You are viewing the site in preview mode

Skip to main content
Download PDF

Time dependent predictors of cardiac inflammatory adverse events in cancer patients receiving immune checkpoint inhibitors

Cardio-Oncology volume 11, Article number: 40 (2025) Cite this article

Abstract

Background

Cardio-inflammatory immune related adverse events (irAEs) while receiving immune checkpoint inhibitor (ICI) therapy are particularly consequential due to their associations with poorer treatment outcomes. Evaluation of predictive factors of these serious irAEs with a time dependent approach allows better understanding of patients most at risk.

Objective

To identify different elements of patient data that are significant predictors of early and late-onset or delayed cardio-inflammatory irAEs through various predictive modeling strategies.

Methods

A cohort of patients receiving ICI therapy from January 1, 2010 to May 1, 2022 was identified from TriNetX meeting inclusion/exclusion criteria. Patient data collected included occurrence of early and later cardio-inflammatory irAEs, patient survival time, patient demographic information, ICI therapies, comorbidities, and medication histories. Predictive and statistical modeling approaches identified unique risk factors for early and later developing cardio-inflammatory irAEs.

Results

A cohort of 66,068 patients on ICI therapy were identified in the TriNetX platform; 193 (0.30%) experienced early cardio-inflammatory irAEs and 175 (0.26%) experienced later cardio-inflammatory irAEs. Significant predictors for early irAEs included: anti-PD-1 therapy at index, combination ICI therapy at index, and history of peripheral vascular disease. Significant predictors for later irAEs included: a history of myocarditis and/or pericarditis, cerebrovascular disease, and history of non-steroidal anti-inflammatory medication use.

Conclusions

Cardio-inflammatory irAEs can be divided into clinically meaningful categories of early and late based on time since initiation of ICI therapy. Considering distinct risk factors for early-onset and late-onset events may allow for more effective patient monitoring and risk assessment.

Introduction

Immune checkpoint inhibitors (ICIs) have significantly improved survival outcomes in cancer patients although its utilization has been associated with a unique class of toxicities. Immune-related adverse events (irAEs) from ICI therapy have wide ranging effects on the body often caused by excessive inflammation stimulated by unchecked immune system activity [1]. Cardiac inflammation, characterized as myocarditis or pericarditis, is a rare irAE occurring in approximately 1% of patients receiving ICI therapy [2, 3]. While some types of irAEs like thyroiditis and dermatitis can have positive survival implications and are readily managed, cardiac irAEs are associated with a dire prognosis including hospitalization for their management and an increased mortality risk [4, 5]. Discontinuing ICI therapy is one approach to management, however this can prevent patients from receiving optimal cancer treatment [6]. The most common intervention is treatment with corticosteroids, which also my limit the efficacy of ICI agents due to immunosuppression [7]. Established labs and tests like troponin levels or ECG readings that might identify people at risk or allow for patient monitoring are not routinely available for a majority of ICI patients, especially those receiving therapy in the outpatient setting over many months [8,9,10]. To this point, some preliminary risk factors for the development of cardio-inflammatory adverse events identified include ICI combination therapy, biological sex, pre-existing cardiovascular comorbidities, and select medications [11,12,13]. Given limited monitoring ability and effective interventions, identification of time-dependent risk factors may allow for better identification of patients who are at risk for these serious irAEs over time.

While the timing of adverse events for other oncology agents appears more predictable, cardiac irAE onset can occur after single infusions of ICI therapy or more than 1 year removed from therapy initiation consistent with irAEs impacting other body systems [14,15,16]. Identifying distinct risk factors for early versus later adverse events may be warranted for a variety of reasons. With more common irAEs like dermatitis or thyroiditis, the risk factors for their occurrence can be significantly influenced by a patients time on ICI therapy [17,18,19,20,21]. Additionally, with certation irAEs like pneumonitis and hepatitis, earlier onset events are associated with poorer overall and progression free survival outcomes compared to later onset events [22, 23]. Preliminary studies suggest distinct physiological and immune pathways associated with irAEs dependent on their timing, with later events more likely to be triggered by dysregulated humoral immune response while early onset events associated with cell-mediated immunity [24, 25]. Lastly, cancer patient populations can vary significantly in their treatment trajectories over time. Patients who experience mortality outcomes within a few months of initiating ICI therapy are likely to have distinct cancer severities, demographic backgrounds, or medical histories making risk factors for cardiac irAEs evolve as treatment progresses. Given these considerations, exploring time-dependent risk factors for cardiac irAEs is warranted.

Literature assessing cardiac inflammatory adverse events predominately focuses on myocarditis, leading to limited generalizability of ICI impact on cardiovascular health. Consideration of pericarditis may provide a more complete picture, with preliminary studies suggesting that pericarditis irAEs lead to increased mortality rates similar to myocarditis [26]. Similar causative pathways and significant relationships between ICI induced inflammation of pericardial tissue and myocarditis have also been identified [27]. Considering myocarditis and pericarditis as cardiac inflammatory adverse events together will allow for more nuanced analysis and better capture the impact of ICIs on cardiovascular health. A challenge in assessing risk factors for cardiac inflammation in ICI patients, especially in a time dependent manner, is having a representative sample [28]. TriNetX is a unique repository of real-world healthcare data with a network of hundreds of health care institutions, allowing for the creation of unique large cohorts of patients receiving ICI therapies [29]. Contemporary studies investigating cardio-inflammatory irAEs feature smaller samples ranging from a select few cases to rarely exceeding 100 total patients [30,31,32]. Leveraging the TriNetX platform, uniquely large cohorts of ICI patients can be utilized to investigate predictive factors of early and later cardio-inflammatory irAEs. Considering myocarditis and pericarditis as cardio-inflammatory irAEs together, in addition to distinguishing between early and later onset events, may allow for identification of novel risk factors predisposing patients to them.

