Systematic surveillance of immune-related adverse events in clinical practice and impact of subsequent steroid medication on survival outcomes

Background Recent advances in immune-checkpoint inhibitors (ICIs) have highlighted the need for effective management of immune-related adverse events (irAEs). This study aimed to conduct a systematic surveillance of real-world development of irAEs for understanding their characteristics and examine the prognostic impact of steroid use for these events. Methods We retrospectively investigated cancer patients treated with ICIs between 2014 and 2021 and collected information about irAEs throughout their development, management, and clinical outcomes. Results Overall, 458 patients (45.4%) developed 670 irAEs. The prevalence of irAEs varied by cancer type, but it was increased in regimens with longer treatment durations. Severe irAEs were more common in the nivolumab + ipilimumab and pembrolizumab + axitinib regimens. Patients who received steroids for irAEs at a dosage of < 2 mg/kg had comparable prognosis to those who did not receive steroids; however, patients who received methylprednisolone pulse therapy, primarily for severe pneumonitis and hepatitis, had shorter overall survival than those who did not receive steroids (7.8 versus 23.4 months, p = 0.016). Furthermore, methylprednisolone pulse therapy for irAEs was a poor prognostic factor in multivariate analysis (hazard ratio: 2.19, 95% confidence interval: 1.34–2.86, p < 0.001). Conclusion Steroid treatment for irAE does not affect prognosis and should thus be used promptly to control inflammation. However, pulse therapy for severe cases is a poor prognostic factor, and early detection remains the key to managing such irAEs. The irAE characteristics in each regimen should be clarified to establish and provide more sophisticated irAE management, and the current findings will be beneficial to this goal.