Popis: |
Acute lymphoblastic leukemia (ALL) is the most common childhood cancer. The five-year event free survival is 85% but the incidence of relapse is high and prognosis after relapse remains poor. To improve outcomes for those patients, precision medicine programs are designed to target specific genomic alterations. Although many biomarkers have been identified, targeted therapies have been less successful than expected. This is primarily a consequence of the aggressiveness of relapse cancers and the limited understanding of functional changes underlying genetic biomarkers. Ideally, the molecular and functional characterization of potential targets could start earlier to prepare for potential relapse. However, there are conflicting reports of how targetable lesions persist through disease progression and little known about progression in ALL proteomes. I hypothesize that the targetable lesions detected in childhood leukemias will be stable through disease progression and the combined genome and proteome analysis will better clarify dysregulated pathways. I first assessed the performance of a recently developed childhood cancer-specific next generation sequencing (NGS) assay in 28 childhood tumour specimens. The childhood cancer-specific assay detected almost 10% more targets than a broad cancer panel and both panels showed high concordance with whole-genome sequencing (WGS). I next investigated genomic stability and persistence of druggable events in paired diagnosis (Dx)-relapse (R) samples from 11 patients treated at BC Children’s Hospital, and whole exome sequencing data from paired Dx-R samples from 69 patients treated at St. Jude’s Hospital. Approximately 64% of patients had at least one druggable target retained between diagnosis and relapse. Six paired specimens were treated in-vitro with variant-matched targeted inhibitors, and although the sensitivity to inhibitors was low, IC50 doses of paired samples were highly correlated (r=0.8486). Similarly, a comprehensive proteome analysis of paired ALL specimens revealed high statistical equivalence (median = 85%) and similar abundance profiles of cancer-associated proteins between diagnosis and relapse. Finally, discovery whole-proteome analysis identified PARP1 as a potential new pan-ALL therapeutic target, and sensitivity to PARP1/2 inhibitors was confirmed via in-vitro drug assays. My thesis indicates that comprehensive interrogation of tumour genomes and proteomes through disease progression may provide support for implementing a prospective precision oncology approach. |