AUTHORS
Ajibike Dorothy Lapite, MD, MPHTM, Hidetaka Uryu, MD, PhD2, James Decuir, MD, MPH3, Koichi Takahashi, MD,PhD4, Katherine Y King, MD, PhD5
AFFILIATIONS
1 Baylor College of Medicine, Houston, Texas, United States
2 MD Anderson Cancer Center, Houston, Texas, United States
3 The Mount Sinai Hospital, New York, New York, United States
4 MD Anderson Cancer Center, Houston, Texas, United States
5 Baylor College of Medicine, Houston, Texas, United States
ABSTRACT
Clonal hematopoiesis (CH) is a phenomenon of aging in which hematopoietic stem cells acquire somatic mutations that confer relative fitness advantage, resulting in overrepresentation of a clonal population in the bone marrow and peripheral blood. CH portends a 2-fold increased risk of hematologic malignancy and increased risk for a number of other disease states including but not limited to cardiovascular disease, cerebrovascular disease, non-hematologic malignancy, and autoimmune/autoinflammatory conditions. Importantly, CH has therapeutic implications, as its presence is associated with disease progress across multiple organ systems and decreased treatment responsiveness. It is imperative to understand drivers of CH emergence and proliferation to improve human health.
Preclinical and clinical studies make clear that infection is a powerful driver of CH. For example, multiple studies have compared people living with HIV (PLWH) with uninfected persons and found that PLWH have increased prevalence of CH as compared to age-matched controls. Our group recently demonstrated a mutation-specific dose-response relationship between number of infections and subsequent risk of CH in a general population. Given that infection burden correlates strongly with geography, we asked: could the mutational landscape of CH vary in matched geographically disparate populations with different infectious exposures.
To address this question, we designed a retrospective cohort study comprised of male and female African American (AA) and Ugandan PLWH and utilized whole-exome sequencing (WES) and targeted sequencing to ensure both breadth and depth in our assessment of CH burden in this study population. We optimized previously validated bioinformatic pipelines to ensure appropriate alignment given the genetic ancestry of our cohort. We utilized descriptive statistics and statistical modeling to assess for differences for those with and without CH in the overall cohort and in the sub-populations (AA compared to Ugandan).
Our study included 177 AA and 172 Ugandan PLWH of which 54% were female; thereby, our study is the first CH investigation in PLWH without a male predominance. The median ages of the AA and Ugandan PLWH were similar at 51 and 50.1 years respectively. HIV was well controlled across the study cohort. We identified CH in 17.5% (n = 61) of the study cohort with similar frequency in AA and Ugandan PLWH. Interestingly, we identified unique mutational patterns of CH that varied with location. CHEK2, EZH2, and SH2B3 mutations were only observed in AA PLWH whereas BAX, GNAS, RUNX1, YLPM1, ZNF318 mutations were only observed in Ugandan PLWH. Many of these mutations would not have been identified without WES as theymutations are not routinely included in targeted sequencing panels.
Consistent with prior studies, AA PLWH with CH were older as compared to AA PLWH without CH (55 years compared to 50 years; p = 0.008). Ugandan PLWH with CH were of a similar age to Ugandan PLWH without CH but were noted to have a higher BMI (24.29 compared to 22.15 kg/m2; p = 0.02). Reported co-infections was available for AA PLWH and increased for males with CH as compared to those without (2 infections compared to 1; p = 0.02). Interestingly, CH was more common for males as compared to females in the AA sub-population but the inverse for the Ugandan sub-population.
Our findings demonstrate that PLWH of African genetic ancestry have a similar CH prevalence to PLWH of European genetic ancestry. Importantly, our study recapitulates that environmental factors impact CH emergence and suggests that mutational landscapes can vary with geographic location.
DISCLOSURES
Ajibike Dorothy Lapite, MD, MPHTM: No relevant conflicts of interest to declare.