HOUSTON, June 8 — More than a decade after the government urged scientists to enroll more women in clinical research projects, females were still significantly under-represented in clinical trials involving nonsex-specific cancers, a review of more than 600 studies showed.

Women accounted for less than 40% of patients in these trials and were significantly lacking in trials involving six of seven nonsex-specific cancers, Reshmi Jagsi, M.D., D.Phil., of the University of Michigan in Ann Arbor, and colleagues reported online in Cancer.

The researchers also found that trials with government funding had significantly more women than privately funded studies.

• Explain to patients that this study showed the fewer women were represented in published high-impact studies of cancer than would be predicted by the sex distribution of the cancer.
• The authors did not address potential reasons for the disparity.

“Only by understanding the forces affecting the sex distribution of study subjects can we as a society succeed in ensuring that our medical research efforts are inclusive and to the benefit of all,” the authors concluded.

The Revitalization Act of 1993 emphasized the need to include women in clinical research projects. In a 2001 report, the Institute of Medicine identified gender as a basic variable in the design and analysis of clinical studies.

Whether women are adequately represented in clinical trials remains an unresolved issue, Dr. Jagsi and colleagues said.

For example, a review of federally funded studies published in medical journals in 2004 showed that women constituted 37% of participants in nonsex-specific trials.

Studies of participation in cancer-specific research have yielded conflicting results.

So, to update the current status of women’s representation in clinical cancer studies, the authors reviewed articles published in 2006 in eight major journals.

They identified 1,534 cancer research articles, including 661 pertaining to prospective clinical studies involving a total of 1,096,098 participants. The prospective studies formed the basis for the primary analyses.

The authors analyzed the studies with respect to seven nonsex-specific cancer types: central nervous system, gastrointestinal, head and neck, hematologic, lung, sarcoma, and urinary.

They compared women participants in the studies to the number of females that would be expected on the basis of sex distribution for that that type of cancer.

The estimated incidence of specific cancers in women was based on data from the U.S. Bureau of Census and the National Cancer Institute.

“For six of seven cancer types analyzed, the overall percentage of women among treatment study participants was significantly (P<0.001) lower than the percentage expected based on general population incidence data,” the authors said.

Overall, women accounted for 38.8% of participants in studies of nonsex-specific cancers. Studies focusing on treatment had an even lower percentage of women compared with studies focusing on other aspects of cancer (36.9% versus 42.2%, P=0.004).

Female participation was higher in government-funded studies: 41.3% versus 36.9% in studies funded by other sources (P=0.005).

Studies whose first or senior author was a woman also had a higher proportion of female participants than studies led by males (41.3% versus 37.3%, P=0.01).

In a multivariate analysis, lack of government funding and a focus on cancer treatment were independent predictors of a lower percentage of female participants (P=0.03).

Limitation of the study included focus on high-impact published studies in a single year, assessment of participants in published articles rather than trial enrollment, inability to evaluate other factors such as race and socioeconomic status, and inclusion of both study as a whole and individual participants as units of measure.

The authors reported no potential conflicts of interest.

Primary source: Cancer

Source reference:
Jagsi R, et al “Under-representation of women in high-impact published clinical cancer research” Cancer 2009; DOI: 10.1002/cncr.24366.