Investigation of variable therapy outcomes in chronic lymphocytic leukaemia using a proteomics approach

Khan, Umair ORCID: 0000-0002-9470-3481 (2022) Investigation of variable therapy outcomes in chronic lymphocytic leukaemia using a proteomics approach. PhD thesis, University of Liverpool.

Text 201206027_Dec2022.pdf - Author Accepted Manuscript Access to this file is embargoed until 1 August 2024. Download (11MB)

Abstract

Huge progress has been made in recent years in elucidating the biology of chronic lymphocytic leukaemia (CLL) through the application of genomic, epigenomic and transcriptomic approaches. Despite this, our understanding of the biological basis of CLL heterogeneity remains incomplete. Given that disease phenotype is ultimately determined at the protein level, we hypothesised that adopting a proteomics approach may provide new insights into variable therapy outcomes in CLL with the potential to identify new drug resistance mechanisms, drug targets and biomarkers. Peripheral blood mononuclear cells (PBMCs) were processed and subjected to Sequential Windowed Acquisition of all Theoretical fragments mass spectrometry (SWATH MS) for protein quantification. A pilot study consisting of three paired, pre- and post-treatment CLL samples from the Liverpool Blood Disease Biobank confirmed the feasibility of the study whilst highlighting the importance of batch effects. Subsequent experiments deconvoluted the source of batch effects in the proteomics workflow and, as a result, the experimental design was optimised. Next, PBMCs from a total of 64 CLL patients treated with regimens containing fludarabine, cyclophosphamide and rituximab (FCR) obtained from the UK CLL Biobank (ARCTIC and AdMIRe trials) and Leicester Haematological Malignancies Tissue Bank were processed using SWATH MS. Two independent proteomic datasets were generated which compared the proteomes of (i) 32 pre-treatment trial samples from patients who achieved minimal residual disease positive or negative responses after therapy, and (ii) 16 paired samples, collected pre-treatment and at disease relapse or progression. Differential protein expression analysis of comparison (i), using cut-offs of 1.2 fold change and p<0.001, identified six proteins of interest. Among these, high expression of PI4K2A protein correlated with inferior progression-free survival; this finding was corroborated at the mRNA transcript level using the International Cancer Genome Consortium database of 476 patients. These findings suggested a potential role of PI4K2A in early disease progression. In addition, Ingenuity Pathway Analysis using differentially expressed proteins (p<0.05) identified from comparisons (i) and (ii) showed EIF2 signalling as the most enriched pathway in both datasets. EIF2 signalling was also identified as the top pathway in a proteomic comparison of fludarabine-sensitive HG-3 cells with HG-3 cells that had acquired fludarabine resistance through prolonged sublethal drug exposure. Pharmacological inhibition of PERK, a key kinase in the EIF2 or eIF2α pathway, increased the fludarabine-induced cell killing of fludarabine-resistant HG-3 cells by 74% (p=0.002), whereas no such sensitisation was observed in the corresponding fludarabine-sensitive cells. Moreover, PERK inhibition did not sensitise fludarabine-resistant HG-3 cells to the BCL-2 inhibitor venetoclax, suggesting that therapy resistance occurred upstream of mitochondria. In summary, this study confirmed the utility of a proteomics approach using SWATH MS in providing further insight into CLL biology and identified PI4K2A and the eIF2α pathway as novel determinants of therapy outcome.

Item Type:	Thesis (PhD)
Divisions:	Faculty of Health and Life Sciences
Depositing User:	Symplectic Admin
Date Deposited:	29 Aug 2023 13:20
Last Modified:	29 Aug 2023 13:21
DOI:	10.17638/03169573
Supervisors:	Pettitt, Andrew Jenkins, Rosalind Falciani, Francesco Phipps, Alex
URI:	https://livrepository.liverpool.ac.uk/id/eprint/3169573