Methods

Study design and data source

We performed a multicenter, multinational electronic health research retrospective cohort study, evaluating all patients receiving ICI therapy within the TriNetX platform from 80 international healthcare institutions providing the data. Files of patient data were obtained on May 15, 2023, including reported encounters, diagnoses, and medications received. Using these files, events occurring after ICI administration and relevant medical histories prior were obtained for subsequent analyses.

Patient population

Medical records for patients receiving ICI therapies from January 1, 2010 to May 1, 2022 were queried from the TriNetX platform. Inclusion criteria for the study were: adult patients 18 years and older at time of ICI initiation, with a history of neoplasm diagnosis, having birthdate and biological sex information, and survival times consistent with ICI initiation times (Supplemental Table 1). Patients with myocarditis/pericarditis diagnoses up to 6 months prior to ICI initiation were excluded from the patient population; this prevented pre-existing conditions from being counted as cardiac adverse events (Supplemental Table 1). Survival times were utilized in selecting appropriate patient populations for prospective analyses. These were defined as the number of days since from ICI therapy initiation until patients were either deceased or until there last encounter in the database.

Index immune checkpoint inhibitor(s) and index date

Index ICI’s for each prospective patient were defined as the agent(s) given for their first administration of ICI therapy. Index ICI’s considered were: anti-PD-1 monotherapy (i.e., nivolumab, pembrolizumab), anti-PD-L1 monotherapy (i.e., atezolizumab), and Combination ICI therapy. The index dates utilized in subsequent analyses were the reported date of administration of a patient’s first ICI therapy (Supplemental Table 1).

Cardiac inflammation adverse events

Cardiac inflammation adverse events were defined by International Classification of Diseases, Ninth (ICD-9) and Tenth (ICD-10) Revision codes representing both myocarditis or pericarditis consistent with prior studies (Supplemental Table 2) [33,34,35]. Adverse events within the study were defined as the occurrence of selected diagnosis codes within a patient’s medical record on or within 365 days after ICI index date. Time to adverse event was defined as the number days removed from ICI index date a patient experienced an event. Early adverse events were defined as occurring within 90 days of index date, while later events were defined as occurring after consistent with other studies [24, 36,37,38]. We respect that “early” and “late” onset event definitions can be subjective with other studies defining them based on other factors like number of cycles received (1 or 2 cycles) or shorter thresholds relative to initiation (i.e. 6 weeks). However, we feel our definition is supported and sufficiently distinguishes acute and chronic exposures to ICI agents. Patients were considered to have a history of cardiac inflammation if they experienced relevant diagnosis codes more than 6 months prior to ICI therapy initiation (patients within 6 months were excluded from the study).

Covariates utilized

A variety factors ranging from patient demographic information, history of comorbidities, and medication history were considered as predictors of cardiac inflammation adverse events (Supplementary Tables 1&2). Age at index was the number of years from date of birth to index ICI date. Due to incomplete reporting on racial and ethnic backgrounds of patients and limited sample sizes, patients were dichotomized as “Non-Hispanic White/Caucasian” or “Non-White/Caucasian.” Patients were considered to have a history of relevant comorbidities if they occurred within 5 years prior to their index date. Specific comorbidities were selected based on their relevance to cardiovascular health. Relevant ICD-9 and 10 codes were selected based on previous work identifying relevant codes to assess patient medical history in insurance claims (Supplemental Table 2) [39, 40]. Patients were considered as having a history use of selected medications or medication classes if they had documented of them up to 100 days prior to their index date (Supplemental Table 1). Select medications were selected if they were generally commonly used and either having cardiovascular indications or have proven associations with increased cardiac inflammation. Descriptive statistics were utilized to describe the prevalence of these features within the entire population, within patients experiencing early events, and within patients experiencing later adverse events.

Logistic regression modeling

Separate logistic regression models were implemented predicting cardiac inflammatory adverse events occurring within 90 days of index date and later adverse events. Models predicting early events included all patients that experienced adverse events and patients having survival times of at least 90 days. Models predicting later events excluded patients experiencing early adverse events and utilized patients having survival times of at least 365 days (Supplemental Fig. 2). Univariate logistic regression analyses were implemented to identify predictors for the occurrence of early and later adverse events. Variables having a P-value less than 0.1 were selected for consideration into multivariate models. These selected variables were utilized in separate baseline logistic regression models predicting early and later adverse events. Following creation of baseline models, Akaike information criterion (AIC) optimized models were created utilizing an exhaustive algorithm starting with the created baseline models. Significance of individual feature(s) for each prospective logistic regression model were assessed based on Wald statistics and their associated P-values with a significance threshold of 0.05. Variance inflation factors (VIF) were calculated to assess multicollinearity in multi-variable models, with a threshold value of 5 utilized.

Cox-proportional hazards modeling

Cox Proportional Hazards (COXPH) models were utilized to determine hazard ratios for variables identified with univariate feature selection relevant to occurrence of early and later adverse events. The COXPH model for early adverse events included all patients meeting initial inclusion/exclusion criteria, while the COXPH model for later adverse events excluded these patients experiencing early adverse events and only included patients having at least 90 days of survival (Supplemental Fig. 2). Significance of individual features of COXPH models were assessed based on Wald statistics and their associated P-values with a significance threshold of 0.05. Censoring due to morality and follow up time month over month was determined. Schoenfeld residuals were evaluated for early and late onset cox-models to ensure appropriate use of this strategy. Variance inflation factors (VIF) were calculated to assess multicollinearity, with a threshold value of 5 utilized.

