Associations of Lineage-Specific Clonal Hematopoiesis with COVID-19 Hospitalization and Mortality

Introduction: Somatic variants indicative of clonal hematopoiesis (CH) are known to increase risks of prevalent inflammatory diseases, hematologic malignancies, and mortality. In recent years, the relevance of CH to COVID-19 has been controversial, with conflicting reports of the association between myeloid clonal hematopoiesis (M-CH) variants and COVID-19 severity. Furthermore, the prevalence of CH variants in lymphoid driver genes (L-CH) has not been investigated in COVID-19 thus far. This study reports prevalences of both M-CH and L-CH in a large Canadian COVID-19 cohort and explores associations with disease severity. Methods: CH variants were detected from whole-genome sequences of 6651 patients with COVID-19 enrolled in the CGEn HostSeq initiative (CGEn- Canada's national platform for genome sequencing and analysis) between 01/2020 and 10/2022. The median age of the cohort was 50 years (range 20-103 years) and 57.9% (n=3851) were female. Patients with a prior diagnosis of leukemia were excluded from these analyses. Variant calling was performed using Mutect2, following a pre-defined list of putative variants in 144 genes associated with myeloid and/or lymphoid malignancies. Additional filters were applied to exclude germline variants and probable sequencing artifacts. Associations of CH with clinical severity status and hematologic laboratory parameters were determined using regression analyses adjusted for age and sex. Results: The total prevalence of CH was 6.6% (n=441/6651). Most patients had M-CH exclusively (81.4%, n=359), though others had L-CH (12.2%, n=54), or both M-CH and L-CH (6.3%, n=28). 456 M-CH variants were found, and were most frequent in DNMT3A (n=125) and TET2 (n=96). Among 88 identified L-CH variants, those in KMT2C were most common (n=20), followed by KMT2D (n=18) and SPEN (n=10). Disease severity status was defined for a subset of 5890 patients as: ambulatory (n=3927), hospitalized with mild disease (with or without oxygen by mask or nasal prongs; n=1180), hospitalized with severe disease (requiring intubation, mechanical ventilation, high-flow oxygen, or other organ support; n=499), or death (n=284). CH was significantly more prevalent among patients hospitalized with mild disease compared to the ambulatory group (odds ratio [OR]=1.70, 95% confidence interval [CI]=1.29-2.25, p=0.0002). This association remained significant when restricting the analyses to patients with M-CH (OR=1.50, 95% CI=1.11-2.01, p=0.007) and L-CH (OR=3.38, 95% CI=1.78-6.40, p=0.0002). Laboratory results of 942 patients hospitalized with mild COVID-19 showed that CH and M-CH were associated with lower hemoglobin (β=-4.97, 95% CI=-8.50- -1.45, p=0.006 and β=-6.02, 95% CI=-9.74- -2.29, p=0.002); and L-CH was associated with increased white blood cell count (β=1.8, 95% CI=0.08-3.43, p=0.04). Differences in CH status between patients with severe versus mild disease were not statistically significant (OR=0.78, 95% CI=0.54-1.13, p=0.18). Notably, L-CH alone was significantly overrepresented among patients with COVID-19-related mortality when compared to the ambulatory group (OR=4.22, 95% CI=1.39-12.78, p=0.01). Conclusions: This study demonstrates an association of CH with COVID-19 hospitalization, and introduces L-CH as a potentially important prognostic factor of the disease. These results support earlier suggestions that antiviral responses may be altered in the presence of CH, including possible influence on the inflammatory cascade that is thought to drive worse outcomes. No associations were found between CH and the need for advanced respiratory support, suggesting that the progression from mild to severe disease may be affected by other biological or clinical factors. This work highlights lineage-specific variants and disease severity status as important considerations in the relationship between CH and COVID-19. These findings may carry relevance for ongoing work looking at CH in the context of other infectious diseases.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.