Mitochondrial Heteroplasmy Is a Novel Predictor of Chronic Lymphocytic Leukemia Risk

Background: Clonal hematopoiesis (CH) is defined as the relative clonal expansion of a hematopoietic stem cell (HSC). A subset of CH characterized by the presence of somatic mutations in cancer-associated genes has been referred to as clonal hematopoiesis of indeterminate potential (CHIP) and has been associated with hematological malignancies. More recently two subtypes of CHIP were described based on the presence of mutations in “myeloid” genes (M-CHIP) and “lymphoid” genes (L-CHIP) and demonstrated significant associations with myeloid and lymphoid malignancies (particularly chronic lymphocytic leukemia [CLL]), respectively. We have previously reported that the presence of somatic mutations in mitochondrial DNA (mtDNA), also known as heteroplasmy, was associated with myeloid malignancies but the effect of heteroplasmy on CLL incidence is not clear. Thus, we aimed to evaluate the association between mitochondrial heteroplasmy as well as L-CHIP and their impact on the incidence of CLL in the UK Biobank (UKB) cohort.

Methods: We used the MitoHPC pipeline to quantify mtDNA heteroplasmy from whole genome sequencing data obtained from buffy coat. We defined a variant allele frequency between 5 and 95% as heteroplasmic and identified predicted deleterious consequences of mtDNA variants using a recently developed local constraint-based score, the modified mitochondrial local constraint score sum (mMSS). CHIP variants were called from whole exome sequencing data. CLL was identified using ICD-10 codes from cancer registry data linked to the UKB. We evaluated the associations of mtDNA heteroplasmy, L-CHIP, and incident CLL using Cox proportional hazards models adjusting for age, sex, smoking, and a history of cancer diagnosis in individuals who did not have a history of CLL or myeloid neoplasms (MN). Individuals with an absolute lymphocyte count over 5,000/uL were also removed, as they likely represented undiagnosed CLL.

Results: In the current study, we included 419,229 individuals from the UKB. The median age was 58 (50, 63) years, 45.8% (n = 191,955) were men and 94.8% (n = 395,888) were self-identified as White. Heteroplasmy was present in 118,515 (28.3%) individuals, while L-CHIP was present in 10,747 (2.6%) individuals. Overall, there were 463 cases of incident CLL with a median follow-up time of 7.6 (5.2, 9.9) years. Heteroplasmy prevalence and mMSS were higher in individuals with L-CHIP compared to those without and enriched in high-risk L-CHIP subsets (VAF ≥ 10% and multiple mutations). Consistent with previous reports, L-CHIP was associated with an increased incidence of CLL (HR = 5.1; 95% CI 3.9–6.7; P < 0.0001).

Within the whole cohort, we found that individuals with heteroplasmy were at significantly higher risk of developing CLL (HR = 1.5; 95% CI 1.3–1.8; P < 0.0001). Additionally, likely deleterious mutations as determined by mMSS, were associated with significantly higher incidence of CLL (HR = 4.2; 95% CI 3.2–5.5; P <0.0001). In individuals without L-CHIP, we also demonstrated that heteroplasmy prevalence was an independent predictor of CLL risk (HR = 1.5; 95% CI 1.2–1.8; P < 0.0001). Moreover, in individuals without L-CHIP, mMSS (HR = 4.4; 95% CI 3.3–5.9; P <0.0001) remained significantly associated with incident CLL. The fact that heteroplasmy and mMSS remained significantly associated with CLL in L-CHIP negative individuals, suggests that this association was not merely a marker of underlying L-CHIP but that it likely plays a biologically significant role in the expansion and/or transformation of the underlying lymphoid clones.

Conclusions: Overall, mitochondrial heteroplasmy appears to be a novel risk factor for CLL development even in a CHIP negative population. Our study also suggests that heteroplasmy is not only a marker of clonal expansion, but altered mitochondrial function may play an important biological role in the expansion and/or oncogenic transformation of the pre-malignant lymphoid clone.