Don't test, do sell: legal implications of inclusion and exclusion of women in clinical drug trials

Evaluation of Drug Trials in High-, Middle-, and Low-Income Countries and Local Commercial Availability of Newly Approved Drugs

IMPORTANCE

Clinical research supporting US Food and Drug Administration (FDA) drug approvals is largely conducted outside the US.

OBJECTIVE

To characterize where drugs were tested for FDA approval and to determine how commonly and quickly these drugs received marketing approval in the countries where they were tested, both overall and by country income level and geographical region.

DESIGN, SETTING, AND PARTICIPANTS

This cross-sectional analysis of trials supporting FDA approval of novel drugs in 2012 and 2014, sponsored by large drug companies, did not involve human participants. The settings were the countries hosting trials supporting US drug approval. Data sources included Drugs@FDA, ClinicalTrials.gov, PubMed, Google Scholar, EMBASE, and drug regulatory agency websites. Data analysis was completed March through September 2020.

MAIN OUTCOMES AND MEASURES

The primary outcomes were the proportion of drugs approved for marketing in the countries where they were tested for FDA approval within 1, 2, 3, 4, and 5 years of FDA approval and the proportion of countries contributing participants to trials supporting FDA approvals receiving market access to the drugs they helped test within 1, 2, 3, 4, and 5 years of FDA approval.

RESULTS

In 2012 and 2014, the FDA approved 34 novel drugs sponsored by large companies, on the basis of a total of 898 trials, 563 of which had location information available. Each drug was tested in a median (interquartile range [IQR]) of 25 (18-37) unique countries, including a median (IQR) of 20 (13-25) high-income countries, 6 (4-11) upper-middle-income countries, and 1 (0-2) low-middle-income country. One drug was approved for marketing in all testing countries within 1 year of FDA approval and 15% (5 of 34 drugs) were approved in all testing countries within 5 years of FDA approval. Of the 70 countries contributing research participants for FDA drug approvals, 7% (5 countries) received market access to drugs they helped test within 1 year of FDA approval and 31% (22 countries) did so within 5 years. Access within 1 year occurred in 13% (5 of 39) of high-income countries, 0 of 22 upper-middle-income countries (0%), and 0 of 9 lower-middle-income countries (0%), whereas at 5 years access rates were 46% (18 of 39 countries), 9% (2 of 22 countries), and 22% (2 of 9 countries), respectively. Approvals were faster in high-income countries (median [IQR], 8 [0-11] months) than in upper-middle-income countries (median [IQR], 11 [5-29] months) or lower-middle-income countries (median [IQR], 17 [11-27] months) after FDA approval. Access was lowest in African countries.

CONCLUSIONS AND RELEVANCE

These findings suggest that substantial gaps exist between where FDA-approved drugs are tested and where they ultimately become available to patients, raising concerns about the equitable distribution of research benefits at the population level.

Exclusion of Women of Childbearing Potential in Clinical Trials of Type 2 Diabetes Medications: A Review of Protocol-Based Barriers to Enrollment

OBJECTIVE

Women of childbearing potential are often excluded from participating in clinical trials owing to concerns about adverse fetal effects of treatment. This study aims to determine the prevalence of fertility-related exclusion criteria in clinical trials of type 2 diabetes medications and to determine whether these criteria are commensurate with drug risk.

RESEARCH DESIGN AND METHODS

ClinicalTrials.gov was queried for trials of type 2 diabetes medications that were phase 2 or 3, were based in the U.S., and enrolled participants 18-40 years old. Six hundred eighty-eight trials met criteria. Information collected about each trial included enrollment, trial length, exclusion and inclusion criteria, trial sponsor, and pregnancy category of drug(s) administered.

