A case of co-mutation of SF3B1 and BCR::ABL1 demonstrating an MDS-phenotype

Abstract INTRODUCTION Co-mutations of BCR::ABL1 and clonal hematopoiesis genes are known to occur in chronic myeloid leukemia (CM)L, but the presence of mutant SF3B1 and BCR::ABL1 is rarely described. A 74-year-old man was referred for a hemoglobin of 91g/L with high MCV (122 fL); platelet and leucocyte counts were normal, as was white count differential. The patient had treated metastatic prostate cancer with good disease control. Bone marrow examination revealed hypercellularity (80%), erythroid predominance, ringed sideroblasts, 1% myeloblasts, and absence of basophilia. A t(9;22) was identified in 18/25 metaphases; molecular testing confirmed the presence of the p210 BCR::ABL1, IS 17.6%. A 36-gene myeloid next-generation sequencing panel revealed a missense mutation p.K666R in SF3B1 (VAF 46%). A diagnosis of concurrent CML and myelodysplastic syndrome (MDS) with mutated SF3B1 was made. Bone marrow sampling repeated 3 months after starting asciminib showed minimal change in cellularity and erythroid predominance. At this time, BCR::ABL1 dropped to 0.0063% IS and SF3B1 VAF was 32%. After a transient improvement in anemia and macrocytosis at 6 months, the hemoglobin dropped to 81, and MCV rose to 116 after 1 year, at which time BCR::ABL1 was undetectable. We performed single cell DNA sequencing to elucidate the relationship between the somatic SF3B1mutation and BCR::ABL1. METHODS Whole genome sequencing using Nanopore was used to identify the t(9;22) breakpoint. Probes were subsequently designed by Mission Bio to include the BCR::ABL1 translocation and the SF3B1 mutation. Using the Tapestri single-cell DNA sequencing platform, cells were isolated into individual droplets where they were lysed, amplified, tagged and sequenced. To obtain the lineage-specific genetic profile, cells were visualized using Uniform Manifold Approximation and Projection and clustered using Hierarchical Density-Based Spatial Clustering of Applications with Noise and manually gated. RESULTS A total of 4065 cells were sequenced. Most cells were erythroid precursors (88.8%, n=3609), followed by lymphocytes (8.45%, n=276), hematopoietic stem and progenitor cells (HSPC) (1.6%, n=65) and monocytes (0.86%, n=35). Among all sequenced cells, 45% (n=1779) harboured both mutant SF3B1 and BCR::ABL1; 25% (n=969) had mutant SF3B1 only. The remaining cells had neither abnormality. Within HSPC, 20% (n=13) were wild type, 24.6% (n=16) only had the SF3B1 mutation and 55.4% (n=36) had the SF3B1 mutation with BCR::ABL1. Erythroid precursors were 26.3% (n=949) wild type, 25.7% (n=926) only had the SF3B1 mutation and 48% (n=1734) had the SF3B1 mutation with BCR::ABL1. Among monocytes, 42.9% (n=15) were wild-type, 42.9% (n=15) had the SF3B1 mutation and 14.3% (n=5) had the SF3B1 mutation with BCR::ABL1. Finally, 97% of lymphocytes were wild type (n=269), 1.5% of lymphocytes only had the SF3B1 mutation (n=4) and 1.5% (n=3) had the SF3B1 mutation and BCR::ABL1.Interestingly, most of the erythroid precursors and HSPC contained the SF3B1 mutation, with or without BCR::ABL1. A smaller subset of monocytes contained both mutations and lymphocytes were mainly wild type. CONCLUSION In this unique case, the presence of mutated SF3B1 likely preceded the emergence BCR::ABL1 and exerted clinical and morphological dominance, yielding an MDS phenotype. At the single-cell level, mutant SF3B1 and BCR::ABL1 were notably absent from lymphoid cells. This contrasts with findings in CML, where BCR::ABL1 is usually present in all cell lineages at diagnosis (Haferlach, BJH, 1996). SF3B1 has been reported to impede lymphocyte differentiation (Mortera-Blanco, Blood, 2018) and seemingly also hampers myeloid proliferation typical of CML, as this case demonstrates. Granulocyte precursors likely experienced compromised viability during sample storage, leading to underrepresentation in the single-cell analysis. Nonetheless, a correlation persists between the reduced granulopoiesis observed in the bone marrow and the findings from single-cell analysis. Single-cell RNASeq is planned to delineate signaling pathways active in SF3B1 and BCR::ABL1 co-mutated cells. To our knowledge, this is the first report mapping the clonal architecture of a case of concurrent CML and SF3B1-mutated MDS using single-cell DNA sequencing. These findings not only provide novel insight into lineage restriction and clonal dominance but also reveal how a concurrent mutation can mask the classic disease phenotype of CML.