Division of Dockets Management
Food and Drug Administration
5630 Fishers Lane, Room 1061 (HFA-305)
Rockville, Maryland 20852
Comments of the National Center for Health Research
on
“Guidance: Food and Drug Administration Safety and Innovation Act Action Plan”
Docket No. FDA-2013-N-0745-0053
The National Center for Health Research appreciates the opportunity to provide comments on the Food and Drug Administration Safety and Innovation Act (FDASIA) Action Plan. We strongly support inclusion of separate analysis of safety and effectiveness for demographic subgroups in clinical trials for drugs and devices. It is essential that methodologically sound data about how drugs and devices work in these groups be conducted and made publicly available.
Our research center along with several other non-profit, research, academic and advocacy organizations provided oral comments at the public hearing held by the FDA on April 1, 2014 on the FDA’s 907 report. We focused our comments on what the FDA can do to ensure greater diversity in the clinical trials submitted to the agency, but underscored the importance of having large enough numbers of major demographic groups so that subgroup analyses could be performed to determine safety and effectiveness for those subgroups. Additionally, we emphasized the importance of using those subgroup analyses as a basis of approval and labeling decisions, and making those results widely available in a user-friendly format to providers, patients, and other stakeholders. Many other speakers expressed similar concerns. We are very disappointed that the FDA’s Action Plan did not incorporate more of the suggestions and recommendations voiced during the public meeting.
The Section 907 report Action Plan outlines 27 action items in three priority areas: quality, participation, and transparency. Taken together, the action items reflect the FDA’s preliminary approach to addressing the need for greater demographic inclusion in clinical trials. However, the plan is too vague and does not adequately focus on the need for meaningful subgroup analyses. It also fails to emphasize the regulatory role of the FDA in ensuring those analyses rather, and instead focuses on the role of the FDA as a bully pulpit with no enforcement power.
Our specific comments are below:
Priority One: Quality–Improve The Completeness And Quality Of Demographic Subgroup Data Collection, Reporting And Analysis
Guidance Document for Medical Products
Section 1.1 of the Action Plan states that CBER and CDER plan to “review, update, and/or finalize, as needed, relevant industry guidance and internal FDA good review practice documents to encourage greater demographic subgroup representation in clinical trials, subgroup analysis and communication of results.” The FDA has been encouraging greater diversity and subgroup analyses for years, with limited success, and there is no evidence that more encouragement will improve the outcome. If the FDA believes it cannot require such analyses, then they should specify that approval for a general population (e.g. not just White males under 65, or White adults 18-65) requires that the company evaluate the major demographic subgroups to determine if the drug or biologic is safe and effective for those various groups.
We support the action item that states the FDA plans to draft “a guidance document on analysis and reporting of ethnicity, race and age in medical device clinical studies. In doing so, FDA plans to explore key barriers and limitations to meaningful data analysis.” This guidance is essential and long overdue. We are concerned, however, about the vague and tentative tone in the use of such phrases as “Begin drafting” and “explore key barriers.”
We also support the action item that encourages “greater demographic subgroup representation in clinical trials, subgroup analysis and communication of results.”
The barriers and limitations to meaningful subgroup data analyses are very clear: Clinical studies for many drugs and most medical devices are too small to be able to assess statistical significance when conducting subgroup analyses. This is a particularly pervasive problem regarding racial and ethnic minority patients, since the numbers of those groups are usually far too small to be analyzed separately for safety and effectiveness outcomes. These small device studies, which often include less than 10 African American patients, should be unacceptable for a PMA process that is limited to the highest risk medical devices.
Recommendations:
- FDA should revise and finalize the three mentioned guidance documents and explicitly require sponsors to include the appropriate subgroup analyses to determine the safety and effectiveness outcomes for each major subgroup.
- Subgroup analyses should analyze all safety and effectiveness outcomes for each major demographic subgroup. Too often, the only subgroup analyses available to the public are to compare the demographic traits of each subgroup; such analyses do not provide meaningful information to determine safety or effectiveness. All subgroup analyses should be publicly available at FDA Advisory Committee meetings and on the FDA web site for each medical product.
- The FDA should make it clear that the agency will not approve medical products for all populations if the product was not tested for safety and effectiveness on all major demographic subgroups with meaningful subgroup analysis.
Training
We support the action item in section 1.3 that states that the FDA will train new clinical trial reviewers, and “offer additional education and training courses for experienced reviewers to better clarify FDA’s expectations for data collection and analysis related to demographic subgroups.”
