Optimizing selection of real-life NSCLC patients for ICI is necessary in order to get more patients to obtain long-term effect. Many NSCLC patients are ≥ 75 years old with tobacco-related comorbidity, poor performance status (PS) and widespread metastatic disease. These factors may increase the risk of cancer-associated venous thromboembolism (VTE), which may lead to premature termination of ICI and affect outcome. Studies on VTE and comorbidity in real-life NSCLC patients undergoing ICI are lacking but warranted.
Retrospective data of 366 incurable stage III-IV NSCLC patients treated with ICI monotherapy in 1st (n=139) and ≥ 2nd line (n=227) at three different Danish Oncologic Departments from 2015-2019 was gathered. Comorbidity, a history of prior known VTE (P-VTE), development of VTE from first ICI until two months after last ICI (D-VTE) and from first ICI until end of follow-up (F-VTE) were registered. For survival analysis Kaplan Meier and cox regression were performed.
Median follow-up time was 29.1 months and 10% developed F-VTE with 82% being pulmonary embolism. P-VTE was found in 12% and D-VTE in 6% (first line: 8%, ≥ 2. Line: 4%). Of those with P-VTE, 15% had D-VTE and 24% had F-VTE. Median time to VTE onset for D-VTE was 2.6 months [IQR 0.9-6.1]. PD-L1 ≥ 50% correlated to radiologic response (r=0.25, p<0.0009), but D-VTE and F-VTE did not. Precise PD-L1 status was obtained for 55% and D-VTE correlated to PD-L1 level ≥ 80 % (r=0.2, p=0.024). Comorbidities were hypertension (HT) (44%), COPD (35%), other prior malignancy (OPM) (18%), atrial fibrillation (14%), diabetes (13%), autoimmunity (9%), myocardial infarction (8%) and peptic ulcer (PU) (4%). In multivariate analysis HT, PU and PS ≥2 were significantly associated to impaired OS and PFS.
Real life NSCLC patients eligible for palliative ICI have substantial comorbidity. Having a P-VTE prior to ICI increases the risk of recurrent VTE considerably. PU and HT seem to increase the risk of impaired outcome from ICI. VTE development during ICI in NSCLC real life patients might be associated to PD-L1 level. Future prospective studies need to explore this and the impact of VTE and comorbidity on outcome.
Legal entity responsible for the study
University of Southern Denmark, Region of Southern Denmark, Department of Oncology Odense University Hospital.
All authors have declared no conflicts of interest.