Participation of Females in Clinical Trials and Gender Analysis of Data in Biologic Product Applications
This research project examined the extent to which females have been included in clinical trials for biological products and to what extent the data from these studies have been analyzed and presented with respect to gender. This study was funded by the FDA Office of Women's Health, the "FDA Scholarship in Women's Health Program."
The project focused on currently marketed products, specifically biologics for which a new product or biologics license application was approved by the Center for Biologics Evaluation and Research (CBER) during calendar years 1995-1999. Data on enrollment, inclusion and exclusion criteria, study type and gender analyses were searched and recorded. The data were obtained from the PLA/BLA review prepared by CBER reviewer(s), summary basis of approval (SBA), product label or clinical trial summaries contained in the PLA.
The results indicate that the inclusion of female subjects into clinical studies for biological products appeared to be similar to that for males. The population enrolled reflected the population that will receive the biological product. Documentation with regard to gender composition and gender analysis was not consistent. Clinical trials were not prospectively designed to evaluate potential gender differences. Gender analysis was by subgroup analysis using gender as demographic variable and occurred only for a small percentage of the total clinical trial summaries reviewed. Data on safety and effectiveness of the product were not presented according to gender. The female specific information included in the product label was generally limited to the pregnancy subsection of the label.
Historically, investigators have been reluctant to include female subjects in clinical trials due, in part, to concerns with potential birth defects. In addition, the 1977 FDA guideline entitled "General considerations for the clinical evaluation of drugs" excluded women of childbearing potential from early drug development studies. This may have further contributed to a general lack of females participating in drug development studies and thus, to a paucity of information about drug and biologic product effects in females.
In order to reverse this real or perceived regulatory barrier to the participation of women of childbearing potential in clinical trials the agency has taken a number of initiatives.
The 1988 document entitled "Guideline for the format and content of the clinical and statistical sections of new drug applications" emphasized the importance of including analyses of demographic data in NDA applications.
The 1993 "Guideline for the study and evaluation of gender differences in clinical evaluation of drugs" provides guidance regarding inclusion of both genders in drug development, analysis of clinical data by gender, and assessment of potential pharmaco-kinetic differences between genders.
In addition to these guidelines, FDA amended its regulations to require effectiveness and safety data for important demographic subgroups, specifically gender, age and racial subgroups (21 CFR Parts 312 and 314, February 11, 1998, Final rule: Investigational New Drug Applications and New Drug Applications). The final rule published in June 2000, permits the agency to place a clinical hold on one or more studies under an IND if men or women with reproductive potential are excluded from participation in an investigation only because of risk or potential risk of reproductive or developmental toxicity from use of the investigational drug (21 CFR 312, June 1, 2000. Final Rule: Investigational New Drug Applications; Amendment to Clinical Hold Regulations for Products intended for Life Threatening Diseases).
The Food and Drug Administration Modernization Act of 1997 (FDAMA Sec. 115 Clinical Investigations. (b) Women and Minorities) amended Section 505 (b) (1) 21 U.S.C. 355 (b) (1) mandating the review and development of guidance, as appropriate, on the inclusion of women in clinical trials. In order to implement this section of the FDAMA the agency has formed the "FDAMA women and minorities working group" with representatives from the agency and the National Institutes of Health. One of the recommendations of this working group was that the agency continues to implement procedures that will enhance the ability to gather, search and evaluate demographic data.
The FDA Office of Women's Health has developed a program to educate agency staff about women's health issues and to advance the women's health agenda within the centers. One critical objectives of this program is to gather information about past and current agency activities related to women's health. This research project was initiated with the objectives to determine
- to what extent females have been included in clinical trials for biological products,
- to what extent gender specific data have been submitted for review,
- to what extent gender specific data have been analyzed either by the sponsor or the FDA,
- whether there is a systematic documentation of results by gender either by the FDA or the sponsor,
- if biological product labeling includes female specific information and/or information specifically pertaining to women.