Recommendations:
- To strengthen this aspect of the Action Plan, training and education of FDA reviewers must be greatly improved and consistent across CDER, CBER, and CDRH.
- The FDA should also provide training on these issues for Members of FDA Advisory Committees and Panels. For many years, through mid-October 2014, most Advisory Committee members have shown little interest in demographic inclusion. We have attached Table 1, which provides a sample of some recent examples of drug and device Advisory Committee votes on products with very limited diversity and lacking in subgroup analyses.
OWH and OMH
In Section 1.5, we support the following action items:
We support the Office of Women’s Health (OWH) plans to target OWH funding to projects that answer specific regulatory research questions and emerging priorities from the product review centers.
We support the Office of Minority Health (OMH) plans to develop research projects to better understand medical product clinical outcomes in racial and ethnic demographic subgroups.
We support the OMH plans to collaborate with National Institute of Health’s (NIH) National Human Genome Research Institute in research into the role of genetics and genomics in health disparities.
Recommendation:
- We agree that additional research is needed to better understand issues related to participation, recruitment and analyses of diverse populations. The Action Plan should specify that OWH and OMH will conduct research to better understand subgroup differences in safety or effectiveness that become apparent as a result of the new FDA policies requiring subgroup analyses of major demographic subgroups. For example, if subgroup analyses indicate a difference in effectiveness for women compared to men or Blacks compared to Whites for one or more diabetes drugs, the OWH or OMH would conduct research to better understand those disparities.
Priority Two: Participation–Identify Barriers To Subgroup Enrollment In Clinical Trials And Employ Strategies To Encourage Greater Participation
We support the FDA collaborating with NIH, other Department of Health and Human Services (HHS) agencies, industry and other stakeholders to broaden diverse participation in clinical research.
We support FDA’s OMH collaborating with agencies of the HHS, NIH, Institute of Medicine and the Human Service’s Office of Minority Health to convene a 2015 meeting to better understand the barriers to participation in clinical trials by communities of color.
We strongly support the action item where FDA’s OWH will collaborate with the NIH Office of Research on Women’s Health (ORWH) on a national campaign to educate and promote clinical trial participation, focusing on women.
However, we do not believe that the major impediment to diversity in clinical trials is the lack of interest of women, people of color, or patients over 65. We are very concerned that the FDA’s statements put the onus on patients to be recruited, rather than on companies to recruit them. In the same way that companies recruit the best possible physicians by providing generous incentives to participate in clinical trials, companies should do the same for a diverse population of patients. Diversity would be greatly improved if companies would improve the incentives for patients to participate in clinical trials: reimbursement for patients’ time and travel for appointments, providing free medical appointments for follow-up medical evaluations, availability of child care for mothers of young children during patient visits and follow-up, and other efforts to reduce financial disincentives and increase financial incentives.
Recommendations:
- The FDA should gather information from companies that are successfully achieving diversity in their trials to determine the strategies they are using, and compare these practices to those of companies that are not accomplishing this goal.
- The FDA should work with individual drug and device companies to discuss the incentives needed to obtain optimal inclusion of major demographic subgroups (gender, race/ethnicity, and age) in clinical trials, as part of their ongoing conversations with the sponsors prior to final applications for approval.
- The FDA’s OMH should collaborate with the HHS’s Office of Minority Health and companies that make medical products to create an education campaign, specifically targeting communities of color, similar to the campaign focusing on women.
Priority Three: Transparency–Make Demographic Subgroup Data More Available And Transparent
We do not support the action items in section 3.1, as currently written. We agree that it is vital to focus on transparency and it is important to make information available to patients, providers and policy makers. However, the action items under this section fall well short of that goal. The draft guidance states that the FDA will post “demographic composition and analysis by subgroup in pivotal clinical studies for FDA-approved medical products,” and for “pivotal studies” for devices.It is unclear what this means, but if it means that each subgroup will be analyzed in terms of their demographic characteristics, rather than analyzed for safety and effectiveness, that is wholly inadequate. It is not always possible to predict how different major subgroups might respond differently to a drug or device, and for that reason is it essential that each group be analyzed to determine if the product is safe and effective for them.
Recommendation:
- The FDA should clarify that subgroup analyses should be conducted to determine safety and effectiveness for major demographic subgroups, such as men and women; Whites, Blacks, and Hispanics; adults under 65 and adults over 65.
