Estudio de la utilidad del análisis de perfiles de micrornas plasmáticos en la clasificación pronóstica en pacientes con síndrome mielodisplásico (smd) y cariotipo normal

Resumen

Abstract Introduction The Myelodisplastic Syndromes (MDS) are a widely heterogeneous group of myeloid neoplasms; the key characteristic feature is a progressive bone marrow failure syndrome with an increased risk to acute myeloid leukemia (AML) transformation. It is a clonal disease of the myeloid hematopoietic stem cell with a repercussion in the shape characteristics of at least one of the different cell lines (granulocytes, monocytes, erythrocytes, platelets). Their estimated incidence in Europe vary from 3 to 20 cases per 100,000 inhabitants, with a mean age of presentation ~ 76 years with an increasing peak with along the elderly, also their can be considered as primary or “de novo” or related to therapy or another conditions. The accepted classification system is the W.H.O. classification of 2008 that includes refractory citopenia with unilineage dysplasia (RCUD) (refractory anemia -RAUD / neutropenia -RNUD / thrombocytopenia - RTUD), RC with multilineage dysplasia (RCMD), RCMD with ringed sideroblast (RCMD+RS), refractory anemia with RS (RARS), RA with excess of blasts-1/2 (RAEB-1/2), MDS unclassifiable (uMDS) and 5q- Syndrome as a provisional entity. The identification of some recurrent cytogenetic or karyotyping alterations in ~50% of cases have improve the diagnosis and help in cases even without clearly dysplasia support the diagnosis of MDS. After the diagnosis, the most important step to settle the management approach to the patient is the prognosis assessment. Along with the improvements on cytogenetic testing and hematopoiesis knowledge, there has been a continuum evolution on prognosis score. One of the biggest steps was the establishment of the cytogenetic risk score of five categories by Schulz et al in 2011. This score permits to assignee a risk value to ~90% of the cytogenetic alterations, distributing them on 4 categories very good, good, bad and very bad, remaining only 10% in the indeterminate or intermediate risk category. This cytogenetic risk score, permitted to improve the international prognosis scoring system (IPSS) created in 1997 by Greenberg et al. The IPSS was based on 3 cytogenetic categories (good, intermediate and bad), the number of citopenia lineages and the number of blast; it was created to works along with the FAB and first WHO classification, and it was also working appropriately with the WHO classification of 2003. The patients are divided in 4 risk categories: low risk, intermediate-1, intermediate-2 and high risk. Regarding therapy, only intermediate-2 and high risk are candidates for therapy otherwise than supportive. However, the use of WHO 2003 and 5-categories of risk cytogenetic classifications led to a revision of the IPSS (IPSS-R) in 2011 by Greenberg et al. The IPSS-R includes the Schanz et al cytogenetic risk score, a more detailed analysis not only the number of lineages of citopenia, but also their severity, and an accurately assessment of the risk derived from the bone marrow blast count. This score establish 5 risk categories: very low, low, intermediate, poor and very poor. It’s important that even after 5 years, the IPSS-R is the best risk assessment score, it is not accepted to define the suitability of patients to therapy. Therapy of the MDS patients, are based on two basic approaches: low risk and high-risk patients. Low risk patients, includes the IPSS low and intermediate-1 groups, and the very low, low and intermediate for IPSS-R. For these patients, there are no indicated therapies in Spain. Different approaches including transfusion support, erythropoietin, granulocyte stimulation agents (G/GM-CSF), hormones, immunosuppression therapy, chelation and other supportive therapies are used. The rest of the patients are considered at high risk and aggressive (natural history modifiers) therapy is preferred, including chemotherapy, allo stem cell transplantation, hypomethilating agents including all the supportive therapies needed. For 5q- syndrome regarding the excellent results in this subset of patients, FDA approves the use of lenalidomide; in Spain still is used as a compassionate drug. The IPSS-R base its major prediction accuracy in the presence of cytogenetic alterations, the fact, that ~50% of cases do not present this alterations implies problems in their prognosis assessment, actually there are groups working