Immune checkpoint inhibitors (ICIs) are pharmacological agents indicated for recurrent and metastatic head and neck squamous cell carcinoma (HNCSCC). Immune-related adverse events (irAEs) have been reported as predictors of therapeutic response to ICIs. However, previous studies have not adequately addressed the immortal time bias. Therefore, we aimed to investigate the association between the onset of irAEs and oncological outcomes, accounting for immortal time bias. We conducted a retrospective study involving 130 patients with HNSCC who were treated with ICIs. The objective response, progression-free survival (PFS), and overall survival (OS) were assessed using logistic regression analysis, the Kaplan-Meier method, and the Cox proportional hazard (PH) model. The immortal time bias was considered using a landmark analysis and an extended Cox (EC) model. The odds ratios for response and disease control were smaller in the landmark than in the naïve analyses. In the landmark analysis, the 1-year PFS rates were 47.6% and 27.2% for irAE+ and irAE- patients, respectively (p = 0.049), and the 1-year OS rates were 85.7% and 66.5%, respectively (p = 0.006). Regarding PFS, the adjusted HRs for irAEs were 0.49 (95% confidence interval (CI) 0.28-0.85) in the PH analysis and 0.75 (95% CI 0.40-1.40) in the EC analysis. As for OS, the adjusted HRs for irAEs were 0.36 (95% CI 0.19-0.66) in the PH analysis and 0.51 (95% CI 0.27-0.95) in the EC analysis. IrAEs were an independent prognostic factor for OS but not PFS. Without considering the immortal time bias, the association between irAEs and oncologic outcomes in patients with HNSCC treated with ICIs was overestimated. Therefore, the balance between the benefits and risks of ICI therapy must be carefully weighed in clinical settings.
Copyright: © 2024 Tamura et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.