April 19, 2017
Food and Drug Administration
5630 Fishers Lane Rm 1061
Rockville Maryland 20852
RE: Manufacturer Communications Regarding Unapproved Uses of Approved or Cleared Medical Products
Submitted electronically via: www.regulations.gov FDA-2016-N-1149
Dear Sir or Madam:
Thank you for the opportunity to provide input on issues related to Manufacturer Communications Regarding Unapproved Uses of Approved or Cleared Medical Products. At the National Pharmaceutical Council (NPC), we believe it is important that all stakeholders work together to understand how treatments work in the real world. This evidence is central to guiding population health decision-makers as they budget, make coverage decisions, and incentivize more efficient and high-quality care. Understanding the constraints on sharing evidence and identifying standards for credible, reliable, truthful, and non-misleading evidence has been at the forefront of NPC’s work for the past several years. Therefore, we welcome the FDA’s comprehensive review of regulations and policies concerning the communication of evidence.
NPC is a health policy research organization dedicated to the advancement of good evidence and science, and to fostering an environment in the United States that supports medical innovation. We are supported by the major U.S. research-based biopharmaceutical companies and focus on the issues of evidence, access, innovation, and the value of medicines for patients. Our research helps inform critical health policy debates and supports the achievement of the best patient outcomes in the most efficient way possible.
We would like to provide comment on three issues:
- Public Health Goals and Health Care Reform Initiatives Require Broader Communication
- Payor and Risk-Bearing Providers Desire More Information; Potential Benefits of Information Outweigh Potential Harms
- Incentives Exist to Develop Adequate Evidence to Inform Coverage and Reimbursement. Many Outcomes Desired by the Changing Health Care System Require Beyond-the-label Information.
- Standards and Best Practices Exist and Can be Relied on to Ensure Scientific Integrity For Researchers and Evaluators
- Pre-Approval Exchange of Information with Payors is Needed for Both Investigational Products and Investigational Uses of Approved Products
1. Public Health Goals and Health Care Reform Initiatives Require Broader Communication
Question 1a: FDA is interested in input from stakeholders on how increased communications from firms about unapproved uses could impact the public health, and on whether the impact would differ across different categories of medical products. For example: What are the benefits for clinical decision making, research, coverage, reimbursement, or other purposes (please specify) if firms communicate to health care professionals, payors, researchers, and/or patients more information, including preliminary or inconclusive information, about unapproved uses of approved/cleared medical products? What are the drawbacks and risks? Are there safeguards or requirements that would effectively mitigate any drawbacks or risks?
To achieve our shared goal of a high-performing value-based health care system, informed and evidence-based decisions are needed. Greater efficiency and better care requires more information exchange—not less—about which treatments work best for whom and under which care settings. However, the laws and regulations regarding what information exchange is permitted are numerous and ambiguous. Although Section 114 of the Food and Drug Administration Modernization Act of 1997 (FDAMA Section 114) created a pathway to enable communications between payors and biopharmaceutical manufacturers, a lack of further guidance created uncertainty among stakeholders as to the types of information that can be exchanged, as well as with whom that information can be shared. It was only recently that Section 3037 of the 21st Century Cures Act offered some additional clarity and detail about the exchange of information.
Consider the following examples:
- An accountable care organization (ACO) needs to understand the overall health care cost and quality implications of including a drug on the formulary that reduces hip and vertebral fractures. Would the ACO be a “formulary committee or other similar entity” under Section 3037 of 21st Century Cures? Would information exchange be permitted?
- A health plan is interested in developing a value-based contract based upon hospital readmission rates for lung disease. Can information on outcomes that are relevant to a health plan, but beyond the information included in the FDA label, be shared?
- An organization of providers is enrolled in the Center for Medicare and Medicaid Innovation Oncology Care Model. They are interested in the impact of different chemotherapy regimens on patient symptoms and quality measures. Would a provider organization developing care pathways be considered to select medications? Would risk-bearing providers responsible for meeting cost and quality of care benchmarks be eligible for information exchange?
- A clinical practice guideline group aims to “include cost-effectiveness/value assessments and recommendations in practice guidelines and performance measures.” Would clinical practice guideline organizations or value assessment bodies be considered “other multidisciplinary entities that review scientific and technology assessments?” Would information exchange be permitted?
