Abstract
Importance There is a paucity of information on the association between clonal hematopoiesis of indeterminate potential (CHIP) and cardiovascular disease (CVD) in patients with cancer, including those with multiple myeloma (MM) undergoing hematopoietic cell transplant (HCT), a population at high risk of developing CVD after HCT.
Objective To examine the association between CHIP and CVD in patients with MM and to describe modifiers of CVD risk among those with CHIP.
Design, Setting, and Participants This was a retrospective cohort study of patients with MM who underwent HCT between 2010 and 2016 at City of Hope Comprehensive Cancer Center in Duarte, California, and had pre-HCT mobilized peripheral blood stem cell (PBSC) products cryopreserved and accessible for CHIP analyses. The study team performed targeted panel DNA sequencing to detect the presence of CHIP (variant allele frequency 2% or more).
Main Outcomes and Measures The primary end point was the 5-year cumulative incidence and risk for developing de novo CVD (heart failure, coronary artery disease, or stroke) after HCT.
Results Of 1036 consecutive patients with MM (580 male [56%]; median age, 60.0 years) who underwent a first autologous HCT, 201 patients had at least 1 CHIP variant (19.4%) and 35 patients had 2 or more variants (3.4%). The 5-year incidence of CVD was significantly higher in patients with CHIP (21.1% vs 8.4%; P < .001) compared with those without CHIP; the 5-year incidence among those with 2 or more variants was 25.6%. In the multivariable model, CHIP was associated with increased risk of CVD (hazard ratio [HR], 2.72; 95% CI, 1.70-4.39), as well as of individual outcomes of interest, including heart failure (HR, 4.02; 95% CI, 2.32-6.98), coronary artery disease (HR, 2.22; 95% CI, 1.06-4.63), and stroke (HR, 3.02; 95% CI, 1.07-8.52). Patients who had both CHIP and preexisting hypertension or dyslipidemia were at nearly 7-fold and 4-fold increased risk of CVD, respectively (reference: no CHIP, no hypertension, or dyslipidemia).
Conclusion and Relevance CHIP was significantly and independently associated with risk of CVD in patients with MM undergoing HCT and may serve as a novel biologically plausible biomarker for CVD in this cohort. Patients with MM and both CHIP and cardiovascular risk factors had an exceptionally high risk of CVD. Additional studies are warranted to determine if cardiovascular preventive measures can reduce CHIP-associated CVD risk.
Introduction
Autologous hematopoietic cell transplant (HCT) is an effective treatment for patients with multiple myeloma (MM), especially for individuals who are medically fit and demonstrate disease response to induction therapy.1,2 In fact, MM is the most common indication for autologous HCT, with more than 7000 patients undergoing transplant annually in the US alone.3 Research into survivorship issues during the past 2 decades has highlighted the high risk for developing cardiovascular disease (CVD) following autologous HCT, especially among patients with lymphoma attributed to pre-HCT exposures to cardiotoxic chemotherapies, HCT-associated conditioning exposures, and de novo modifiable cardiovascular risk factors (eg, hypertension, diabetes, dyslipidemia) that emerge after HCT.4–6 Yet, there is a paucity of information on the risk of CVD after HCT7 for patients with MM, including the role of emerging biologic modifiers of CVD risk in the general population.
Clonal hematopoiesis of indeterminate potential (CHIP) refers to clonal expansion of blood cells driven by somatic leukemogenic variants in otherwise healthy or patients or individuals without leukemia.8 In nononcology populations, CHIP has been associated with aging-related health conditions, including coronary artery disease (CAD),9,10 heart failure,11 and stroke.12 Accumulating data suggest CHIP is more prevalent in patients with cancer after cancer treatment exposures compared with age-matched controls, attributed in part to clonal selection that results from myelotoxic stressors.13
Despite the growing body of literature on the risk of CVD associated with CHIP in nononcology populations, there is limited information on the association between CHIP and CVD risk among patients with cancer, a population with high rates of CHIP that may be driven by the cancer itself, associated comorbidities, or cancer treatment exposures. Therefore, we examined the association between pre-HCT CHIP and CVD after HCT in a demographically diverse population of patients with MM and further explored the interaction between modifiable cardiovascular risk factors at the time of HCT and CHIP on long-term cardiovascular outcomes.
