Skip to Main Content
Centers for Medicare & Medicaid Services

View Public Comments for CED Public Solicitation

Haenlein, Kelly

January 17, 2012

Louis Jacques, MD
Coverage and Analysis Group
Centers for Medicare & Medicaid Services
Department of Health and Human Services
7500 Security Boulevard
Baltimore, MD 21244


Re: Coverage with Evidence Development Public Solicitation

Dr. Jacques:

Genentech welcomes the opportunity to submit comments on the Coverage with Evidence Development (CED) process in response to the November 7, 2011 public solicitation issued by the Centers for Medicare & Medicaid Services (CMS).1 Founded more than 30 years ago, Genentech is a leading biotechnology company that discovers, develops, manufactures, and commercializes medicines to treat patients with serious or life-threatening medical conditions. The company, a member of the Roche Group, has headquarters in South San Francisco, California. Our products are used by Americans of all ages, ethnicities, and income levels, many of whom are Medicare beneficiaries and are directly impacted by coverage decisions. We appreciate CMS’ public outreach on this important issue.

In the remainder of this letter, we will elaborate upon the following three topics to CMS:

  1. Application of CED to drugs and biologics;
  2. Defined circumstances for CED within the NCD process; and
  3. Process improvements to ensure an effective, transparent and accountable CED process.

I. Application of CED to drugs and biologics

As a state-of-the-art research organization, Genentech develops clinical evidence for evidence-based decision-making by the Food and Drug Administration (FDA), providers and patients. We support CMS’ interest in the appropriate generation and dissemination of evidence to help physicians make the most effective treatment decisions for their patients. We also support the underlying premise of CED in the national coverage determination (NCD) process, which we understand is to expand access to promising technologies and treatments for Medicare beneficiaries when additional evidence is needed. However, we wish to emphasize some important characteristics of drugs and biologics that have historically made them a rare subject of CED, which we contend should continue to be the case.

In the public solicitation for comments on CED, CMS states, “many new technologies are developed with insufficient attention to addressing the needs of the Medicare beneficiary population.”2 Although this statement might be true for some technologies, it is seldom true for drugs and biologics used by the Medicare population. These therapies are subject to a rigorous FDA approval process, and their approved prescribing information clearly indicates the population for which each therapy is approved and the data upon which that approval was granted. Genentech believes that CED is rarely appropriate for the FDA-approved uses of drugs and biologics that are approved for use in the Medicare population, including the disabled and patients older than age 65. Moreover, the Medicare statute’s definition of “drugs or biologics” requires that each drug or biological be included or approved for inclusion in the United States Pharmacopoeia or be “approved by the pharmacy and drug therapeutics committee (or equivalent committee) of the medical staff of the hospital furnishing such drugs and biologics for use in such hospital.”3 These requirements, combined with FDA approval, provide ample assurance that the therapy has been thoroughly reviewed by independent experts prior to coverage.

CED also should not be used for drugs and biologics that are used for “medically accepted indications,” as defined by the statute, or pursuant to longstanding Medicare guidance. Under the statute, “medically accepted indications” of drugs or biologics used in anti-cancer chemotherapeutic regimens include the FDA-approved uses as well as uses that are listed in certain compendia or are supported by peer-reviewed literature.4 Medicare also has long granted its contractors authority to determine that unlabeled uses of other drugs are “medically accepted” based on “the major drug compendia, authoritative medical literature and/or accepted standards of medical practice.”5 By using authoritative compendia and medical literature to define “medically accepted indications,” the statute and Medicare’s guidance protect beneficiaries’ timely access to drugs and biologics while also ensuring that Medicare’s coverage policies are truly evidence-based. In circumstances where the use of a drug or biologic qualifies as a medically accepted indication under these evidence-based standards, there is no need to restrict coverage to beneficiaries who are willing and able to enroll in a clinical study, and no justification for doing so.

Genentech invests heavily in research and development to discover and develop innovative treatments, conduct clinical trials to establish efficacy and safety, shepherd them through the rigorous FDA regulatory approval process, and ultimately manufacture them on a large scale to meet the needs of the patient population. The Roche Group’s research and development investment in 2010 reached $9.65 billion.6 Such investment is undertaken with reasonable assurance that upon FDA approval, beneficiaries of programs such as Medicare will have access to therapies as a covered treatment option. As CMS stated in its 2006 Guidance on CED,7 CED is a coverage policy tool that should be used infrequently in the NCD process, in the rare instance that data are not available upon which to make an ordinary coverage determination. CED should rarely apply to drugs or biologics for the reasons stated above. However, Genentech understands that there are circumstances when additional data collection by Medicare may be necessary for coverage of a product. The specific circumstances in which CED could be appropriate are outlined in the next section.

II. Defined Circumstances for CED within the NCD Process

Genentech suggests that CMS should employ CED when there is a clear opportunity to allow coverage for a promising item or service, and when the value of the information generated outweighs the cost of collecting the additional evidence and restrictions imposed on beneficiary access.

