Pre-Existing Cardiovascular Disease Increases Risk of Atrial Arrhythmia in Cancer Patients Treated with Ibrutinib

Background: Ibrutinib is a Bruton’s tyrosine kinase inhibitor used in the treatment of hematological malignancies. The most common cardiotoxicity associated with ibrutinib is atrial arrhythmia (atrial brillation and utter). It is known that patients with cardiovascular disease (CVD) are at an increased risk for developing atrial arrhythmia. However, the rate of atrial arrhythmia in patients with pre-existing CVD treated with ibrutinib is unknown. Objective: This study examined whether patients with pre-existing CVD are at a higher risk for developing atrial arrhythmias compared to those without prior CVD. Methods: A single-institution retrospective chart review of patients with no prior history of atrial arrhythmia treated with ibrutinib from 2012 to 2020 was performed. Patients were grouped into two cohorts: those with CVD (known history of coronary artery disease, heart failure, pulmonary hypertension, at least moderate valvular heart disease, or device implantation) and those without CVD. The primary outcome was incidence of atrial arrhythmia, and the secondary outcomes were all-cause mortality, risk of bleeding, and discontinuation of ibrutinib.


Introduction
Ibrutinib is a Bruton's tyrosine kinase (BTK) inhibitor that has been approved for the treatment of multiple hematological malignancies either as a single agent or as a combination till date. Ibrutinib had shown durable single agent e cacy in patients with relapsed or refractory mantle cell lymphoma which led to its rst approval in November 2013 (1). Till date, this drug has been approved and indicated for treatment of chronic lymphocytic leukemia (CLL)/small lymphocytic lymphoma (SLL), marginal zone lymphoma, Waldenstrom macroglobulinemia, and chronic graft versus host disease (GvHD) (1,2,3,4). Furthermore, it exhibited signi cant activity in patients with CLL with 17p del or TP53 mutation (5).
Ibrutinib has well-documented side effect pro le including bleeding, infections, cytopenias, hypertension, tumor lysis syndrome, and cardiac arrhythmias. The cardiac arrhythmias include ventricular tachyarrhythmias and, most commonly, atrial brillation. Atrial brillation was documented in 4% of patients in clinical trials and 5.77 per 100 person-years (6). Another study has demonstrated 11% incidence of atrial arrhythmia over 5-year follow-up (7). Ibrutinib has an off-target inhibitory effect on Tec protein tyrosine kinase (TEC). Both BTK and TEC are expressed in cardiac tissue with an increased concentration in atrial tissue (8).
Atrial brillation is generally more common in older adults, affecting 11-12% of people greater than 85 years of age (9). It is also associated with multiple modi able risk factors including obesity, hypertension, diabetes, obstructive sleep apnea, alcohol consumption, smoking, and sedentary lifestyles. Coronary artery disease (CAD) has also been associated with atrial brillation, with the prevalence of CAD in patients with atrial brillation ranging from 17-46.5% (10). Incidence of arrhythmia has been shown to increase by up to 40% after coronary artery bypass surgery (10). Chronic in ammation and ischemia seen in CAD trigger myocyte degeneration, activating necrotic or apoptotic signals and triggering reparative brosis (11). Studies have shown increased collagen deposition in patients with lone atrial brillation compared to controls with sinus rhythm (12). Although some of these risk factors have a more pronounced contribution to the incidence of atrial brillation, ibrutinib was shown to increase the incidence of atrial brillation independently of these risk factors (8). Since both cardiovascular disease and the use of ibrutinib are associated with an escalated risk for atrial brillation, our study aimed to investigate the incidence of new-onset atrial brillation after ibrutinib initiation in patients with underlying cardiovascular disease (CVD).

Methods
This study's primary outcome is to assess if prior CVD increased the risk of atrial brillation incidence after receiving ibrutinib. The secondary outcomes were all-cause mortality, bleeding, and discontinuation of Ibrutinib. We obtained approval and appropriate oversight from the Institutional Board Review prior to data collection initiation. We retrospectively collected data on consecutive patients who received ibrutinib from November 2012 to September 2020 at a single institution. We chose this study period because it was the rst initiation date of ibrutinib for a patient to the last known follow-up. We divided the patients into two cohorts 1) patients with prior cardiovascular disease (CVD) and 2) patients without CVD.
Patients with known CAD (with [12%] or without [6.5%] revascularization), known chronic heart failure with reduced ejection fraction (4.6%) or chronic heart failure with preserved ejection fraction (10.1%), pulmonary hypertension (1%), pacemakers (1.4%), implantable cardioverter de brillators (0.5%), ventricular arrhythmia (1.8%), or at least moderate valvular heart disease (8.3%) were included in the prior CVD group. Since one of the outcomes of interest was a new diagnosis of atrial brillation, we excluded patients with an existing diagnosis of atrial brillation before ibrutinib initiation. We also excluded patients with missing data for medical history or the ibrutinib initiation date. We also excluded patients who received Ibrutinib for GVHD. The clinical variables are either self-reported by the patients, documented diagnoses from clinicians, or con rmed through review of relevant cardiac testing including, but not limited to, electrocardiogram (ECG) or echocardiography.
Categorical variables were analyzed with the chi-square test and expressed as percentages. The continuous variables were analyzed with the Mann-Whitney U test after evaluating the normalcy of data (using the Shapiro-Wilk test) and expressed as medians and interquartile range. A binary logistic regression was created to predict CVD as an independent predictor for new-onset atrial brillation after adjusting for age, chronic kidney disease, diabetes, and initiation dose of the medications. The nding was reported as an odds ratio with a 95% con dence interval. We also created a Cox proportional hazards model to factor in age, cancer diagnoses, post-chemotherapy atrial brillation, and bleeding complications for CVD to predict mortality expressed as hazard ratio with 95% con dence intervals. The follow-up time for this analysis was calculated from initiation to the last known follow-up date in the chart. The survival probability was estimated using the Kaplan-Meier curve and the log-rank test was used to compare the prior CVD and without CVD groups. The statistical analysis was conducted on SPSS (IBM Corp., Amrock, New York) and SAS 9.4 (SAS Institute, Cary, North Carolina).

