The Overexpression of the Hsa-miR99b-5p/Let-7e Cluster Identifies High-Risk Myelodisplastic Patients. an IPSS-R Stratification Combined Model Approach

Introduction: The Myelodisplastic syndromes (MDS) are a heterogeneous group of clonal hematological neoplasms characterized by a progressive bone marrow failure with an increased risk for transformation to acute myeloid leukemia (AML). The WHO 2008 classification and the revised international prognosis-scoring system (IPSS-R) had improved prognosis assessment and therapy. The IPSS-R is based on the cytogenetic risk scoring system focus into the presence of cytogenetic alterations. However more than 50% of MDS patients are diagnosed without cytogenetic abnormalities (diploid Cytogenetics, DC-MDS), different to their counterparts with abnormal karyotype (AK-MDS) they face problems for prognostication. Epigenetic tests like microRNAs (miRNAs) studies can found biomarkers for diagnostic, prognosis and potentially target therapy. The miRNAs implications on hematopoiesis are wide recognized. After an exploratory work of miRNAs implication on MDS patients, a cluster located inside the chromosome region 19q13.2-19q13.4 appears to be related with the outcomes and risk for transformation. Here there is described novel data incorporated into a combined model to improve prognostication of patients with DC-MDS.

Aims: To explore the usefulness of miRNAs determination to improve prognosis assessment in MDS patients.

Patients and Methods: Based on previous works (Andres-Coduras, et al, ELN 2012, Andrade-Campos et al, Blood 2015: 126(23): 1642), a miRNAs profile, composed by 19 miRNAs identified in 40 patients with MDS compared to controls was analyzed in 242 plasma samples obtained at diagnosis from patients with MDS, ethic committee and inform consent according Helsinki declaration signature were required. The miRNA profile included: hsa-miR-26a, hsa-miR-451, hsa-miR-99b, hsa-miR-24, hsa-miR-625, hsa-miR-15b, hsa-miR-19b, hsa-let-7e, hsa-miR-16, hsa-miR-140-3p, hsa-miR361-3p, hsa-miR-378 and hsa-miR-942; The 2-d(dCT) method and expression over 1.3 (-10Log) times were considered as overexpression.

Results: from 242 samples, 141 (62.4%) were males, mean age of the entire cohort: 73.3 (29-91 years). For DC-MDS patients (n=154), mean age 74 (33-91) years, 94 (61.0%) males. For WHO, Cytogenetic and IPSS-R risk classification see table 1. During follow-up (median 29 (2-76) months), 47 progressions (24 on DC-MDS group) and 108 deaths (64 in DC-MDS) were registered. For a median progression free survival (PFS) of 49.4 (42.4-56.3) months and 58.0 (52.0-64.5) months for AK-MDS and DC-MDS respectively, and overall survival (OS): 36.0 (29.6-42.6) months and 45.4 (40.9-49.8) months respectively.

During the multivariate analysis with the objective to assess the risk of progression/death, the cytogenetic risk, IPSS-R and overexpression of miRNAs Let-7e-5p, miR-99b-5p and miR-140-3p showed significant differences among all subgroups. Regarding miRNAs only Let-7e and miR-99b showed a normal distribution into the entire cohort, and considering that both are part of a cluster, a combined expression model analysis was carried out. The hazard ratio (HR) for risk to progression of the combined overexpression (Let-7e + miR-99b) was 2.9 for the entire cohort, HR 1.3 for AK-MDS, but for the DC-MDS group: HR: 4.5 confirmed by Cox regression (p=0.034), for OS the results were also significant (p=0.008) for the DC-MDS, but not for the AK-MDS, in which IPSS-R remain as the more predictive tool. Considering the fact that >50% of patients present DC-MDS with problems for the stratification, a combined model adding the overexpression of Let-7e+miR-99b-information to the IPSS-R was created in order to split the categories as follows: 1. Very low with miRNAs overexpression 2-Very low without miRNAs overexpression, 3-Low with miRNAs overexpression, 4-Low without miRNAs overexpression, 5- Intermediate with miRNAs overexpression, 6- Intermediate without miRNAs overexpression (see table 2). With this approach the predictive value of the combined system (IPSS-R & miRNAs) improves the AUC for ROC analysis= 0,786 for PFS and AUC (ROC)=0.752 for OS (both p<0.001), identifying patients at risk for progression or death independently of the IPSS-R category, with especial repercussion among the low and intermediate risk groups.

Conclusions: this model offers for the first time the possibility to improve the accuracy of IPSS-R and identified high-risk patients.

This work has been partially supported by FEHHA.