Response to Comments: MolDX: Breast Cancer IndexTM (BCI) Gene Expression Test

I am Professor of Medicine at Yale University, Director of Breast Cancer Translational and Co-Director of Genomics and Genetics Program at the Yale Cancer Center and a practicing medical oncologist for 22 years. My research focuses on breast cancer biomarkers and we have published over 250 manuscripts in this field. Multi-gene diagnostic assays have made a significant contribution to the clinical care of breast cancer patients. The first generation of these assays addressed the question which estrogen receptor positive (ER+) breast cancers have excellent prognosis with 5 years of adjuvant endocrine therapy alone and therefore would not require adjuvant of chemotherapy. More recently we learned that the risk of metastatic recurrence for ER+ breast cancers extends over a very long time and extended adjuvant endocrine therapy beyond 5 years improves survival, but only by 2-3 percentage points. This implies that if all patients are treated with extended adjuvant endocrine therapy the risk of overtreatment is very high. However, every ER+ patient who completes 5 year of adjuvant endocrine therapy (approximately 150,000 women per year) has to make a decision about stopping or continuing treatment. I am writing this letter in support of the BCI assay because I believe that it represents the best validated molecular assay to assist with this decision.

The Breast Cancer Index is the only commercially-available genomic assay that has been validated to predict whether extended endocrine therapy benefits a patient or not. This distinguishes BCI from the all other multi-gene assays that have only been shown to predict risk of recurrence over a 10-year period.

The BCI has been validated in 3 pivotal studies to stratify patients based on the overall (0-10 year) risk of distant recurrence, as well as to identify patients who are at high vs low risk of late recurrence five or more years after diagnosis. BCI has achieved level 1b evidence for prognostication of overall 10-year and late distant recurrence, a key validation measure of clinical utility for a biomarker, through rigorous prospective-retrospective analyses of the Stockholm Trial and Trans-ATAC (Arimidex, Tamoxifen, Alone or in Combination) randomized control trial (RCT) cohorts. BCI has the prognostic ability to identify a significant proportion of patients who are at a low 10-year risk of 4.8% and a low risk of late distant recurrence of 2.8%-3.5% that was shown in the Stockholm and Trans-ATAC RCTs. In addition, the predictive ability of BCI for benefit from tamoxifen and aromatase inhibitor endocrine therapies has been demonstrated in 3 blinded, prospective-retrospective studies within randomized controlled trial cohorts (Stockholm Tamoxifen RCT, Trans-ATAC RCT, and NCIC-CTG MA.17 RCT) in which BCI Predictive (HoxB13/IL17BR; H/I) demonstrated a statistically significant interaction (p<0.05) between H/I and various endocrine therapies. BCI was validated as a predictive tool in the extended endocrine therapy setting in the seminal NCIC-CTG MA.17 RCT where patients categorized as high H/I experienced a 67% reduction in risk of recurrence (P=0.007), whereas patients with low H/I did not show a significant decrease in cancer recurrence. To date, no other clinical, pathologic, or genomic test has been shown to predict benefit from extended adjuvant endocrine therapy.

The decision impact of the BCI results were also examined in clinical studies. Physician recommendations for extended endocrine therapy changed for 26 % of patients after considering BCI results, with a net decrease in recommendations for extended endocrine therapy from 74 % to 54 %. Equally importantly after testing, fewer patients reported decision conflict and anxiety also decreased.

Based on the available study results, my assessment is that the BCI has achieved a level of validation for the prognostic and predictive test results that are aligned with the intended use of the draft LCD. Clinical studies also showed that incorporation of BCI results into risk/benefit discussions of extended endocrine therapy resulted in changes in treatment recommendations and improved patient satisfaction. Based on these results and on my personal experience with the test I fully support Medicare coverage for the assay.

The BCI assay feels an important unmet clinical niche and making it more broadly available to our patients would increase quality of care. Please feel free to contact me if you need any clarifications or documentations for the above statements.

Thank you for the comment. We hope that this coverage policy will help ensure that Medicare beneficiaries have access to reasonable and necessary testing to help guide treatment of breast cancer.

