20 February 2009. From clothing to computer chips, U.S. products are increasingly being made and tested overseas. The trend also applies to big pharma drugs, suggests a study published yesterday in the New England Journal of Medicine (NEJM). In their analysis of U.S.-based pharmaceutical companies, researchers led by Kevin Schulman at Duke University in Durham, North Carolina, found that lower research costs and access to larger patient populations have sent more and more clinical trials abroad, often to developing countries. The findings raise safety concerns and call for increased scientific rigor and harmonization of international standards for clinical research, the authors say.
Much has been written about the globalization of clinical trials in recent years. “We wanted to put our arms around to what extent this is actually occurring,” said lead author Seth Glickman in an interview with ARF. Glickman is a physician who does research in healthcare quality and health policy at University of North Carolina at Chapel Hill School of Medicine. He and colleagues mined the clinicaltrials.gov registry for Phase 3 clinical trials being conducted by the 20 largest U.S.-based pharmaceutical companies as of November 2007. In that analysis, about a third of the trials (157 of 509) were being run entirely outside of American soil, and more than half of the trial sites (13,521 of 24,206) lay outside U.S. borders.
When they reviewed 150 articles reporting clinical trial results in top-tier medical journals (NEJM, the Lancet, and the Journal of the American Medical Association) in 1995, the researchers found that 16.7 percent of trials were multinational. In a similar analysis of trials published in 2005, that figure had jumped to 29.3. Between 1995 and 2005, the number of countries serving as trial sites outside the U.S. more than doubled.
The report offered several reasons for this trend. One is the sheer cost savings of running Phase 2 and Phase 3 trials in developing countries where human labor is vastly cheaper. Another is the lure of quicker recruitment in highly populated countries with more potential research participants. Meanwhile, increasingly complex regulations governing the conduct of clinical trials are placing a higher administrative burden on researchers in many wealthy countries, further driving up the cost and time to run trials.
As more and more trials enter developing countries, there is also concern for decreased transparency. The authors suggest that investigators in developing nations may be less likely to publish trial results because of inexperience with international guidelines regarding study design and data access. Results dissemination may warrant concern even in developed countries. In a retrospective analysis comparing Phase 1 trials to larger, more definitive studies (Phase 2-4), researchers found that Phase 1 trials were more likely to be initiated and completed and more often produced confirmatory results, yet were still far less likely to be published. The findings emerged from a review of several hundred protocols approved in 1994 by ethics committees in France. Lead investigator Francois Chapuis of the University of Lyon, France, and colleagues reported the results in this month’s issue of the open-access journal PLoS Medicine (Decullier et al., 2009).
Aside from concerns that overseas trial data may not reach the public domain, Glickman and colleagues argue in the NEJM paper that the results themselves could be hard to interpret due to regional differences in medical infrastructure. “In the U.S. and other high-income countries, there is a robust healthcare system that ensures patients have access to healthcare during the course of the trial,” Glickman said. “The healthcare systems in developing countries are entirely different.” This could mean that drugs are being tested in certain patients with no prior treatment for the indicated condition, and in others who enter the trial having already taken other medications.
Trial results could also hinge upon genetic variants that may be more prevalent in certain parts of the world. “These genetic polymorphisms can affect drug metabolism and how effectively a drug might work,” Glickman told ARF. “It's entirely conceivable that the same drug being tested in the U.S. versus in a developing country could produce markedly different results.”
Others agree that cultural, healthcare, and genetic differences can shift trial outcomes. “I think that many of the concerns raised in this article are reasonable and appropriate,” said Paul Aisen, University of California, San Diego, who heads the NIA-funded Alzheimer’s Disease Cooperative Study.
Regional differences in life expectancy, for instance, become important in clinical trials of AD, a disorder whose primary risk factor is age. In the recent Phase 2 trial of Dimebon conducted in Russia (see ARF related news story), where men typically die in their sixties (compared to U.S. men who on average live to almost 80), the research participants had a mean age significantly lower than in U.S. and Western Europe trial sites. “That’s a difference that has to be examined and considered,” Aisen said.
However, he believes other issues highlighted in the new study “go too far.” One of these concerns the mismatch, in some cases, between the populations testing a compound and the intended market for that drug. Drugs for common diseases are increasingly being tested on people in developing countries but are primarily sold in wealthy nations. Conversely, “we didn't find many trials being conducted for diseases like tuberculosis or malaria, which disproportionately affect the developing world,” Glickman said.
Aisen does not see these discrepancies as deal breakers. “I don't believe it is unethical to use international trial sites that do not accurately reflect the ultimate market for the drugs being tested,” he told ARF. Aisen pointed out that though the U.S. Food and Drug Administration generally requires that some research subjects be in the U.S., there are no strict guidelines regarding what proportion. Beyond that, he contends that clinical research carries many benefits that do not depend on trial outcome. The participants receive extra care and attention, as well as new information, simply by taking part in a trial. Local physicians learn state-of-the-art management. In addition, global trials promote exchange of knowledge and experience between investigators in different countries. “International conduct of clinical trials benefits everyone,” Aisen said. “It benefits the local communities, and it benefits global health by supporting the development of new treatments.” He and the study authors agree that centralized oversight and improved international collaborations between investigators could help ensure that clinical trials adhere to global standards of scientific rigor and ethical behavior.—Esther Landhuis.
Glickman SW, McHutchison JG, Peterson ED, Cairnes CB, Harrington RA, Califf RM, Schulman KA. Ethical and Scientific Implications of the Globalization of Clinical Research. N Engl J Med 2009 Feb 19;360(8):816-23. Abstract