Risk of venous thromboembolism in resectable non-small cell lung cancer treated with neoadjuvant and adjuvant immunotherapy: A self-controlled case series.

Abstract

e20100 Background: In October 2021, atezolizumab was approved for adjuvant treatment of non–small cell lung cancer (NSCLC) with PD-L1 expression ≥1%. Since then, pembrolizumab and nivolumab have also been approved for neoadjuvant and adjuvant treatment in resectable NSCLC. The approval of immunotherapy (IO) to treat NSCLC has increased survival and improved outcomes in this population. However, the relationship between IO and VTE risk remains incompletely characterized. Here, we studied the risk of VTE in patients with resectable NSCLC receiving PD-1 or PD-L1 inhibitors in the neoadjuvant and adjuvant settings. Methods: We conducted a retrospective study at an urban academic center that included patients with resectable NSCLC who received neoadjuvant or adjuvant PD-1 or PD-L1 inhibitors between January 2022 and December 2025. Statistical analyses were performed using GraphPad Prism (10.6.1) and IBM SPSS Statistics (30.0.0). Fisher’s Exact and t-test analyses were conducted as indicated. This study utilized a self-controlled case series design in which each patient served as their own control to minimize confounding factors. A control window of one year prior to IO initiation was compared with a risk window of the year following initiation. Results: Twenty-six patients were included who received PD-1 or PD-L1 inhibitors in the neoadjuvant or adjuvant setting. VTE occurred in 11.5% of patients after IO initiation. Incidence of VTE following IO had an incidence rate ratio of 1.5, with incidence rate of 7.69 per 100 person-years prior to IO and 11.54 per 100 person-years following start of IO. The mean time to VTE was 267 (SD = 64.26) days. Though not reaching statistical significance, patients with neoadjuvant IO exposure had higher odds of VTE development compared to those with adjuvant exposure (OR = 9.50; 95% CI [0.83, 142.60]; p = 0.123). Average Khorana score (KS) in those without VTE development was 1.48 (SD = 0.67) while average KS in those who developed VTE was 2.33 (SD = 1.53). Conclusions: This study is novel in its exploration of VTE incidence in NSCLC treated with neoadjuvant or adjuvant IO. The rate of VTE development in this study was consistent with previously reported real-world VTE incidence (between 6% to 13%) in NSCLC patients with platinum-based doublet chemotherapy, but higher than reported VTE rate in major phase 3 clinical trials (between 2% to 4%). These results suggest a signal between neoadjuvant IO use and increased VTE risk which warrants further investigation in a larger prospective study. Additionally, the results demonstrate that utilization of KS could assist in identification of patients at higher risk for VTE development in this population. Future studies are needed to assess whether those at highest risk for VTE development may benefit from thromboprophylaxis prior to IO initiation.