Discussion
In this study, leveraging a large, well-characterized, and demographically diverse cohort of patients with MM undergoing autologous HCT with available pre-HCT DNA, we found that CHIP was highly prevalent and associated with a significantly higher incidence of CVD after HCT compared with those without CHIP. In the adjusted model, CHIP was associated with a nearly 3-fold risk of developing any CVD and this association remained consistent by CVD subtype. The incidence of CVD after HCT was highest in patients with CHIP and hypertension or dyslipidemia (30% each), corresponding to nearly 7-fold and 4-fold increased risk of CVD, respectively, compared with those without CHIP and these modifiable risk factors. The findings from this study highlight the potential role of pre-HCT CHIP as a novel biomarker to better define CVD risk in patients with MM prior to HCT.
The current study is in line with the growing body of literature demonstrating that CHIP is a commonly occurring somatic mutation in patients with cancer exposed to cytotoxic therapies,25,26 given that all patients had received induction treatment prior to referral to HCT. That said, the associations between HCT-CI (comorbidity burden),27 complete remission status, and CHIP are unique. The association between high HCT-CI score and CHIP may be explained, in part, by the chronic proinflammatory state attributed to underlying comorbidities, an association also reported in patients living with HIV28 or those who develop premature menopause due to chronic illness.29 The association between complete remission status and CHIP may be attributed to a higher disease burden at HCT and associated challenges of maintaining durable response prior to HCT.
The presence of CHIP prior to HCT was significantly and independently associated with increased risk of de novo heart failure, CAD, stroke, as well as composite CVD. Sensitivity analysis censoring follow-up at the time of relapse/recurrence further confirmed these associations. There was also a graded association between the number of CHIP variants and incidence of CVD, suggesting a dose-dependent and additive effect. Variants in ASXL1 were most significantly associated with risk of CVD. We speculate this association may be driven in part by that the high (more than 50%) proportion of CVD cases that were heart failure. ASXL1 has been previously identified as uniquely associated with risk of having reduced left ventricular ejection fraction in a large nononcology population-based CHIP study11 and there is preclinical evidence to suggest a potential causative relationship between ASXL1-mediated clonal hematopoiesis and heart failure.30
Despite the strong association between CHIP and CVD, to our knowledge, to date, there are no CHIP-targeted interventions to modulate the associated CVD risk. Modifiable risk factors, such as hypertension, diabetes, and dyslipidemia, are established targets for CVD risk reduction in the general population but have not been studied in the context of CHIP. While there was no statistically significant interaction between these risk factors and CHIP, we believe our combined analyses reflect clinically meaningful subgroups of patients with MM. In our study, patients with CHIP and hypertension had nearly 7-fold risk of developing CVD and the risk was nearly 4-fold in patients with CHIP and dyslipidemia, compared with patients without CHIP and these individual risk factors. These results underscore the exceedingly high risk of CVD in the subset of patients with MM who have both CHIP and modifiable CVD risk factors, which may provide the rationale for resource allocation to develop more personalized interventions, such as early screening and treatment of these modifiable risk factors (eg, intensive blood pressure control and lipid-lowering strategies),31 in coordination with cardiologists to decrease the morbidity and mortality associated with CVD after HCT.
Conclusions
In summary, we demonstrate a significant association between CHIP and incident CVD after HCT in patients with MM, with the strongest association noted among patients undergoing HCT with modifiable risk factors, such as hypertension and diabetes. These findings may inform more targeted approaches to screening for CVD risk, allowing for improved CVD risk prediction prior to HCT and to guide the implementation of personalized preventive measures to improve health outcomes in at-risk patients. Additional studies are needed to further delineate if the association between CHIP and CVD exists in other cancer populations (eg, solid tumors, nonleukemic hematologic cancers) and to interrogate the specific gene-environment interactions that modulate long-term CVD risk.