- If a biopharmaceutical company compares the effectiveness of two migraine treatments on outcomes that matter to employers and patients, will the biopharmaceutical company be able to communicate the results in the same manner as the public, not-for-profit Patient-Centered Outcomes Research Institute?,
Payors and risk-bearing providers are eager for more, and better, information. A survey of payors and risk-bearing providers found that between two-thirds and three-quarters of payors and providers feel at least somewhat limited by the amount, type, and the quality of information they receive. (Section 2 Appendix A and Appendix B)
The potential benefits associated with sharing information outweigh the potential harms. Payors and risk-bearing providers were most likely to describe benefits of broader information exchange which is consistent with, but not included in, the FDA label to enhance decision making. Among payors, 76% of the surveyed respondents rated three or more potential benefits as very/extremely important. Among providers, 84% rated three or more benefits as very/extremely important. Both payors and providers felt these benefits are much more likely to occur than not.
In contrast, nearly a quarter of the surveyed payors and providers cited “no potential harms” when asked unaided. Payors were concerned with questionable data that might impact patient outcomes and both payors and providers were concerned that inadequate disclosures on the limitations of the information would be provided. (Section 3 Appendix A and Appendix B) Despite these concerns, sharing information beyond-the-FDA Label, was perceived to have a net benefit on patient care.
The nation’s public health goals and health care reform initiatives require broader communication with payors and population health decision makers — not less — about which treatments work best for whom and under which care settings.
Question 2a and 2b: FDA is aware of changes happening in the health care system that are outside of FDA’s role, which may provide an impetus for the development of high-quality data to fully assess the risks and benefits of new uses of medical products. 2a. To what extent do changes occurring in the health care system that give payors and formulary committees more influence on prescribing decisions (including by denying, limiting, or endorsing coverage of unapproved uses of approved medical products) provide incentives for firms to generate the high-quality data needed to demonstrate safety and effectiveness for new uses? 2b. To what extent do these changes affect (to preserve, enhance, or suppress) incentives for firms to seek FDA approval/clearance of new uses?
The changing health care system includes a wide variety of evolving organizations responsible for financing or reimbursing treatments, selecting drugs, and managing formularies. Many stakeholders making population health recommendations, delivering care, and managing costs did not exist five years ago. For example, new care delivery models such as accountable care organizations — virtually unheard of 10 years ago — now provide care for nearly 28 million lives. As clinical practice guideline groups and value assessment framework organizations evolve, they are expected to have a greater influence on providers, health plans and health systems. New delivery models and entities are likely as care is coordinated and the lines separating traditional health care organizations from payors are blurred.
These scenarios provide an impetus for manufacturers to develop high-quality data to inform population health decisions. Payors use sophisticated approaches and make coverage policies based on the drug indication. A recent analysis of over 301 drug-indication coverage decisions by the 17 largest health plans found that 32% of all coverage decisions were more restrictive than the FDA-approved label (e.g., plans either do not cover, or the policies are more restrictive than the FDA label). Coverage restrictions often require patients to try prior therapies or restrict coverage to patient populations narrower than the approved label. Coverage policies followed the FDA-approved label for 58% decisions. However, for many treatments such as cancer drugs, orphan drugs, or pediatric indications payors may need additional information such as contained in compendia or in the medical literature to inform coverage, review individual exception policies, and address provider and patient requests. Therefore, while information from the FDA-approved label is desired, it is not sufficient to meet payor information needs. Without additional information on the safety, effectiveness, and costs of new indications and products, reimbursement and coverage are likely to be restricted.
Payor and population health decisions should be informed by evidence fit for the decision. Although randomized controlled trials may be able to answer some questions, other study designs, such as real-world studies, may be a better fit to address other health care system issues.
Payors and risk-bearing providers want information about multiple outcomes when making drug selection and coverage decisions. Between 60% and 90% of payors felt it was “extremely important” or “very important” to understand each of the following: the clinical benefits of treatments; which side effects might occur; how a treatment compares to alternatives based on information from either a clinical trial or high-quality, real-world study; whether patients reach quality outcome measures (e.g., asthma control or lipid goals); whether medication use is associated with changes in the utilization of other health care services; and the projected budget impact of treatment. Among payors, over half of the respondents found six or more criteria as very or extremely important to understand. Over 75% of the providers surveyed found seven or more criteria very or extremely important for treatment decisions. In the next three to five years, payors and providers expect comparative information, outcome measurement and comparative costs to be more impactful in decision making. Many of these outcomes desired by payors and risk-bearing providers are not typically included in the FDA-approved label. (Section 1 Appendix A and Appendix B)
In addition to the outcomes listed above, payors and risk-bearing providers recognize the increasing importance of information on treatment performance against quality measures. Quality measures help payors benchmark performance and are central to accountable care organizations and innovative payment programs such as bundled payments and risk-sharing agreements. Payors and providers were mixed on whether a pharmaceutical product’s performance against quality measures (e.g., reaching asthma goal etc) was important and impactful for today’s decisions. However, both groups agreed that a product’s performance against quality measures would influence coverage decisions in the next three to five years. Among payors, 53% agreed “completely” or “very much,” and 12% did “not agree very much” or “not at all.” Among providers, 78% agreed “completely” or “very much,” and 3% “did not agree very much” or “not at all.” (Section 1 Appendix A and Appendix B)
Incentives remain for biopharmaceutical manufacturers to develop adequate evidence to inform coverage and reimbursement. Many outcomes desired by the the changing health care system require beyond-the label information.
