The World Health Organization’s most recent data from 2021 has shown that cancer in general accounted for around 2 million of the nearly 10 million deaths worldwide (1). Therefore, it might be viewed as a significant health burden that can be reduced through early detection, precise diagnosis, and improved management and care. However, the pharmaceutical agents used to treat cancer, either alone or in combination therapy, have been linked to the emergence of toxicities affecting several vital organs, including the heart and vessels. In the growing field of cardio-oncology, which serves as a link between cardiologists and oncologists, the impact of cancer treatment on the heart and the management of cardiotoxicity are the main topics of interest (1, 2). New therapeutic agents have emerged over the past decade to improve cancer treatment and lessen the toxic side effects, including immune checkpoint inhibitors (ICIs), a special type of immunotherapy subclass whose cardiotoxicity potential was found to be lower than that of conventional chemotherapeutic drugs (3).

The development of these agents has been a revolutionary milestone that was associated with remarkable benefits and resulted in long-lasting tumor responses. They are now widely accepted as a key component of therapeutic strategies in cancer management (4, 5). The ICI subclasses include monoclonal antibodies that block programmed cell death receptors or their ligand (PD-1/PD-L1) and cytotoxic T-lymphocyte-associated antigen 4 (CTLA-4). PD-1, PD-L1, and CTLA-4 immune checkpoints are markedly expressed in cancer cells and contribute to the inhibition of T-cell activation and are thought to represent one of many tumoral adaptive responses to escape from the immune system (6). The first successful use of CTLA-ICI therapy on mice was reported in 1996; then, in 2000, the first human CTLA-4 ICI, i.e., IPILUMAB, was introduced that got FDA approval in 2011 (7, 8). In 2006, nivolumab became the first PD1 ICI used in patients and also the first to obtain FDA approval in 2014 for melanoma (9) followed by pembrolizumab, another PD1 ICI (9). Nevertheless, their use also led to increased occurrences of different types of side effects (irAE), characterized by autoimmune reactions in various tissues that were rare but had serious side effects on the heart (10). In recent years, several studies have consistently reported the incidence of cardiac-related toxicities such as myocarditis, athero-cardiovascular disease, and heart failure in the setting of PD1 ICI therapy among cancer patients (11–14).

The possible theories explaining the rationale of the occurrence of cardiac adverse events with ICI therapy among cancer patients have previously been explored in experimental animal models. It was found that PD1 and PDL1 upregulation had a cardioprotective role for cardiomyocytes; therefore, their blockades or inhibition with PD1 ICI would favor cardiomyocyte damage. Many research papers also reported that there can be an increased likelihood of rare but aggressive cardiotoxic events associated with the use of the PD1 inhibitor subclass of ICI compared with other ICIs. However, because PD1 ICI drugs were found to be safer in terms of the occurrence of other high-grade iRAes, they are being more frequently used than the rest. This gave rise to the need for establishing a balance between this novel oncology treatment efficacy and the burden of drug-adverse cardiac effects among treated patients. However, to our knowledge, no studies have yet been published directly to compare the risk of cardiotoxicity events between nivolumab and pembrolizumab, the two most commonly encountered PD1 ICI drugs used in cancer management. Logically, identification of the drug with a higher tendency for cardiac toxicities than others would be essential to helping specialist caregivers to select the better alternative for patients, taking into account both efficacy and cardiotoxicity indexes in their management and thereby possibly lessening the burden of PD1 ICI-induced cardiotoxicity.

This meta-analysis was conducted in accordance with the MOOSE reporting guidelines (15).

Our analyses found that the proportion of cardiotoxicity credited to nivolumab was the highest among ICI drugs; moreover, the probability of developing cardiotoxicity for someone treated with pembrolizumab was slightly lower than if the drug was nivolumab, but the difference was not statistically significant. However, although the results of the overall meta-analysis seemingly depicted a trend that favored pembrolizumab over nivolumab, it remained statistically insignificant even after performing subgroup analysis. The results remained inconclusive, and we observed that the considerable heterogeneity noted can have been because of the quality of studies, their designs, the great variation of outcome effect size, sample size, and other confounding factors such as cardiovascular history and type of cancers. With the exception of a subset of patients whose cardiotoxicity type was myocarditis, it can be observed that pembrolizumab therapy had a greater risk for cardiotoxicity than nivolumab.