Results
We identi ed 300 patients taking ibrutinib from November 2012 to September 2020. Of these, 51 patients had missing data for medical history, 7 patients had (GvHD), 18 patients had missing initiation dates, and 7 patients had atrial brillation before chemotherapy initiation. After excluding these patients, the nal cohort included 217 patients for further analysis. The follow-up time was available for 189 patients (87%), and the median follow-up time was 1.1 years, with an IQR of 0.4 to 2.6 years. The entire cohort's median age was 74 years (IQR: 64-81 years). Males comprised 64% of the entire cohort and majority of the patients were White (93%).
Chronic lymphocytic leukemia (CLL) and small lymphocytic lymphoma (SLL) were the most common forms of cancer (72%), followed by mantle cell lymphoma (16%) and Waldenstrom macroglobulinemia (5%). Marginal zone lymphoma, diffuse large B-cell lymphoma, and other types of hematologic malignancies constituted the remaining 6%.

Baseline Characteristics
Prior CVD cohort had 69 patients (32%) and without CVD cohort had 148 patients (68%). The median ages for the prior CVD and without CVD cohorts were 74 and 70 years (p = 0.02). There were no statistical differences in gender or race between the two cohorts. The patients with prior CVD had a higher burden of comorbidities, such as hypertension, diabetes, dyslipidemia, and chronic obstructive lung disease (Table 1). There was no statistical difference for cerebrovascular events, liver disease, chronic kidney disease, tobacco use, or illicit drug use between the two cohorts. Higher proportions of patients in the prior CVD cohort were on statins and neurohormonal blocking drugs (e.g., beta blockers, angiotensinconverting enzyme inhibitors) and aspirin. However, there was no statistical difference for CYP3A inhibitors or inducers, anticoagulants, or diuretics between the two groups (Table 1).   25; p = 0.02) of developing atrial brillation than those without prior CVD. Secondary outcomes including bleeding (19% with prior CVD vs. 16% without CVD; p = 0.65) and discontinuation of ibrutinib (44% with prior CVD vs. 31% without CVD; p = 0.08) were not statistically signi cant between the two cohorts, indicating CVD status was not signi cantly associated with either bleeding events or discontinuation of ibrutinib. However, the mortality rate was signi cantly higher in patients with prior CVD (39%) vs. without CVD (23%) (p = 0.01) over the follow-up period. After adjusting for the following confounders age, new-onset atrial brillation, and cancer diagnoses (CLL/SLL and other cancers combined were treated as two groups), a hazard ratio for mortality with CVD as a predictor was 1.90 (95% con dence interval: 1.06-3.41). The estimated survival probability was 43% with prior CVD and 54% without CVD over the follow-up period (log-rank test p = 0.0484) (Fig. 1).

