Abstract

Introduction: Clonal Hematopoiesis (CH) refers to age-associated somatic mutations in hematopoietic stem cells. CH may be incidentally discovered in oncology patients via next-generation sequencing. Cancer therapies such as cytotoxic chemotherapy and radiation can accelerate the development or expansion of CH, particularly in genes associated with therapy-related myeloid neoplasms. Certain CH variants are also linked to increased cardiovascular (CV) risk. Understanding which patients are at higher risk of poor CV outcomes is an underrecognized threat to long-term cancer survivorship.

Methods: This retrospective cohort study of Avera Sequencing and Analytics Protocol (ASAP; NCT05142033) patients used Tempus xF+, a cell-free DNA liquid biopsy panel (~1.8 Mb) that detects single-nucleotide variants and insertions/deletions across 523 genes, including all tests completed prior to March 2025. CH classification was based on mutations in 17 commonly associated genes. Data on demographics, CV outcomes, (e.g., stroke, hypertension, angina, myocardial infarction, heart failure, arrhythmias, valvular heart disease, and venous thrombosis), cancer characteristics, and labs were extracted from the EMR using ICD-10-based SQL queries. Statistical analyses evaluated differences by presence of CH.

Results: A total of 456 patients were included in the final analysis. 197 (43%) patients were identified as CH-positive. The most frequently mutated genes were DNMT3A, TP53, and PPM1D. Results supported that CH is an age-associated condition, with prevalence rising from 11% in patients under 50 years old, to 30% among those aged 50-59, 42% in the 60-69 age group and reaching 56% in patients over 70. CV events were observed in 72.6% of CH patients, compared to 57.5% of non-CH patients. Among CH-positive individuals, the average VAF was 8.4% (0.3% – 77.0%) for those with CV events, versus 7.6% (0.3% – 51.6%) for those without. Consistent with existing literature, TET2 and ASXL1 mutations conferred the highest probabilities of developing CV outcomes. Pancreatic (OR 5.9, p = 0.2), lung (OR 4.1, p = 0.02), and esophageal (OR 3.5, p = 0.6) cancers exhibited strong but variably significant associations with CV risk, potentially limited by sample size.

Conclusion: Our findings are consistent with previously reported data showing higher rates of CV outcomes in patients with CH in this oncology patient cohort. Some limitations of this study include the retrospective, single cohort design, the use of correlational data, and potential selection bias for study inclusion. This data supports the need for development of cardiology referral guidelines to reduce morbidity and mortality of CV outcomes in oncology patients where CH mutations are identified by liquid biopsy testing.

VIEW THE PUBLICATION