[Skip to Navigation]
[Skip to Navigation Landing]
Views 154
Citations 0
Comment & Response
December 13, 2018

Inappropriate Grading of Adverse Events in Cancer Clinical Trials—Reply

Author Affiliations
  • 1Department of Oncology, Lausanne University Hospital, Lausanne, Switzerland
  • 2Statistics Department, EORTC Headquarters, Brusells, Belgium
JAMA Oncol. 2019;5(2):269-270. doi:10.1001/jamaoncol.2018.5871

In Reply We thank Dr Zhang and colleagues for their comment regarding the grading of adverse events in the phase 3 PETACC-3 trial, which compared biweekly infusional fluorouracil/leucovorin alone or in combination with irinotecan as adjuvant treatment of stage 3 colon cancer.1 We agree with Dr Zhang that the quality of adverse event grading is a critical issue in clinical trials, especially when evaluating factors influencing toxic effects. However, in contrast to what Zhang and colleagues assume, the investigators of the PETACC-3 trial1 graded alopecia correctly as mentioned in our article2; adverse events were graded according to the National Cancer Institute of Canada Common Toxicity Grading expanded common toxicity criteria (version revised in May 1991, which was detailed in the PETACC-3 protocol but is no longer available for reference). In this version, alopecia is graded from grade 1 (mild hair loss) to grade 3 (total body hair loss). Therefore, no correction is needed; in 2974 patients receiving chemotherapy, 1.4% of women vs 0.4% of men presented with an alopecia grade greater than 2.2 It is in the Common Toxicity Criteria version 2.0 and later that alopecia was graded as either grade 1 (mild hair loss) or grade 2 (pronounced hair loss).3 Moreover, we would like to mention that, regardless of the number of categories in which alopecia is graded, a bias in the comparative assessment of the severity of alopecia between men and women is likely to be introduced by differences in baseline condition, with men more often having alopecia at baseline. Furthermore, sex differences in the subjective perception of the severity of alopecia might also contribute to a bias because women might pay more attention to their hair loss. However, the significant differences in neutropenia described in our article clearly indicate that significant sex differences are present in chemotherapy toxic effects, including objectively measurable toxic effects.