Various steps in data analysis were conducted in the R-programing platform version 4.1.1. Data cleaning and creation of descriptive statistics utilized base R-language and the DPLYR package. The comorbidity package within R aided in assessment of patient comorbidity history. Logistic regression models were created with the stats, bestglm, and caret packages. Lastly, we utilized survival and survminor packages to create COXPH models.

Results

Patient population

A total of 66,068 patients receiving ICI therapy were identified from the TriNetX database meeting study inclusion/exclusion criteria (Supplemental Fig. 1). A total of 368 (0.56%) patients experienced cardiac inflammatory adverse events, with 193 (0.30%) experiencing early adverse events, 175 (0.26%) experiencing later adverse events, and 135 (0.20%) patients having a history of relevant myocarditis/pericarditis diagnoses in their medical history (Table 1). The average patient age at index was 65 years of age, 76.8% of patients were identified as Caucasian or White, and 57.4% of patients were male. ICI therapy utilized at index was an anti-PD-1 agent for 75.1% of patients, a anti-PD-L1 agent in 14.5% of patients, and a combination ICI therapy in 7.0% of patients (Table 1). Prevalence of different patient characteristics of interest included: history myocardial infarction (MI) in 6.7%, congestive heart failure (CHF) in 9.6%, peripheral vascular disease (PVD) in 18.4%, cerebrovascular disease (CEVD) in 11.9%, hypertension (HTN) in 49.0%, and diabetes mellitus (DM) in 19.3% of patients (Table 1). Medication use histories are also reported (Table 1). Additionally, the distribution of these characteristics amongst patients not experiencing a cardio-inflammatory irAE, patients experiencing an early adverse event, and patients experiencing later events are also reported (Supplemental Table 3).

Table 1 Descriptive statics for immune checkpoint inhibitor patient population
Full size table

Univariate logistic regression analysis and feature selection for multivariate models predicting early and later cardiac inflammatory adverse events

The following variables had significant associations with the occurrence of early cardiac inflammatory adverse events in univariate analysis: anti-PD-L1 agent at index, combination ICI therapy at index, prior history of CHF, history of PVD, history of CEVD, history of HTN, history of aspirin, history of anti-coagulant, and history of atypical antipsychotic (Supplemental Table 4). No additional variables met the threshold of significance having a P-value less than 0.1 for inclusion in multivariate models predicting early adverse events (Supplemental Table 4). Variables having a significant association with development of later adverse events included: history of diagnosis myocarditis and/or pericarditis, history of CHF, history of CEVD, history of HTN, history of DM, history of ACE-Inhibitor (ACE-I), history of beta-blocker, statin, and non-steroidal anti-inflammatory (NSAID) use (Supplemental Table 4). The following features met the threshold for inclusion in multivariate models predicting later adverse events with P-values less than 0.1: combination therapy at index, history of loop diuretic, and history of anti-coagulant (Supplemental Table 4).

Baseline and optimized multi-variate regression analyses predicting early and later adverse events

Within the baseline logistic regression models predicting early adverse events, the following variables were significant: combination ICI therapy (OR:4.66, 95% CI = 2.79–7.99) at index and a history of peripheral vascular disease (OR:1.54, 95% CI = 1.09–2.15). The optimized logistic regression model predicting early events had 2 significant features including combination therapy at index (OR:4.67, 95% CI = 2.79–7.99) and history of peripheral vascular disease (OR:1.71, 95% CI = 1.23–2.34) (Table 2). The optimized model additionally kept anti-PD-1 agent at index (OR:1.48, 95% CI = 0.97–2.37), history of anticoagulant (OR:1.30, 95% CI = 0.97–1.75) and history of atypical antipsychotic (OR:1.70, 95% CI = 0.86- 3.00) as predictors although they were not significant (Table 2). No features had variance inflation factors above 5, indicating no issues with multi-collinearity.

Table 2 Coefficients and P-values for baseline and optimized logistic regression models predicting early and later cardio-inflammatory IrAEs
Full size table

Within the baseline logistic regression model predicting later adverse events, the following variables were significant: history of myocarditis and/or pericarditis (OR:8.19, 95% CI = 1.95–23.1), history of CEVD (OR:1.99, 95% CI = 1.34–2.88), history of ACE-I (OR:1.60, 95% CI = 1.02–2.43), and history of NSAID (OR:1.67, 95% CI = 1.18–2.32) (Table 2). The optimized logistic regression model predicting later adverse events had 5 significant features including: History of myocarditis and/or pericarditis (OR:8.67, 95% CI = 2.08- 24.00), history of CEVD (OR:2.06, 95% CI = 1.40- 3.00), history of DM (OR: 1.42, 95% CI = 1.00–2.00), history of ACE-I (OR:1.68, 95% CI = 1.10–2.48), and history of NSAID (OR:1.71, 95% CI = 1.22–2.37) (Table 2). Combination ICI therapy at Index (OR:1.54, 95% CI = 0.92–2.43) was included in the optimized model but was not a significant predictor of later adverse events (Table 2). No features had variance inflation factors above 5, indicating no issues with multi-collinearity.

Cox-proportional hazards models identifying significant risk factors in developing early and later cardiac inflammatory adverse events

For the risk factors identified by the COXPH model in development of early adverse events, the following features were significant: anti-PD-1 agent at Index (HR:1.6, 95% CI: 1.03–2.5), combo ICI therapy at Index (HR:4.8, 95% CI:2.85–8.1), history of PVD (HR:1.6, 95% CI:1.14–2.3), history of anti-coagulant (HR: 1.4, 95% CI:1.01–1.8), and history of atypical antipsychotic (HR:2.0, 95% CI:1.07–3.06) (Fig. 1). The COXPH model assessing risk factors in the development of later adverse events had the following significant features: history of myocarditis and/or pericarditis (HR:7.20 95% CI:2.27-23.0), history of CEVD (HR:2.00, 95% CI:1.36–2.90), and history of NSAID (HR: 1.60, 95% CI:1.18–2.30) (Fig. 1). All other features in COXPH models were not significantly associated with the risk of the development of adverse events (Fig. 1). For reference, month over month patient censoring over time due to mortality and/or lack of follow up time is detailed in supplemental Table 5. The Schoenfeld residuals test for individual features and overall models for Cox proportional hazards analyses identified no significant violations of the proportional hazards assumption. No features for early or late-onset models had variance inflation factors above 5, indicating no issues with multi-collinearity.

