This drug kills patients! Avoiding sensationalism in clinical trials

Apr 19, 2010Michael O'Riordan

Durham, NC - Reporting of important clinical trials needs to avoid unnecessary hype and fear-mongering and proceed in a scientific manner that helps define the risks and benefits of new therapies [1]. These are the conclusions from a group of researchers addressing the recent public sensationalism of clinical trials, particularly the emergence of unexpected cancer findings with ezetimibe (Zetia, Merck/Schering-Plough Pharmaceuticals) and prasugrel (Effient, Eli Lilly/Daiichi Sankyo).
"For those of us who conduct clinical trials, over the past several years there have been a number of sensational items in the lay and medical press about associations and unusual findings," Dr Magnus Ohman (Duke Clinical Research Institute, Durham, NC), the lead author of the commentary, told heartwire. "These unusual findings have always occurred in clinical trials, and we're going to continue to see these events, but what has changed is that it has become very difficult to separate the wheat from the chaff."
Publishing the report online April 19, 2010 in the American Journal of Medicine, Ohman and coauthors Drs Matthew Roe (Duke Clinical Research Institute), Paul Armstrong (University of Alberta, Edmonton), Keith Fox (Royal Infirmary, Edinburgh, UK), and Dorairaj Prabhakaran (Center for Chronic Disease Control, New Delhi, India) point specifically to the premature release of data in the ezetimibe studies.
As reported extensively by heartwire, results from the Simvastatin and Ezetimibe in Aortic Stenosis (SEAS) showed a significant increase in the risk of cancer, a finding that triggered an analysis of two large ongoing studies, the IMPROVE-IT and SHARP trials. Sir Richard Peto (Clinical Trials Service Unit, Oxford, UK), an expert in clinical-trial meta-analyses and cancer epidemiology, performed an analysis of SEAS, IMPROVE-IT, and SHARP and found there was no overall risk of developing cancer with ezetimibe/simvastatin. He concluded the findings were likely due to chance and should not divert anybody from taking the drug.
Despite the assurances from the analysis, Ohman noted that some clinical sites dropped out of the trial, and with the public release of the "cancer concern," some patients may have stopped other evidence-based lipid-lowering therapies. Still, the ezetimibe story is an example where the scientific community responded to the issues in an appropriate manner, he said. With the cooperation of the data safety and monitoring boards (DSMBs) from all three clinical trials, researchers were able to "better define the risk/benefit ratio" of ezetimibe.
"We already have the structures in place to safeguard patients and to get good results from well-designed clinical trials," Ohman, referring to the DSMB, told heartwire. "When these issues come up, we need to alert the DSMB about these risks and to ask them about it. If the monitoring board feels there is no risk or the findings are due to chance, we need to respect their decision. We have to create some sort of way to recognize the value of the DSMB."

The prasugrel experience

In their commentary, the authors also highlight the higher rate of colonic neoplasms in patients treated with prasugrel compared with those treated with clopidogrel in the Trial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibition with Prasugrel (TRITON-TIMI 38). However, a closer analysis and clarification by site investigators reduced the number of cancers diagnosed since the start of the study medication, and when this was done the difference between prasugrel and clopidogrel was no longer statistically significant. Not all media or statements by medical experts accurately reported the statistically nonsignificant nature of the increased risk, however, and this led to some confusion about the risk/benefit trade-offs with prasugrel.
In their commentary, Ohman, Roe, Armstrong, Fox, and Prabhakaran write that there was no shortage of differences in medical opinion with regard to ezetimibe and prasugrel but stress that researchers and clinicians need to avoid causing undue alarm with the public and can do this by using the word "significant" in a scientific way (ie, when p<0.05).
In addition, clinicians, the media, and the public should be aware that risk factors for coronary heart disease, such as age, inactivity, obesity, and poor diet, are similar to the risk factors for cancer. To heartwire, Ohman noted that this issue is particularly important for the TRILOGY ACS trial, a study of prasugrel vs clopidogrel in medically treated acute coronary syndrome patients. This has led investigators to prospectively collect cancer information at baseline and throughout the course of the trial with a detailed case-report form. The intention, said Ohman, is to "scientifically address the question, because in the past few years it hasn't been addressed categorically."
Ohman said that there is an increased demand for medical news, and researchers are asked more and more to prepare and present their findings for the media. Most of this research, including an appropriate understanding of a drug's risks and benefits, can't be explained in a headline, he said. As a result, doctors should provide some insight and explanation about the findings so that reporters can properly explain the results, including the significance of uncovered risks, to their readers.
Ohman disclosed receiving research and or grant support from Bristol-Myers Squibb, CV Therapeutics, Daiichi Sankyo, Datascope, Eli Lilly, Sanofi-Aventis, Schering-Plough, and the Medicines Company; consulting fees from Abiomed, AstraZeneca, CV Therapeutics, Datascope, Gilead Sciences, Liposcience, Northpoint Domain, Pozen, Response Biomedical, the Medicines Company, and WebMD. Disclosures for the coauthors are listed in the paper.

Fuente: The Heart.org