Programmed cell death receptor inhibitor 1 is a protein found on T cells, which can bind to another ligand called PDL-1, thus preventing the T-cell-mediated destruction of other cells (18). Based on this mechanism, inhibitor agents for PD-1 and PDL-1 proteins have been manufactured to increase the T cells’ ability to fight cancer cells and increase overall survival (19, 20). Furthermore, there has been substantial objective improvement in cancer outcomes with the increased use of these novel agents, and it has now become a trend in the oncology field (21–23). ICI drugs from the PD1 receptor inhibitor subclass are well known to be associated with side effects such as pneumonitis, pruritus, and neurological, endocrine, and gastrointestinal adverse effects, although it has been observed that the frequency of the irAEs was relatively lower than with the other subclasses (23–25, 51). However, the PCD receptor inhibitor subclass has recently been identified in previous studies as bearing a certain cardiotoxicity adverse effect potential, albeit accounting for<5% of all adverse effects.

The cardiotoxicity spectrum has in majority been represented by myocarditis (26, 27), but other cardiac conditions can also be observed such as arrhythmias, pericardial effusion, heart failure, coronary events, pericarditis, and heart block (28). Existing case studies and pharmacovigilance data showed that the irAEs mainly affect cardiac conduction and myocyte function, which would then lead to heart damage (29, 30). It also appears that cardiotoxicity incidences are more observable among patients undergoing combination therapy (31). The routinely and broadly used PD1 ICI are pembrolizumab and nivolumab. Hence, the core objective was to identify whether there were any differences between them regarding cardiac adverse effects despite belonging to the same subclass. Can those differences also be related to their clinical efficacy, structure, mechanisms of action, spectrum, and frequency of side effects? Moreover, with regard to the cardiotoxic adverse effects, which one can be better than the other? The two drugs belong to the same subclass and are used for similar therapeutic indications, the most common being non-small cell lung cancers, melanoma, and metastasis (20). However, from a previous report, it seemed as though nivolumab was potentially a more cardiotoxic ICI, as roughly 60% of cardiotoxicity reports in cancer-treated patients were associated with nivolumab (32, 33) in mono or combination therapy regimen, but this was observed more in studies in which the quality on the NOS scale was not satisfactory; however, recent cohort and case–control studies (also included in this meta-analysis) rather reported an increased risk associated with the use of pembrolizumab (see Zachary et al., 2022). The results after pooling all the included studies showed that although there were more reports of cardiotoxicity for nivolumab, the difference in cardiotoxicity odds ratio between pembrolizumab and nivolumab was not statically significant. This was consistent with some previously reported results. One study that aimed to compare the efficacy of the two drugs (50) reported in their survival analysis that despite the observation of a higher objective response rate with pembrolizumab than nivolumab, there was no significant difference in the progression of free survival between the patients treated with pembrolizumab and those treated with nivolumab (22). Another investigation on the overall incidence and risk of irAE between the two drugs (95% CI: 0.97–1.79) indicated that the difference was not statistically significant (33, 34). Moreover, there were no significant discernible differences in the mechanism of action of the two drugs: pembrolizumab and nivolumab are both humanized IgG4 monoclonal antibodies against PD-1, but with the distinction that they do not induce antibody-dependent cellular cytotoxicity, as would be the case for normal IgG antibodies; hence, their toxicities and side effects have been more characterized as immune-related than cytotoxic (35). Analysis of the included studies showed that nivolumab had the highest proportion of induced cardiotoxicity among all ICI-induced cardiotoxicities (13). In addition, the majority of patients with reports of cardiotoxicity had other factors such as conventional cardiovascular risk factors or cardiovascular history. Although this may explain the frequency of reports of cardiotoxicity, it cannot explain the large disparity in the proportions of cardiotoxicity between the two drugs, given that the distributions of cardiovascular risk factors or the history of the two groups were comparable (36). In this study, cardiotoxicity refers to any cardiac damage irrespective of extent or severity. The cardiotoxicity odds ratio between pembrolizumab and nivolumab was not statistically significant; however, this does not imply that there would not be any difference at all. Therefore, after consideration of the descriptive proportions reported for both drugs, it can be suggested that nivolumab is associated with more cardiotoxicity events than pembrolizumab. One plausible explanation for this finding can be that nivolumab was introduced first and gained recognition first before pembrolizumab. Another possible theory would be that because in general, the overall number of cases with cancer expressing high PDL1 (expression: >50%) seems to be lower than that of cases with a low PDL1 expression (expression: 1–49%). Per standard recommendation, pembrolizumab is known to be effective in tumors expressing PDL1 at >50% and nivolumab in tumors expressing PDL1 at >1%. Therefore, there would naturally be more cases (low PDL1 expression) of cancer treated with nivolumab than with pembrolizumab. All these could have led to more reports of cardiac toxicity events with nivolumab than with pembrolizumab. Additionally, the small number of included studies and the high degree of heterogeneity between studies may have also played a role in the lack of statistical significance from the pooled odds ratio of cardiotoxicity effect between the two drugs. A few case series on ICI-induced cardiotoxicity also highlighted that cardiomyopathy, myocarditis, and conduction abnormalities were being underreported, which could have as well influenced the overall effect result of the current analysis (28). Another possibility can be that those considered to be developing cardiac toxicity were patients with cardiac clinical symptoms, resulting in non-involvement of the subclinical cases, which could have significantly influenced the results.