Fig. 1
figure 1

Forrest plots reporting hazard ratios for Cox-proportional hazards models in early and later cardio-inflammatory adverse events. Plot A represents reported hazard ratios for early adverse events, while plot B represents reported hazard ratios for later adverse events. From left to right, each plot reports variable names, hazard ratios, hazard ratio plots, and P-values for each prospective variable

Full size image

Discussion

Requisition of data from a large multicenter, multinational database such as TriNetX followed by implementation of strict inclusion/exclusion criteria allowed for creation of a uniquely large cohort of ICI patients. Adoption of a multi-step predictive modeling workflow identified distinct factors significantly associated with early developing and later cardiac inflammatory irAEs. For early events, combination ICI therapy, anti-PD-1 therapy, and a history of PVD were significant in baseline and optimized logistic regression models in addition to significant risk factors in COXPH models. For later events, history of myocarditis/pericarditis, history of CEVD, and history of NSAID were significant in all 3 multivariate models implemented. While these select few variables were statistically significant with all modeling efforts, many other variables related to medication histories and comorbidities showed strong associations with the development of these adverse events. These discovered associations provide insights to identify patients most at risk of experiencing these serious events.

Combination therapy as a predictor dependent on time removed from therapy initiation

Our study supports the growing body of literature demonstrating a significant association between combination ICI therapy and the development of irAEs and their timing. Generally, combination ICI therapy is suspected to be more toxic than monotherapy due activation multiple immune pathways simultaneously [41, 42]. This causes a more robust immune response which in turn means more unwanted immune system activity will be more likely to cause irAEs. Multiple studies have demonstrated that combination ICI therapy significantly increases the likelihood of development of not only cardio-inflammatory adverse events like myocarditis and pericarditis, but other more common events like dermatitis, thyroiditis, and colitis as well [21, 43,44,45,46,47]. Combination ICI therapy has also been shown to be significantly associated with the occurrence of earlier adverse events (within 90 days of therapy initiation), as opposed to later events in comparison to ICI monotherapy [48, 49]. Our study further supports these relationships within the context the development of myocarditis and pericarditis in ICI patients, demonstrating combination therapy having a strong association with the development of early cardio-inflammatory events and not as much in later events. While combination therapy was by far the strongest predictor of early cardio-inflammatory adverse events, there were several factors significantly more predictive of later developing cardio-inflammatory adverse events relative to ICI combination therapy. Continued evaluation of distinct risk factors associated with the development of early versus later adverse events will allow clinicians additional insight in monitoring these patients.

Medication history as a predictor

Evaluation of patient medication histories and their associations with cardio-inflammatory adverse events provided some unique insights. While none of these variables were significant in all models predicting early adverse events, some showed strong associations including usage history of anticoagulants and atypical antipsychotics. While there is no obvious documented association with anticoagulant agents themselves and cardiac inflammation, the medication may be a surrogate for patients with serious comorbidities that would leave them at risk for such events [50]. Atypical antipsychotics usage however, has a documented history of increased risk of such events in a variety of settings [51, 52]. Various studies have highlighted why atypical anti-psychotics have cardiotoxic properties suggesting that there interruption of membrane ion channel function, adrenoreceptor affinity, and lipophilic properties allowing cardiac cell membrane permeation are all contributing factors [51, 53]. For later adverse events, history of NSAID was a significant predictor with all modeling strategies implemented. Observation of NSAID increasing the likelihood of the development of cardiac adverse events is consistent with prior studies [54]. Observed cardiotoxicities with NSAIDs are often attributed to the disruption of the balance prostaglandins throughout the body and potential increases in blood pressure from taking these agents [54]. Follow up evaluations of these agents usage during ICI therapy as opposed to prior, particularly atypical antipsychotics and NSAIDs, might better inform clinicians on the extent they increase risk of serious cardio-inflammatory adverse events.

Patient comorbidities as a predictor

Comorbidities related to cardiovascular health also had significant association with cardio-inflammation adverse as events as expected. For early events, history of PVD was significant in all predictive models while history of CEVD was similarly significant in all approaches implemented for later events. Peripheral vascular disease has well documented associations with a variety of serious cardiac events, with their underlying causes heavily tied to inflammation triggered by blood vessel occlusion [55]. The direct influence of cerebrovascular disease with cardiac inflammation outcomes appears less obvious, however it’s associations with vascular health and cardiovascular function may indirectly represent a risk factor for patients [56]. We investigated associations of diabetes and relevant therapeutic agents due to established links to between them and cardiovascular outcomes [57, 58]. With respect to later adverse events, DM was strongly associated with their likely development in our analysis. While we hoped to observe the influence of antidiabetic therapies on risk of cardio-inflammatory events, our patient sample was simply limited and did not suggest any significant associations.

