High prevalence of clonal hematopoiesis in fanconi anemia revealed by whole exome sequencing Free

Fanconi anemia (FA) is an inherited bone marrow failure syndrome (IBMFS) characterized by hypersensitivity to DNA damage and a markedly increased risk of myelodysplastic syndrome (MDS) and acute myeloid leukemia (AML). Clonal hematopoiesis (CH) is frequently observed in FA and precedes overt malignant transformation. However, the somatic drivers of CH in FA remain inadequately defined. Importantly, most clinical surveillance strategies rely on sequencing panels designed for sporadic CH, which may not reflect the profile of somatic driver mutations specific to FA and other IBMFSs.

To address this unmet need, we performed whole-exome sequencing (WES) of bone marrow and germline tissue (bone marrow fibroblasts) from 28 subjects with FA (15 male, age 3-26 years) to define the prevalence and somatic mutational landscape of FA-associated CH. We identified 54 putative somatic driver mutations (defined as present at ≥2% variant allele frequency (VAF); range 3.7-45.5%), with 89% (25/28) of FA patients found to bear at least one mutation. The spectrum of somatic mutations was broad, with candidate drivers identified in genes that regulate metabolism, mRNA processing, cellular signaling, and transcription, with minimal overlap with canonical drivers of sporadic MDS/AML that are typically included on screening panels. There was a positive association of number of mutations with age. There was no association of CH, number of somatic mutations, or maximum VAF with clinical bone marrow failure (BMF), MDS, requirement for hematopoietic stem cell (HSC) transplantation, complementation group, or cytogenetic abnormalities.

To begin to validate somatic drivers of clonal expansion, we utilized our human induced pluripotent stem cell (iPSC) model of FA-associated BMF. In this model, FANCA-deficient human iPSCs bear a doxycycline-inducible wild-type FANCA cDNA so that treated iPSCs retain genomic stability and pluripotency. During directed differentiation to a target lineage, FANCA expression can be modulated by continuation or discontinuation of doxycycline to yield isogenic FANCA-proficient (FANCA^+/+) or -deficient (FANCA^-/-) cells for use in disease models. For modeling, we prioritized a putative gain-of-function mutation in ALDH9A1 (p.S11Y) linked to clonal expansion in our WES dataset given that increased or decreased endogenous aldehyde detoxification are known to mitigate or amplify, respectively, FA-associated BMF. Using CRISPR/Cas9 with homology-directed repair, we introduced ALDH9A1^S11Yinto our FA iPSC model in the heterozygous state and generated clones that we confirmed to lack copy number variation. We additionally generated ALDH9A1^-/- FA iPSCs as a benchmark for the phenotype of complete ALDH9A1 deficiency. Following directed differentiation of iPSCs to hematopoietic progenitor cells (HPCs), we analyzed the fitness of HPCs of defined FANCA and ALDH9A1 status in colony-forming unit (CFU) assays. As expected, compared to benchmark FANCA^+/+ HPCs, FANCA^-/- HPCs showed markedly decreased clonogenesis, consistent with recapitulation of the clinical BMF phenotype. Remarkably, the ALDH9A1^+/S11Y variant in the FANCA^-/- background promoted a near complete rescue of HPC clonogenesis with maintenance of the full diversity of myeloerythroid colony outcomes, suggestive of a gain-of-function effect to enhance aldehyde detoxification. In contrast, depletion of ALDH9A1 in the FANCA^-/- background resulted in poor clonogenesis.

Collectively, our work highlights the unmet need of better understanding clonal evolution IBMFSs. Since IBMFS-associated CH occurs in the context of germline mutations that are detrimental to the self-renewal or differentiation of HSCs, this pathobiology departs from that of sporadic, age-associated CH that occurs in (for the most part) otherwise healthy individuals. Our results posit that commonly used sequencing panels for surveillance of sporadic CH/MDS/AML are likely to underestimate CH in IBMFSs, thus emphasizing the need to better understand the drivers of CH in these conditions. In support of this idea, we demonstrate that a putative gain-of-function missense mutation in the gene encoding the detoxification enzyme ALDH9A1 may adaptively confer increased aldehyde clearance to act as a disease-specific driver of clonal expansion in FA. Future research will focus on the identification and functional validation of both adaptive and maladaptive somatic drivers of CH in IBMFSs.