Characterization by Next Generation Sequencing of mutations of spliceosome genes and bone marrow mesenchymal cells in patients with myelodysplastic syndromes

Abstract

A two-step approach for sequencing spliceosome-related genes as a complementary diagnostic assay in MDS patients with ringed sideroblasts Our study aimed to analyze the presence of mutations in SF3B1 and other spliceosome-related genes in myelodysplastic syndromes with ringed sideroblasts(MDS-RS)by combining conventional Sanger and next-generation sequencing(NGS)methods,and to determine the feasibility of this approach in a clinical setting. 122 bone marrow samples from MDS-RS patients were studied. Initially, exons14 and 15 of the SF3B1 gene were analyzed by Sanger sequencing. Secondly, they were studied by NGS covering besides SF3B1, SRSF2, U2AF1 and ZRSR2 genes. An 86%of all patients showed mutations in the SF3B1 gene. Six of them, which were not identifiable by conventional sequencing in the first diagnostic step,were revealed by NGS. In addition,19.5%of cases showed mutations in other splicing genes: SRSF2, U2AF1, and ZRSR2. Furthermore, 8.7%of patients had two mutations in SF3B1, SF3B1 and SRSF2, and SF3B1 and U2AF1, while 5.7% showed no mutations in the four spliceosome-related genes analyzed. The combined use of conventional Sanger and NGS allows the identification of mutations in spliceosome-related genes in almost all MDS patients with RS. This two-step approach is affordable and could be useful as a complementary technique in cases with an unclear diagnosis. Clinical, biological and prognostic implication of SF3B1 concomitant mutations in very low/low and intermediate-risk MDS patients SF3B1 is one of the most frequently mutated genes in MDS patients, related to a specific subtype and parameters of good prognosis in MDS without excess blasts. However, more than 40% of MDS patients carry at least 2 mutations and little is known about the impact of concurrent mutations in the outcome of MDS patients. We aimed to analyse the co-occurrence of SF3B1 with other mutations and reveal their clinical, biological and prognostic implications in very low/low and intermediate-risk of MDS patients. Mutational analysis of 102 SF3B1mut patients showed that 19.6% of cases carry isolate SF3B1 mutation, while 80.4% of patients presented additional mutations in other genes, with a median of 2 additional mutations per patient (range 0-5). The genes most frequently mutated concomitant to SF3B1 were: TET2 (39.2%), DNMT3A (25.5%), SRSF2 (10.8%), CDH23 (5.9%) and ASXL1, CUX1, KMT2D (4, 9% each). We observed that presence of ≥2 concomitant mutations to SF3B1 had an adverse impact on survival as compared to those with SF3B1 + <2 additional mutation (median of 54 vs. 87 months, respectively, p=0.007). Furthermore co-occurrence of SF3B1 with SRSF2 mutations was associated with shorter overall survival as compared to isolated SF3B1 mutation with SRSF2 wt (median of 27 vs. 75 months, respectively, p=0.001). The similar adverse effect was observed with concomitant IDH2 mutations, (median OS of 11 vs. 75 months, respectively, p=0.001), and BCOR (median OS of 11 vs. 71 months, respectively, p=0.036,). Interestingly, SF3B1 with NUP98, and SF3B1 with STAG2 had also a negative effect on patients prognosis (medians of 27 and 11 vs. 71 months, respectively, p=0.008 and p=0.002). In summary, Therefore our results suggest that more complete NGS study in MDS SF3B1mut patients would be recommend to better estimation of the evolution of the disease discarding the influence of additional mutations that could interfere in the good prognosis associated with SF3B1 mutations and would allow proposing more adequate therapeutic options for each patient.