Numerous studies have shown that despite being relatively uncommon and few, the majority of myocarditis cases would present during the acute phase of therapy, with a propensity for seriousness and mortality, or the development of MACEs such as cardiac arrest, cardiac death, or stroke, but respond well to high corticosteroids for remission when administered in a timely manner (37–39). Although cardiac irAEs with ICIs are uncommon, the increased rate of mortality seen is an important factor to consider, as was also highlighted in another similar analysis (27). According to Dolladile et al.’s 2020 study, heart failure with left ventricular systolic dysfunction was seen among cases as a late adverse event. Therefore, patients treated with ICI should be monitored for at least 304 days. Additionally, because silent toxicities are possible (toxicity that manifests slowly before symptoms become obvious), such patients should also undergo routine cardiovascular screening for early detection of any abnormalities, especially for those aged >65 years and presenting at least two conventional cardiovascular risk factors or cardiovascular history (40). Nevertheless, the presumed advantages of early detection of cardiotoxicities through active screening, serial electrocardiograms, troponins, BNP, and echocardiography (which are helpful tools for the detection of subclinical cardiotoxicity during oncology therapy) should take into account the cost of testing as well as the possibility of false results, incorrect interpretation, and other related errors (31, 41). In a study, the highly sensitive troponin’s prognostic value showed that a value higher than 14 ng/L before the administration of pembrolizumab was significantly associated with a high incidence of MACE, including stroke and cardiac death (42). However, this does not necessarily mean that highly sensitive troponin should be considered an eligibility criterion for pembrolizumab or nivolumab therapy, but rather it can be useful as a predictor of cardiotoxicity risk.

This study has some limitations. The included studies were observational in nature, with some data collected from electronic and registered databases, implying a high susceptibility to information and selection biases (43). Selection bias was a concern in the (39) study, which was of a retrospective design, because there was no prospective cardiovascular screening protocol across all sites, and screening for cardiac biomarkers and other tests was left to the discretion of each individual care provider. (26) study was distinguished by the small sample size, which resulted in confounder interference and reporting bias. The criteria for control groups in two of the included studies were dubious, making it difficult to select a group of cancer patients with similar cardiovascular comorbidity and who underwent adequate testing to exclude cardiac pathology, as controls (28, 44). The criteria for selecting pembrolizumab or nivolumab for patients were not detailed or obvious in the results of the included studies. The risk of bias across the included studies could not have negated the evidence found in each study, but it could have led to an underestimation of association or effect size. According to two VigiBase analyses, pharmacovigilance analyses generate hypotheses that must be tested, ideally in prospective studies. Adverse Reaction Reporting System databases may be biased given the significant overlap of Individual Case Safety Reports (ICSRs) between databases. In addition, some cases may not have been reported to State drug enforcement authorities (34). The only included prospective cohort study had a short follow-up period, which may have led to a limited number of events and biased interpretation of results (25). In addition, randomized clinical trials on cardiotoxicity issues were not available at the time of the search, resulting in a limited number of studies, small sample sizes, and low statistical significance. Moreover, unpublished records on the topic as well as articles published in languages other than English cannot be evaluated. To our knowledge, this is the first meta-analysis to directly compare the two drugs, and hence, these results are still important, because they highlight the need for additional, in-depth research on this topic in multiethnic, large-center settings to provide oncology patients with the best possible care while reducing the likelihood of cardiotoxicity.

Immune checkpoint inhibitors have revolutionized the treatment of advanced-stage cancers including metastases; however, the potential danger to vital organs (52) such as the heart cannot be overlooked. Therefore, it is paramount to look at every strategy with the potential to limit, reduce, or control the magnitude of this issue. Previous studies regarding the cardiotoxicity risks of ICIs and comparisons between anti-PD1 and anti CTLA4 (45) were made; however, this paper addressed the direct comparison between nivolumab and pembrolizumab cardiotoxicity potentials. Contrary to what was recently reported (46, 47), the descriptive proportions described herein have provided a clear indication that nivolumab-induced cardiotoxicities are reported more in the literature over the past years than pembrolizumab-induced cardiotoxicities. The discrepancy between these findings and previous ones highlights the need for a prospective analysis on a larger sample cohort. However, the consensus on the need for proper cardiac screening before and after remains strong among researchers (48, 49). Therefore, the importance of multidisciplinary collaboration between oncologists, immunologists, and cardiologists in the management of cancer patients cannot be overemphasized.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding author.

This manuscript was compiled and written by FN under the direction, guidance, and supervision of Z-QW. Proofreading and editing were done by CM. All authors contributed to the article and approved the submitted version.