For patients diagnosed with early stage hormone receptor-positive (HR+) breast cancer, a key challenge is that the risk of cancer recurrence continues beyond their disease-free 5-year anniversary. Due to this persistent risk, termed late recurrence risk, an increasing proportion of HR+ patients are recommended to extend their primary adjuvant treatment to include more than 5 years of endocrine therapy (ET). In addition, clinicopathological approaches to select women at greatest risk of breast cancer recurrence and/or in which patients will the longer duration therapy be most efficacious are limited and not individualized.

Several clinical trials (NSABP B-33, ATioM/ATLAS, MA.17(and MA.lR), and NSABP-842) have demonstrated improved outcomes for 3-5% of women from extending adjuvant endocrine therapy from 5 versus 10 years. However, extended endocrine therapy (EET) can adversely impact quality of life (e.g., hot flashes, sexual side effects), and increase the risks of potentially serious conditions (e.g. endometrial cancer and thromboembolic events with extended tamoxifen, and new osteoporosis and bone fractures with extended aromatase inhibitors). Given the modest efficacy of EET in an unselected population from the clinical trials and the low adherence rate, Medical Oncologists rely on clinicopathologic estimations of risk (age, node status, tumor size, etc.) and the symptomology of our patients to make decisions for EET. Subgroup analyses of these pivotal trials, however, reveal no consistent clinicopathologic characteristics that predict the patients who benefit from EET, leading to the overtreatment of a great number of women based on clinicopathologic factors alone. Therefore, a critical unmet need exists for better toots that can identify which patients are most at risk for recurrence of breast cancer after five years and who will benefit (or who will not) from prolonged ET.

The Breast Cancer Index (BCI) allows us to more appropriately select patients who have a higher probability of benefitting from EET. For the past three years, myself and my colleagues in the Medical Oncology Department at the Siteman Comprehensive Cancer Center in St. Louis, have utilized SCI to help individualize decision s for EET in early-stage, HR+ patients in order to prevent late distant breast cancer recurrence. Prior to integration of BCI into our practice, patients experienced significant distress and conflict with the decision to continue ET.

This meant that we spent much more time during appointments coaching our patients to take their prescriptions and more focus on mitigating ET­ associated side effects. BCI provides an individualized readout of the tumor biology that allows us to balance the risk versus benefit of continuing therapy beyond 5 years. The individualized approach to patient management allows us to frequently de-escalate ET in patients who would have otherwise been recommended to extend therapy based on clinicopathologic features alone. We have also used BCI results to extend therapy for patients in whom we would have not otherwise based on clinicopathology alone, thus preventing undertreatment in this critical subset of patients.

BCI is a validated assay with the appropriate clinical utility to address the critical unmet needed to identify patients most likely to benefit versus those who may not benefit from EET, thus preventing overtreatment of patients exposed to the side effects of prolonged endocrine therapy.

Thank you for the comment. We hope that this coverage policy will provide access to a test to allow Medicare beneficiaries and their physicians make well informed decisions as to whether or not to use endocrine therapy in the treatment of breast cancer.

I am writing in support of the draft Medicare policy (DL37794) for Breast Cancer Index (BCI). I am of the opinion that the proposed coverage is aligned with the published evidence and clinical utility of the assay in the management of patients with hormone receptor positive breast cancer who have been prescribed adjuvant systemic hormonal therapy.

Assessment of long term risk of metastatic disease recurrence in these patients as a significant percentage of recurrences occur five or more years following diagnosis. This stands out in juxtaposition with the perceptions of patients that there is a "five year mark" after which they are relatively free of recurrence risk. The BCI is critical to our effort to identify those patients for whom this misconception can have significant adverse consequences.

Medication compliance is a significant clinical challenge in this patient population. For a wide variety of reasons, a recommendation for extended adjuvant hormonal therapy can meet with considerable patient resistance and can be associated with significant late adverse effects. In my experience, BCI can assist with adjuvant systemic therapy decision making because it independently identifies patients with good outcomes from years 0-10 with 5 years of endocrine therapy alone.