Thirty three (33) of a total of 63 product license applications for new biological products approved by CBER during calendar years 1995 - 1999 were searched. The search excluded PLAs for further manufacture, test kits, plasma products and imaging reagents. The documents reviewed included the medical officer's review, statistician's review, pharmacologist's review, summaries of clinical trials submitted with the PLA, the product label, advisory committee transcripts, and SBAs. The type of document(s) available for a specific product application varied. A data extraction sheet was created on which information about the product, indication, number of studies conducted under the PLA, type of studies conducted, percent of females and males enrolled, age, exclusion criteria, gender analyses and label information was recorded. This information was transferred to a data base created in Microsoft Accessâ. The database was analyzed to determine if females were included in clinical trials, if gender specific information was submitted, the extent to which these data were analyzed either by the FDA or the sponsor and if gender specific information was included in the product label.
There were a total of 63 new products approved by CBER between calendar years 1995-1999, 23 in the Office of Therapeutics Research and Review (OTRR), 24 in the Office of Blood Research and Review (OBRR) and 16 in the Office of Vaccines Research and Review. Excluded from the search were PLAs for further manufacture, test kits, plasma products and imaging reagents. Included in the search were 17 approved therapeutic products, 9 approved vaccine products, and 7 approved blood products. Sixteen of the 17 therapeutic products that were included in the search were indicated for both, male and females. One product was indicated for female and male pediatric patients. One product was for treatment of metastatic breast cancer and was studied exclusively in female patients. All of the 7 blood products were indicated for both, male and females. Three of the 7 approvals were for treatment of hemophilia A and B. One blood product approval was indicated for female and male pediatric patients. All of the 9 vaccine products were indicated for males and females, 4 of these were indicated for the female and male pediatric population only.
In general, the population enrolled in the pivotal clinical trials reflected the population for which the biological product was indicated.
Studies with gender specified
Gender information was collected from 109 clinical studies summaries for the 9 vaccine products, 87 clinical study summaries for the 17 therapeutic products and 22 clinical study summaries for the 7 blood products. Information about gender was available for 14%, 55% and 63% of the clinical trial summaries reviewed for blood, therapeutics and vaccines products, respectively (Figure 1). These data were further analyzed to determine the study type, i.e., Phase 1 and 2 and Phase 3 for which this information was collected. Results show that gender was documented for 33%, 67% and 62% of the Phase 3 studies conducted for blood, therapeutic and vaccine products, respectively (Figure 2). Gender was documented for 11%, 48% and 64% of the Phase 2 studies conducted for blood, therapeutic and vaccine products, respectively (Figure 3).
Participation of females in clinical trials
The data in Figure 4 show the average percentage of females and males included in clinical trials by individual PLA and across all PLAs (n=24) for which information about gender composition was documented. The average percentage of females and males enrolled into the clinical studies across all PLAs was 45.3% and 54.6%, respectively. However, some products were predominantly studied in men, such as a recombinant plasminogen activator for management of acute myocardial infarct and a recombinant PDGF for the treatment of lower extremity diabetic ulcers. One product was exclusively studied in females and was indicated for treatment of metastatic breast cancer. Overall, the representation of males and females in clinical trials appeared to be similar. Pregnancy was the major exclusion criteria defined resulting in exclusion of females. In some studies, lactation was defined as additional exclusion criteria. In a total of three studies, 1 conducted for a vaccine product and 2 conducted for therapeutic products, women of childbearing potential were excluded, if they did not practice adequate means of birth control (Figure 5).
Gender analysis by sponsor
In all cases documented, gender analysis was conducted as subgroup analysis in which gender was used as a demographic variable. Clinical trials were not prospectively designed to detect gender differences. Information with regard to gender analysis by sponsor was obtained for 8 of 9 vaccine products, 12 of 17 therapeutic products and 0 of 7 blood products (Figure 6). Next, the percentage of studies for each product category containing information about gender analysis by sponsor was determined. This information was available for 12%, 24% and 0% of the clinical trials conducted for vaccine, therapeutic and blood products, respectively (Figure 7). Figure 8 shows the percentage of Phase I/II and Phase III studies for the various product categories. For vaccines, 12% of all studies were Phase III trials and 88% were Phase II trials. For therapeutic products, 32% of the studies were Phase III trials and 68% were Phase I/II trials. For blood 14% were Phase III trials and 86% were Phase I/II trials. Figures 9 and 10 illustrate that information about gender analysis was available for 60% and 37% of the Phase III studies for vaccine and therapeutic products, respectively. Four percent ( 4%) and 20% of the Phase I/II trial summaries for the vaccine and therapeutics, respectively, contained documentation about gender analysis by sponsor. There was no documentation for gender analysis by sponsor for any of the clinical trial summaries reviewed for the blood products.
