Quality of reporting of phase II trials in oncology in highly ranked oncology journals.

Abstract

6582 Background: Phase II trials represent an essential step in the development of anti-cancer drugs. The aim of this study was to assess the quality of their reporting in highly-ranked oncology journals, to investigate predictive factors of quality and to develop better reporting guidelines for authors. Methods: We reviewed the tables of contents of 8 peer-reviewed oncology journals published between January 2011 and December 2011 and with a 2011 impact factor (IF) >4: Br J Cancer, Eur J Cancer, Oncologist, J Clin Oncol, Lancet Oncol, JNCI, Clin Cancer Res and Ann Oncol. Two reviewers assessed the quality of each report by using a 44-point overall quality score (OQS; range, 0 to 44 points) inspired from the revised Consolidated Standards of Reporting Trials statement. Primary endpoint definition, justification of sample size and definition of the evaluable population for each endpoint, were assessed separately because of their crucial methodological importance using a 3-point key methodological score (KMS; range, 0 to 3). Exploratory analyses were used to identify predictive factors associated with the different scores. Results: 156 articles were included. Agreement between the reviewers for each item was good (kappa coefficient range: 0.62-1). The median OQS was 28 (range: 9-35). OQS sub-score analysis showed that reporting of statistical methods was particularly low with a mean of 2.5 (6 items). The median KMS was 2 (range 0-3). Primary endpoint definition, justification of sample size and definition of the evaluable population were reported only in 107 (68.6%), 121 (77.6%), and 52 (33.3%) cases, respectively. On multivariate analysis, reporting of clinicaltrial.gov registration was associated with improved OQS, OR = 3.2 (95CI, 1.5 to 7.1). No predictive factor for KMS were identified. Conclusions: Phases II trials reporting is still poor even in journals with strict editorial policies. This may lead to biased interpretation of phase II trial results. We have developed a checklist for use by authors, reviewers, and editors to improve reporting of these studies. As well as using a checklist during the preparation of their manuscript, we recommend that authors provide reviewers and readers with the last version of the study’s protocol.