Pulmonary immune-related adverse events post-immune checkpoint inhibitors (ICI): Real-world experience.

Abstract

11166 Background: Immunotherapy (IT) has transformed cancer treatment and is often better tolerated than chemotherapy; however, immune-related pulmonary adverse events (Pulm-AE) have emerged as clinically significant and potentially treatment-limiting toxicities that may impact morbidity, treatment continuity, and survival. Real-world comparative data on Pulm-AE risk remain limited. We evaluated the incidence, cumulative risk, severity, and clinical outcomes of Pulm-AEs in patients receiving IT versus non-IT systemic therapies. Methods: We conducted a retrospective cohort study of 5,556 adult cancer patients treated at the Cleveland Clinic between 2010–2022. Included cancers were bladder (stage II–IIIA n = 476; IV n = 158), endometrial IV (n = 157), esophageal/gastric (II–III n = 797; IV n = 590), head & neck IV (n = 36), renal (III n = 88; IV n = 223), metastatic melanoma (n = 45), and lung cancers: non-squamous NSCLC (IIIB n = 206; IV n = 1,609), squamous NSCLC (IIIB n = 182; IV n = 366), and SCLC IIIB–IV (n = 623). Patients receiving IT (± chemotherapy or radiation) were compared with non-IT regimens and followed from treatment initiation to Pulm-AE, death, or last encounter. Pulm-AEs included pneumonitis, pleural effusion ± pleuritis, and interstitial lung disease. Data were collected on demographics, CTCAE grade, timing, management, and outcomes. Incidence rates per 100 person-years (py) were estimated using Poisson regression, cumulative incidence (CIR) using Kaplan–Meier methods, and adjusted hazard ratios (HRs) using time-dependent Cox models. Results: Of 5,556 patients, 2,133 (38%) received IT, including 1,291 (60%) as first-line therapy; 99.9% of IT exposure was ICI. The cohort was 38% female, 87% White, with a median age of 66 years. Over a median follow up of 14 months (CI: 6-35), 127 Pulm-AEs occurred, including 51 among first-line IT recipients. Pulm-AE incidence was higher with IT than non-IT (0.2 vs 0.06 per 100 py; P < 0.05), driven primarily by pneumonitis (0.16 vs 0.04 per 100 py; P < 0.05). At 24 months, CIR for Pulm-AEs was 5.5% with first-line IT versus 2.0% without (HR 3.50, P < 0.001); pneumonitis CIR was 4.3% versus 1.3% (HR 5.21, P < 0.001). Among pneumonitis cases, 86% were CTCAE grade 2–4, 81% required steroid therapy, and 28% resulted in permanent IT discontinuation, frequently necessitating long-term immunosuppression. Other Pulm-AEs in the IT & non-IT groups were pleural effusion (n = 10, 24), pleuritis without effusion (n = 0, 2), and interstitial lung disease (n = 0, 2). Conclusions: Receipt of IT was associated with a significantly increased risk of Pulm-AEs, primarily clinically significant pneumonitis. These toxicities were common (~7% by 24 months), frequently severe, and often led to irreversible damage and treatment cessation, underscoring the need for vigilant monitoring, early recognition, multidisciplinary management, and prompt intervention to minimize morbidity.