Safety and efficacy of immune checkpoint inhibitors in patients with pre-treatment reduced left ventricular function

Abstract

Aims

Immune checkpoint inhibitors (ICIs) have revolutionized cancer treatment outcomes. However, the response varies across different populations, and their use may lead to life-threatening cardiovascular (CV) events. While pre-treatment reduced left ventricular ejection fraction (LVEF) is considered a marker for high-risk cardiotoxicity and a contraindication for anthracycline and HER2-targeted therapies, there is limited evidence on the safety and efficacy of ICIs therapy in patients presenting with pre-treatment reduced LVEF. The study aims to evaluate the safety and efficacy of ICIs therapy in patients with pre-treatment reduced LVEF.

Methods

Retrospective single center cohort of patients treated with ICIs therapy, who performed pre-treatment LVEF assessment. The primary endpoint was to evaluate the safety of ICIs among this population, assessed by CV events (composite of myocarditis, acute coronary syndrome, heart failure, and arrhythmias). The secondary endpoint was to evaluate the efficacy of ICIs, assessed by all-cause mortality and progression-free survival (PFS).

Results

The cohort included 307 patients, with 30 (10%) presenting with pre-treatment reduced LVEF, with a mean LVEF of 39 ± 7%. While a significantly higher incidence of CV events was observed in the reduced LVEF group (37% vs. 14%, p = 0.004), following a multivariate Cox regression analysis including baseline CV diseases and risk factors, pre-treatment reduced LVEF did not remain a significant independent predictor (p = 0.358). No significant differences were observed between the groups regarding all-cause mortality and PFS.

Conclusions

Pre-treatment reduced LVEF was not identified as an independent marker for clinical outcomes in patients treated with ICIs therapy.

Graphical abstract

Safety and Efficacy of ICIs in pre-treatment reduced LVEF. ICIs = Immune checkpoint inhibitors, LVEF=left ventricular ejection fraction, CV=cardiovascular, HF=heart failure, ACS=acute coronary syndrome, HR=hazard ratios.

Introduction

Immune checkpoint inhibitors (ICIs) have significantly improved clinical outcomes across various cancer types [1] and are currently used to treat metastatic and earlier-stage diseases [2,3,4]. These therapies play a pivotal role in activating the patient’s immune system against the tumor cells [5] by impeding the tumor cells’ ability to evade the immune system, reducing the tumor’s survival. Several ICIs are currently approved by the Food and Drug Administration (FDA), including cytotoxic T lymphocyte antigen-4 (CTLA-4) inhibitors (ipilimumab, tremelimumab), programmed cell death protein 1 (PD-1) inhibitors (nivolumab, cemiplimab, pembrolizumab), and programmed death-ligand 1 (PD-L1) inhibitors (atezolizumab, avelumab, durvalumab).

ICIs-related adverse events arise when pro-inflammatory T cells overreact upon the introduction of ICIs, resulting in damage to healthy tissues. These events are common and can lead to significant morbidity and mortality as well as cessation of effective life-saving therapy [6, 7]. While the reported prevalence of cardiovascular (CV) events is relatively low [8], it is believed to be underdiagnosed and potentially fatal [9,10,11]. Although myocarditis is the most recognized [7], other CV events are increasingly reported, including acute coronary syndrome (ACS), heart failure (HF), and arrhythmias [7].

The Heart Failure Association–International Cardio-Oncology Society (HFA-ICOS) baseline CV toxicity risk stratification has defined baseline reduced left ventricular ejection fraction (LVEF) < 50% to be classified as high risk in patients treated with anthracycline and human epidermal receptor 2 (HER2)-targeted therapy [12]. However, there is no reference to the role of baseline reduced LVEF in patients treated with ICIs, mainly due to the limited evidence evaluating its safety and efficacy in this population. While the European Society of Cardiology (ESC) cardio-oncology guidelines endorse pre-treatment transthoracic echocardiography (TTE) assessment in patients with prior left ventricular (LV) dysfunction and CV diseases [13], the role and management of pre-treatment reduced LVEF is not specified.

Our study aimed to evaluate the safety and efficacy of ICIs therapy, as assessed by CV events, all-cause mortality, and progression-free survival (PFS) in patients presenting with reduced LVEF pre-treatment.

Methods

Study population

We conducted a retrospective, single-center, observational study at Tel-Aviv Sourasky Medical Center, a tertiary cancer center in Israel. Consecutive records of all patients treated with ICIs between 2015 and 2021 were reviewed. The treating oncologist chose the specific therapy protocol according to current best practices. Our cohort included patients who performed pre-treatment LVEF assessment within 12 months before initiating ICIs therapy. Exclusion criteria included age under 18 years and previous treatment with anthracyclines due to their cardiotoxicity long-term potential, which may be a confounding factor [13]. The study was approved by the local Tel Aviv Sourasky Helsinki regulator ethics committee (Identifier: 0228-16-TLV).