In general, Genentech believes CMS should initiate CED only as an alternative to otherwise limiting coverage. Specific circumstances under which we believe CED is appropriate include:

  • When the alternative is national non-coverage based on limited evidence but the direction of that evidence shows clinical benefit;
  • When there is considerable non-coverage at the local level, creating a de facto national non-coverage policy;
  • When the final NCD will be more restrictive than the current FDA-approved labeled indication, as signaled by the draft NCD; or
  • Prior to removing coverage for an item or service that was previously covered by Medicare.

Moreover, CED policies should continue to be issued only within the NCD process. We do not believe it is appropriate for CED to be implemented by local Medicare contractors due to challenges with carrying it out at this level, including small sample sizes from which to power studies adequately and limited resources. It could also lead to duplicative trials. Additionally, given that a CED decision may impact a class of products, it is also important that such a decision be made at the national level to ensure that all relevant stakeholders have the opportunity to participate in the comment process. The NCD process, with its statutorily defined public comment periods and process requirements, provides stakeholders with the opportunity to monitor and provide evidence to inform coverage policy making and that will impact them.

III. Process improvements to ensure an effective, transparent and accountable CED process

As CED is carried out in the future, Genentech suggests the following process refinements to the CED process to ensure the effectiveness of data collection efforts, accountability of the effort to stakeholders, and transparency to the public. We recommend that CMS clearly define/improve the CED process as follows:

  1. Conduct a gap analysis of ongoing or planned public and private evidence generation activities before applying CED

To justify the application of CED, CMS should demonstrate that a research gap exists that frustrates efforts by physicians and patients to identify the most medically appropriate therapy. Without determining that a research gap exists, CMS risks duplicating time and resource-intensive data collection efforts already ongoing within the industry. When issuing the decision memo calling for CED, we recommend that CMS include the results of the gap analysis and clearly outline the evidentiary gap that must be addressed through the generation of additional evidence.

CMS should also work directly with stakeholders via an open and transparent process to determine whether there are ongoing or planned private and/or public evidence generation activities, such as by product sponsors or FDA that will provide answers to well-defined research questions identified by CMS, or to determine the best methods for gathering the necessary evidence. This will ensure that the agency’s CED efforts complement and do not duplicate other efforts. This may best be accomplished by convening a multi-stakeholder Technical Expert Panel at the point CMS considers applying CED. Genentech suggests that in addition to the stakeholders, CMS publicly request nominations for the Technical Expert Panel. A Technical Expert Panel would clearly delineate ownership and accountability between the Agency and involved stakeholders and ensure that (if CMS ultimately adopts a CED policy); it is structured to generate the appropriate evidence needed to make a coverage decision. The Technical Expert Panel should also include stakeholders with experience in implementing data collection and rolling out such studies. We discuss the Panel’s role in the implementation of CED in the following section.

  1. Require that a clear implementation pathway for CED be established in the final decision, including a vetted study design, timeframes for reconsideration of evidence, and a sustainable funding mechanism.

The existing process fails to require that, as part of the NCD process for a CED decision, there be a clear pathway for ensuring that the CED is successfully implemented following the final decision. Such a pathway should include details on the study design, analysis plan, timeframes for reconsideration of the evidence, and the funding mechanism for the data collection and analysis. The Technical Expert Panel suggested above could serve a valuable role in ensuring that each of these factors is clearly addressed in the final decision. If these factors cannot be addressed in the NCD timeframe, the final decision should identify a reasonable timeline for when these points will be addressed and publicly posted.

  1. Study Design

The most successful cases of CED have been where the agency has articulated the key research questions that must be addressed and the essential components of the study design in the final decision memo. CMS’ 2010 decision on allogeneic hematopoietic stem cell transplantation is such an example. In the decision memo, CMS identified the key research questions, outcomes of interest for each question, and a list of standards that a study had to meet to be approved for CED.8 In other cases, such as the 2004 CED for anticancer chemotherapy for four colorectal cancer drugs, the research questions were not clearly identified.9 Of the nine National Cancer Institute clinical trials approved for CED in the decision, there was significant variation in the treatment regimen, primary endpoints, length of follow-up, treatment regimen, and patient population, making it unclear how CMS would use the evidence generated to inform its coverage policy.10 Therefore, Genentech requests that the new guidance address this important issue to avoid similar situations and to ensure that CED is appropriately designed and implemented to generate the evidence needed to answer well-defined questions. Genentech recommends that CMS engage in a continuing collaborative process to create such guidance, with further opportunities for public input.

Genentech appreciates that CMS provides stakeholders with the opportunity to meet with the agency to discuss clinical trial protocols and receive guidance outside of the NCD process. In that spirit, we encourage CMS to work closely with the relevant stakeholders to proactively clarify study designs and data collection requirements that will allow CMS to decide whether an item or service is reasonable and necessary. This will ensure that at the conclusion of the study the evidence will be sufficient to make a coverage determination. Genentech and our research partners conduct well-designed clinical trials and prospective observational studies, and have tremendous insight to offer. Genentech strongly encourages CMS to engage the relevant stakeholders in identifying appropriate studies for CED through the Technical Expert Panel process outlined above. This will allow CMS to benefit from the insights of all stakeholders and experts in an appropriately transparent process.