Dependency of data outcome on source document
It should be stressed that these results are dependent on the source document(s) available for the search. This is illustrated in Figure 11. For example, documentation of gender analysis by sponsor was available for 8 of 9 approved vaccine products. However, when only CBER review document(s) were searched, information about gender analysis by sponsor was present for only 5 of the 8 vaccine products. For the additional 3 products, information about gender analysis was found by searching the clinical trial summaries submitted by the manufacturer to the PLA. Documentation about gender analysis was available for 12 of the 17 therapeutic products. This information was retrieved from 11 CBER reviews and 1 SBA. However, in general, SBAs and the product label for blood, vaccine and therapeutic products did not contain information on gender or gender analysis. This may explain, why no results on gender analysis are available for blood products. For these products, the only source documents available were SBAs and the product label.
Gender analysis CBER
There was no documentation about gender analysis by the CBER reviewers for the blood and vaccine product approvals searched. For one vaccine product, it was documented that CBER had requested that the sponsor conduct a gender analysis for the efficacy trials. Of the 17 therapeutic products there were 4 products for which CBER reviewers performed a subgroup analysis using gender as demographic variable. For these products, the CBER gender analysis was done on results derived from the pivotal trials and PK studies (Figure 12).
Overall, the documented results of gender analyses by sponsor or CBER showed no differences in the efficacy or safety outcomes with regard to gender. There were a few products for which gender differences were noted, however the significance level for these findings is questionable. In general, both the sponsor and the CBER concluded that the studies were not sufficiently powered to detect gender differences, that is, small sample sizes impacted on the significance level of the findings.
Female specific information contained in product label
In general, the female specific information in the product label was restricted to the pregnancy subsection of the label. The majority of products were classified as pregnancy category C. For the purpose of this search any additional information specific for females other than the pregnancy category C was documented. Sixteen of the 33 product labels contained information and/or instructions for females in addition to or instead of the pregnancy category C label, 3 for vaccine products, 3 for blood products and 10 for therapeutic products (Figure 13). This information included additional precautionary statements for females of childbearing potential, pregnant or lactating mothers; a description of adverse effect on the fetus and pregnancy; designation of the product as pregnancy category B; or pertained to instructions as to when to administer the product. Some therapeutic product labels contained a description of the PK study results, i.e., PK parameters analyzed by gender.
Of the 33 product approvals searched, pregnancy registries were implemented as a phase 4 commitment for 2 products. These were prophylactic vaccine products indicated for healthy subject populations. The target population for these vaccines includes females of childbearing potential.
Performance of pre-clinical reproductive toxicity studies
Pre-clinical reproductive toxicity studies were performed for 10 of the 33 approved products, 8 of these studies were performed prior to licensure for therapeutic products. There was one blood product and one vaccine product for which data from reproductive toxicity studies are available. The reproductive toxicity study performed for the vaccine product was done under a phase 4 commitment (Figures 14 and 15).
The inclusion of female and male subjects into clinical studies for biological products approved by the CBER during calendar years 1995-1999 appeared to be similar. In general, the population enrolled was reflective of the population for which the product is indicated. Documentation of gender composition was not consistent and information about gender analysis was available for only a small percentage of the total clinical trial summaries. Gender analyses was limited to exploratory analyses using gender as demographic variable. No statistically significant gender differences in terms of efficacy and safety of the products reviewed were found, however, trials were not prospectively designed to detect gender differences. There was no systematic documentation of results by gender. The female specific information included in the product label was mostly limited to the pregnancy subsection of the label.
The results suggest the implementation of procedures to ensure consistency in the documentation of information pertaining to gender composition, gender analysis and gender specific data. Further discussions may be needed to determine if and for what products and/or product indications gender analyses should be performed and during what stage in clinical development this information should be collected.