Significant Impact of the Molecular Profile on the Prognosis of Patients with Myeloproliferative Neoplasms and Splanchnic Vein Thromboses

Introduction: Myeloproliferative neoplasms (MPN) are the most frequent underlying causes of splanchnic vein thromboses (SVT), including Budd-Chiari syndrome (BCS), portal, splenic and mesenteric vein thromboses. This subgroup of MPN patients with SVT (MPN/SVT) has been shown to have distinct clinical and molecular features like younger age and lower JAK2V617Fmutant allele burden (the most predominant driver mutation in MPN/SVT patients) and are often considered as early stages of MPN. However, there is no study that has investigated the contribution of NGS data to the risk stratification of these MPN/SVT patients. Main objective of this study was to evaluate the impact of the presence of additional mutations to the driver mutation in the long term outcome of MPN patients with SVT.

Patients and methods: Over the past 10 years, a total of 286 patients with SVT have been referred to our center for diagnostic assessment of an underlying MPN. Among them, a diagnosis of MPN (WHO criteria) was finally made in 197. Full molecular analyses by NGS was available in 61 of these MPN/SVT patients. The molecular profiling of patients was performed using a Capture-based custom NGS panel provided by Sophia Genetics. This panel comprised the coding sequences of 36 genes involved in MPN. Bioinformatics analysis was performed by Sophia Genetics, with a detection limit of variants set at 1%. We compared (i) MPN/SVT patients to our local cohort of 1371 MPN patients without SVT with full NGS data available; (ii) MPN/SVT patients with and without additional mutation to the driver mutation.

Results: Median follow-up of MPN/SVT patients was 11 years, 62% were females, 33% had BCS and 67% portal vein thrombosis (Table 1). All patients received anticoagulants and 58 (95%) a cytoreductive therapy after the diagnosis of MPN. The type of MPN was polycythemia vera in 69%, essential thrombocythemia in 25% and myelofibrosis in 6%. The driver mutation was JAK2V617F in 57 (93%), and CALR mutation in 4 patients, respectively, and no patient had MPL mutation. By NGS, an additional mutation was identified in 27 (44%) patients, while 34 patients had only the driver mutation. The most frequent additional mutations were in TET2 (23%), ASXL1 (8%), DNMT3A (7%), IDH1/2 (5%), and LNK (5%) genes. EZH2, SRSF2 and TP53 mutations were found in only 1 patient, respectively. When compared to NGS data from 1371 MPN patients, there was no significant difference in the frequency of each additional mutation, except a slightly lower incidence of ASXL1 mutations in MPN/SVT patients (8% vs 23% in MPN patients, p=0.07). Of note, MPN/SVT patients were younger at MPN diagnosis than the control MPN cohort (44 vs 61 years, p< 10^-5). Clinical characteristics were not significantly different between MPN/SVT patients with and without additional mutations (Table 1), including age at SVT or at MPN diagnosis, type of MPN or type of SVT, use of cytoreductive therapy, and median follow-up (11 years in both groups). In contrast, MPN/SVT patients with additional mutations had a significantly higher median JAK2V617F mutant allele burden than those with only JAK2 mutation (34% vs 11%, p=0.003, Figure 1); and higher risk of hematological transformation to myelofibrosis or acute leukemia or death (30% vs 6%, p=0.017). Only few patients experienced recurrence of thrombosis during follow up, with no difference according to the mutational profile (3 (11%) and 6 (17%) in patients with and without additional mutations, respectively).

Conclusion: This is to our knowledge the first study assessing the impact of molecular abnormalities in the long term outcome of MPN/SVT patients showing a possible prognostic role for NGS data in these patients. Our results suggest that MPN/SVT patients with only JAK2V617F mutation and a low mutant allele burden (below 50%) are at low risk of adverse hematological outcome. In contrast, NGS may identify a group of patients with additional mutations at very high long term risk of unfavorable hematological evolution (30% transformation or death after 11 years of median follow-up, that is unexpected in such a young MPN population). In this group, a disease modifying therapy like interferon or other targeted therapies should be proposed when possible to reduce this risk of transformation.

View largeDownload slide

View largeDownload slide

Close modal