  1. Timeline

Because the existing CED process lacks clarity in terms of timelines, it creates uncertainty among stakeholders (including manufacturers, providers, and the public). Genentech requests that CMS clearly define a timeframe in which it will reconsider coverage, by evaluating evidence generated via CED, which timeframe is outlined in the final decision memo to ensure accountability. Genentech believes that this timeline may be different for various products or services, and should be part of the ongoing dialogue between CMS, stakeholders, and appropriate expert advisors, such as clinical epidemiologists and scientists. At the conclusion of the pre-determined timeframe for data collection, a rigorous analysis of the evidence, pursuant to a previously agreedupon statistical analysis plan, should be conducted to inform the new NCD. In specific instances, a defined extension of data collection may be appropriate if involved stakeholders agree that such an extension would generate the evidence needed to make the coverage decision. For example, if sufficient patients cannot be identified during the planned study timeframe, an extension would allow for generation of sufficient data upon which a coverage decision can be made while upholding scientific rigor and soundness of the analyses.

  1. Financing

The current lack of a designated funding source for CED efforts results in inconsistency, confusion, and most importantly, delay in the initiation of data collection efforts. The Medicare Payment Advisory Commission (MedPAC) discussed this in its July 2010 report to Congress as a significant shortcoming of the CED process.11 Genentech believes that the financing and management of CED may be different for various items or services, and should be part of the dialogue between CMS and the manufacturer. The resulting funding arrangements for data collection and analysis should be explicitly stated up front in the final decision memo, before the data collection begins.

Another important aspect of CED financing involves the additional time and burden imposed on providers that are ultimately responsible for the submission of evidence. Providers’ ability to afford this added responsibility, and willingness to participate, affects both the viability of a strong data collection effort and patient access to important treatments (as only patients participating in the study are covered). Consequently, CMS should ensure that providers participating in data collection related to CED are reimbursed for the extra time required for data submission, either through existing channels or by a new CMS-proposed mechanism to achieve this goal.

  1. Maintain local coverage during the CED process

To safeguard patient access, we also urge CMS to maintain local coverage as an option during the CED process if local coverage is already available. Local coverage should not be disrupted, as many patients may not be eligible to participate in the studies initiated through CED, either due to exclusion criteria or inability to travel to the study sites. CMS should clearly state in the final decision memo that local contractors have discretion to continue to cover an item or service outside of the CED requirements. To avoid confusion, CMS should also restate this in its implementation guidance and any instructions sent to contractors. Additionally, CMS should clarify that local contractors do not have the authority to conduct CED; rather, it is a coverage tool to be used exclusively through the NCD process. Since local and regional Medicare Advantage (MA) plans must generally follow the coverage offered in their plans’ jurisdictions, we would appreciate CMS clarifying the issue of MA coverage when Part B coverage is contingent upon patient participation in a CED-sanctioned registry or study.

Again, Genentech appreciates CMS’ outreach to the public in advance of drafting new guidance for the CED process and welcomes the opportunity to work closely with CMS to define a process that will improve patient access to important treatment options and ensure that high quality care is delivered to Medicare beneficiaries. If we can be of any assistance, or if you have any questions on the above recommendations, please contact Stephanie Dyson, Sr. Director, Public Policy and Reimbursement, at or (202) 296-7272.


Evan L. Morris, Esq.
Vice President, Government Affairs
Genentech, Inc.

1 Centers for Medicare & Medicaid Services. CED Public Solicitation. Issued November 7, 2011 ( aspx?MCDId=8&McdName=National+Coverage+Determinations+with+Data+Collection+as+a+Condi tion+of+Coverage%3a+Coverage+with+Evidence+Development&mcdtypename=Guidance+Documents&M CDIndexType=1&bc=BAAIAAAAAAAA&).
2 CED Public Solicitation.
3 Social Security Act (SSA) § 1861(t)(1). CMS guidance interpreting this provision has essentially updated the list of compendia in the statute and references USP-Drug Information, USP-National Formulary, and the American Dental Association Guide to Dental Therapeutics. Medicare Benefit Policy Manual Chapter 15, § 50.1.
4 SSA § 1861(t)(2)(B).
5 Medicare Benefit Policy Manual, ch. 15, § 50.4.2.
6 Sanford C. Bernstein & Co., LLC Global Pharmaceuticals: Bernstein Monthly Drug Pipeline Report for US & European Drug Companies - November 2011.
7 CMS, “NCDs with Data Collection as a Condition of Coverage: Coverage with Evidence Development,” (Baltimore: CMS, July 2006).
8 CMS. Decision Memo for Allogeneic Hematopoietic Stem Cell Transplantation (HSCT) for Myelodysplastic Syndrome. (CAG-00415N, August 2010)
9 Carino, et al. (2006). “Medicare’s Coverage of Colorectal Cancer Drugs: A Case Study in Evidence Development And Policy.” Health Affairs, 25 (5), 1231-1239.
10 Ibid.
11MedPAC Report to Congress (Ch. 1, 2010)