Next Generation Sequencing (NGS) identify somatic mutations in driver genes in up to 90% of patients with MDS. Intensive Chemotherapy (IC) is commonly used in patients with High-Risk Myelodysplastic syndromes (HR-MDS), especially those who are candidates to receive allogeneic Hematopoietic Stem Cell Transplantation (HSCT).
To evaluate the impact of somatic mutations in patients with HR-MDS receiving intensive chemotherapy.
We retrospectively analysed 35 HR-MDS patients, including Acute Myeloid Leukaemia (AML) with <30% of blast (FAB RAEB-T), treated with IC. DNA from bone marrow samples from diagnosis was screened for somatic mutations by NGS, using a NextSeq platform (Illumina). Funding sources: PI17/01741; GRS 1349/A/16; GRS 1179/A/15; CM17/00171.
Median age of our patients was 50 years (p25-p75 40-58); 60% were male. According to WHO 2008 classification 16 (45.7%) were AML, 15 (42.9%) RAEB-2, 3 (8.6%) RAEB-1 and 1 (2.9%) RCMD. Among patients with calculated R-IPSS (27 of 35 patients) 7 (25.9%) were intermediate risk, 11 (40.7%) were high risk and 9 (33.3%) were very high risk. 26 patients (74.3%) underwent HSTC. Karyotype was good in 17 (48.6%), intermediate in 5 (14.3%), poor in 3 (8.6%), very poor in 3 (8.6%) and unknown in 7 patients (20%); 4 patients (11.4%) had a complex kariotype. 19 patients (54.3%) had a Complete Response without Minimal Residual Disease (CR MRD-) prior to transplant or after chemotherapy; 16 patients (45.7%) had any detectable disease.
Regarding mutational status, median number of mutated genes per patient was 2. Only 6 patients (17.1%) did not present any mutations; 8 patients (22.9%) had 1 mutated gene, 7 (20%) had 2 mutated genes, 7 (20%) had 3 mutated genes, 5 (14.3%) had 4 mutated genes and 2 (5.7%) had 5 mutated genes. Most frequently mutated genes were: RUNX1 in 6 patients (18.6%), SF3B1, SRSF2, U2AF1, IDH1, STAG2 and NPM1 in 4 (11.4%) and TET2, DNMT3A, NF1, TP53, WT1 and NRAS in 3 (8.6%); other mutated genes were presented with a lower frequency.
We found a significant association between mutational status (mutated vs. non-mutated) and response prior to transplantation or after chemotherapy: 100% of CR MRD- in non-mutated patients vs. 44.8% in those with detectable disease (p = 0.022). Only mutations in U2AF1 were significantly associated with absence of MRD- (p = 0.035).
We analyse samples from 12 patients prior to transplant in order to determine in chemotherapy is able to “clear” somatic mutations: 8 of them were non-mutated pre HSCT and 7 had any mutated gene. We didn’t found any relationship between “clearing” of mutations and disease status prior to transplant.
After a median of follow up for survivors of 4.35 years, median Overall Survival (OS) and Disease Free Survival (DFS) were 6.24 and 0.83 years, respectively. Non-muted patients had a significantly better DFS than patients with any mutation at diagnosis (83.3% vs. 21.4%, p = 0.019). Univariate analysis determined that mutational status (mutated vs. non-mutated), complex karyotype and somatic mutations in SF3B1, SRSF2 and TP53 significantly influenced on DFS. In the multivariate analysis only mutations in SRSF2 (HR 6.42) and TP53 (HR 8.77) retained their signification.
We conclude that the presence of somatic mutations at diagnosis in patients with HR-MDS receiving intensive chemotherapy could have a significant impact on DFS and could be related with response.