Data collection

A thorough baseline medical history (including age, sex, diabetes mellitus (DM), hypertension (HTN), hyperlipidemia, ischemic heart disease (IHD), HF, atrial fibrillation (AF), and chronic kidney disease (CKD)), medical treatment (mainly cardio-protective medications, including angiotensin-converting enzyme inhibitor (ACEi), angiotensin receptor blocker (ARB), and mineralocorticoid receptor antagonist (MRA)), current malignancy status, and cancer therapy were obtained from the Tel Aviv Sourasky Medical Center electronic database, diagnosed by the treating physician and updated in the medical chart.

Echocardiography

A Standard TTE was performed according to the American Society of Echocardiography (ASE) guidelines [14], using state-of-the-art imaging equipment (iE33 or Epiq, Philips Medical Systems, Bothell, WA, USA). All echocardiograms were performed by trained sonographers and were read by expert echo cardiologists. Visual assessment of LVEF was initially performed, and in cases where the expert reader suspected abnormality, a biplane Simpson’s method was applied for a more detailed quantitative evaluation of LVEF. Reduced LVEF was defined as LVEF < 50%, according to the accepted literature [13]. Early trans-mitral flow velocity (E), late atrial contraction (A) velocity, and early diastolic mitral annular velocity (septal and lateral e’) were measured in the apical 4 chamber (4 C) view to provide an estimate of LV diastolic function [15]. The peak E/e’ ratio (average mitral E/e’ ratio) was calculated from the average of at least 3 cardiac cycles, and the deceleration time (DT) of the E wave was also measured. The left atrial volume index (LAVI) was calculated using the biplane area length method at end-systole [16].

Study endpoints

The primary endpoint was to evaluate the safety of ICIs therapy, assessed by the incidence of CV events in patients with pre-treatment reduced LVEF (Reduced LVEF group), compared to patients with pre-treatment preserved LVEF (Preserved LVEF group). CV events were defined as the composite of myocarditis, ACS, HF exacerbation, and arrhythmias (including AF, atrial flutter, and ventricular tachyarrhythmia), and were determined by the treating physician and recorded in the electronic medical charts. Myocarditis was diagnosed based on the criteria established by Bonaca et al. [17]. These endpoints were selected based on the most commonly reported CV events related to ICIs therapy [13].

The secondary endpoint was to evaluate the efficacy of ICIs therapy, assessed by the incidence of all-cause mortality and PFS, between the Reduced and Preserved LVEF groups. All-cause mortality was extracted from the population registry bureau, and PFS was determined by an oncologist who reviewed the electronic medical charts.

Statistical analysis

Categorical variables were expressed as frequency and percentages. A chi-square test was used to evaluate the association between these variables. The distribution of continuous variables was assessed using Kolmogorov–Smirnov test. Normally distributed continuous variables were described as mean and standard deviations (SD), these variables were compared using the independent samples t-test. Non-normally distributed continuous variables were expressed as median and interquartile ranges (IQR), Mann–Whitney U test was used for comparison. CV events cumulative incidence curves and survival curves were plotted using the Kaplan–Meier method. We used Cox regression analysis to predict CV events and all-cause mortality. The models were adjusted for variables with known clinical significance and included all cardiac risk factors and baseline CV diseases that might have increased the risk for CV events and mortality (age, sex, HTN, IHD, HF, AF). Adjusted hazard ratios (HR) with 95% confidence intervals (CI) were reported for all variables. The median follow-up time for all-cause mortality was calculated using the reverse Kaplan-Meir method. A two-tailed p-value of < 0.05 was considered significant for all analyses. All analyses were performed with IBM SPSS 29.0 software (SPSS Inc., Chicago, IL) and The R statistical package (version 3.3.1) (R Foundation for Statistical Computing, Vienna, Austria).

Results

Baseline patient characteristics

From 2015 to 2021, we identified 310 patients treated with ICIs therapy who performed pre-treatment LVEF assessment within 12 months before ICIs initiation. Three patients did not meet the exclusion criteria due to prior treatment with anthracyclines, leaving 307 patients in our cohort.

Baseline characteristics are presented in Table 1. Our cohort’s mean age was 67 ± 12 years and was predominantly male (60%). The most common cancer diagnosis was lung cancer (39%), followed by gastrointestinal (GI) (11%) and renal cell carcinoma (RCC) (9%). The majority of the patients presented at stage IV (81%). ICIs therapy included pembrolizumab (54%), nivolumab (29%), avelumab (2%), atezolizumab (9%), durvalumab (4%). A combination therapy of ipilimumab and nivolumab, which the guidelines consider as a high-risk cardiotoxicity therapy, was used in only 4% of the patients. Overall, 21% of the patients were treated with combined chemotherapy, 52% were previously exposed to chemotherapy, and 9% were treated with chest radiation. The median number of ICIs cycles was 5 [IQR: 3–14].

