Immune checkpoint inhibition for advanced cutaneous squamous cell carcinoma in patients with concomitant chronic lymphocytic leukemia: Integrated single-center and national Cosmos analyses.

Abstract

e21565 Background: Patients with chronic lymphocytic leukemia (CLL) have immune dysregulation and increased risk of aggressive cutaneous squamous cell carcinoma (cuSCC), yet prospective ICI trials largely excluded this population. We combined detailed single-center chart review with a national real-world dataset to characterize safety, hematologic trajectory, imaging patterns, and overall survival (OS) with PD-1 blockade in cuSCC with concomitant CLL. Methods: We combined a single-center retrospective chart review (2017-2025) of adults with CLL and advanced cuSCC treated with PD-1 blockade and a national Cosmos EHR cohort of ICI-treated cuSCC (2017-2025) comparing patients with vs without CLL comorbidity. Patients with other cancer types were excluded. Primary endpoint was overall survival (OS) from ICI start. Within the Cosmos dataset, propensity matching on age and sex and COX regression were used. Data used in this study came from and is available through Epic Cosmos, a dataset created in collaboration with a community of health systems using Epic representing more than 300 million patient records from over 1,800 hospitals and 40,000 clinics as of December 2025. Within the single-center cohort we also looked at longitudinal CBCs, irAEs, CLL treatment initiation, and FDG-PET/CT patterns. Results: In Cosmos, 10,540 ICI-treated cuSCC patients were identified; 226 (2.1%) had CLL comorbidity. CLL patients were older (mean 80.20 vs 71.88 years) and more often male (85.0% vs 70.9%). ICI selection differed (Cemiplimab 45.1% with CLL vs 21.3% without; pembrolizumab 54.9% vs 78.7%). Unadjusted mortality was 46.9% with CLL vs 44.0% without (p = 0.417). In an age/sex propensity-matched sample (n = 452), OS was similar between groups (Cox HR 1.06, 95% CI 0.80–1.39; p = 0.696). In the single-center cohort (n = 10; median age 80.5), any irAE occurred in 5/10, including 2 grade ≥3 events, with no treatment-related deaths. Longitudinal CBCs and hematologic Rai groupings were generally stable without a consistent pattern suggestive of leukemic acceleration; 3/10 initiated CLL-directed therapy after ICI. FDG-PET/CT (9/10) typically showed intensely FDG-avid cuSCC lesions with lower-grade uptake in background CLL adenopathy, though occasional SUV overlap required clinicopathologic correlation. Conclusions: This integrated analysis provides, to our knowledge, the largest description of ICI-treated advanced cuSCC in patients with concomitant CLL. In a national real-world cohort, concomitant CLL was not associated with inferior OS after age/sex matching. Single-center demographics mirrored the national cohort (older, predominantly male), supporting generalizability of the detailed toxicity, hematologic, and imaging findings while emphasizing biopsy when SUVs overlap.