Mutations in CHIP-associated genes at myeloid neoplasm diagnosis and risk of cardiovascular/cerebrovascular events.
Clonal hematopoiesis of indeterminate potential (CHIP)-associated somatic mutations are associated with increased cardiovascular and cerebrovascular risk in the general population; whether mutations detected at myeloid neoplasm (MN) diagnosis predict incident cardiovascular and cerebrovascular events (CCVEs) is unclear.
We retrospectively studied 203 adults with newly diagnosed myelodysplastic neoplasms, myelodysplastic/myeloproliferative neoplasms, or acute myeloid leukemia unfit for intensive chemotherapy (2016-2023) who underwent pretreatment next-generation sequencing. Cause-specific hazard models adjusted for cardiovascular risk factors assessed associations between CHIP-associated gene mutations and incident CCVEs, including age-stratified and age-by-mutation interaction analyses.
Median age was 68 years (IQR, 61-75); 127 (62.6%) were male; 141 (69.5%) carried ≥1 CHIP-associated gene mutation, most commonly ASXL1, TET2, TP53, or DNMT3A. Over median 12-month follow-up, 74 (36.5%) experienced a first CCVE. Any CHIP-associated gene mutation was associated with CCVE risk in univariable analysis; this association was not significant after multivariable adjustment (adjusted hazard ratio [aHR], 1.58; 95% CI, 0.87-2.88; p=0.13). In joint age-mutation analyses, mutation-positive patients aged ≥70 years had the highest CCVE risk (aHR, 2.77; 95% CI, 1.17-6.56), whereas younger mutation-positive patients showed a nonsignificant trend (aHR, 2.16; 95% CI, 0.90-5.20; p=0.085); interaction was not significant (p=0.29). In multivariable models, CHIP-associated gene mutations, higher-risk MN, and incident CCVEs were each associated with higher all-cause mortality.
CHIP-associated gene mutations at MN diagnosis were not independently associated with incident CCVEs. Older mutation-positive patients had higher observed CCVE risk, predominantly heart failure, with limited precision. These exploratory findings warrant prospective multicenter validation.
We retrospectively studied 203 adults with newly diagnosed myelodysplastic neoplasms, myelodysplastic/myeloproliferative neoplasms, or acute myeloid leukemia unfit for intensive chemotherapy (2016-2023) who underwent pretreatment next-generation sequencing. Cause-specific hazard models adjusted for cardiovascular risk factors assessed associations between CHIP-associated gene mutations and incident CCVEs, including age-stratified and age-by-mutation interaction analyses.
Median age was 68 years (IQR, 61-75); 127 (62.6%) were male; 141 (69.5%) carried ≥1 CHIP-associated gene mutation, most commonly ASXL1, TET2, TP53, or DNMT3A. Over median 12-month follow-up, 74 (36.5%) experienced a first CCVE. Any CHIP-associated gene mutation was associated with CCVE risk in univariable analysis; this association was not significant after multivariable adjustment (adjusted hazard ratio [aHR], 1.58; 95% CI, 0.87-2.88; p=0.13). In joint age-mutation analyses, mutation-positive patients aged ≥70 years had the highest CCVE risk (aHR, 2.77; 95% CI, 1.17-6.56), whereas younger mutation-positive patients showed a nonsignificant trend (aHR, 2.16; 95% CI, 0.90-5.20; p=0.085); interaction was not significant (p=0.29). In multivariable models, CHIP-associated gene mutations, higher-risk MN, and incident CCVEs were each associated with higher all-cause mortality.
CHIP-associated gene mutations at MN diagnosis were not independently associated with incident CCVEs. Older mutation-positive patients had higher observed CCVE risk, predominantly heart failure, with limited precision. These exploratory findings warrant prospective multicenter validation.