Question 3: FDA recognizes that information about medical products, including information about unapproved uses of approved/cleared medical products, is now broadly available from a wide variety of sources (e.g., academic and governmental organizations, scientific journals, professional societies, compendia) in both traditional and new communication vehicles and platforms, particularly electronic communication platforms (e.g., the Internet). What is the impact of the increasing availability of this information on firms’ incentives to communicate information about unapproved uses of approved/cleared products? FDA is also interested in input on other factors that firms may consider when making decisions about providing information about unapproved uses of their approved/ cleared medical product, including financial considerations.
Health plans and risk-bearing providers use many sources of information on a daily, weekly, and monthly basis. Over 75% of the payors and over 66% of the providers we surveyed used six or more information channels at least monthly. The internet was the most highly cited source and was used more frequently than 6 “curated” channels such as clinical practice guidelines, compendia, or continuing medication information. The breadth and frequency of new communication vehicles and electronic or internet platforms are expected to remain consistent over the next 5 years. (Section 5 Appendix A and Appendix B)
2. Standards and Best Practices Exist and Can be Relied on to Ensure Scientific Integrity For Researchers and Evaluators
Question 4b: Given the importance of the scientific integrity of the information that may be relied on in making decisions about the use of medical treatments, FDA is interested in input from stakeholders on the standards that should apply to unapproved use communications to minimize the potential of these communications to be misleading or otherwise cause harm. 4b. What criteria should the Agency consider in determining whether a study or analysis that is the basis of a firm’s communication is scientifically appropriate to support the presentations or conclusions in the communication?
Over the past five to ten years, there has been a proliferation of best practices and standards. A literature review published in 2016 found nine standards exist for conducting high-quality observational research studies. These standards and best practices were developed by a variety of disciplines (e.g., epidemiology, pharmacovigilance, and health economics) and organizations (e.g., professional societies, cross-stakeholder efforts, and governmental and non-governmental organizations). (These nine standards and best practices are arrayed in Appendix C)
Among these initiatives, are standards and best practices in which NPC has had an opportunity to be involved. For example, NPC funded and a multi-stakeholder group developed the Good ReseArch for Comparative Effectiveness (GRACE) Principles and GRACE checklist to help decision-makers assess the quality and usefulness of observational studies., The GRACE effort has been cited by several organizations including: the International Society of Pharmacoepidemiology, the Journal of Managed Care and Specialty Pharmacy, and United Kingdom National Institute for Health and Clinical Excellence. Similarly, NPC collaborated with the Academy of Managed Care Pharmacy (AMCP) and ISPOR, professional societies of experts in their respective areas, to create online tools and checklists to help population health decisionmakers evaluate the relevance and credibility of new research methods. These CER Collaborative checklists include techniques to assess if observational, modeling and network meta-analysis studies are sufficiently credible based on good practices agreed upon by academic experts, industry researchers and payors.,, These techniques, when followed with sound data and high-quality methods, can develop truthful and non-misleading evidence, but would not typically meet the substantial evidence threshold relied upon by the agency for communication.
NPC believes that standards and best practices exist and can be relied on to ensure scientific integrity. More information that meets these standards should be accessible to sophisticated users — not less.
Question 4c and 4d: What do health care professionals generally understand about the quality and utility of different kinds or levels of scientific evidence related to unapproved uses? Can the same information be misleading to some audiences of health professionals and not others? What information is most important to health care professionals and other entities in allowing them to judge the validity and utility of firms' communications about unapproved uses, including the level of uncertainty of the evidence, and why? Does the answer to this question differ depending on the recipient's purpose—e.g., making treatment decisions for an individual patient, informing the direction of further research, making formulary or institutional supply chain contracting decisions, or making coverage determinations?
