Abstract
Background: Analysis of genomic alterations can play an important role in clinical and pathological evaluation of patients with MDS. Most studies have assessed the role of only a few genes. It is difficult to analyze a large number of genes in a large patient cohort in the context of a clinical molecular diagnostic laboratory.
Aim: We sought to characterize the role of 56 candidate genes in MDS outcomes in 451 consecutive patients referred to one institution.
Methods: DNA extracted from bone marrow aspirates were analyzed in a clinical molecular diagnostic laboratory underCLIA regulations between 9/2012 and 1/2014. Mutation analysis of 56 cancer-related genes was performed using a next generation sequencing (NGS) panel. Briefly, DNA extracted from bone marrow sample was tested for mutations in 56 genes using Illumina TruSeq chemistry on MiSeq (Illumina Inc.) NGS instrument. Sequence variants were detected using MiSeq Reporter software and verified by pathologists’ review. Summary statistics were used to describe the study population. Overall survival (OS) was calculated as number of months from diagnosis to death or last follow-up date. Patients alive at their last follow-up were censored on that date. Event-free survival (EFS) was calculated as the number of months from diagnosis to relapse, transformation to AML, death, or last follow-up date. The Kaplan-Meier product limit method was used to estimate the median OS for each clinical/demographic factor. Univariate Cox proportional hazards regression was used to identify association with each of the variables and survival outcomes. For these univariate analyses, some variables were transformed using the natural log scale. If the transformed variable showed a narrowed 95% CI (confidence interval) for the hazard ratio (HR), the natural log of the variable was used for analysis.
Results: Median age was 66.6 (18-90) years. Fifty-nine patients (13%) were considered to have MDS/MPN, and 128 (28%) patients received prior therapy at the time analysis. IPSS-R cytogenetics were very good for 10 patients (2.4%), good for 198 (47%), intermediate for 98 (23%), poor for 16 (4%), and very poor 96 (23%). Most patients (62%) received HMA-based therapy. Complete response to any therapy was seen in 82 patients (38%). The median number of mutations was 1 (range 0 to 6), and 228 patients (50%) had no mutations, 154 (34%) had 1 mutation, 45 (10%) had 2 mutations, and 24 (5%) 3 or more mutations. Only 16 out 56 genes had frequencies >5: ASXL1, DNMT3A, EZH2, IDH1, IDH2, JAK2, KRAS, NOTCH1, NPM1, NRAS, PTPN11, RUNX1, TET2, TP53, CEBPA, and FLT3-ITD. These genes were used for further analysis. There was no significant association between any clinical or demographic parameters and mutations in JAK2, NOTCH1, RUNX1, or CEBPA. All significant associations between clinical or demographic parameters and mutated genes are summarized in Fig 1. To further characterize the role of genetic mutation in outcomes, we created a “gene score” parameter based on univariate Cox regression models for each of the 16 genes. The gene score was the sum of all mutational contributions for a given survival parameter. Using stepwise backward selection in a multivariate model, we identified age (HR=1.08, p<0.001), bone marrow blast percentage (HR=1.44, p=0.012), and the gene score (HR=2.47, p=0.001) as associated with worse event-free survival. Peripheral blood blast percentage (HR=0.92, p<0.001), log of white blood cell count (HR=0.76, p<0.001), and hemoglobin (HR=0.64, p<0.001), were significant factors for better event-free survival. Age (HR=1.05, p<0.001) and gene score (HR=2.77, p<0.001) were associated with worse OS, and hemoglobin (HR=0.77, p=0.001), log of platelet count (HR=0.7, p=0.005), and peripheral blood blasts (HR=0.98, p=0.023) were associated with better OS.
Conclusions: We identified a group of 12 genes that are significantly associated with clinical and demographic variables in a large cohort of MDS patients. Further characterization of the results of analysis and role of gene score as predictor of response may allow for more accurate stratification of patients based on genomic parameters.
Abstract 4658. Table 1 ASXL1 DNMT3A EZH2 IDH1 IDH2 KRAS NPM1 NRAS PTPN11 TET2 TP53 FLT3-ITD Age BM blasts PB blasts Mean WBC Mean neutrophils Mean platelets Mean Hgb IPSS CG risk Complex CG Prior chemotherapy Prior radiotherapy Diagnosis
Disclosures
No relevant conflicts of interest to declare.
Differential expression of circulating exosomal microRNAs in refractory intracranial atherosclerosis associated with antiangiogenesis