Lumbar Artificial Disc Replacement (LADR)

June 24, 2007

Jyme Schafer, MD, MPH
Lead Medical Officer, Division of Medical and Surgical Services
Centers for Medicare & Medicaid Services
Office of Clinical Standards and Quality
Coverage and Analysis Group
7500 Security Boulevard
Baltimore, MD 21244-1850

Re: Lumbar Artificial Disc Replacement (LADR) CAG-00292R

Dear Dr. Schafer:

Medtronic's Spinal and Biologics business is the leader in the development of spinal therapies throughout the world, and the evolution of our therapies over the past ten years has improved the quality of life for thousands of patients with spine conditions. Medtronic has a long history of working with CMS to ensure appropriate patient access to medical technologies and we have engaged with both the coverage and payment groups on numerous issues affecting spine therapies, including coding for spinal procedures; new-technology add-on payments and DRG reassignments; and coverage decisions and MCACs involving spine fusion and other spinal interventions. We appreciate the opportunity to comment on the first reconsideration of the National Coverage Analysis (NCA) for lumbar artificial disc replacement (CAG-00292R).

Medtronic remains committed to improving patient care by conducting clinical research that adds value and knowledge to the overall body of science. Though Medtronic's artificial lumbar disc technology 'known as MAVERICK™ Total Disc Replacement' has not yet been approved by the FDA, the proposed decision memorandum references the device and its IDE study in a number of instances. We would like to take the opportunity to clarify some points about MAVERICK™ and our clinical trial, as well as address some larger issues raised by the proposed decision memorandum.

While we agree with the proposed decision's continuation of local Medicare coverage of lumbar artificial disc replacement for Medicare beneficiaries under the age of 60, we are disappointed that the proposed policy would exclude coverage of lumbar artificial disc replacement technologies for beneficiaries age 60 and over.

MAVERICK™ Total Disc Replacement

MAVERICK™ is a total disc replacement device for the lumbar spine. The pre-market approval (PMA) application for MAVERICK™ is currently under FDA review. The device has a two piece, metal-on-metal, posterior center of rotation design that is intended to restore and maintain disc height while also providing stability, motion preservation, and relief of pain. The posterior center of rotation offered by MAVERICK™ most naturally replicates the spinal kinematics.

The IDE clinical trial on MAVERICK™ was a multi-center, prospective, randomized, controlled design conducted to evaluate the disc for treatment of lumbar degenerative disc disease (DDD) at one level from L4 to S1. The total sample size in the study was 577 patients, with a distribution of 405 investigational and 172 control patients, treated at 31 investigational sites. The primary objective was to evaluate the overall success rate associated with the use of the investigational implant. Overall success is defined as meeting all of the following conditions: 1) pain/disability success (as defined by a '15-point Oswestry Disability Index improvement compared to preoperative); 2) neurological status success; 3) disc height success; 4) no serious adverse event classified as 'implant associated' or 'implant/surgical procedure associated'; and 5) no additional surgical procedure classified as a 'failure'. If the overall success rate for the investigational treatment is statistically noninferior to the overall success rate for the control group at 24 months following surgery, the disc will be considered to be safe and effective. In addition, if the investigational treatment is found to be statistically noninferior to the control treatment, statistical superiority of the investigational treatment to the control treatment will also be evaluated.

The proposed decision memo stated that CMS received an abstract presenting single-site data from the MAVERICK™ Total Disc Replacement randomized clinical trial. While the information presented in the abstract is accurate regarding that site, certain points about the study in general' and about CMS' characterization of the study ' need to be clarified.

First, the age range for enrollment in the MAVERICK trial at large was 18 to 70 years. Though no one over the age of 55 was included in the single-site data sample submitted to CMS, the protocol for the MAVERICK trial did in fact extend to individuals over the age of 60. While there was never expected to be a large number of patients from this age group who would qualify for the study, the patients over the age of 60 who did meet the study's inclusion criteria were included. This is different than the CHARITE™ and ProDisccÂ-L clinical trials, which did not enroll any patients over the age of 60.

