Cardiovascular events in patients receiving immuno-oncology agents, in phase I clinical trials: CAVIO study.

Abstract

e24009 Background: Among immunoncology agents (IO), immune checkpoint inhibitors are the most used, demonstrating survival benefit across multiple cancer types. Adverse events are often reported with IO; cardiotoxicity is rare, but potentially fatal. CAVIO study evaluated the occurrence of cardiovascular events (CVE) in patients receiving IO in phase I clinical trials (ph1CT). Methods: We retrospectively reviewed clinical and cardiological data of patients consecutively enrolled, from Jan2016 to Dec2021, in ph1CT with IO alone or in combination. Any type and grade of CVE occurred on treatment, related or not to IO, was evaluated. Fisher’s exact tests were used to compare variables. Cox regression and Kaplan Meier methods were used for multivariable analyses and estimating overall survival (OS). Hazard ratio (HR) and 95% confident interval (CI) were calculated. Results: 211 patients were included; median age was 57 years. The most frequent site of primary tumor was breast (25%). 43 of 211 patients experienced ≥1 CVE of any grade (G) during the treatment, of which < 1% G3 and none ³G4. The most frequent ones were hypertension (n = 20, 9%) and thrombosis (n = 15, 7%), only one myocarditis (2%). Table 1 shows the significant differences observed between patients experiencing CVE or not. The occurrence of CVE was not associated with pre-existing cardiovascular risk factors nor with number of prior therapy lines. The median progression free survival (mPFS) and mOS were 2 and 9 months respectively, regardless the CVE occurrence. Patients using anticoagulant at the end of ph1CT showed worse PFS and OS (1.8 and 5.4m) than those who did not (2.5 and 11.1m), similarly was for PS ECOG 1 vs 0. The result was consistent at the multivariate model adjusted for the primary site of tumor, PS ECOG, previous lines of treatment and age, in which PS ECOG 1 and the use of anticoagulant remained statistically significant associated with negative OS (for PSECOG: HR:1.84, 95% CI 1.31-2.57, p < 0.001; for anticoagulant use: HR: 1.72, 95%CI:1.20-2.47, p:0.003). Conclusions: The occurrence of CVE did not change the outcomes of patients treated with IO in ph1RC. Severe CVE were rare, hypertension being the most frequent. Further data are needed on the correlation of anticoagulant use, IO and patients prognosis. [Table: see text]