Changes in primary outcome and sample size measures after initiation of accrual among trials supporting approval of drugs for hematological malignancies by the US food and drug administration

Changes in primary outcome and sample size measures after onset of patient accrual affects the scientific basis of evidence-based medicine. The FDA website was searched for trials supporting new hemato-oncology drug approvals from January 2010 to December 2017. Matching ClinicalTrials.gov entries were compared to identify modifications. Associated publications were reviewed for reporting of these changes. Of 69 included trials, five (7%) had post accrual modifications in primary outcome and 30 (43%) had modifications in planned sample size. Sample size was increased in 24 trials (median 66 patients, IQR 35–95, median relative increase of 40% from initial sample size) and was decreased in 6 trials (median 38 patients, IQR 27–60, median relative decrease of 25%). None of the primary outcome modifications and only 53% of the sample size modifications were reported in the related publications. Further improvement is warranted in order to achieve complete transparency in reporting landmark studies.