Second, patients in the control arm of the MAVERICK™ study received an anterior lumbar interbody fusion procedure with the INFUSE™ Bone Graft/LT-CAGE® Lumbar Tapered Fusion Device, a technology that Medtronic and many others in the spine clinical community consider to be the standard of care for the patient population and that has consistently demonstrated excellent clinical outcomes in relieving pain, improving function, and improving quality of life. This cohort data from the INFUSE™/LT-CAGE® PMA in 2002 represent the best randomized, controlled trial data available specifically for the diagnosis of DDD. The Oswestry Disability Index scores from that IDE study are better than the scores reported for either of the lumbar artificial discs reviewed by CMS.

Thus, the clinical trial for MAVERICK included a broader age-range of patients and, we believe, a tougher standard of comparison than the CHARITE™ and ProDisccÂ-L clinical trials. Though Medtronic believes the data from the MAVERICK™ clinical trial are favorable, we are unable to share the full results at this time because they are currently undergoing review for FDA regulatory approval and they await publication in a U.S. peer-reviewed journal. We intend to notify and meet with you when the data are published or fully released publicly.

Comparison to Fusion

CMS expressed concern that the full clinical impact of lumbar disc arthroplasty technologies could not be assessed by the clinical trials that have been conducted thus far because thecontrol groups in those trials used lumbar spinal fusion as the comparator. CMS believes the effectiveness of spine fusion for DDD is not well-established in comparison to non-surgical treatment and therefore cannot be used as a benchmark to determine the clinical effectiveness of lumbar artificial disc replacement. To support this position in the proposed decision memo, CMS referenced the November 30, 2006 MCAC meeting and the draft AHRQ technology assessment (TA) prepared for that meeting on spinal fusion.

Medtronic believes the references to the AHRQ technology assessment on spine fusion are problematic. First, a coalition of spine professional societies has raised several issues with the draft of the technology assessment and its presentation at the MCAC meeting. The societies have commented on these issues directly to AHRQ and believe a revision of the TA is forthcoming. The societies' concerns included: 1) conflicting conclusions within the TA executive summary, the TA manuscript body, and the information presented at the MCAC meeting; 2) the limited diagnosis targeted for the MCAC review; and 3) the patient entry criteria for the European randomized clinical trials (RCTs) referenced in the TA. Based on the concerns raised to AHRQ about the conclusions in the technology assessment, Medtronic believes CMS should reserve judgment on the value of spinal fusion until a final version of the TA is issued reflecting AHRQ's response to the concerns raised by the spine societies.

Second, as described above, the control intervention in the MAVERICK trial was spine fusion using INFUSE™ Bone Graft/LT-CAGE® Lumbar Tapered Fusion Device. This type of fusion surgery has the best RCT findings of any fusion procedure studied to date.

Noninferiority Study

In the proposed decision memo, CMS expressed concern that the artificial disc studies were conducted under a noninferiority trial design. CMS stated that it considers noninferiority trials only to be an option 'when the investigational agent is believed to have comparable efficacy to the control, but has other advantages, for example fewer adverse events or less cost.' The proposed memorandum also cited the BCBS TEC TA entitled Artificial Vertebral Disc Replacement to conclude that at this time, no such advantage has been demonstrated for LADR.

Again, safety and efficacy of the MAVERICK™ disc were established using the highest scientific threshold of INFUSE™ Bone Graft/LT-CAGE® Lumbar Tapered Fusion Device as the comparator. Advanced fusion technologies, such as fusion with this device, have demonstrated through clinical trials effectiveness in relieving pain, improving function, and improving quality of life. Showing non-inferiority to such a control is appropriate not only for the regulatory approval, but also for establishing clinical evidence of improved patient outcomes with the MAVERICK™ disc.

The statistical analysis of this study was based on both non-inferiority and superiority evaluations. Nonetheless, although the primary objective of the MAVERICK™ RCT was to establish non-inferiority, superiority was also predefined in the protocol as the secondary objective. It is methodologically appropriate to examine and claim superiority (if existing), after non-inferiority is established.

Conclusion

Medtronic believes that lumbar disc arthroplasty is an effective clinical option for certain, well-selected Medicare patients. Patient-specific conditions and circumstances, in addition to the clinical evidence supporting use, should be the most important factors in determining which treatments are appropriate and most likely to provide the greatest clinical benefit for an individual beneficiary.