IrAE timing considerations

Our investigation into different predictive factors for cardio-inflammatory adverse events in distinct time windows is important given developing evidence related to irAE timing. Multiple studies have demonstrated irAEs occurring later in the course patient therapy significantly predictive of improved treatment outcomes, while earlier developing events do not have the same implications [24, 59]. The distinct significant predictors for early and later events identified our study further support consideration of timing when considering predictive factors of irAEs might offer additional nuance in managing ICI patients. Clinicians administering ICI combination therapy can focus on monitoring patients for cardiac labs over there first few months of therapy. Furthermore, in patients having several different risk factors like ICI combination therapy or histories of different cardiovascular comorbidities, medications like atypical antipsychotics and NSAIDs can be avoided. There are some markers potentially predictive of cardiac inflammatory irAEs that have been identified (i.e. ECG test or troponin), however monitoring all ICI patients for them is not feasible. Patients having multiple risk factors that we have identified here, like certain medication histories with concurrent comorbidities, could be monitored with these tests in an effort to create a practical selective approach [30, 60]. While more work needs to develop optimal methods for identifying at risk patients, we feel our efforts evaluating time dependent risk factors demonstrates that potential risk factors can be identified over the course of treatment trajectories.

Study limitations (Requested as separate section)

The nature of our study does have important limitations that need to be addressed. First, relying on claims data from a database introduces various biases. Additional patient demographic factors that could not be captured that may influence the observed outcomes; future studies assessing risk can explore relationships with key data points like BMI, smoking history, or familial history. Predictive modeling strategies were also influenced by the rarity of cardio-inflammatory adverse events, with many features having very wide confidence intervals describing their influence on the outcome. Future studies can implement different modeling approaches or study designs (i.e. machine learning or population matching), to address the rare nature of this outcome. Furthermore, to ensure no violations of statistical modeling assumptions, past medical histories and medication usage were only collected based on data occurring up to the patients index date. Future studies can evaluate these incidences and/or exposures over time. Lastly, over the course of a year the make-up of an initial cancer patient population identified can significantly change over time due to censoring events (either lack of follow up or mortality). Future studies can focus on more nuanced populations to better address this. Even with these limitations, our study does provide important insight into the development of early and later cardio-inflammatory adverse events in ICI patients.

Conclusion

As demonstrated with various predictive modeling and risk assessment strategies, there are distinct risk factors for the development of early and later immune related adverse events. While early adverse events are heavily associated with initial ICI therapy administered, later events had stronger associations with different patient comorbidities and medication histories. Consideration of early and later events as having distinct risk factors may allow clinicians to better monitor and identify patients most likely to deal with these challenging irAEs.

Data availability

The data underlying this study were collected within TriNetX electronic health records, by means of institutional access through the University of California Irvine. The TriNetX policy states their data can only be only be accessed “by researchers that are either part of the network or have a collaboration agreement with TriNetX.”

Abbreviations

ACE-I:

Angiotensin Converting Enzyme Inhibitor

AIC:

Alkaike information criterion

CEVD:

Cerebral vascular disease

CHF:

Congestive heart failure

COXPH:

Cox-proportional hazards model

HTN:

Hypertension

ICI:

Immune-checkpoint Inhibitor

irAE:

Immune-related adverse event

MI:

Myocardial infarction

NSAID:

Non-steroidal anti-inflammatory drugs

PD-1:

Programmed cell death protein 1

PD-L1:

Programmed Cell Death Ligand 1

PVD:

Peripheral vascular disease

References

  1. Choi J, Lee SY. Clinical characteristics and treatment of immune-related adverse events of immune checkpoint inhibitors. Immune Netw Feb. 2020;20(1):e9. https://doiorg.publicaciones.saludcastillayleon.es/10.4110/in.2020.20.e9

    Article  Google Scholar 

  2. Kuhnly NM, Coviello J. Immune checkpoint inhibitor-related myocarditis: recognition, surveillance, and management. Clin J Oncol Nurs Feb. 2022;1(1):54–60. https://doiorg.publicaciones.saludcastillayleon.es/10.1188/22.cjon.54-60

    Article  Google Scholar 

  3. Gong J, Drobni ZD, Zafar A, et al. Pericardial disease in patients treated with immune checkpoint inhibitors. J Immunother Cancer Jun. 2021;9(6). https://doiorg.publicaciones.saludcastillayleon.es/10.1136/jitc-2021-002771

  4. Liu Y, Wu W. Cardiovascular immune-related adverse events: evaluation, diagnosis and management. Asia Pac J Clin Oncol Aug. 2020;16(4):232–40. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/ajco.13326

    Article  Google Scholar 

  5. Wu Y, Xu Y, Xu L. Drug therapy for myocarditis induced by immune checkpoint inhibitors. Front Pharmacol. 2023;14:1161243. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fphar.2023.1161243

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  6. Komiya K, Nakamura T, Abe T, et al. Discontinuation due to immune-related adverse events is a possible predictive factor for immune checkpoint inhibitors in patients with non-small cell lung cancer. Thorac Cancer Sep. 2019;10(9):1798–804. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/1759-7714.13149

    Article  CAS  Google Scholar 

  7. Celsa C, Cabibbo G, Fulgenzi CAM, et al. Characteristics and outcomes of immunotherapy-related liver injury in patients with hepatocellular carcinoma versus other advanced solid tumours. J Hepatol Mar. 2024;80(3):431–42. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jhep.2023.10.040

    Article  CAS  Google Scholar 

  8. Power JR, Alexandre J, Choudhary A, et al. Electrocardiographic manifestations of immune checkpoint inhibitor myocarditis. Circulation Nov. 2021;2(18):1521–3. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/circulationaha.121.055816

    Article  Google Scholar 

  9. Jaffe AS. Biomarkers for immune checkpoint inhibitor-induced myocarditis: caution needed. JACC CardioOncol. 2022;5:701–3.

