Neutrophil‐to‐Lymphocyte Ratio Dynamics and Progression‐Free Survival in Advanced Cutaneous Squamous Cell Carcinoma Treated With Cemiplimab

Background Patients with locally advanced (la) and metastatic (m) cutaneous squamous cell carcinoma (CSCC) for which curative surgery or radiation is not available may benefit from treatment with anti–programmed cell death 1 (PD‐1) monoclonal antibody cemiplimab. The predictive and prognostic role of hematological markers in such patients remains unclear. Type of Study A real‐life retrospective cohort study was conducted to evaluate the role of hematological inflammatory parameters as predictive and prognostic factors in patients with laCSCC and mCSCC who underwent first‐line treatment with cemiplimab. Methods Lymphocyte–monocyte ratio (LMR), neutrophil–lymphocyte ratio (NLR), platelet–lymphocyte ratio (PLR), albumin, creatinine, systemic immune–inflammation index (SII), and LDH were measured prior to cemiplimab therapy and subsequently at the first follow‐up visit. To evaluate independent prognostic factors, the 12‐month time‐dependent performance was assessed using the incident/dynamic receiver operating characteristic (ROC) framework implemented with a nearest‐neighbor estimator. Model performance metrics, including Harrell’s C‐index and the time‐dependent area under the curve (AUC), were quantified with 95% confidence intervals. Results A total of 43 CSCC patients including 37 (86.1%) with laCSCC and 6 (14.0%) with mCSCC were retrospectively enrolled. After a median of 33.1 weeks (8.4 months) of exposure to cemiplimab, ORR was 72% (31/43), including 41.9% (13/31) complete responses and 58.1% (18/31) partial responses; stable disease was observed in 9.7% (3/31). The median progression‐free survival (PFS) was 6.4 months (IQR, 3.9–14.7 months). Overall, 7% of patients discontinued therapy due to adverse events (Grade 3 renal insufficiency and liver toxicity). We identified a significant inverse association between NLR change from baseline to first on‐treatment visit (ΔNLR) and PFS (Coeff = −0.25; 95% CI jackknife = −0.45–0.05, p = 0.014), showing higher ΔNLR score in patients with shorter PFS. The ΔNLR was associated with a 36% higher chance of progression (HR = 1.36, 95% CI = 0.97–1.89, p = 0.073), adjusted for age, systemic inflammatory index, and sex. Decision curve analysis showed a modest net benefit of the ΔNLR‐based model across a broad range of threshold probabilities (approximately 7%–49%), without identification of a clinically optimal decision threshold. Conclusion Our study suggests that ΔNLR may serve as an exploratory prognostic indicator in patients with laCSCC and mCSCC.