in mutation scores similar to the molecular prognosis assessment used in the management of AML. Based on this, in this work, we approach a new way to found useful prognosis information on MDS in general and in normal or diploid karyotype MDS patients using epigenetics studies, in our case microRNAs quantification. The microRNAs or miRNAs are small (18-25 nucleotides) single chain RNA molecules implicated in different cells process like maturation, differentiation, hematopoiesis en general an so on. Early after their discovery were named as oncoMIRs due to their identification/implication in different oncologic process. Their act mainly as suppressors of the translation process, but also as promoters, either interacting directly to the promoters regions of different genes in the nucleus or in the translation level of the mRNA at cytoplasm. They can be isolated from cells, plasma or other biological fluids and quantified by PCR. Hypothesis and Objectives The aims approached in this thesis were: 1- to identified a miRNAs profile in patients with MDS; 2- to compare the profile in samples from cell or plasma from peripheral blood (PB cells, plasma); 3-to explore the useful of the miRNA profile as a diagnostic tool; 4- to compare the miRNAs profile in MDS patients with diploid cytogenetics vs. with karyotype alterations and 5- to explore the useful of miRNAs profile as a prognostic marker in diploid MDS patients. Patients and Methods To answer these aims, a multistep process was conducted. A primary exploratory pilot study were developed in 2009-2011 using 40 samples (PB cells and plasma) from MDS patients (AR:10, ARSA: 2, CRDM: 8, CRDM+SA:2, AREB-1:4, AREB-2:4, LAM: 7, LMMC:3) and 40 age and gender matched controls, mean age 67 (19-86) and 70 (21-80) years respectively. A screening of 754 miRNAs using Megaplex Human Plasma A and BTM platform was conducted in both population and both different samples. The initial results permit us to identify a profile of 14 miRNAs differentially expressed in MDS patients respect to controls. Also it permits to compare the expression between PB cells and plasma, except Let-7e, the rest of the miRNAs profile show a higher expression in plasma samples, selecting this source for the further steps. After this partial analysis the miRNA profile was defined including 19 miRNAs that showed different expression, some of them with reported implication in the hematopoiesis process by other groups. The final profile was composed by: 1.- Has-miR-625-5p 2.- Has-let-7a-5p 3.- Has-let-7e-5p 4.- Has-miRNA-140-3p 5.- Has-miRNA-15a-5p 6.- Has-miRNA-15b-5p 7.- Has-miRNA-17-5p 8.- Has-miRNA-19b-3p 9.- Has-miRNA-24-3p 10.- Has-miRNA-25-3p 11.- Has-miRNA-26a-5p 12.- Has-miRNA-30b-5p 13.- Has-miRNA-361-3p 14.- Has-miRNA-378a-3p 15.- Has-miRNA-378a-5p 16.- Has-miRNA-451ª 17.- Has-miRNA-625-3p 18.- Has-miRNA-942 19.- Has-miRNA-99b-5p In an extension phase, after PCR technique optimization 242 plasma samples from MDS patients from the cooperative group INBIOMED were processed, for miRNA expression the formula 2-dd(CT) and log 10 expression were used to adjust the expression founded according controls and housekeeping gene, in our case GeNorm program selected miR-16 to be used as housekeeping gene. For the general / clinical analysis the FAB, WHO 2008, IPSS, WPSS, IPSS-R and five categories cytogenetic risk scores were used. The cases in which a diploid karyotype and absent of alteration by FISH study were considering as carries of normal or diploid karyotype. General demographic data and transformation to AML or death were recorded in median follow-up of 29.5 months. Results and Discussion From de 242 samples, 141 (62.4%) were males and 87 (38.6%) females (14 missing data). According WHO 2008: CMML: 5 (2.2%), AREB-1: 31 (13.6%), AREB-2: 19 (8.3%), ARSA: 13 (5.7%), RCUD: 22 (9.6%), RCMD: 109 (47.8%), hipoplastic MDS: 2 (0.9%), uMDS: 6 (2.6%), 5q- syndrome: 15 (6.6%) and AML: 6 (2.6%). Cytogenetic five-category risk: very good: 11 (4.8%), good: 169 (72.8%) (Normal karyotype 154 – 66.4%), intermediate: 29 (12.6%), poor: 14 (4.2%), very poor: 10 (4.2%) (in 11 patients only cytogenetic risk assessment for IPPS were founded but not for this score). IPSS: low risk: 113 (47.3%), intermediate-1:80 (33.5%), intermediate-2: 32 (13.4%) and high risk: 14 (5.8%). For IPSS-R: very good: 59 (25.9%), good: 93 (40.8%), intermediate: 37 (16.2%), poor: 27 (11.8%), very poor: 12 (5.3%). A predominance of good risk categories were founded, similar to other published cohorts. For