I have utilized Breast Cancer Index at the time of initial diagnosis to provide individualized information about the anticipated duration of treatment. Early assessment of risk and benefit is critical to the open discussions between doctor and patient that are critical to patient compliance with effective adjuvant hormonal interventions. The improved decision-making from BCI provides information beyond traditional clinical features, helping insure that we prolong exposure to adverse events only in those instances where the risks and benefits warrant extended therapy.

I support the proposed coverage in the draft LCD for BCI based on:

• the published evidence validating BCI for prognosticating long-term risk of recurrence,
• the ASCO clinical practice guidelines that recommend BCI as a tool to assist in decisions regarding adjuvant systemic therapy, and
• the clinical utility of BCI to identify patients with favorable long-term outcomes.

Thank you for the comment. We agree that the published evidence supports the use of BCI, and we hope that this coverage policy will allow Medicare beneficiaries to have access to reasonable and necessary testing to help guide decisions regarding endocrine therapy in the treatment of breast cancer.

I believe the proposed coverage is in line with the published evidence and with the clinical utility of determining long-term risk, thus aiding in developing a comprehensive management plan for ER+ breast cancer patients.

Determining the long-term risk of metastatic disease recurrence for my ER+, early-stage breast cancer patients is crucial. While these patients have a relatively good prognosis as a group, their risk of metastatic recurrence continues long after the first 5 years of primary adjuvant endocrine therapy, with more than half of recurrences occurring 5 or more years after diagnosis. While clinicopathologic features like tumor size, tumor grade, and nodal status are somewhat prognostic and commonly used when risk­ stratifying patients, genomic tests such as BCI provide information beyond clinicopathologic features and help individualize patient management. This value to genomic assays has been recognized by various clinical practice guidelines. In the case of BCI the 2016 ASCO Biomarker guidelines for women with early-stage invasive breast cancer acknowledged BCI's utility for aiding in adjuvant systemic therapy decision making by identifying patients who could expect good outcomes in years 0-10, with only 5 years of endocrine therapy and no other systemic therapy.

I have used BCI close to time of diagnosis to help my patients formulate an individualized long-term treatment plan. I also found that BCI allowed me to give my patients a better sense of what duration of treatment they may be facing based on their risk from the time of diagnosis and likelihood to benefit from endocrine therapy after 5 years. Determining a patient's genomic risk and potential response to therapy early with BCI has allowed me to optimize their long-term treatment and to balance treatment benefit and side effects by directing treatment to only those patients that would benefit.

I support the proposed coverage in the draft LCD for BCI based on evidence validating BCI for prognosticating long-term risk of recurrence, ASCO guideline recommendation of BCI as a tool to for assisting in adjuvant systemic therapy decision-making, and on the clinical utility of BCI to identify patients with favorable outcomes.

Thank you for the comment. We hope that this coverage policy will help ensure that Medicare beneficiaries have access to reasonable and necessary testing to help guide treatment of breast cancer.

We recommend updating the coding information in the LCD prior to finalization to reflect the new BCI Category I CPT code and the new Z-code identifier for BCI. BCI was assigned code 816X0 in October 2017, which has since been finalized to 815181 and will take effect January 1, 2019. The new BCI Z-code identifier is [Z-code redacted].

In the CY 2019 Clinical Laboratory Fee Schedule (CLFS) Preliminary Determinations, the Centers for Medicare & Medicaid Services (CMS) proposes to crosswalk the new BCI CPT code to code 81519 (Oncology (breast), mRNA, gene expression profiling by real-time RT-PCR of 21 genes, utilizing formalin-fixed paraffin embedded tissue, algorithm reported as recurrence score).

Biotheranostics has submitted public comments in support of this proposed crosswalk. Once this crosswalk is finalized, the new BCI code 81518 will be nationally priced on the 2019 CLFS effective 1/1/19.

