Abstract
Introduction: Clonal hematopoiesis of indeterminate potential (CHIP) is a situation where asymptomatic clonal genetic alterations are present in the blood and bone marrow of otherwise healthy individuals. CHIP is associated with old age and increased age-adjusted risks for the development of acute myeloid leukemia (N Engl J Med 2014; 371:2488-98) and cardiovascular disease (N Engl J Med 2017; 377:111-21) . Deficiency of TET2 is one of the hallmarks of CHIP and causally related with accelerated atherosclerosis (Science 2017 355:842-7). In a population study, we investigated the rates and risks of heart disease-related mortality (HDRM) and cerebrovascular disease-related mortality (CVARM) in myeloid neoplasms that are frequently affected by TET2 mutations: chronic myelomonocytic leukemia (CMML, ~50% TET2- mutated) and myelodysplastic syndromes (MDS, ~25% TET2 -mutated).
Methods: All 18 Surveillance Epidemiology and End Results (SEER) registries were queried for CMML, MDS, breast and prostate cancer cases, information on survival and cause of death using the custom-built SEERaBomb program. Breast and prostate cancers were used as controls because they are not associated with TET2 mutations or risk factors for cardiovascular disease. HDRM and CVARM were identified using ICD8-10 codes as defined by SEER. Relative risks (RR) for HDRM and CVARM, adjusted for age, sex and year, were calculated by comparing these rates in SEER cohorts to the general population using US background population data from the Centers of Disease Control's Wonder dataset.
Results: Cohorts of 4,699 CMML patients, 38,304 MDS patients, 1,053,780 breast cancer patients and 1,082,390 prostate cancer patients were included. Over the full duration of follow-up, 8.6% and 1.4% of CMML patients, 10.7% and 1.5% of MDS patients, 9.9% and 2.0% of prostate cancer patients and 6.4% and 1.7% of breast cancer patients deceased of HDRM and CVARM, respectively. When compared relative to mortality rates from the US background population, CMML and MDS patients had increased adjusted risks of HDRM (CMML RR for years 1-10: 2.98 [95% CI 2.69-3.29] P= 6*10-90; MDS RR: 2.61 [2.53-2.70] P < 1*10-100; Fig. 1a), while breast and prostate cancer patients had decreased HDRM risks (breast cancer RR: 0.99 [0.98-1.00] P= 0.03; prostate cancer RR: 0.90 [0.90-0.91] P < 1*10-100). The relative risks for CVARM showed similar trends (Fig 1b). In CMML and MDS patients, the RRs for HDRM and CVARM were highest in the first year after cancer diagnosis whereafter these risks decreased, but they remained elevated at two to three times the adjusted risk relative to the background population until at least 10 years after cancer diagnosis.
Discussion: This is the first study to investigate rates and risks of CVD-related deaths in CMML and MDS, two types of myeloid neoplasms frequently affected by TET2 -mutations, at the population level. Risks of HDRM and CVARM were markedly increased in CMML and MDS patients, especially in the first year after diagnosis, suggesting that upon cancer diagnosis these patients already had cardiovascular disease. This epidemiologic evidence supports the conclusions of earlier prospective case-control studies (N Engl J Med 2017; 377:111-21) at the population level and strongly argues for screening of newly-diagnosed patients with CMML and MDS for cardiovascular disease and aggressive management of risk factors for cardiovascular disease in these patients.
Sekeres: Celgene: Membership on an entity's Board of Directors or advisory committees.