very low and low risk categories, medians of progression free survival (PFS) were not reached, means were 67.5 (64.0-79.0) and 56.0 (52.4-59,6) months, for intermediate risk group median PFS was 64 (1.7-126.2) months, for poor risk: 19.0 (0.1-43.2) months and very poor: 8.0 (5.1-10.8) months. Medians for overall survival (OS) for very good risk group was not reached, for good risk was 60 (49.2-70.7), for intermediate: 27 (19.6-34.3) months, for poor risk: 14 (10.3-17.6) months and for very poor: 11 (3.0-18.9) months. In the 19 miRNAs profile, globally miR-26a, miR-625 and miR-24 were the most expressed respect controls, independently of the MDS subtype, all of them would be used as biomarkers for MDS diagnosis, this profile has not published elsewhere and our experiment is different from others miRNA studies due to the screening platform and the miRNAs source used. More information about this is presented in the discussion section. Comparing the expression between the group with cytogenetic alterations vs. normal karyotype, miR-99b, Let-7a-5p, miR-15a-5p, miR-24-3p, miR-378-3p and miR-378a-5p showed an statistically different expression between cohorts, also their expression were higher respect to controls, this led us to think about their usefulness as diagnostic biomarkers. Focusing in the normal karyotype cohort (154 patients), it was found that miR-625-3p/5p, miR-24-3p, miR-26a-5p, miR-30b-5p, miR-378-3p and miR-99b-5p showed the highest expression inside the group, until now, the longer cohort ever published. During survival analysis, either PFS or OS, it was found that the higher expression of miR-99b, Let-7e and miR-24-3p independently were associated with bad outcomes (p<0.001, p<0.001 and p=0.005 for PFS and p<0.001, p<0.001 and p<0.009 for OS respectively). The distribution in both cohorts were regular for miR-99b and Let-7e, but not for miR-24-3p, for that reason, only the first two were selected to be combined with the IPSS-R data and create a mixed prognosis score. The combined prognosis score (miR-99b+Let-7e+IPPSS-R) were created splitting the IPSS-R categories according the presence of combined higher expression of miR-99b+Let-7e. The categories were as follow: 1. Very low with miRNAs overexpression 2. Very low without RNAs overexpression 3. Low with miRNAs overexpression 4. Low without RNAs overexpression 5. Intermediate with miRNAs overexpression 6. Intermediate without RNAs overexpression Comparing the prediction value by ROC curves, it was found that for the combined model the ROC = 0.786 for PFS and 0.752 for OS, higher than the obtained by the IPSS-R or miRNAs (miR-99b+Let-73) overexpression alone. This permits to improve the prognosis classification, See table 1. Table 0.1. Risk assessments by IPSS-R and combined model. IPSS-R categories OS IPSS-R (months) OS combined model (months) PFS IPSS-R (months) PFS combined model (months) Very good (N=52, 34,7%) 54,0 (42,7-60,5) 54,0 (33,9-74,1) 53,9 (51,2-56,7)* 100% SLP 46,0 (17,7-80,7) 88% SLP Good (N=71, 47,3%) 55,0 (35,2-74,9) 55,0 (46,1-63,9) 56,5 (52,6-60,4)* 59,5 (56,1-62,9) 32,0 (20,5-43,4) 43,7 (35,5-51,8) Intermediate (N=17, 11,3%) 26,0 (15,2-36,7) 26,0 (14,9-37,1) 64,0 (9,4-118,5) 56,8 (35,5-78,1) 24,0 (6,5-41,5) 24 (0,1-49,1) * Means, medians not reached. As is presented, this new prognosis approach permits us to improve the accuracy of the IPSS-R prognosis value. Led it to identified patients initially included in the very good and good categories but with OS bellow 3 years, also split the intermediate group identifying patients with 65% risk to progression to AML; this two initial observation could permit to modifying their treatment approaches. The miRNAs Let-7e and miR-99b are part of a cluster with mir-17, the three localized in the chromosome 19. Their expression have been described independently and related with different cellular process, however this is the very first time describing their relation with MDS patients, opening the possibilities to explore more in the knowledge of their roles on MDS pathophysiology. This study is, in the best of my knowledge, the first prognostic tool that improves the accuracy of IPSS-R led in us to go an step ahead in the personalizing medicine way. The aims have been fulfilled and it can be said, that based on this data, miRNAs are good epigenetic biomarkers, useful for diagnosis and prognosis assessment in patients with MDS, especially in the normal karyotype subset in which their expression can supply the absence of cytogenetic data.