Analysis of genetic variants according to the GIPSS prognostic model in patients with Myelofibrosis.

CAMACHO MF ; HELLER PG; ENRICO A; MOIRAGHI B.; CASTRO RIOS MA; SACKMANN F; BENDEK G; VALLEJOS V; VARELA A; MONTIVERO R; DE LUCA G; GUTIERREZ M; FLORES G; PATRICIO P; BELLI C; LARRIPA I

Congreso; LXVII Reunión Anual de la Sociedad Argentina de Investigación Clínica; 2022

Myelofibrosis (MF) is characterized by stem cell-derived clonal myeloproliferation, associated with bone marrow fibrosis. Most patients present one driver mutation in JAK2, CALR or MPL genes, which are mutually exclusive. The majority acquire others high molecular risk (HMR) in genes affecting epigenetic regulation (ASXL1 and IDH1/2) and splicing machinery (SRSF2 and U2AF1), that may be combined. The Genetically-Inspired Prognostic Scoring System (GIPSS) includes three genetic risk factors: very high risk (VHR) or unfavorable karyotype, HMR mutations and the absence of type 1/like CALR mutations.Our aim was to analyze the karyotypes, driver and HMR genetic variants according to the GIPPS criteria without other clinical parameters.The cohort included 84 patients (56% females) with MF diagnosed according to the 2016 WHO criteria. Driver mutations were 45% JAK2 (p.V617F), 20% Type 1 CALR (p.L367fs*46), 5% Type 2 CALR (p.K385fs*47), 8% MPL (p.W515L/K) and 21% triple-negatives. Genomic DNA samples were analyzed using allele-specific-primers for IDH1/2 (exon 4), Sanger sequencing for ASXL1 (exon 12-13) and high-resolution melting confirmed by Sanger sequencing for SRSF2 (exon 1) and U2AF1 (exon 2). HMR variants were detected in 32 patients (38%), 7 (8%) of them with ≥2 variants. ASXL1 was characterized by frameshift or nonsense variants (n23) while the remaining by missense changes in IDH1 (n1, p.R132H), IDH2 (n5, p.R140Q), SRSF2 (n7, p.P95H/L) and U2AF1 (n4, p.Q157P/R). Abnormal karyotypes were identified in 10 (12%) patients: 6 VHR and 4 unfavorables.The overall survival was 86 months with a median follow-up of 25 months (1-182). The GIPSS score differentiated: High, Intermediate and Low risk (median survival: non reached vs 86 vs 27 months, respectively, p=0.029). No differences were observed between intermediate findings (Int-1 vs Int-2, p=0,866). Our results support the GIPSS model as a prognostic tool to risk-adapt therapy in our MF population.