Do corticosteroids affect immunotherapy efficacy in malignancy? – A systematic review

Abstract Background: Early studies indicated that corticosteroids may limit the survival benefit from immunotherapy. We conducted this systematic review to evaluate the effect corticosteroids have on immunotherapy in patients with malignancy, when adjusted for potentially confounding effects of corticosteroids given for palliative indications. Methods: 3 electronic databases (PubMed, Embase, and Medline) were searched on 1st of February 2023. Studies that measured response or survival to immunotherapy in people receiving corticosteroids for non-cancer indications compared to either no corticosteroids or corticosteroids for cancer-related indications were included. Studies exclusively evaluating the effect of corticosteroids administered for immune-related adverse events (irAE) were excluded to avoid immortal time bias. Pooled odds and hazard ratios with 95% confidence intervals (CI) were calculated using a random effects model. Study heterogeneity was assessed using the I2 statistic, and publication bias was evaluated by funnel plot and Egger's regression model. Results: 8426 titles were identified on our search. 8 studies met our inclusion criteria for meta-analysis. Administration of corticosteroids does not have a statistically significant effect on survival and response to immunotherapy when administered for non-cancer related indications, with a pooled odds ratio for overall response rate 0.92 (95% CI 0.58-1.44); pooled hazard ratio (HR) for progression free survival 0.87 (95% CI 0.68-1.12); and pooled HR for overall survival 0.79 (95% CI 0.59-1.05). Conclusion: This systematic review indicates that administration of corticosteroids does not affect response to immunotherapy nor survival outcomes, when removing confounding palliative corticosteroid indications. These results are limited by the retrospective nature of the studies included, small sample sizes, lack of information about corticosteroid dosing the inclusion of irAE in 2 of the studies which could bias the results.