Diarrhea in Placebo Arms of Cancer Studies

Abstract

Background/aim: Diarrhea is among the most common adverse events in early oncology clinical trials, and drug causality may be difficult to determine.

Materials and methods: This is a systematic literature review of placebo arms of randomized cancer trials.

Results: Anemia was reported in 95 of 127 placebo monotherapy cohorts. Publications involving healthy volunteers and cancer prevention studies reported lower frequencies than those with cancer patients. The average reported frequency of diarrhea grade 1 or higher among studies in cancer patients was 15%. The maximal reported frequencies for grades 1, 2, 3, 4, 5 were 56, 24, 6, 2, and 0%, respectively.

Conclusion: When higher diarrhea frequencies than those are observed in treatment arms of clinical trials, then drug causality is likely.

Keywords: Meta-analysis; adverse events; cancer; oncology; placebo; randomized; review; systematic review.

Figure 1. Frequency of diarrhea reporting in published randomized clinical cancer trials by grade. Open circles: treatment arms, filled circles: placebo arms. The reported frequency of diarrhea grades 1 and 2 correlated closely with the frequency of grade 0. Those were excluded from the regression-based model used for imputation.

Figure 2. Frequency of diarrhea randomized clinical trials. Open circles: treatment arms, filled circles: placebo arms. Although the correlation between grade 1 and 2 versus grade 3 and 4 was mathematically detectable (R=0.51), it was not close enough to be used for imputation of missing variables.