We urge CMS to use caution when initiating NCDs for evolving technologies early in the development process as these coverage restrictions place a huge limitation on the amount of clinical evidence development and vetting that naturally occurs for a new product within the physician community. We believe that CMS should allow local Medicare contractors the discretion to develop appropriate local policies, based on the indications for use, evidence supporting the effectiveness of these devices, and other appropriate criteria they develop in collaboration with trained physicians and other healthcare providers in their jurisdictions.

We also urge CMS not to view all spinal fusion cases as having the singular diagnosis of degenerative disc disease. Several presenters at the November 30, 2006 MCAC meeting pointed out that very rarely do these patients only have DDD, instead experiencing a cascade of issues and diagnoses. These additional diagnoses can cause a multitude of problems for the patient, each requiring a different treatment plan in order to provide relief.

Again, we thank you for the opportunity to comment and look forward to continued dialogue with CMS on the clinical outcomes offered by MAVERICK™ Total Disc Replacement. If you have any questions, please contact me at (901) 396-3133.

Sincerely, Michael McCormack
Reimbursement Planning

Less

June 21,2007
Steve Phurrough, M.D.
CMS
7500 Security Blvd.
Baltimore, MD 21244

Dear Dr. Phurrough,

This letter is written of behalf of the Professional Society Coalition on Lumbar Fusion,representing the North American Spine Society (NASS), the American Association ofNeurological Surgeons (AANS), the American Academy of Orthopaedic Surgeons (AAOS), theCongress of Neurological Surgeons (CNS) and the Scoliosis Research Society (SRS). We wish to express our concern with regard to the proposed Coverage Decision Memorandum for lumbarartificial disc replacement issued May 25,2007. This document contains a number of referencesto the November 2006 Medicare Coverage Advisory Committee (MCAC) meeting which we believe inaccurately represent the data that was presented as well as the discussion that occurredat that meeting. In particular, we are concerned about the use of the draft Technology Assessment, generated prior to the MCAC meeting, and made available for public comment earlier this year.

In a previous communication (Attachment A), we expressed our concern that the executive summary and conclusions of the draft Technology Assessment did not accurately reflect the evidence presented within the document. Further, the conclusions are not consistent with thepresentation given by the Technology Assessment authors at the MCAC meeting. We are concerned that propagation of this draft document, as a reference in the lumbar artificial disc NCD,would be contrary to the development of an improved evidence-based approach for evaluating lumbar fusion procedures.

In addition, we believe that consideration of the Technology Assessment or MCAC panel voteoutside of the specific MCAC context could lead to significant misinterpretation. A primaryreason that the MCAC committee felt that the evidence for fusion for low back pain in theMedicare population was weak was because fusion is generally not performed for low back painin the elderly population (it is performed as an adjunct to decompression in patients with stenosis or to correct deformity). Since the procedure in question is rarely if ever performed, studies are lacking and the panel was forced to conclude that there was little evidence to support the practice of fusion for low back pain in the Medicare population. While true, this statement is irrelevant and potentially misleading for the non-Medicare population of patients who do undergo fusion for low back pain.

Attachment A
Chuck Shih, MHS
Center for Outcomes and Evidence
Agency for Healthcare Research and Qual~ty
540 Gaither Road
Rockville, MD 20850

Re: Society Comments on AHRQ Spinal Fusion Report

On behalf of the medical societies listed below and our approximately 28,OOO members, we would like to thank the AHRQ for the opportunity to comment on the Technology Assessment prepared for the November 30 MCAC meeting on fusions. We appreciate the effort put forth on the tech assessment and the guiding principle of providing the best care. We feel that we all have the same goals.

In terms of process, the coalition of spine societies is concerned about the possible disconnect between the CMS process and the AHRQ process. We recognize that CMS commissioned the tech assessment on spinal fusion for degenerative disc disease in the over-65 population for the MCAC panel and worked inconjunction with AHRQ.In terms of the substance of the tech assessment, we have concerns with the following:

Executive Summary: The executive summary as currently written does not accurately represent the manuscript body of the tech assessment content, nor the information as presented by Douglas C McG-ory,MD, MHS, who provided oral comments at the MCAC meeting. We feel they gave a more logical interpretation of the information than the written material alone.