Table 1 Baseline characteristics

Cardiac risk factors were relatively common, as expected, among this elderly population and included HTN (45%), hyperlipidemia (45%), DM (25%), and CKD (8%). Overall, 24%, 10%, and 10% of the patients had a diagnosis of IHD, HF and AF at baseline. A total of 27%, 14%, 40%, and 55% of the patients were treated with ACEi, ARB, beta-blockers, and statins, respectively. Furthermore, 30% and 15% of the patients were treated at baseline with aspirin and anti-coagulation therapy, respectively.

Complete blood count, creatinine levels, and mean high-sensitivity troponin I (hs-TnI) were within the normal range.

Pre-treatment mean LVEF was 50 ± 16%. Mean diastolic function parameters were within the normal range and included e’ lateral 8.6 ± 3 m/s, E/e’ lateral 8.9 ± 4.1, LAVI 33 ± 12 ml/m^2, and systolic pulmonary artery pressure (SPAP) 33 ± 12 mmHg.

Reduced vs. preserved LVEF groups

Overall, 30 (10%) patients were included in the Reduced LVEF group, with a mean LVEF of 39 ± 7%, compared to a mean LVEF of 59 ± 3% in the Preserved LVEF group, p < 0.001. There were no significant differences between the Reduced and Preserved groups regarding age, sex, cancer type, cancer therapy, and staging. As expected, baseline cardiac risk factors and CV diseases were significantly higher in the Reduced LVEF group, including HTN (67% vs. 43%, p = 0.020), IHD (63% vs. 19%, p < 0.001), HF (43% vs. 7%, p < 0.001) and AF (23% vs. 9%, p = 0.034) (Table 1). Accordingly, baseline cardio-protective therapy was significantly higher as well among the Reduced LVEF group, including ACEi (60% vs. 24%, p < 0.01), beta-blockers (67% vs. 35%, p < 0.01), MRA (23% vs. 3%, p < 0.001), Sacubitril/valsartan (10% vs. 0%, p < 0.001) and statins (83% vs. 52%, p = 0.002). Of note, hs-TnI was within the normal range (<50ng/L) for both groups, with no significant differences. While median creatinine values were significantly higher in the Reduced LVEF group, they were within the normal values (1.0 [0.8, 1.2] vs. 0.8 [0.7, 1.1], p = 0.031). Besides lower LVEF, the only other pre-treatment echocardiography parameter showing a significant difference was higher LAVI among the Reduced LVEF group (39 ± 14 vs. 32 ± 11, p = 0.006) (Table 1).

Primary and secondary endpoints

During a median follow-up of 49 months [IQR: 36–66], 51 (17%) patients developed 66 CV events; 3 (1%) myocarditis, 8 (2%) ACS, 26 (9%) HF, and 29 (9%) arrhythmias (Table 2). A significantly higher incidence of CV events was observed in the Reduced LVEF group (37% vs. 14%, p = 0.004) (Table 2), which was mainly driven by significantly higher HF exacerbation events (37% vs. 5%, p < 0.001). As presented in Table 3, following a multivariate Cox regression analysis that included baseline cardiac risk factors and CV diseases, pre-treatment reduced LVEF did not remain an independent significant predictor for CV events development (adjusted HR = 1.44 (95% CI 0.66–3.14), p = 0.358) (Fig. 1). The only significant predictors observed for CV events development were age (adjusted HR = 1.05 (95% CI 1.01–1.08), p = 0.008) and baseline AF (adjusted HR = 4.15 (95% CI 1.96–8.8), p < 0.001).

Table 2 Outcomes

Table 3 Cox regression model for predicting CV events

Fig. 1

Cumulative Incidence of CV events. Kaplan-Meier Curves Comparing CV events Cumulative Incidence between the two study groups, showing a higher incidence of CV events in the Reduced LVEF group

Among the patients developing CV events, permanent ICIs discontinuation and all-cause mortality were higher in the Reduced LVEF group (18% vs. 15% and 27% vs. 17%, respectively). Only two patients renewed ICIs therapy; both were in the Reduced LVEF group.

Overall, 236 (77%) patients died during follow-up, with a higher incidence in the Preserved LVEF group (78% vs. 67%); however, this did not reach statistical significance (Table 2; Fig. 2). PFS was similar for both groups, p = 0.416 (Fig. 3). As expected, in the Reduced LVEF group, patients who developed CV events had higher all-cause mortality compared to patients who did not develop events (82% vs. 58%).

