Abstract Background Immune checkpoint inhibitors (ICIs) are transformative treatments in oncology, yet they are also associated with a range of cardiovascular (CV) events. Purpose To identify the incidence of CV events (including subclinical) in patients receiving ICIs for solid cancer, either alone or in association with other cancer therapies (e.g., chemotherapies, target therapies). Objectives and methods The Adverse Myocardial and vascular side effects of Immune Checkpoint Inhibitors(AMICI) study prospectively assessed the incidence of CV changes by means of a thorough multimodal assessment. This included repeated CV imaging (transthoracic echocardiograms (TTE), cardiac magnetic resonance imaging (CMR)), serum biomarkers (troponin, brain natriuretic peptides in a single laboratory), and ambulatory ECG monitoring, before and during (at 6 weeks and at 6 months) ICI therapy (Figure). Cardiovascular events included myocarditis, pericarditis, acute coronary syndrome, heart failure (HF), high degree conduction abnormalities (defined as new grade 2 or 3 atrioventricular block) or sustained ventricular arrhythmias (defined as 3 or more ventricular beats at a rate ≥ 120/min), CV death as a composite endpoint, validated by an independent expert panel Results Out of 130 solid cancer patients considered for ICI therapy, 49 patients (38 (77.6%) male; mean age 64.3 (SD 11.0) years old) were included (June 2020 -December 2021). 22 (55.1%) suffered from kidney or bladder cancer; 7 (14.3%) had CV disease prior to cancer diagnosis; 4 (8.2%) received PD1 and CTLA4 inhibitors. Early CV events were observed in 9 (18.4%) patients at mean 40.1 (SD 5.9) days, with heart failure (HF) in 5 (10.2%), ventricular arrhythmias or new conduction disorders in 4 (8.2%) patients. History of AF (HR 4.49 (CI 1.11-18.14), P=0.035) predicted early CV events at 6 weeks (Table). At 6 months follow-up, 18 CV events were observed in 15/49 (31%) patients, with 6 (12.2%) HF events, 5 (10.2%) significant ventricular arrhythmias or conduction disorders and 4 (8.2%) AF. There was a significant decline in LVEF (P<0.001) in patients even in the absence of CV event (P=0.003) or HF (P=0.003). Higher creatinine at inclusion (HR 0.99 [0.98-1.00], P=0.006) predicted HF. Conclusions Cardiovascular complications arising during cancer treatment with ICIs are primarily due to the exacerbation of pre-existing HF or the onset of non-inflammatory heart rhythm disorders. The majority of these CV changes manifest shortly after starting ICIs and do not need hospital admission. Our results suggest that more close follow-up may be needed in patients with underlying CV conditions when starting ICI therapy, by means of CMR and ambulatory ECG holter monitoring.