Transcriptional Effects of Cytarabine Based Regimens in Patients with Acute Myeloid Leukemia By RNA-Seq Analysis

• Published in: Blood 2024-11, Vol.144, p.6150-6150
• Main Authors: Mora, Alba, Álvarez, Noemi, Colmenares, Rafael, Morales, Maria Luz, Garcia-Vicente, Roberto, Carreño, Gonzalo, Rapado, Inmaculada, Martínez-Lopez, Joaquín, Linares, María, Rosa-Rosa, Juan Manuel, Ayala, Rosa
• Format: Article
• Language: English
• ISSN: 0006-4971
• DOI: 10.1182/blood-2024-210005

INTRODUCTION Acute myeloid leukemia (AML) is the most common acute leukemia in adults. Despite the use of different therapeutic strategies, based on Cytarabine regimens, achieve partial remission in a substantial proportion of patients, almost 50% of young patients and 80% of older patients die primarily because of refractoriness or relapse after treatment. For this reason, identification and characterization of molecular mechanisms involved in these failures is urgently required. To address this purpose, we evaluated differential gene expression with RNAseq, which has become an essential tool to study most of the hematologic malignancies. METHODS For this study we included 25 patients with AML diagnosis treated in our hospital between 1999 and 2019 with Cytarabine based regimens. We performed retrospective transcriptomic analysis by RNAseq to compare paired frozen bone marrow patient's samples at diagnosis and after therapy. Comparative groups were done according to patients who responded (partial response, PR; n=8) to therapy after 1 cycle versus patients with relapse (RE) or refractoriness (RF) AML (n=11 and n=7, respectively). Samples from patients who achieved complete responses were not available at the moment of this study. Samples were prepared with KAPA RNA HyperPrep Kit and analyzed by Illumina Platforms. RNA sequencing data from healthy donors (n=8) obtained from Sequence Read Archive (SRA) database were included as controls. Analysis was performed using Salmon v1.10.2 and Deseq2 Library from R software. The analysis was aimed to identify the differential expression of transcripts. RESULTS We first identified those transcripts that were differentially expressed in our samples with an unsupervised analysis and 166 transcripts with the highest statistically significant fold-change were selected. Samples at diagnosis were grouped together or closer to its paired post-treatment sample in the majority of cases, but we observed some cases which had similar patron expression to patients who achieve partial response or who relapsed after therapy. Interestingly, samples from patients with refractoriness AML were grouped together in a cluster and showed a remarkably lower expression of these transcripts. The relative expression of selected transcripts in post-treatment samples were relativized to that obtained in pre-treatment samples and grouped according to response to therapy (PR, RE, RF). One-hundred five transcripts were down-regulated in PR compared