    Article  Google Scholar 

  10. Puzanov I, Subramanian P, Yatsynovich YV, et al. Clinical characteristics, time course, treatment and outcomes of patients with immune checkpoint inhibitor-associated myocarditis. J Immunother Cancer Jun. 2021;9(6). https://doiorg.publicaciones.saludcastillayleon.es/10.1136/jitc-2021-002553

  11. Yousif LI, Screever EM, Versluis D, et al. Risk factors for immune checkpoint inhibitor-mediated cardiovascular toxicities. Curr Oncol Rep Jul. 2023;25(7):753–63. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s11912-023-01414-4

    Article  Google Scholar 

  12. Mitsuboshi S, Hamano H, Niimura T, et al. Association between immune checkpoint inhibitor-induced myocarditis and concomitant use of thiazide diuretics. Int J Cancer Oct. 2023;15(8):1472–6. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/ijc.34616

    Article  CAS  Google Scholar 

  13. Zamami Y, Niimura T, Okada N, et al. Factors associated with immune checkpoint inhibitor-related myocarditis. JAMA Oncol Nov. 2019;1(11):1635–7. https://doiorg.publicaciones.saludcastillayleon.es/10.1001/jamaoncol.2019.3113

    Article  Google Scholar 

  14. Palaskas N, Lopez-Mattei J, Durand JB, Iliescu C, Deswal A. Immune checkpoint inhibitor myocarditis: pathophysiological characteristics, diagnosis, and treatment. J Am Heart Assoc. 2020;21(2):e013757. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/jaha.119.013757

    Article  CAS  Google Scholar 

  15. Niimura T, Miyata K, Hamano H, et al. Cardiovascular toxicities associated with anaplastic lymphoma kinase inhibitors: a disproportionality analysis of the WHO pharmacovigilance database (VigiBase). Drug Saf Jun. 2023;46(6):545–52. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s40264-023-01300-9

    Article  CAS  Google Scholar 

  16. Nishiuchi S, Yagi K, Saito H, et al. Investigation of drugs for the prevention of doxorubicin-induced cardiac events using big data analysis. Eur J Pharmacol Aug. 2022;5:928:175083. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.ejphar.2022.175083

    Article  CAS  Google Scholar 

  17. Poels K, van Leent MMT, Boutros C, et al. Immune checkpoint inhibitor therapy aggravates T cell-driven plaque inflammation in atherosclerosis. JACC CardioOncol Nov. 2020;2(4):599–610. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jaccao.2020.08.007

    Article  Google Scholar 

  18. Reynolds G. Rheumatic complications of checkpoint inhibitors: lessons from autoimmunity. Immunol Rev Sep. 2023;318(1):51–60. https://doiorg.publicaciones.saludcastillayleon.es/10.1111/imr.13242

    Article  CAS  Google Scholar 

  19. Del Gaudio A, Di Vincenzo F, Petito V, et al. Focus on immune checkpoint inhibitors-related intestinal inflammation: from pathogenesis to therapeutical approach. Inflamm Bowel Dis Oct. 2023;6. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/ibd/izad229

  20. Hercun J, Vincent C, Bilodeau M, Lapierre P. Immune-mediated hepatitis during immune checkpoint inhibitor cancer immunotherapy: lessons from autoimmune hepatitis and liver immunology. Front Immunol. 2022;13:907591. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fimmu.2022.907591

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. Iyer PC, Cabanillas ME, Waguespack SG, et al. Immune-related thyroiditis with immune checkpoint inhibitors. Thyroid Oct. 2018;28(10):1243–51. https://doiorg.publicaciones.saludcastillayleon.es/10.1089/thy.2018.0116

    Article  CAS  Google Scholar 

  22. Suresh K, Psoter KJ, Voong KR, et al. Impact of checkpoint inhibitor pneumonitis on survival in NSCLC patients receiving immune checkpoint immunotherapy. J Thorac Oncol Mar. 2019;14(3):494–502. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jtho.2018.11.016

    Article  CAS  Google Scholar 

  23. Xu M, Xu L, Hao Y, Shao K, Song Z. Clinical characteristics and prognostic implications of immune-related hepatitis in patients with advanced non-small cell lung cancer treated with immune checkpoint inhibitors: a retrospective study. J Thorac Dis Mar. 2024;29(3):1900–10. https://doiorg.publicaciones.saludcastillayleon.es/10.21037/jtd-23-1684

    Article  Google Scholar 

  24. Hsiehchen D, Naqash AR, Espinoza M, et al. Association between immune-related adverse event timing and treatment outcomes. Oncoimmunology. 2022;11(1):2017162. https://doiorg.publicaciones.saludcastillayleon.es/10.1080/2162402x.2021.2017162

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  25. Khan S, von Itzstein MS, Lu R, et al. Late-onset immunotherapy toxicity and delayed autoantibody changes: checkpoint inhibitor-induced Raynaud’s-like phenomenon. Oncologist May. 2020;25(5):e753–7. https://doiorg.publicaciones.saludcastillayleon.es/10.1634/theoncologist.2019-0666

    Article  CAS  Google Scholar 

  26. Chen DY, Huang WK, Chien-Chia Wu V, et al. Cardiovascular toxicity of immune checkpoint inhibitors in cancer patients: a review when cardiology meets immuno-oncology. J Formos Med Assoc Oct. 2020;119(10):1461–75. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jfma.2019.07.025

    Article  CAS  Google Scholar 

  27. Gan L, Liu D, Ma Y, et al. Cardiotoxicity associated with immune checkpoint inhibitors: current status and future challenges. Front Pharmacol. 2022;13:962596. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fphar.2022.962596

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  28. Zamami Y, Hamano H, Niimura T, et al. Drug-repositioning approaches based on medical and life science databases. Front Pharmacol. 2021;12:752174. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fphar.2021.752174

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  29. Data from: Trinetx Global Health Data Platform. 2023. Deposited 2013.