Discussion
Given the widespread use of ibrutinib, patients and clinicians must be aware of the cardiotoxicity of the medication. Atrial brillation is the most commonly encountered cardiac arrhythmia and contributes to myriad complications. Our single-institution study examined the incidence of new-onset atrial arrhythmia in patients with and without pre-existing CVD treated with ibrutinib for various hematologic malignancies, most commonly CLL. Hypertension, diabetes, hyperlipidemia, and COPD were the most signi cant comorbidities in our cohort of patients with prior CVD and remain amongst the most common causes of major modi able risk factors for CVD across the general population (Table 1). While the association between ibrutinib use and the development of atrial brillation has been previously cited in the literature as high as 6-16% (13,14,15), our data demonstrated that patients with pre-existing CVD prior to initiation of ibrutinib were nearly three-times more likely to develop atrial brillation than those without prior CVD. A previous study demonstrated that an elevated Framingham Heart Study-AF score was associated with increased incidence of atrial brillation in those receiving ibrutinib therapy (16). Survival probability estimates also differed signi cantly for those patients with a history of CVD versus those without CVD, with an absolute difference of 11% in those patients with prior CVD vs. those without CVD within our follow-up. Previous studies have demonstrated the time-to-onset of atrial brillation after ibrutinib initiation to occur within the rst year of ibrutinib initiation (17,18), making our mean follow-up time of 1.1 years to be adequate.
The pathophysiology of atrial brillation is multifactorial, and various etiologies have shown association with the development of atrial brillation. In particular, cardiac conditions associated with persistent in ammation and ischemia have been implicated in the development of atrial brillation. Valvular heart disease, congestive heart failure, and coronary artery disease also increase the likelihood of atrial brillation. Elevated lling pressures within the atria, either due to hemodynamically signi cant valvular disease or diminished cardiac function, leads to eventual chamber dilation and resultant brosis. This cardiac remodeling provides the foundation for the electrical disturbances, primarily through ion channel dysfunction and myocyte uncoupling associated with atrial brillation (19). On a molecular level, pro brotic growth factors, such as transforming growth factor-beta1 (TGF-B1) and platelet-derived growth factor/vascular endothelial growth factor (PDGF/VEGF), are upregulated as a result of various cardiac injury (20).
Disruptions in molecular signaling also provide insights into the association of ibrutinib and atrial brillation. Through the binding to cysteine 481 residue of BTK, ibrutinib inhibits the dysregulated B-cell receptor signaling responsible for proliferation and survival in B-cell malignancies (21,22). BTK is expressed in human cardiac tissue and appears to be expressed greater in patients with atrial brillation compared to those in normal sinus rhythm (23,24), which could explain the pro-arrhythmogenic effects of ibrutinib in atrial dysrhythmias. Similar sequela of atrial dysfunction has been seen in those with genetic mutations of tyrosine kinase pathways, such as the KCNA5 mutation which encodes the ultrarapid delayed recti er potassium channel that then in turn modulates tyrosine kinase signaling (25).
Other in vivo animal studies have suggested that ibrutinib induces atrial brillation through structural remodeling and dysregulated calcium handling within atrial myocytes (26).
Our study demonstrates the increased rate of atrial brillation and decreased survival in patients with preexisting CVD initiated on ibrutinib for hematological malignancies. No formal guidelines exist for how to monitor and treat this population, although some strategies have been proposed through the assessment of hemodynamic stability, ECG and echocardiograph ndings, and careful assessment of drug-drug interactions in rate/rhythm control pharmacotherapy (26). While most studies on the management of atrial brillation are in patients without malignancy, cancer patients with atrial brillation are at an increased risk of both heart failure and thromboembolism (27). In addition to anticoagulation medications, rate and rhythm control medications also interact with ibrutinib through the CYP450 CYP3A metabolic pathway, making management of atrial brillation di cult (28). Surveillance electrocardiography in order to identify left atrial abnormalities has been identi ed as a simple clinical tool, especially in the early stages of treatment (29). Routine blood pressure monitoring should also be part of regular surveillance metrics as worsening hypertension has also been associated with ibrutinib use (30,31). More studies are warranted to help create sound clinical guidelines for the management of atrial brillation in patients being treated with ibrutinib with careful consideration to stroke risk and other cardiac complications.

Limitations
Our study was limited due to a relatively small sample size and inherent nature of retrospective studies. Additionally, our search for patients on ibrutinib was limited to our institutional EMR only. From the chart review, it was di cult to locate exact dates or timing of atrial brillation, and "time-to-event" was not possible. Our decision to include patients with various cardiac conditions was based on our referral patterns, but future studies should focus on the impact of speci c cardiovascular conditions on ibrutinib patients. Many echocardiographic variables, such as atrial size, are not available since the majority of patients did not undergo routine echocardiography before and while on ibrutinib. Therefore, we couldn't adjust for these variables.
Our study highlights important implications for management of patients on Ibrutinib. While our study did not reveal statistically signi cant differences in bleeding events between the two cohorts, predisposition to bleeding is a well-known side effect of ibrutinib (32,33). This must be taken into consideration when either initiating, continuing, or terminating anticoagulation therapy for stroke and thromboembolic prevention in patients receiving ibrutinib therapy, as certain groups of patients may already be receiving antiplatelet therapy. This also presents a clinical conundrum in the decision of whether to discontinue ibrutinib at the expense of progressive oncologic disease or to continue ibrutinib treatment with the risk of cardiovascular compromise. Our study did not reveal a difference in the rate of ibrutinib discontinuation in our two cohorts; however, atrial brillation remains the most common cause of ibrutinib discontinuation in treated patients (34,35).

Conclusions
Cardio-oncology has emerged as a quickly growing specialty to help manage patients with malignancy and cardiovascular comorbidities. We highly encourage hematology-oncologists to work together with cardio-oncologists in a multi-disciplinary setting for effective management of patients on ibrutinib therapy. Through early identi cation of cardiac disease, assessment of cardiovascular structure and function, optimization of cardiovascular risk factors, and the utilization of surveillance testing such as serial electrocardiogram during treatment (Fig. 2), cardio-oncologists play a vital role in the prevention and management of cardiotoxicity in cancer patients.

Consent for publication: Not applicable
Availability of data and materials: The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.
Competing interests: the authors declare that they have no competing interests.