Safety of immune checkpoint inhibitors for cancer treatment: real-world retrospective data analysis from Qatar (SAFE-ICI-Q study)

Introduction Immune checkpoint inhibitors (ICIs) have significantly improved the therapeutic landscape of multiple malignancies. It becomes critical to understand the incidence, profile, and consequences of immune-related adverse events (irAEs) within real-world populations. Aim We aimed to assess the safety profile of ICIs in adult cancer patients at the National Center for Cancer Care and Research (NCCCR), Qatar, and explore the factors associated with irAEs, including the impact of irAEs on the survival outcomes. Methods This retrospective study included adult cancer patients who received at least one dose of an ICI between January 1, 2015, and January 1, 2020. Data was collected from electronic health records and institutional adverse drug reaction (ADR) reporting systems. irAEs were graded using Common terminology criteria of adverse events, version 5 (CTCAE v5). Logistic regression analysis was used to evaluate factors associated with irAEs. Kaplan–Meier and landmark analysis assessed associations between irAEs and progression-free survival (PFS) and overall survival (OS). Approvals were obtained from HMC IRB (MRC-01-20-251) and Qatar University IRB (073/2025-EM). Results A total of 236 patients (median age 57 years, 72% male) were included. Most patients had advanced solid tumors, with thoracic malignancies being the most common. Pembrolizumab was the predominant agent used. irAEs occurred in 55.9% of patients, with the most frequent side effects being endocrine (26.4%), dermatologic (13.5%), and hepatic (12.4%) toxicities. Sixteen patients (6.8%) experienced fatal irAEs, with pneumonitis being the most common cause of death. The median time to onset of irAEs was 55 days (IQR 16‐129.5 days). Most events occurred in the acute phase (21–180 days post-treatment). Resolution rates of irAEs varied, with gastrointestinal irAEs resolving in 92% of cases, compared to 40% for hematological events. Pulmonary irAEs were associated with the highest rate of treatment discontinuation. Factors associated with irAEs included a higher number of ICI treatment cycles (p=0.019), lower baseline and six-week platelet counts (p=0.015 and p=0.012, respectively), and elevated baseline TSH (p=0.048). In multivariable regression analysis, the only factor that remained statistically significant was the number of treatment cycles ( p = 0.004). Dermatologic irAEs were significantly more common among patients aged ≥65 years (17.9% vs. 7.1%, p=0.018). Patients with poor performance status (PS ≥ 2) experienced a significantly higher rate of cardiac irAEs compared to those with good PS (10.9% vs. 1.7%, p=0.036). In the 30-day landmark analysis, patients who developed irAEs had significantly worse PFS (3.3 vs. 7.1 months, p=0.0085) and OS (4.37 vs. 9.0 months, p=0.0004) compared to those without irAEs. These finding were confirmed using adjusted landmark analysis where irAEs were associated with worse OS (HR 2.13, 95% CI 1.34–3.3, P = 0.001) and PFS (HR 1.88, 95% CI 1.22–2.87, P = 0.004). Additionally, time-dependent Cox regression also demonstrated worse OS (HR 1.86, 95% CI 1.23–2.79, P = 0.003) and PFS (HR 1.96, 95% CI 1.41–2.72, P = 0.001). Conclusion In this real-world cohort, irAEs were frequent and clinically diverse. Using adjusted landmark analysis and time-dependent Cox regression, early-onset irAEs were associated with inferior survival in our cohort. Poor baseline PS was linked to an increased risk of cardiac irAEs. Older adults were at a higher risk of dermatological irAEs. Some factors such as higher number of ICI treatment cycles, thrombocytopenia and elevated TSH at baseline may aid in risk stratification. These findings reinforce the need for timely detection and multidisciplinary management of irAEs to optimize ICI safety and effectiveness.