  30. Vasbinder A, Chen Y, Procureur A, et al. Biomarker trends, incidence, and outcomes of immune checkpoint inhibitor-induced myocarditis. JACC CardioOncol Dec. 2022;4(5):689–700. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jaccao.2022.11.004

    Article  Google Scholar 

  31. Wang F, Sun X, Qin S, et al. A retrospective study of immune checkpoint inhibitor-associated myocarditis in a single center in China. Chin Clin Oncol Apr. 2020;9(2):16. https://doiorg.publicaciones.saludcastillayleon.es/10.21037/cco.2020.03.08

    Article  Google Scholar 

  32. Nishikawa T, Inoue T, Otsuka T, et al. Prevalence and characteristics of immune checkpoint inhibitor-related myocardial damage: a prospective observational study. PLoS ONE. 2022;17(11):e0275865. https://doiorg.publicaciones.saludcastillayleon.es/10.1371/journal.pone.0275865

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  33. Gunturu KS, Pham TT, Shambhu S, Fisch MJ, Barron JJ, Debono D. Immune checkpoint inhibitors: immune-related adverse events, healthcare utilization, and costs among commercial and medicare advantage patients. Support Care Cancer May. 2022;30(5):4019–26. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s00520-022-06826-9

    Article  Google Scholar 

  34. Jain P, Gutierrez Bugarin J, Guha A, et al. Cardiovascular adverse events are associated with usage of immune checkpoint inhibitors in real-world clinical data across the United States. ESMO Open Oct. 2021;6(5):100252. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.esmoop.2021.100252

    Article  CAS  Google Scholar 

  35. George S, Bell EJ, Zheng Y, et al. The impact of adverse events on health care resource utilization, costs, and mortality among patients treated with immune checkpoint inhibitors. Oncologist Jul. 2021;26(7):e1205–15. https://doiorg.publicaciones.saludcastillayleon.es/10.1002/onco.13812

    Article  CAS  Google Scholar 

  36. Tao Y, Li X, Liu B, et al. Association of early immune-related adverse events with treatment efficacy of neoadjuvant toripalimab in resectable advanced non-small cell lung cancer. Front Oncol. 2023;13:1135140. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fonc.2023.1135140

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  37. Ksienski D, Wai ES, Croteau N, et al. Efficacy of nivolumab and pembrolizumab in patients with advanced non-small-cell lung cancer needing treatment interruption because of adverse events: a retrospective multicenter analysis. Clin Lung Cancer Jan. 2019;20(1):e97–106. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.cllc.2018.09.005

    Article  CAS  Google Scholar 

  38. Yang Y, Li L, Tian J, et al. Delayed immune-related adverse events profile associated with immune checkpoint inhibitors: a real-world analysis. Front Pharmacol. 2024;15:1453429. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fphar.2024.1453429

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  39. Gasparini A. An R package for computing comorbidity scores. Journal of Open Source Software; 2018.

  40. Quan H, Li B, Couris CM, et al. Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am J Epidemiol Mar. 2011;15(6):676–82. https://doiorg.publicaciones.saludcastillayleon.es/10.1093/aje/kwq433

    Article  Google Scholar 

  41. Liu L, Li S, Wang G, et al. Dynamic toxicity landscape of immunotherapy for solid tumors across treatment lines. J Natl Cancer Cent Sep. 2023;3(3):186–96. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.jncc.2023.07.002

    Article  Google Scholar 

  42. Ma X, Zhang Y, Wang S, Wei H, Yu J. Immune checkpoint inhibitor (ICI) combination therapy compared to monotherapy in advanced solid cancer: a systematic review. J Cancer. 2021;12(5):1318–33. https://doiorg.publicaciones.saludcastillayleon.es/10.7150/jca.49174

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  43. Naqash AR, Moey MYY, Cherie Tan XW, et al. Major adverse cardiac events with immune checkpoint inhibitors: a pooled analysis of trials sponsored by the National Cancer Institute-cancer therapy evaluation program. J Clin Oncol Oct. 2022;10(29):3439–52. https://doiorg.publicaciones.saludcastillayleon.es/10.1200/jco.22.00369

    Article  Google Scholar 

  44. Wu Z, Chen Q, Qu L, et al. Adverse events of immune checkpoint inhibitors therapy for urologic cancer patients in clinical trials: a collaborative systematic review and meta-analysis. Eur Urol Apr. 2022;81(4):414–25. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.eururo.2022.01.028

    Article  CAS  Google Scholar 

  45. Nguyễn T, Maria ATJ, Ladhari C, et al. Rheumatic disorders associated with immune checkpoint inhibitors: what about myositis? An analysis of the who’s adverse drug reactions database. Ann Rheum Dis. 2022;2:e32.

    Article  Google Scholar 

  46. Nazzaro G, Buffon S, Giacalone S, Maronese CA, Marzono MA. Skin manifestations associated with checkpoint inhibitors. 2 ed. Journal of the European Academy of Dermatology and Venereology; 2022. pp. 73–87.

  47. Losurdo G, Angelillo D, Favia N, et al. Checkpoint inhibitor-induced colitis: an update. Biomedicines May. 2023;22(5). https://doiorg.publicaciones.saludcastillayleon.es/10.3390/biomedicines11051496

  48. Tang SQ, Tang LL, Mao YP, et al. The pattern of time to onset and resolution of immune-related adverse events caused by immune checkpoint inhibitors in cancer: a pooled analysis of 23 clinical trials and 8,436 patients. Cancer Res Treat Apr. 2021;53(2):339–54. https://doiorg.publicaciones.saludcastillayleon.es/10.4143/crt.2020.790

    Article  CAS  Google Scholar 

  49. Johnson DB, Nebhan CA, Moslehi JJ, Balko JM. Immune-checkpoint inhibitors: long-term implications of toxicity. Nat Rev Clin Oncol Apr. 2022;19(4):254–67. https://doiorg.publicaciones.saludcastillayleon.es/10.1038/s41571-022-00600-w

    Article  Google Scholar 

  50. Umerah CO, Momodu. II. Anticoagulation. StatPearls. StatPearls publishing copyright © 2023. StatPearls Publishing LLC.; 2023.

