Background/Purpose: Systemic sclerosis (SSc) is an autoimmune disease characterized by widespread fibrosis, vasculopathy, autoantibody presence, and high mortality in its diffuse subtype with limited effective therapies. Genome-wide association studies (GWAS) have identified candidate genes that contribute to SSc development, but do not fully explain the drivers of heterogeneity between subtypes. Preliminary evidence suggests somatic (acquired) mutations contribute to SSc, and that there is elevated age-related clonal hematopoiesis (ARCH) variant burden in SSc versus healthy individuals. We hypothesized that immune dysregulation of ARCH contributes to the pathogenesis of the increased inflammatory phenotype of diffuse SSc.

Methods: Our cohort consists of limited cutaneous SSc (lcSSc) (n = 1611, mean = 55.9 ± 13 years, non-White = 6.82%, Female = 90.2%), diffuse cutaneous SSc (dcSSc) (n = 1240, mean = 50.47±13 years, non-White = 18.1%, Female = 82.3%), and healthy control (HC) (n= 984, mean = 46.7 ± 15.4 years, non-White = 5.7%, Female = 54.9%) whole blood exome sequences (WES) (median depth = 39X). We performed somatic variant calling with Mutect2 and a panel of normals generated from a subset of our HC sequences. We filtered germline and technical artifacts and annotated variants with ANNOVAR. We adhered to best practices of sequence-based filtering (DP >20, minAD ≥ 3, F2R1 ≥ 1, variant allele fraction (VAF) > 2%) and curated ARCH variants for all samples against a list of 74 CH driver genes. Logistic regression and Fisher’s exact test were used to compare CH between sample groups and clinical features when suitable and linear regression was used for inspection of CH rate increases.

Results: We observe a significantly higher likelihood of ARCH in dcSSc compared to lcSSc after adjusting for age and sex, with confirmation in the UK Biobank (n=454,787). Additionally, the rate of CH with age rises faster in dcSSc than lcSSc or HC (R2 values = 0.86, 0.65, 0.85 respectively). The most frequently mutated genes are DNMT3A, TET2, and ASXL1 and the median VAF of all ARCH variants is 12.3%. PPM1D variants are elevated in dcSSc and occur in younger patients (n= 7, mean = 55.3 years) compared to lcSSc (n= 2, mean = 76.0 years). PPM1D variants are also associated with shorter disease duration in dcSSc patients (mean = 4.12 years, total mean of dcSSc ARCH variants = 9.17 years). We observe modest odds of anti-RNA polymerase III autoantibody presence and ARCH in dcSSc (OR = 1.20), and moderate association between ARCH and an impaired lung function threshold of % predicted diffusing capacity of the lung for carbon monoxide (DLCO) < 40 (p = 0.1, OR = 2.02 adjusted for disease duration). However, larger clone ARCH variants with VAF > 10% confer significant likelihood of having % predicted DLCO < 40 within dcSSc subtype (p = 0.0495, OR = 1.53 adjusted for disease duration).

Conclusion: Higher prevalence of ARCH in dcSSc versus lcSSc and gene-specific patterns of ARCH variant occurrence in dcSSc suggest a biological mechanism driving clonal selection between subtypes and possibly influencing fibrotic progression patterns. Understanding the timing and gene-specific nature of these mutations provides insight into selective pressures underlying SSc subtype heterogeneity.

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/accelerated-and-gene-specific-patterns-of-clonal-hematopoiesis-distinguish-subtypes-of-systemic-sclerosis/