Year: 2019
Type: dissertation
Abstract: This thesis is comprised of three studies demonstrating the application of different statistical and bioinformatic approaches to address distinct challenges of implementing precision medicine strategies for hematological malignancies. The approaches focus on the analysis of next-generation sequencing data, including both genomic and transcriptomics, to deconvolute disease biology and underlying mechanisms of drug sensitivities and resistance. The outcomes of the studies have clinical implications for advancing current diagnosis and treatment paradigms in patients with hematological diseases. Study I, RNA sequencing has not been widely adopted in a clinical diagnostic setting due to continuous development and lack of standardization. Here, the aim was to evaluate the efficiency of two different RNA-seq library preparation protocols applied to cells collected from acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL) patients. The poly-A-tailed mRNA selection (PA) and ribo- depletion (RD) based RNA-seq library preparation protocols were compared and evaluated for detection of gene fusions, variant calling and gene expression profiling. Overall, both protocols produced broadly consistent results and similar outcomes. However, the PA protocol was more efficient in quantifying expression of leukemia marker genes and drug targets. It also provided higher sensitivity and specificity for expression-based classification of leukemia. In contrast, the RD protocol was more suitable for gene fusion detection and captured a greater number of transcripts. Importantly, high technical variations were observed in samples from two leukemia patient cases suggesting further development of strategies for transcriptomic quantification and data analysis. Study II, the BCL-2 inhibitor venetoclax is an approved and effective agent in combination with hypomethylating agents or low dose cytarabine for AML patients, unfit for intensive induction chemotherapy. However, a limited number of patients responding to venetoclax and development of resistance to the treatment presents a challenge for using the drug to benefit the majority of the AML patients. The aim was to investigate genomic and transcriptomic biomarkers for venetoclax sensitivity and enable identification of the patients who are most responsive to venetoclax treatment. We found that venetoclax sensitive samples are enriched with WT1 and IDH1/IDH2 mutations. Intriguingly, HOX family genes, including HOXB9, HOXA5, HOXB3, HOXB4, were found to be significantly overexpressed in venetoclax sensitive patients. Thus, these HOX-cluster genes expression biomarkers can be explored in a clinical trial setting to stratify AML patients responding to venetoclax based therapies. Study III, venetoclax treatment does not benefit all AML patients that demands identifying biomarkers to exclude the patients from venetoclax based therapies. The aim was to investigate transcriptomic biomarkers for ex vivo venetoclax resistance in AML patients. The correlation of ex vivo venetoclax response with gene expression profiles using a machine learning approach revealed significant overexpression of S100 family genes, S100A8 and S100A9. Moreover, high expression ofS100A9was found to be associated with birabresib (BET inhibitor) sensitivity. The overexpression of S100A8 and S100A9 could potentially be used to detect and monitor venetoclax resistance. The combination of BCL-2 and BET inhibitors may sensitize AML cells to venetoclax upon BET inhibition and block leukemic cell survival.
Source: Työväentutkimus Vuosikirja
Author: Ashwini Kumar
Language: English
Subfield: Hematology
Field: Medicine
Domain: Health Sciences
Open Access status: green
OpenAlex ID: W2979719902