Relationship between hemodynamic forces based on CMR and major adverse cardiac events in cancer patients treated with anti-PD-1/PD-L1 immune checkpoint inhibitors
摘要
Immune checkpoint inhibitors (ICIs) have significantly improved the management of many malignancies, but immunotherapy-mediated major adverse cardiac events (MACE) should not be ignored. We aim to explore the relationship between hemodynamic forces (HDFs) of cardiovascular magnetic resonance (CMR) parameters and MACE in cancer patients treated with ICIs.
MethodsWe prospectively recruited cancer patients who planned to receive ICIs treatment in three hospitals between January 2022 and December 2023. Echocardiograms of all patients before treatment showed normal cardiac function. The patients were planned to undergo CMRs approximately at 12 weeks after starting one cycle of ICIs therapy. All patients were followed up for MACE, including myocarditis, pericarditis, atrial fibrillation, acute coronary syndrome, cardiogenic shock, new-onset complete heart block, cardiac arrest, heart failure hospitalization and sudden or cardiac death. The CMR parameters after ICIs therapy were analyzed. The univariable and multivariable Cox regression analysis of CMR parameters were examined. The incremental prognostic value of HDFs for MACE were evaluated by comparing area under the receiver operating characteristic curve (AUC) values of different models. Survival plots were obtained via Kaplan-Meier analysis.
ResultsOf Two hundred-twenty-five cancer patients were included, 21 (9.3%) experienced MACE with an average time of 161.7 days. Lower HDFs in the apex–base (A-B) direction at the systolic phase, diastolic phase and entire heart cycle were all associated with high risk of MACE after adjustment confounding factors (P < 0.001 for all comparisons). In analysis that compared tertile 3 (high) of HDF A-B with tertile 1 (low), the hazard ratios were 0.18 (95% CI, 0.05 to 0.64) for entire heart cycle, 0.05 (95% CI, 0.01 to 0.28) for the diastole phase, and 0.01 (95% CI, 0 to 0.23) for the systole phase. Each increment of 1% in HDF A-B of entire heart cycle, diastolic phase and the systolic phase were associated with an 31%, 48% and 32% decrease in cardiovascular risk. Adding HDF to the baseline risk model improved the predictive value of MACE in cancer patients treated with anti-PD-1/PD-L1 ICIs (net reclassification improvement [NRI]: 0.5434 [0.2546–0.8322], p = 0.00023; integrated discrimination improvement [IDI]: 0.1916 [0.0622–0.3210], p = 0.00371). The Kaplan-Meier curves showed that the risk of MACEs was lower as the HDF A-B (%) tertiles increased.
ConclusionIn cancer patients treated with ICIs, the HDF was significantly associated with the incidence of MACE. It may be a useful predictor of the incidence of MACE induced by ICIs in cancer patients. The addition of the HDF to the baseline risk model had an incremental effect on the predictive value of MACE in cancer patients treated with ICIs.
Graphical Abstract