Fig. 2

Overall Survival. Kaplan-Meier curves show no statistically significant difference in overall survival. between the Reduced and Preserved LVEF groups. OS = Overall survival

Fig. 3

Progression-Free Survival. Kaplan-Meier curves depict no statistically significant difference in PFS between the two groups. PFS = progression-free survival

Discussion

To our knowledge, this is the first study to evaluate the safety and efficacy of pre-treatment-reduced LVEF in patients undergoing ICIs therapy. We found that pre-treatment reduced LVEF had no significant independent impact on clinical outcomes, in patients treated with ICIs.

The introduction of ICIs therapy has significantly altered the field of cancer therapy, with nearly 50% of patients diagnosed with cancer being found eligible for immunotherapy in 2019¹⁸. Despite the effectiveness of these therapies [18, 19], ICIs therapy carries the potential of adverse events, manifested by overactivation of the host’s immune system against various organs and tissues, with CV events notably considered as particularly devastating [20]. Although myocarditis is the most well-known CV event [7, 11], it is still underdiagnosed and represents only one facet of a spectrum of diverse CV events, encompassing ACS, HF, and arrhythmias [21]. Notably, in our cohort, myocarditis presented in only 1% of the patients, while HF exacerbation and AF showed a higher incidence (9% each).

In cardiotoxic cancer therapies, mainly anthracycline and HER2-targeted therapy, patients with pre-treatment reduced LVEF are contraindicated for therapy due to the higher risk for cardiotoxicity [12]. However, this data is not available for ICIs, and evidence regarding the outcomes of ICIs therapy in this group is limited. Our study observed a notably higher incidence of CV events in the Reduced LVEF group, primarily driven by a heightened frequency of HF exacerbation. Following a multivariate Cox regression analysis, including baseline CV diseases and risk factors, pre-treatment reduced LVEF did not persist as a significant independent predictor for the development of CV events. This implies that this group’s elevated incidence of CV events can be attributed to its higher baseline CV diseases. Consequently, our findings suggest that ICIs should not be avoided in patients based only on Reduced pre-treatment LVEF, as with anthracyclines and HER2-targeted therapy.

After multivariate analysis, baseline AF emerged as a significant independent predictor of CV event development. This finding suggests the potential utility of constructing a risk stratification score, enabling the identification of patients at higher risk for CV outcomes. Such a score should asses also other modalities, such as vascular calcification.

Notably, there were no significant differences in all-cause mortality and PFS. The similar overall mortality between the groups could be related to the higher use of cardio-protective therapy in the Reduced LVEF group and to the course of cancer as the leading cause of death in these patients. These findings indicate that ICIs appear effective for patients with reduced LVEF before treatment.

The latest ESC 2022 guidelines on cardio-oncology [13] classified baseline LVEF assessment as a class I(B) indication for high-risk patients and a class IIb(C) for all patients treated with ICIs. Although in our study, pre-treatment reduced LVEF was not an independent predictor for CV events, it is important to highlight the value of baseline LVEF assessment. The decline in LVEF during ICIs therapy may suggest ICIs-induced myocarditis [13], which may necessitate a temporary or permanent discontinuation of ICIs therapy. Baseline LVEF assessment can assist in discerning whether the decline in LVEF is novel or pre-existing, significantly impacting the management of these patients and allowing the continuation of a lifesaving therapy.

Study limitations

Our study has several limitations. First, it is a single-center study, and thus, the generalization of our results is limited. However, since Tel Aviv Sourasky is a tertiary referral cancer center in Israel, the study population includes a variety of ethnic origins from all over the country and, therefore, represents a wide range of populations in Israel. Second, this is a retrospective study, and thus, our results are subject to the effects of potential confounders inherent to the nature of such studies and may be biased by their design. Third, we recognize that the relatively small number of patients with pre-treatment reduced LVEF, reduces the statistical power of our results, and therefore, larger trials are needed. Fourth, only patients with pre-treatment LVEF assessment were included in the study. While pre-treatment LVEF assessment is currently the routine protocol in our facility, in the earlier years, it was not. Hence, there is a possibility for potential bias in the patients referred for LVEF assessment, either due to baseline CVD or symptoms.

Conclusions

As ICIs therapy becomes increasingly prominent as a critical tool for the treatment of various cancer types, its utilization may be constrained by the development of severe adverse events, notably CV. In contrast to anthracyclines and HER2-targeted therapies, we revealed that pre-treatment reduced LVEF was not found to play an independent marker for clinical outcomes in patients treated with ICIs therapy. Prospective larger and longer follow-up trials are imperative to integrate these findings comprehensively.