FDA Requires New Drug Applications Present Effectiveness
and Safety Data for Gender, Racial, and Age Subgroups

By Ronald L. Scott

The prototype subject for clinical research in the United States has been the single white male. Because of our uneasiness with race and ethnicity, medicine has been reluctant to address the relevance of race or ethnicity to the diagnosis and treatment of disease, except in the most obvious cases. Specialists note, however, that diseases and conditions once thought to be ethnically neutral express themselves differently and vary in incidence among ethnic groups. There is medical evidence that certain drugs respond quite differently in some racial and ethnic groups. To the limited extent such knowledge exists, a physician must consider race and ethnicity when making treatment decisions. For example, African Americans with high blood pressure typically respond to diuretics much better than whites, and Asians respond to some anti-psychotic drugs at a dosage of one-tenth that required for whites. Similar differences in response to drugs may occur because of gender and/or age. However, the exclusion of children, women and minorities from clinical trials has resulted in huge gaps in medical knowledge, with physicians increasingly uncomfortable with being forced to extrapolate from all-white-male data when making treatment decisions.

On February 11, 1998, The Food and Drug Administration (FDA) issued a proposed final rule to explicitly require that sponsors of investigational new drug applications (INDs) and new drug applications (NDAs) submit effectiveness and safety data by gender, age, and racial subgroups. The proposed rule is an important, but tentative step forward by the FDA, and has several limitations. The proposed rule would also require IND sponsors to disclose in their annual reports the number of subjects in a clinical trial according to age group, gender, and race. The FDA recognizes the current law does not fully reflect the need to present the safety and effectiveness data by subgroup and omits important subgroups, including gender and racial groups. The amendment to the IND regulations would allow the FDA and sponsors of drugs to monitor the sponsorís success in studying the populations likely to use the drug once it is approved. The current IND regulations require sponsors to submit an annual report that includes information on the total number of subjects intended to be included in the clinical trial, the number entered to date, the number whose participation was completed as planned, and the number who dropped out of the study. The proposed amendment would add a requirement that the annual report include the number of subjects entered into the study "tabulated by age group, gender, and race."

The amendment to the NDA regulations requires sponsors to provide evidence to support the dosage and administration sections of the labeling, including support for dose interval recommended. Data on efficacy is required to be presented by gender, age, and racial subgroups. Further, the rule requires safety data is to be presented by gender, age and racial subgroups.

The proposed rule does have significant limitations. It does not regulate the conduct of clinical studies, and according to the FDA, would not require sponsors to conduct any more studies than they have already conducted. The proposed rule would not require the inclusion of particular numbers of individuals from specific subgroups in any clinical trial. It refers only to the disclosure of data already collected. However, it would be a mistake for pharmaceutical companies to believe the proposed amendments lack substance, or that they only impose additional reporting requirements. If a sponsor fails to provide adequate subgroup data, the FDAís Center for Drug Evaluation and Research may "refuse to file" an NDA on the grounds that it is not submitted in the form required or that it is incomplete because on its face it lacks safety and efficacy data required by the Food Drug and Cosmetic Act. Omission of critical data, information, or analyses needed to evaluate efficacy and safety, or provide adequate directions for use, is an appropriate basis for a "refusal to file" decision by the agency. One ground for refusal to file is a "clearly inadequate evaluation for safety and/or effectiveness of the population intended to use the drug, including pertinent subsets, such as gender, age, and racial subsets." Therefore, the FDA could refuse to file NDAs where data is not presented with regard to gender, age, and racial groups. The agency will have broad discretion in when to refuse an NDA on the basis of inadequate subgroup analysis or data. It seems doubtful the agency would refuse an NDA for a badly needed drug proven safe and effective in the majority population but lacking adequate racial information, perhaps instead requiring post-market evaluations to remedy such lack of data. The rule refers to race rather than ethnicity, and does not define racial categories. A sponsor is given no guidance on how to present the racial information in the labeling for the drug. The final rule is proposed to become effective August 10, 1998.