Background: Despite their anti-tumor efficacy, immune checkpoint inhibitors (ICIs) are associated with a variety of immune-related adverse events (irAEs). Grade ≥2 irAEs require ICI discontinuation. The decision to resume ICI treatment often remains challenging. Methods: We retrospectively studied 1051 adult patients with stage IV NSCLC treated with ICIs at West Cancer Center between January 2015 and December 2020. 99 (9.4%) patients with grade≥2 irAEs necessitating treatment interruption were identified. 40 patients underwent retreatment (rechallenged group), while 59 discontinued treatment (discontinued group). All patients were rechallenged with the initial ICI. Results: The baseline characteristics of patients in the two groups were similar in terms of age, gender, ethnicity, ECOG, smoking status, cancer histology, type of ICI, and PD-L1 status. No differences in types of irAEs or timing of the initial irAEs were seen. Initial irAEs were less severe in the rechallenged group, with fewer grade 3 irAEs (20% vs. 62.7%, p < 0.001), no grade 4 irAEs (0% vs. 6.8%, p < 0.001), less IV corticosteroid use (2.5% vs. 23.7%, p = 0.01), and less frequent hospitalizations for irAEs (7.5% vs. 42.4%, p < 0.001). After rechallenging, 24 of 40 (60%) patients had recurrence of the same or de-novo irAEs. 20 (50%) developed second grade≥2 irAEs with 13 (32.5%) grade 2, and 7 (17.5%) grade 3. No grade 4 irAE or irAE-related death occurred after rechallenging. Among the 24 patients with second irAEs, 6 (25%) still had residual grade 1 irAEs upon rechallenging, while none of the 16 patients without second irAEs had residual irAEs (p = 0.03). With a median follow-up of 572 days, no statistically significant differences in overall survival (p = 0.67, HR 1.15 with 95% CI 0.60-2.22) or PFS (p = 0.57, HR 0.83 with 95% CI 0.43-1.59) were noted between the two groups, which were adjusted for age, gender, ethnicity, ECOG, PD-L1 status, grade of initial irAEs, and best objective response (OR) prior to initial irAEs. Conclusions: In real-world practice, clinicians tend to rechallenge patients with less severe irAEs. Our study showed that rechallenge was associated with a relative high risk of second grade≥2 irAEs. The risk was less if the initial irAEs were resolved. The second irAEs were not more severe than the first. Rechallenge can be performed under close monitoring, ideally, after the initial irAEs have completed resolved. The survival outcomes of ICI rechallenge and discontinuation were similar. ICI discontinuation is one feasible strategy after grade≥2 irAEs, especially severe irAEs.