More Transparency, Standards Urged for Research

Even though thousands of clinical trials are listed on public websites, and sponsors are posting results from more key studies, consumer advocates continue to press for broader public disclosure of information on research involving human subjects. Clinicaltrials.gov operated by the National Institutes of Health (NIH) now holds data on more than 130,000 trials, plus about 6,000 reports on study results. But that does not include many Phase I and exploratory studies or trials conducted overseas. Sponsors, moreover, don't have to report the results of studies for products that never gain approval from the Food and Drug Administration. Consumer and women's health groups feel that these "loopholes" allow pharma companies to hide safety problems and publish biased reports.

To remedy the situation, Rep. Edward Markey and several prominent Democrats have introduced legislation, the Trial and Experimental Studies Transparency (TEST) Act (HR 6272), that defines "applicable clinical trials" as all interventional biomedical trials involving human subjects. That extends registration to very early trials and foreign studies cited in applications to FDA. The measure further clarifies that results have to be posted within one year of trial completion, or two years for a new therapy, even if the product never comes to market. Similar legislation proposed in May 2012 aimed to expand registration of trials funded by NIH by threatening to withhold federal grants to any entity failing to report study results.

Pharma companies claim that registration of early trials and foreign studies provides little benefit for US patients, and that some details on trial protocols and informed consent documents should remain confidential. Sponsors also object to disclosing results from studies on products that never gain FDA approval. Instead, industry has sought to address concerns about unethical research practices and hiding unfavorable results by establishing voluntary standards for research transparency and ethical publication practices. The Medical Publishing Insights and Practices Initiative (MPIP), funded by nine leading pharma companies, issued recommendations last May to raise the credibility of pharma clinical research. Among other things, the guidelines call for sponsors to report all clinical research results, including unfavorable data and adverse events, and to make statistical analytical methods fully transparent.

Industry critics contend that the MPIP policy lacks teeth and doesn't go far enough to ensure ethical research practices. Backers of the Markey legislation claim that much useful information can be gleaned from clinical trials that experience safety problems and fail to support product registration, and that full disclosure can prevent other investigators from repeating similar unsafe and unproductive studies.

Without bi-partisan backing and support from the biomedical research community and patient and disease groups, the measure is unlikely to gain passage in the near future. Yet, any new scandal involving drug safety or research practices is sure to boost public interest in proposals to boost transparency in clinical trials.

Clarifying outcomes research

The drive for transparency in biomedical research is also shaping methods for conducting health outcomes and treatment research, particularly with the rise in government funding for such studies by the Patient Centered Outcomes Research Institute (PCORI). To help the research community use the best methods in designing and conducting real-world outcomes studies and in reporting results, the PCORI Methodology Committee recently issued a draft proposal for 60 standards that reflect the patient perspective in shaping research approaches. The standards address deficiencies or inconsistencies in how studies are conducted in practice, with an eye to reducing bias. Additional methodological reports will examine these issues from the perspectives of clinicians, payers, and policymakers.

Methods for dealing with heterogeneity of treatment effect are discussed, along with the need to define subgroup variables and expected treatment effects in protocols. The report also describes methods for preventing and handling missing data, a problem more common in large outcomes research than in controlled trials. There is guidance on selecting appropriate study designs and for planning and designing adaptive clinical trials, which may encourage broader use of such research methods in outcomes research.

By and large, the outcomes studies funded by PCORI will not be randomized controlled trials of the sort conducted to meet FDA standards for approval, explains Robert Dubois, Chief Science Officer at the National Pharmaceutical Council, which closely follows PCORI activities. Yet FDA approval of a new drug "doesn't mean that anyone's going to pay it," Dubois comments, "and the PCORI methods guide gets to that."