To date, the CER Certificate tools mentioned above have more than 3,800 users among pharmacy, biopharmaceutical, and managed care professionals. In addition, a 19 credit hour training program has been developed to ensure sophisticated professionals have the confidence and ability to critically evaluate the credibility of these new types of studies for coverage, reimbursement, and care pathway decisions. Learners in the program have reported improvements in their capabilities to evaluate various CER studies and identify study design flaw (27-60% improvement). Additionally, most learners used these tools to evaluate these study designs to guide for at least 1-2 decisions a month.
Question 5: FDA is interested in input from stakeholders on factors that the Agency should consider in evaluating whether firms’ communications about unapproved uses of approved/cleared medical products are truthful and non-misleading, including what information firms should disclose in these communications to help ensure audiences are not misled, and on general considerations related to the audience for these communications and on communication vehicles and techniques. For example: 5a. What information should firms communicate to make audiences aware that the medical product is unapproved for the use discussed and to otherwise distinguish between the approved/ cleared use(s) of the medical product and the unapproved use? How could the means of communication affect a recipient’s ability to distinguish between unapproved and approved/ cleared uses or otherwise impede the disclosure of necessary contextual information?
Contextual information about a study is important to allow readers a full understanding of the information relevance and credibility. However, disclosures need to be consistent with the format in which the information is presented. In a recent survey, 65% of payors reported disclosure of study limitations to be “valuable” or “extremely valuable.” Just over half of the payor and provider respondents found disclosures to be “valuable” or “extremely valuable” when the study or the information shared was not included in the FDA-label. (Section 4 Appendix A and B)
However, the disclosures may not apply to all communications with all audiences and all formats. For example, a study assessing patient persistence would not report the perspective of the analysis (e.g., health care system vs. society perspective). The credibility of a network meta-analysis is based on the search criteria to select studies for inclusion and the analytic methods to address potential differences in study design and populations — not the source of cost estimates. Disclosure requirements meant to ensure that information is not false or misleading should not be overly prescriptive. The format for dissemination may vary. Appropriate disclosures should be accessible (e.g., as a cover page to the information with a link to the study publication, protocol, or analysis plan, or other relevant information), but not a part of the specific communication.
3. Pre-Approval Exchange of Information with Payors is Needed for Both Investigational Products and Investigational Uses of Approved Products
Payors need ready access to truthful and non-misleading evidence to aid them as they plan, budget, and forecast for new products and new indications. Payors routinely request information from manufacturers about investigational products and indications. However, less than half of the payors receive this information. Even when information is provided, it is often too late to inform the review process.
Timely information is needed well before product approval. For example, health plans are planning now for investigational products and supplemental indications with a potential 2018 FDA approval. Because insurance rates for the calendar year 2018 must be submitted 6-9 months in advance (e.g., spring 2017), lack of information or awareness may result in two unintended consequences: first, plans may underestimate the utilization and cost of new treatments resulting in premiums which do not cover costs; and second, plans may overestimate the utilization and cost of new treatments resulting in high premiums which discourage consumer decisions to purchase coverage.,
The rationale for communications for investigational products in the Draft Guidance on Drug and Device Manufacturer Communications with Payors, Formulary Committees, and Similar Entities applies equally to new indications for approved medications. It is not uncommon for products to be approved for one indication (e.g., advanced melanoma) and be under investigation for supplemental indications (e.g., non-small cell lung cancer, head and neck cancers, etc.). In 2014, the number of approvals for supplemental new indications was nearly equal to the number of new product approvals. Most, if not all, of the information outlined in the draft guidance is specific to the indication rather than the product (e.g., indication sought, clinical data, and anticipated timeline for possible FDA-approval). If the new indication involves a different dose, frequency, or use (e.g., combination vs. mono-therapy), then product pricing information for the first indication will be inaccurate. Because product utilization and medication costs for supplemental indications may exceed those of the initial approval, plans need information at the indication- rather than the product-level.
NPC agrees with the recommendations that emerged from the AMCP Partnership Forum on Enabling the Exchange of Clinical and Economic Data Pre-FDA Approval. The multi-stakeholder group noted that preapproval information exchange was desired for both new medications and new indications of existing medications.
NPC recommends that the FDA treat communication to payors regarding unapproved uses for approved products consistent with communications to payors regarding unapproved uses for investigational products.
In closing, NPC shares the goal of the FDA: to ensure that communication is done in a way that promotes public health and supports innovative developments in science and technology, medicine, and health care delivery. Meeting these goals requires transparent, reproducible, and ongoing evidence generation and communication. Improved communication includes sharing what is known about the benefits and risks