RESULTS

Most studies (59%) included one or more fertility-related exclusion criteria, most often excluding current pregnancy (55%) and breast-feeding (44%). Trials of medications with increased fetal risk were not more restrictive: trials of category C drugs (evidence of fetal risks in animals) were less likely to exclude pregnancy compared with trials of category B drugs (no known human or animal fetal risks) (45.6% vs. 69.8%, odds ratio [OR] 0.37 [95% CI 0.20, 0.65], P = 0.0005) or to require contraceptive use (29.9% vs. 57.1%, OR 0.32 [95% CI 0.18, 0.56], P = 0.001).

CONCLUSIONS

In clinical trials of type 2 diabetes medications, exclusion criteria affecting women of childbearing potential are often disproportionate to risk to the participant and fetus. These criteria have the potential to impede young women's access to clinical trials and may hinder the acquisition of clinical knowledge critical for improving the care of women with diabetes.

Gender bias in clinical research, pharmaceutical marketing, and the prescription of drugs

This thesis is part of the studies of gender bias in health which together with the paradigm of evidence-based medicine shares the empirical assumption that there are inaccuracies in medical practice, in addition to a lack of rigour and transparency. It worked with the distinction between the concepts of sex and gender and between the concepts of sex-related differences and gender inequalities, in terms of applying a gender perspective in the study design and the subsequent analysis. This PhD review presents the research process conducted in Spain, which can provide an example for future research. Study I described a review of 58 clinical trials (CTs) of etoricoxib to assess its compliance with the Recommendations of Evaluation of Gender Differences in the Clinical Evaluation of Drugs. In Study II, key informants from professions related to different areas in drug development and pharmacovigilance held a working meeting to reach a consensus document on recommendations for the study and evaluation of gender differences in CTs in Spain. In Study III, the websites of the eight best-selling hormone replacement therapy drugs in Spain on Google first page of results were analysed. In Study IV, a logistic regression analysis was performed to compare analgesic prescription by sex in regions with a higher or lower Gender Development Index (GDI) than the Spanish average. Gender biases identified in this thesis limited the legitimacy of medicine, which is not based on the best possible evidence. The results also demonstrate the existence of inequalities between men and women that are not due merely to biological differences, but are gender inequalities stemming from the social differences that exist between both sexes.

[Clinical trials with rofecoxib: analysis of the information from the gender perspective]

BACKGROUND AND OBJECTIVE

There is evidence if the low rate of participation or even exclusion of women in clinical trials (CT), and that sex-differences are not considered in the design and analysis of the CT. The objectives of the study were to determine whether women are properly represented in the CT with rofecoxib and to analyze the information of CT with rofecoxib from a gender perspective.

MATERIAL AND METHOD

Twenty eight rofecoxib CT in adults have been reviewed, all indexed in Medline and published between 1999-2001. The FDA Guideline for the Study and Evaluation of Gender Differences in the Clinical Evaluation Drugs was used to analyze the information.

RESULTS

An 80% of the trials do not describe efficacy results by sex, and only one reports side effects by sex. A 78.3% does not report stratified analysis by sex. In the discussion the possible difference by sex of the results is mentioned in 3 occasions. Only 8% of the CT considers the influence of hormonal variation in the results. The pharmacokinetics issues related specifically to women are poorly followed: in 60% of the CT it is not specified the influence of oral contraceptives in the results of the trial, and in 88.9% of CT it is not specified the influence of estrogen treatment in the results of the trial. Pregnancy as exclusion criteria is only considered in 50% of the trials.

CONCLUSIONS

CT with rofecoxib has included more women than men. Important information on specific situation related to gender, recommends by FDA Guideline for the Study and Evaluation of Gender Differences in the Clinical Evaluation Drugs, have not been followed.

Informed consent and the history of inclusion of women in clinical research

The purposes of this paper are to (a) discuss the troubled history of informed consent for research on women and its ramifications for women's participation in clinical trials; (b) interrogate current informed consent practices as to their accountability and justice in the treatment of women; and (c) recommend to nurse researchers and clinical nurses ways of improving the practice of informed consent in research with women.