Background on BCI

BCI addresses the critical need to identify and select which breast cancer patients may be appropriate candidates for extended endocrine therapy (EET) by providing a genomic-based estimate of distant recurrence risk and endocrine responsiveness. BCI is a proprietary, gene expression-based reverse transcription polymerase chain reaction (RT-PCR) assay for patients with early-stage, HR+ breast cancer. BCI is differentiated by its ability to 1) prognosticate the cumulative (0-10 year) risk of distant metastatic recurrence as well as the risk of late (post 5- year) distant metastatic recurrence; and 2) predict the likelihood of benefit from EET.

BCI has established analytical and clinical validity and clinical utility for two distinct time points in the continuum of care to guide decisions on adjuvant systemic therapy: the decision with regard to addition of adjuvant chemotherapy at the time of diagnosis based on prognostication of overall cumulative 10-year distant recurrence risk and the decision of whether to extend endocrine therapy from five to ten years based on the prognostication of late (post-5 years after diagnosis) distant recurrence risk and prediction of likelihood of benefit from EET. The BCI clinical development program evaluated both components of BCI (prognostic and predictive) in multiple prospective, randomized controlled trial cohorts, and BCI is included in several clinical practice guidelines.

Support Proposed Coverage Criteria for BCI

Based on this extensive body of clinical evidence, Noridian has issued Proposed LCD DL37822, under which BCI will be covered for postmenopausal women with invasive breast cancer when the following criteria are met:

• Pathology reveals invasive carcinoma of the breast that is ER+ and/or PR+ and HER2-; and
• Patient has early-stage disease (T1-3, pN0, M0); and
• Patient is lymph node negative;
• Patient has no evidence of distant breast cancer metastasis (i.e., non-relapsed); and
• Test results will be used in determining treatment management of the patient for chemotherapy and/or extension of endocrine therapy.

Biotheranostics agrees with the proposed coverage criteria for BCI. Like Noridian, we believe that BCI is reasonable and necessary to address one of the most critical and urgent clinical needs in precision oncology for early-stage, hormone receptor-positive breast cancer – the genomic assessment of a patient’s individualized risk of late distant metastatic recurrence, and prediction of the likelihood of benefiting from EET. We encourage Noridian to finalize the coverage criteria for BCI without revisions.

Recommend Coding Update in Final LCD

We also wish to notify you of two coding updates that should be reflected in the final LCD. On October 11, 2017, Biotheranostics was informed by the AMA CPT Editorial Panel that a BCI code change application had been approved to add a new Category I Multianalyte Assay with

Algorithmic Analysis (MAAA) code 816X0 to report testing for BCI. This specific code, which has been finalized to 81518, will be assigned to BCI effective January 1, 2019. 81518 will replace 814792 as the correct code for BCI.

We respectfully request that the final LCD be updated to list 81518 in the coding information for BCI instead of 81479. This will avoid provider confusion and ensure that claims are payable once the final LCD becomes effective. We understand that under the new chapter 13 of the Medicare Program Integrity Manual, section 13.5.1, CPT codes will no longer be included in LCDs and instead will be placed in related billing and coding or policy articles. To the extent this policy is in effect when DL37822 is finalized, we request that the relevant article include 81518 instead of 81479.

Finally, as required by MolDX, we have been assigned Z-code identifier ZB68J for BCI. Please include the Z-code identifier in the LCD or article, as applicable.

Coordinate Retiring of Existing LCD with Effective Date of L37822

Lastly, Noridian J-E currently has in effect the LCD L36314 for BCI, which provides more limited coverage than proposed in DL37822. We are concerned that the continued effectiveness of L36314 after the effective date for L37822 could create confusion as to the applicable coverage criteria. As such, we respectfully request that Noridian synchronize the retirement of the existing LCD L36314 with the effective date of the final L37822.

Thank you for the comment. We will cover this test under the new CPT code, which means that it will be paid on a CLFS price once finalized. We are unable to put a Z-code into the LCD, since Z-codes are the intellectual property of Change Healthcare. Since Biotheranostics is the only lab billing for this test and knows the Z-code, we do not believe that this will be a problem.