Self-reported Conflicts of Interest and Trial Sponsorship of Clinical Trials in Prostate Cancer Involving Radiotherapy

OBJECTIVES To examine the association between trial sponsorship and conflicts of interest (COI) with clinical trial conclusions for prostate cancer trials related to radiotherapy. MATERIALS AND METHODS The MEDLINE database was searched for all prostate cancer clinical trials published between 2004 and 2013 and identified 1396 studies. Two investigators independently identified trials published in the English language of ≥30 patients, and extracted relevant data. Clinical trials were classified according to trial characteristics, sponsorship source and type, COI, and study conclusion, and analyzed by univariable and multivariable logistic regression. RESULTS Of 240 eligible trials, 160 (67.5%) evaluated drugs without radiotherapy, 60 (25%) involved radiotherapy, and 18 (7.5%) involved procedures without radiotherapy. Of the 60 radiotherapy trials eligible for analysis, positive sponsorship and potential COI were present in 58.3% and 20% of trials, respectively. Study conclusions were positive, negative, or neutral in 78.3%, 5%, and 16.7% of trials, respectively. No association was found between positive conclusions and either industry support of potential COI. Positive conclusions were reported in 86.7% and 83.3% of trials with sponsorship and COI, respectively, as compared with 75.6% and 77.1% of those without sponsorship (P=0.37) and COI (P=0.64). Sponsorship was significantly associated with radiotherapy trials combined with drugs (odds ratio 5.5, P=0.01) and higher-risk disease (odds ratio 4.71, P=0.01). CONCLUSIONS The presence of sponsorship was associated with radiotherapy trials involving drugs or studying higher-risk prostate cancer. However, there were no identified associations between study conclusion and sponsorship type or COI.