  51. Daniel P, Rajaree KM, Rudy L, et al. Myocarditis in patients on long-term antipsychotics -mechanism, management and recent updates. Heliyon Mar. 2023;9(3):e13930. https://doiorg.publicaciones.saludcastillayleon.es/10.1016/j.heliyon.2023.e13930

    Article  Google Scholar 

  52. Li XQ, Tang XR, Li LL. Antipsychotics cardiotoxicity: what’s known and what’s next. World J Psychiatry Oct. 2021;19(10):736–53. https://doiorg.publicaciones.saludcastillayleon.es/10.5498/wjp.v11.i10.736

    Article  Google Scholar 

  53. Scigliano G, Ronchetti G. Antipsychotic-induced metabolic and cardiovascular side effects in schizophrenia: a novel mechanistic hypothesis. CNS Drugs Apr. 2013;27(4):249–57. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s40263-013-0054-1

    Article  CAS  Google Scholar 

  54. Varga Z, Sabzwari SRA, Vargova V. Cardiovascular risk of nonsteroidal anti-inflammatory drugs: an under-recognized public health issue. Cureus Apr. 2017;8(4):e1144. https://doiorg.publicaciones.saludcastillayleon.es/10.7759/cureus.1144

    Article  Google Scholar 

  55. Paquissi FC. The role of inflammation in cardiovascular diseases: the predictive value of neutrophil-lymphocyte ratio as a marker in peripheral arterial disease. Ther Clin Risk Manag. 2016;12:851–60. https://doiorg.publicaciones.saludcastillayleon.es/10.2147/tcrm.s107635

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  56. Hannawi Y. Cerebral small vessel disease: a review of the pathophysiological mechanisms. Transl Stroke Res Oct. 2023;21. https://doiorg.publicaciones.saludcastillayleon.es/10.1007/s12975-023-01195-9

  57. Ritchie RH, Abel ED. Basic mechanisms of diabetic heart disease. Circ Res May. 2020;22(11):1501–25. https://doiorg.publicaciones.saludcastillayleon.es/10.1161/circresaha.120.315913

    Article  Google Scholar 

  58. Brown JM, Everett BM. Cardioprotective diabetes drugs: what cardiologists need to know. Cardiovasc Endocrinol Metab Dec. 2019;8(4):96–105. https://doiorg.publicaciones.saludcastillayleon.es/10.1097/xce.0000000000000181

    Article  CAS  Google Scholar 

  59. Sayer MR, Mambetsariev I, Lu KH, et al. Predicting survival of NSCLC patients treated with immune checkpoint inhibitors: impact and timing of immune-related adverse events and prior tyrosine kinase inhibitor therapy. Front Oncol. 2023;13:1064169. https://doiorg.publicaciones.saludcastillayleon.es/10.3389/fonc.2023.1064169

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  60. Talukder S, Ghose A, Chakraborty T et al. Evolving cardiac biomarkers for immune checkpoint inhibitor related myocarditis in cancer patients. Int J Cardiol Heart Vasc. 2023;49:101278.

Download references

Acknowledgements

We would like to express my gratitude to Dr. Wendy Cozen, Professor, Division of Hematology/Oncology, School of Medicine, University of California, Irvine, for her contributions to this project. Her insightful feedback and support have been appreciated.

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Author information

Authors and Affiliations

  1. School of Pharmacy & Pharmaceutical Sciences, University of California, 802 W Peltason Dr, Room 106A, Irvine, CA, 92617, USA

    Michael Sayer & Aya F. Ozaki

  2. Department of Pharmacy, Okayama University Hospital, Okayama, Japan

    Hirofumi Hamano & Yoshito Zamami

  3. Division of Hematology and Oncology, University of California, Irvine, CA, USA

    Misako Nagasaka

  4. Department of Pharmacy, University of California Irvine Health, Orange, CA, USA

    Benjamin J. Lee & Jean Doh

  5. Division of Cardiology, Department of Medicine, University of California, Irvine, CA, USA

    Pranav M. Patel

Authors
  1. Michael Sayer
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Hirofumi Hamano
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Misako Nagasaka
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Benjamin J. Lee
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Jean Doh
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Pranav M. Patel
    View author publications

    You can also search for this author inPubMed Google Scholar

  7. Yoshito Zamami
    View author publications

    You can also search for this author inPubMed Google Scholar

  8. Aya F. Ozaki
    View author publications

    You can also search for this author inPubMed Google Scholar

Contributions

Michael Sayer was involved with study design, conceptualization, designing, data curation, analysis, visualization, writing- draft and review and editing. Hirfumi Hamano and Yoshito Zamami were involved with conceptualization, analysis, and writing- draft and review and editing. Ali Naquvi, Pranav M. Patel, Jean Doh, and Benjamin J. Lee provided supervision and were involved with the interpretation of the results, and provided critical feedback to the manuscript. Aya F. Ozaki supervised and lead the study, and was involved with conceptualization, designing, and writing- draft and review and editing.

Corresponding author

Correspondence to Aya F. Ozaki.

Ethics declarations

Competing interests

The authors declare no competing interests.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary Material 1

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Sayer, M., Hamano, H., Nagasaka, M. et al. Time dependent predictors of cardiac inflammatory adverse events in cancer patients receiving immune checkpoint inhibitors. Cardio-Oncology 11, 40 (2025). https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s40959-025-00331-8

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doiorg.publicaciones.saludcastillayleon.es/10.1186/s40959-025-00331-8

Keywords

Cardio-Oncology

ISSN: 2057-3804

Contact us