Background: Corticosteroid is effective in treating immune checkpoint inhibitor (ICI)-induced immune-related adverse events (irAEs). However, whether its earlier use prevents irAEs remains unknown. In real-world practice, non-small cell lung cancer (NSCLC) patients receiving combined immunochemotherapy are often given corticosteroids right after ICI injection to reduce subsequent chemotherapy-related adverse events. It is thus an ideal model to examine whether this earlier use of corticosteroid also reduce irAEs following immunotherapy. Methods: NSCLC patients receiving at least one cycle of ICI with or without chemotherapy were enrolled. Patients receiving ICI with corticosteroid corticosteroid premedication exposure were categorized as “corticosteroid group (CS)”, whereas patients receiving ICI without corticosteroids premedication exposure were categorized as “non-corticosteroid group (non-CS)”. Frequency of irAEs prompting systemic corticosteroid or leading to ICI discontinuation were compared between the two groups. Exploratory survival analysis was performed. Results: Among 252 eligible patients, 137 patients were categorized as CS group, and 115 patients as non-CS group. 21.9% of patients in the CS group and 30.4% of patients in the non-CS group developed irAEs prompting systemic corticosteroid use (Odds ratio [OR], 0.64; 95% confidence interval [CI], 0.35 to 1.18; p=0.15). 5.8% of patients in the CS group and 15.7% of patients in the non-CS group developed irAEs leading to ICI discontinuation (OR, 0.34; 95% CI, 0.12 to 0.85; p=0.013). In the exploratory survival analysis, early short-course corticosteroids was not associated with a higher risk for death in first-line patients (hazard ratio [HR], 1.06; 95% CI, 0.56–2) or patients receiving immunochemotherapy (HR, 0.72; 95% CI, 0.37–1.45). Conclusions: Early short-course corticosteroids following each ICI injection may reduce the rate of irAEs that leads to ICIs discontinuation. The use of prophylactic corticosteroids on irAEs in cancer patients receiving ICIs should be evaluated in future prospective studies.