Prevalence and dynamics of clonal hematopoiesis caused by leukemia-associated mutations in elderly individuals without hematologic disorders

Abstract

Clonal hematopoiesis is frequently observed in elderly people. To investigate the prevalence and dynamics of genetic alterations among healthy elderly individuals, a cohort of 50 people >80 years was genotyped for commonly mutated leukemia-associated genes by targeted deep next-generation sequencing. A total of 16 somatic mutations were identified in 13/50 (26%) individuals. Mutations occurred at low variant allele frequencies (median 11.7%) and remained virtually stable over 3 years without development of hematologic malignancies in affected individuals. With DNMT3A mutations most frequently detected, another cohort of 160 healthy people spanning all age groups was sequenced specifically for DNMT3A revealing an overall mutation rate of 6.2% (13/210) and an age-dependent increase of mutation prevalence. A significant difference (p = 0.017) in the DNMT3A expression pattern was detected between younger and healthy elderly people as determined by qRT-PCR. To evaluate the selection of clonal hematopoietic stem cells (HSCs), bone marrow of two healthy individuals with mutant DNMT3A was transplanted in a humanized mouse model. Xenografts displayed stable kinetics of DNMT3A mutations over 8 months. These findings indicate that the appearance of low-level clones with leukemia-associated mutations is a common age-associated phenomenon, but insufficient to initiate clonal selection and expansion without the additional influence of other factors.

Fig. 1. Dynamics of somatic mutations detected in 50 healthy elderly individuals during a 3-year follow up period.

Two subjects with mutations in splicing factor genes SF3B1 and U2AF1 died. For two patients (#15 and #35) the 3-year follow-up samples were not available.

Fig. 2. Age-associated increase of DNMT3A mutations in healthy individuals.

Frequency of DNMT3A mutations was analyzed within five age cohorts (0–19; 20–39; 40–59; 60–79; >80 years) by a highly sensitive DNMT3A-specific deep next-generation sequencing assay.

Fig. 3. DNMT3A mRNA expression in healthy individuals.

a Age-associated decrease of DNMT3A expression (median value depicted; *p = 0.017). b A marginal albeit not significant difference between DNMT3A^wt and DNMT3A^mut expression (median value depicted; p = 0.14). Statistical analysis was performed with the Mann–Whitney nonparametric test. wt wild type; mut mutant.