September 25, 2009

Letter Requesting a Negative National Coverage Decision for the Menaflex Collagen Scaffold Medical Device

Director, Coverage and Analysis Group
Centers for Medicare and Medicaid Services
7500 Security Blvd.
Baltimore, MD 21244

We are writing to request that the Center for Medicare & Medicaid Services (CMS) issue a National Coverage Determination (NCD) that would deny Medicare reimbursement for ReGen Biologics’ Menaflex Collagen Scaffold (MCS).  This product is marketed for the reinforcement and repair of the medial meniscus in patients who are undergoing a partial meniscectomy (cartilage removal) during arthroscopy for a damaged medial meniscus.  The implant is intended to function as a lattice for regrowth of new tissue as the device is assimilated and resorbed over time.  In 2007, Medicare beneficiaries underwent over 100,000 partial meniscectomy procedures at a cost of roughly $550 per procedure.[1]

Although the device was cleared for this purpose by the Food and Drug Administration (FDA) on December 18, 2008,[2] an internal FDA report made public yesterday demonstrates that the clearance process was highly irregular and contaminated by congressional intervention, unusual involvement of the former FDA commissioner and marked violations of customary FDA advisory committee processes.[3] Indeed, the agency is now reviewing whether the device should have been approved at all.  At a minimum, therefore, we ask that you do not issue a positive NCD until this second review is complete.

Moreover, even if the FDA were ultimately to conclude that the device should have been cleared, CMS has stated that such clearance does not automatically guarantee reimbursement under the Medicare program.  Indeed, while under the Food, Drug, and Cosmetic Act a device must demonstrate a “reasonable assurance of … safety and effectiveness”[4] before it can be marketed,[*] the Social Security Act provides for reimbursement under Medicare only if the device is “reasonable and necessary.”[5] As CMS has made clear on multiple occasions, while it typically defers to the FDA’s determinations of safety and efficacy,[†] it determines eligibility for reimbursement under a different, more restrictive standard:

Whereas the FDA must determine that a product is safe and effective as a condition of approval, CMS must determine that the product is reasonable and necessary as a condition of coverage under section 1862(a)(1)(A) of the [Social Security] Act. CMS adopts FDA determinations of safety and effectiveness, and CMS evaluates whether or not the product is reasonable and necessary for the Medicare population. Although an FDA-regulated product must receive FDA approval or clearance … for at least one indication to be eligible for Medicare coverage, … FDA approval/clearance alone does not generally entitle that device to coverage.[6]

As we shall demonstrate, the device has never met the FDA’s standard for approval, much less CMS’s more rigorous one.

Randomized Clinical Trial

Expecting to directly demonstrate the safety and effectiveness of the MCS, ReGen designed a two-year, multicenter randomized controlled trial (RCT) comparing partial meniscectomy with MCS implantation to partial meniscectomy alone (the standard of care).[7],[‡] In testimony before the Orthopedics and Rehabilitation Devices Panel of the FDA on November 14, 2008, we outlined the inadequacies in the data supporting the safety and effectiveness of the MCS, and urged the panel to not endorse the device.  Here, we summarize those data and incorporate the entire testimony by reference.[8]

In this study, 311 patients 18-60 years old were randomized and stratified a priori into acute and chronic arms, based on the number of previous meniscal procedures.[§] There were three prespecified primary clinical endpoints assessed at two years (visual analog scale pain score, Lysholm knee score and patient’s self-assessment), and three prespecificied surrogate outcomes assessed at one year (arthroscopic, histologic and radiographic improvement).  In July 2008, the findings from this RCT were reported in the Journal of Bone and Joint Surgery (JBJS).[9],[**]  These results are presented in the Table. 

Primary clinical outcomes for the ReGen Menaflex [10],[11]

		Visual analog pain scalea– meanchange (% change)		Lysholm knee scoreb– mean change (% change)		Patient’s self-assessmentc– mean change (% change)
		Control	MCS	MCS	Control	MCS	Control
2 yrs	Acute	n/a	n/a	26 (41)	32 (53)	n/a	n/a
	Chronic	-30 (77)	-27 (73)	19 (30)	23 (41)	n/a	n/a
5 yrs	Acute	-21 (78)	-16 (76)	26 (41)	28 (48)	0.9 (36)	1.1 (41)
	Chronic	-18 (46)	-18 (49)	16 (25)	22 (39)	0.7 (27)	0.9 (30)

No p-values were reported in JBJS publication; however, the FDA-calculated p-values for the chronic-arm patients at two years are 0.3932, 0.5587 and 0.889, respectively.[12] 

n/a: not able to calculate

a: Measured on a scale of 0-100, with 100 indicating the worst possible pain.  Success was defined as at least 20% improvement compared to pre-operative score

b: Measured on a scale of 0-100, with 100 representing the best possible knee function.  If pre-operative score was <80, success was defined as at least 20% improvement compared to pre-operative score. If pre-operative score was greater than or equal to 80, success was a greater than or equal to 95 final score.

c: Measured on a scale of 1-4.  1=normal; 2=nearly normal; 3=abnormal; 4=severely abnormal. Success was defined as improvement of at least one grade from pre-operative score, or no decrement if pre-operative score was “normal” or “nearly normal.”

