Skip to Main Content
Centers for Medicare & Medicaid Services

View Public Comments for CED Public Solicitation

Fleury, Mark
Science Policy and Legislative Affairs Committee
American Association for Cancer Research (AACR)

January 20, 2012

Louis Jacques, MD
Director, Coverage and Analysis Group
Office of Clinical Standards & Quality
Centers for Medicare and Medicaid Services
7500 Security Blvd
Baltimore, Maryland 21244

The American Association for Cancer Research (AACR) is pleased to offer comments regarding the Centers for Medicare and Medicaid's (CMS) Coverage with Evidence Development (CED) policy. As the world's largest and oldest organization dedicated to cancer research, the AACR has special interest in this policy, which has the potential to impact future research leading to innovation and improved patient care.

Oncology is perhaps one of the most quickly evolving medical fields, as our knowledge of the biology underlying the hundreds of different types of cancer continues to provide new opportunities to tailor treatment to individuals. Cancer is diagnosed, treated and monitored with a variety of diagnostics, devices, procedures and medications that undergo continual refinement through the research performed by AACR members. As one indication of this rapid innovation, seven of the 35 new drugs approved in 2011 by the FDA were for cancer, more than any other disease area.

Demographics indicate that in the U.S. the population over 65 years of age in other words, the Medicare population is set to swell in the coming decades. As a disease mainly of older individuals, cancer incidence is expected to rise with the aging of the population. Therefore, at no time in the history of the country has it been more important to thoroughly understand which interventions are effective in preventing, arresting, or curing cancer. A properly designed CED policy should have the effect of more and better research on the Medicare-aged population and the specific health issues that this group faces.

In the field of cancer, the most likely candidates for CED are procedures and in-vitro diagnostics, as there is an existing statutory process governing reimbursement for off-label drug use, and the number of implantable devices used in oncology is somewhat lower than in other fields. The AACR feels that a properly designed CED process has the potential to expand and accelerate access to promising medical interventions, and that it can also be a way to provide crucial support to late-stage research. Care must be taken, however, as a poorly designed CED process risks the prospect of restricting access to promising therapies or presenting hurdles that could discourage innovation and translational research. Each year CMS issues approximately 10 to 15 National Coverage Determinations (NCDs), and since 2006 only four of those have resulted in a decision requiring further evidence development (CED). If CMS intends to significantly increase the usage of NCDs or of CED, it is important that guidance clearly spells out the interventions targeted and the process of review in order to avoid uncertainty can stifle the translation of important discoveries into usable therapies. This guidance should further ensure that evaluation of new interventions occurs early in the product cycle rather than restricting coverage after wide diffusion has already begun.

General principles that the AACR feels should be followed to help ensure that the CED policy improves access to novel interventions and provides much-needed evidence are:

  1. CED should be used judiciously and emanate from clear guidelines
    1. Selection criteria for the types of interventions subject to CED should be proactively developed in a transparent manner with input from all stakeholders so that the likelihood for any intervention to be subject to CED decision is well understood in advance.
    2. Interventions considered for CED should be amenable to studies with clear and measurable indicators.

  2. The collection and evaluation of additional evidence through CED should be timely and flexible, but scientifically rigorous.
    1. Study endpoints and goals should be clearly defined and communicated.
    2. The half-life of any technology being considered for CED should factor into both the designed duration of any projected CED study as well as the decision whether to employ CED at all.
    3. Any scientifically-valid research method should be considered (not just RCTs and Registries).
    4. Valid evidence obtained outside of CED-specified studies should be considered, and, if sufficient to render a determination, should result in termination of the CED process regardless of the progress of any studies in the original CED determination.
    5. Secondary research should be encouraged using the data from the CED-specified studies
    6. Clinical trials should be registered in and adhere to the standards associated with NIH-sponsored trials, including:
      1. Following the Common Rule
      2. Ensuring patient confidentiality
      3. Promotion of a single IRB
    7. CMS should also give careful consideration to patient access to CED therapies for patients unwilling or unable to consent to research (cognitive impairement or unwillingness to provide data for research.)

  3. Clear jurisdictions and close communication should be developed between federal regulatory and research agencies.
    1. The CED process should not duplicate FDA efforts.
    2. Close cooperation should be developed between FDA, NIH, AHRQ and CMS on research efforts to maximize the value that research performed for one agency has for the other agencies, especially the inclusion of patients and issues relevant to Medicare, where feasible. *See case study in reference 1 for an example of poor coordination.
    3. Conditional coverage decisions requiring additional data collection should be accompanied by infrastructure support from federal research agencies so that additional data burdens are not prohibitive to small innovators.
    4. CMS should be clear and consistent about their policies with regard to how a drug or device approval from FDA relates to coverage under Medicare.
  4. Transparency and stakeholder involvement should be built into the CED process to ensure legitimacy.
    1. Stakeholders should be involved in prioritization of topics and design of studies.
    2. Opportunities for public comment should occur at multiple stages of individual NCD/CED decisions.
    3. While stakeholders should be a part of issue prioritization and study design, the governance and evaluation of studies should protect against any conflict of interest that might skew findings.
    4. As has been the case in the past, only data published in peer-reviewed journals should be considered in coverage determinations in order to promote the full dissemination of findings.

The CED process holds the promise of providing missing data that are needed for proper evaluation of new interventions. The evidence upon which CMS based its coverage reviews was only found to have achieved a rating of “good” in 15% of the cases from 1999-2007 according to independent analysis (2). Clearly many important policy decisions are being made on weak or insufficient data, and CED represents an opportunity for CMS to address this shortfall through improved research and collaboration, which are central tenets of the AACR's mission to prevent and cure cancer.

In addition to the comments offered above, the AACR stands ready to provide any further assistance to CMS as new CED guidance is developed. If you have questions, you may feel free to contact the AACR through Mark Fleury, the associate director for science policy. He can be reached at 215-446-7147 or

Thank you for your consideration of AACR's comments.


William S. Dalton, M.D.,
Chairperson, AACR Science Policy and Legislative Affairs Committee

Margaret Foti, Ph.D., M.D. (h.c.),
Chief Executive Officer, AACR

(1) Miller R. Does coverage of renal intervention hinder trials? Stakeholders disagree. Health News Daily. 2007. (2) “Medicare’s National Coverage Decisions for Technologies, 1999-2007,” P. Neumann, M. Kamae, and J. Palmer, Health Affairs, 27, no.6(2008), pp. 1620-1631.