Section 3.2
We do not support the action item in section 3.2 because it is too vague. It states that the FDA intends “to work with industry, advocacy groups, risk communicators (including FDA’s Risk Communication Advisory Committee) and other stakeholders to explore potential methods for communicating meaningful information on demographic analyses to the public. It does not identify specific work that FDA will do and it does not define “meaningful information.”
Recommendations:
- The FDA should make this section more concrete and describe the work it intends to do and define key terms.
- Inclusion of demographic information should be required, standardized, easy-to-understand sections on the label, so that patients and doctors can quickly find this information. If it does not exist, there should be an excellent justification that is clearly explained.
- The FDA should require subgroup-specific analyses in language that is understandable by both health professionals and patients.
Conclusions
The FDA’s Action Plan is too vague and it is unclear whether the proposed changes would have a meaningful impact. Nowhere in the Action Plan does the FDA state that it will require diversity in clinical trials or require subgroup analyses that evaluate safety and effectiveness. The agency should take a firm stance of not approving medical products unless they have been adequately analyzed for safety and effectiveness on major subgroups, instead of simply recommending it. The FDA should let sponsors know that drugs, biologics, and devices will only be approved for the general population if the major demographic subgroups were large enough to analyze separately. If the FDA does that, sponsors will find ways to greatly improve recruitment and all medical products that are required to have information on safety and effectiveness would provide that information for all major demographic subgroups.
The National Center for Health Research
Table 1- National Center for Health Research: Analysis of Sample FDA Advisory Committee Meetings (2014)
Product Name | Company | Purpose of product | Pivotal Trial Sample size % (n) | Subgroup Analysis Conducted? | Advisory Committee Vote | FDA approval? | |
Black | Hispanic | ||||||
Liraglutide | Novo Nordisk | Chronic weight management drug | All phase 3 trials – 11% (n=575)
U.S. trial – 18% (n=476) |
All phase 3 trials – 10% (n=551)
U.S. trial – 11% (n=309) |
N | 14-1 Overall benefit-risk assessment is favorable to support approval | Pending |
Nebivolol/
valsartan |
Forrest Laboratories, Inc | Treatment of hypertension | Safety – 10% (n=411)
Efficacy – 30% (n=245) |
Safety – 41% (n=1,684)
Efficacy – 24% (n=197) |
Y
Safety & efficacy subgroup analyses: Black vs non Black; Hispanic vs non Hispanic; no difference in BP reduction for Black patients on new drug FDC 20/320 compared to patients on nebivolol 40 mg |
6-4 in favor of approval | Pending |
TissuGlu | Cohera Medical | Reducing fluid build-up after abdominoplasty | 21% (n=21) received product
22% (n=11) control |
7 Hispanics total
5% (n=5) received product 4% (n=2) control |
N | 6-4 (one abstention) in favor – benefits outweigh the risks | Pending |
VBLOC MAESTRO system | EnteroMedics Inc. | Weight loss | 5% (n=8) received device
4% (n=4) control |
4% (n=6) received device
8% (n=6) control |
N | 6-2 (one abstention) in favor – benefits outweigh the risks | Pending |
Serelaxin | Novartis Pharmaceutical Corp. | Improve the symptoms of acute heart failure | 5% (n=29) received drug
4% (n=23) placebo |
10% (n=57) received drug
10% (n=57) placebo |
N | Unanimous vote against approval | Pending |
Inspire II Upper Airway Stimulator | Inspire Medical Systems, Inc | Treat moderate to severe obstructive sleep apnea | (n=0) | (n=1) | N | 12-0-1 (yes, no, abstain) that benefits outweigh the risks | Approved |
Cangrelor | The Medicines Company | Reduction of thrombotic cardiovascular events | 3% (n=156) received drug | 4% (n=196) received drug | N | 7-2 against approval | Approval denied |
Sivextro | Cubist Pharmaceuticals | Treat acute Bacterial Skin and Skin Structure Infections | 12% (n=77) received drug | 28% (n=182) received drug | Y
Safety & Efficacy subgroup analyses only for Blacks; Efficacy analysis only for primary end point >20% reduction in lesion size, not for any secondary end points |
14-0 That applicant provided substantial evidence of
safety and effectiveness |
Approved |
Dalvance | Durata Therapeutics, Inc. | Treat acute Bacterial Skin and Skin Structure Infections | 9% (n=28) received drug | Could not determine the number from the data presented | Y
Efficacy for all racial subgroups but small n |
12-0 That applicant provided substantial evidence of safety and effectiveness | Approved |