Dottorato in INNOVATION IN IMMUNO-MEDIATED AND HEMATOLOGICAL DISORDERS

Titolo della tesi: DEFINITION OF BASELINE GENE SIGNATURES AND CORRELATION WITH RESPONSE TO FRONTLINE HYPOMETHYLATING AGENTS IN HIGH-RISK MYELODYSPLASTIC SYNDROMES AND ACUTE MYELOID LEUKEMIA

Myelodysplastic syndromes (MDS) are blood disorders characterized by ineffective haematopoiesis and progressive marrow failure with final transformation into acute leukemia. Acute myeloid leukemia (AML) is the most common type of leukemia in the elderly. Usually, both diseases have specific cytogenetic defects and recurrent mutations. The DNA methyltransferase inhibitor 5-azacitidine (AZA) is the unique pharmacologic option for unfit patients not candidate to intensive chemotherapy. However, the dynamic of response is slow and responses were usually obtained after a minimum of four cycles. The molecular mechanisms underlying the different degrees of responses are still poorly understood. In MDS and AML, the use of Next-Generation DNA Sequencing (NGS) has allowed a sub classification based on genetic and epigenetic profiles. The aim of this project is to identify baseline different gene signature in responders vs non-responders to AZA by testing patients in NGS. Our proposal is to prospectively enrol patients with intermediate-2/high risk according to IPSS MDS and patients with AML, treated with this drug. We tested DNA samples from 69 patients in the last three years. All patients were analysed by NGS using a SOPHIA myeloid panel which allows the analysis of 30 genes. DNA extraction was performed starting from EDTA anticoagulated bone marrow samples and analysed by NGS. NGS was performed according to standard procedures. The pooled libraries were sequenced on a micro flow cell with V2 chemistry on a MiSeq instrument. After generation of FASTQ files, sequence alignment to the reference genome was performed using Sophia-DDM Software. We analysed 57 patients with AML at diagnosis and 12 patients with MDS at diagnosis, treated subsequently with hypomethylating agent. 19 patients had previously received an intensive chemotherapeutic regimen, 50 patients were treatment-naïve (35 in the AZA cohort and 15 for the DEC cohort). The median number of HMA cycles was 7 (range 1-50). The median age was 71 years (range 31–86). A female predominance was observed (male/female 32%/68%). To determine the characteristics that might predict response to AZA we performed analyses on bone marrow sample at baseline. Among 69 patients 196 gene mutations were detected. The median number of gene mutations was 3 (range 0–7). The most frequent mutations in the entire cohort of patients were observed in ASXL1 (33.3%), RUNX1 (21.7%), SRSF2 (23.2%); FLT3 and IDH2 (17.4% each), DNMT3A (15.9%), NPM1 and TET2 (14.5% each), TP53 and SF3B1 (11.6% each). Four patients (6%) had no mutations and 3 of them are < 60 years old. We found some correlation between gene signature and baseline features: mutations in SRSF2 (p=0.033), IDH2 (p=0.010) and NPM1 (p=0.017) were associated to a higher median age at diagnosis; mutated NRAS (p=0.008) and PTPN11 (p=0.031) with a low level of Hb; EZH2, (p=0.038), ZRSR2 (p=0.038) mutation was associated to a high bone marrow blast count and SF3B1 (p=0.04) and NRAS (p=0.015) with high platelet count. In terms of survival, mutations of TP53 (p=0.038) were associated with reduced overall survival. (OS). Other mutations did not significantly influenced OS: ASXL1 (p=0.736), CEBPA (p=0.578), IDH2 (p=0.573), RUNX1 (p=0.657), DNMT3A (p=0.116). Focusing on possible impact of mutational profile in terms of progression free survival (PFS), a significant inferior of PFS was recorded in patients who harbored FLT3 mutation than wild type FLT3 (median PFS 13.6 months vs 7.8 months, respectively), (p=0.024). Identification of somatic mutations has profound prognostic implications predicting treatment response and survival outcomes in MDS and AML patients. Our results suggest that the presence of some mutations can influence the outcome and the response to the drug. We need to enlarge the population to be examined because it will be essential to clarify the association between gerarchic combination of mutations with the final aim to propose a solid “myeloid signature” panel to be used prior to hypomethylating agents.