, a patient with elevated creatinine but a normal amount of circulating dd-cfDNA typically means there is no damage to the transplant and the patient may be monitored closely, rather than undergo an invasive biopsy.

Conversely, if a creatinine level is normal but the dd-cfDNA is elevated, this may be an early warning sign that there is a need for more investigation, sometimes including a kidney biopsy, to catch problems early and allow for the best treatment before there is further kidney damage. Because dd-cfDNA testing plus serum creatinine is more accurate than just checking serum creatinine, and because it only requires a blood draw, dd-cfDNA tests have become a critical tool for monitoring a patient’s kidney transplant and to help determine if they need a kidney biopsy.

AKF has heard from patients about the importance of dd-cfDNA tests to their care, and how results from the tests helped their doctors decide to adjust their immunosuppressive medications to prevent rejection. Kidney transplant is considered the best treatment option for most people with kidney failure because it increases a patient’s chances of living a longer, healthier life. Donor-derived cell-free DNA tests are a significant improvement to the kidney monitoring tools that have been used before, and they can help a patient get as many years as possible out of their transplant. Given the struggles that many patients encounter in their journey to receiving a transplant, they deserve access to diagnostic and therapeutic innovations that help protect their transplant and improve their lives.

References were provided for review.

Thank you for your comment. Please see Response #1.