The Study: This retrospective study specifically focused on the effect of baseline corticosteroids onboard at the initiation of immunotherapy. The cohort consisted of patients from MSKCC and Gustave Roussy receiving PD-L1 inhibitors for non-small cell lung cancer (NSCLC). A big advantage of this study was the quantification of steroid doses. At the start of immunotherapy, nearly 15% of patients were receiving at least a 10 mg prednisone equivalent (e.g., 2 mg of dex). This level of steroids was associated with one-third the response rate and half the median overall survival time compared with patients taking no steroids. In addition, there appeared to be continual disease-response relationships with both steroid doses as well as temporality of steroid use with starting immunotherapy. But let’s not forget the huge possible confounders in this study. At least some of the negative survival effect is probably related to the fact that patients on steroids had negative reasons for needing them: They were more likely to have brain mets, worse symptoms, and poor performance statuses.
Bottom Line: Notwithstanding the causation versus correlation question mark, patients who require steroids at initiation of immunotherapy have worse response to therapy and worse survival. | Arbour, J Clin Oncol 2018