Transcript: Given the significant time and effort involved in the MCAC meeting, it would be a disservice not to have the transcript containing the authors' comments available along with the tech assessment as the final report: as accepted by AHRQ. The potential for significant disparities between the final tech assessmentand the CMS MCAC meeting minutes could lead to confusion and misunderstandings between CMS andAHRQ. This could be averted by incorporating that information now.

Diagnosis: Perhaps the most important issue is the imprecision of the diagnosis as currently applied in ICD-9-CM codtng. Rarely is this condition, lumbar degenerative &c disease (755.221, seen in isolation in thispatient population. Consequently, there is great heterogeneity in the patient population, treatment regimens and outcomes. When this much heterogeneity is present, the generalizability of narrowly defined RCI's is limited. In the diagnosis where there is the hlghest concordance, which is acquired spondylolisthesis (738.4 (ie., degenerative spondylolisthesis)), the outcomes are quite good with surgical treatment. In addition, it appears that the effect size of surgical treatment is as good as or perhaps better in the over-65 patientpopulation than the under-65 population.

Nonsurgical Treatment and Surgical Treatment Equality: There is an inherent weakness of the tech assessment because of the limitations of the existing literature. The tech assessment a priori presumes the value of nonsurgical treatment without reviewing the nonsurgical literature, yet surgery seems to be held to anasymmetrically higher level of evidence scrutiny compared to nonsurgical treatment. There are few if any nonsurgical versus placebo trials for this diagnosis and none of them have two- year follow up data. Therefore, all would be excluded from the tech assessment indicating no evidence or support for this form of treatment. The concept of minimum clinically important difference was inappropriately applied in the review. Specifically, comparison of group means and loolung for an MCID level of change is not a recognized or validated use of this concept. Points made about applicability of evidence selected in the tech assessment and the generalizability to the heterogeneous population being treated are valid and perhaps could be enhanced further in additional submissions.

RCTs: The European RCTs were disproportionately weighted in the tech assessment simply because they were RCTs. However, we feel that the entry criteria of these studies were not the typical surgical criteria used in the United States. We understand more weight was attributed to the RCTs, however the RCT's in the review were far from ideal and were all conducted outside of North America As such, the generalizability ofthese studies is not known. In terms of conclusions from the RCTs, in spite of the differences in entry criteria, the primary outcome variables in the RCTs showed surgeryto be beneficial, even when an "intent totreat" analysis was used instead of an "as mated" analysis. Further, the nonsurgical regimen of the Fairbanks study, a prolonged inpatient stay, is not avdable in a US setting. At best, RCT's demonstrate efficacy butpractice studies are required to demonstrate treatment effectiveness. These studies are completely lacking to date. Such studies are challenging to accomplish and may require future dedicated funding if they are to be correctly done and used for health care policy decision-making.

Tech Assessment Inclusion Criteria: The minimum cutoff of 50 patients in the tech assessment eliminated the study by Schwender et al [Minimally invasive transforaminal lumbar interbody fusion (TLIF): technical feasibility and initial results. J Spinal Disord Tech. 2005 Feb;l8 Supp1:Sl-6. PMID: 156997931. This studyshowed the greatest effect size of treatment data using the current state of the art technology. This study included 49 patients in a prospective, consecutive cohort multi-center series using prospectively collected patient reported outcomes to assess the treatment effect.

Patient Management
Typically, surgical options are offered to patients only after nonsurgical treatment has failed Management of patients who have failed nonsurgical management and the level of evidence supporting persistence in fdure has to be addressed as well.

In summary, we stand ready to assist HHS in identlfylng and discussing issues in the spine field We seek the best care for our patients and we believe a partnership with HHS can only benefit our mutual understanding about what is best for the patients we serve. We appreciate the chance to participate in this continuingdialogue.

Less

I may have a somewhat unique perspective on this issue, as i am former bioengineering reviewer for the FDA, have been an Army orthopaedic spine surgeon for many years, and am now in acedemia. I have performed approximately two doxen Charite LADR's since its approval in the US, on both soldiers and civilians. My experience of my patients' results roughly mirrors that of the published US trials. I think this remains a very reasonable option for a select number of patients with discogenic back pain and no facet arthrosis. It is extremly unlikely that patients over 60 years old will fall into this category. Thus, I would tend to agree with this coverage decision.

Less