There was no statistical difference between the MCS-implanted and control groups for any of the three primary clinical outcomes at two or five years in the acute and chronic groups.  Thus, the MCS offers no clinical benefit to patients over partial meniscectomy alone.  Even the authors of the paper (the first author is a ReGen executive) acknowledged that “it may be that the collagen meniscus does not provide any positive benefits over partial meniscectomy in acutely injured patients.”[13]  Moreover, the findings of the trial are not generalizable to the Medicare population because patients 60 years and older were not included in the trial.

The surrogate outcomes were difficult to interpret because of several methodological problems, including lack of control and blinding for both arthroscopic and histologic outcomes.[14]  In addition, details of the radiographic outcomes were not reported in the JBJS article because of variable radiographic quality between study sites.[15]

FDA’s clearance of the device was instead based, in large part, on performance measures such as suture retention strength and tensile strength.  These outcomes are less important than the clinical endpoints generated in the randomized, controlled trial.  And, even if the FDA found them convincing or felt that their review processes required them to be considered, CMS should pay them little heed.  The fact is that the “bell” of the randomized, controlled trial has rung, indicating no clinical benefit, and it cannot be unrung.  

Company-Claimed Benefits

Although the RCT failed to demonstrate efficacy on any primary clinical endpoint, the JBJS authors put forth two clinical findings in the chronic subset of patients as evidence of effectiveness.  The FDA reviewing officers refuted both claims. 

The first finding, a statistically significant improvement in the Tegner Index (TI) at five years (42% versus 29%, p=0.02), is of questionable relevance.  TI has not been reported in the medical literature previously.[16]  Moreover, it appears to be a post hoc mathematical manipulation of a prespecified secondary outcome in the RCT, the Tegner Activity Score.  According to the FDA, the Tegner Activity Score was negative at two years,[17] a detail omitted from the JBJS publication.  Regardless, because of an inflated risk of false-positive results, post hoc observations should always be treated with skepticism,[18] especially when they are generated from negative trials.[19]

Second, the authors reported a statistically lower reoperation rate for MCS-implanted patients at five years (9.5% versus 22.7%; p=0.04).[20]  However, reoperation rate was not a prespecified primary or secondary outcome.   Moreover, ReGen narrowly defined reoperation to exclude important procedures.  The study protocol required relook arthroscopy at 12 months for MCS-implanted patients only.  The company excluded any additional procedures performed during this relook arthroscopy.[21]  For example, additional meniscus excised during the relook arthroscopy was not counted as a reoperation.  However, amongst control group patients who did not undergo relook arthroscopies, such a procedure would have counted as a reoperation, potentially biasing the results in the MCS’ favor.  Using a more conservative definition that included these not-per-protocol procedures if they were also accompanied by symptoms, there was no difference in the reoperation rate between the two groups (32.0% versus 30.3%).[22]

Finally, the safety of the MCS cannot be established based upon the information from the RCT.  Serious adverse events[††] occurred more frequently in the MCS-implanted group than in the control group. In addition, the MCS had to be surgically removed six times during the trial for reasons ranging from mechanical failure to infection.[23]  Even though the difference in adverse events did not reach statistical significance, no risk is acceptable for devices that are not proven to be effective. 

In summary, based on a gold-standard randomized controlled trial, we are left with a device that is no more effective than the standard of care (partial meniscectomy alone). The trial failed to achieve statistical significance on all of its primary clinical endpoints and the primary surrogate endpoints were either based on uncontrolled, unblinded assessments or rejected by the sponsor itself for lack of data quality.  Furthermore, the two findings that the sponsor asserts as evidence of superiority were convincingly refuted by the FDA reviewing officers.

Summary

With Medicare beneficiaries undergoing over 100,000 partial meniscectomy procedures yearly between 2005 and 2007, at a cost of almost $60 million per year, partial meniscectomies were among the top 200 most costly CMS expenditures.[24],[25],[26]  Reimbursing for the MCS would only add to this cost without adding any clinical benefit.  Just as a physician would be hard-pressed to justify implanting a device not proven to have any clinical benefit, CMS should be hard-pressed to justify reimbursing for it.  We therefore urge CMS to deny reimbursement for this ineffective device.

Sincerely,

Jonas Hines
Research Associate

Peter Lurie, M.D., M.P.H.
Deputy Director
Health Research Group at Public Citizen

---