Advancing understanding of prognosis in sarcoma through quantitative proteomic analysis

Abstract

Sarcoma is a rare and heterogeneous group of malignancies, often associated with poor prognosis. Advances in quantitative mass spectrometry-based technologies now enable cancer proteomes to be studied on an unprecedented scale, offering opportunities to address unmet clinical needs in sarcoma and uncover novel biological insights. Aiming to identify patients at greatest risk of relapse, this thesis describes the development of a bioinformatic method to define prognostic proteomic signatures. A method was developed using two cohorts of localised prostate cancer tissue samples and identified a five-protein signature prognostic for clinical relapse. This signature was independently prognostic and enhanced the predictive performance of an established clinical risk classifier. The method was then refined and applied to three cohorts of leiomyosarcoma samples. A four-protein signature prognostic for metastatic relapse in localised leiomyosarcoma was identified in one cohort and evaluated in two independent cohorts. The signature complemented established prognostic clinicopathological features and provided robust discrimination of metastatic relapse in composite models. To determine whether proteomic analysis could uncover biological pathways associated with prognosis that transcend histological subtype, 500 bone and soft tissue sarcoma samples across three cohorts were analysed. Consensus clustering in each cohort identified a good prognostic group, and pathway analyses revealed congruent biology across cohorts. These findings were independent of histological subtype, underscoring their subtype-agnostic nature. The study suggests that differences in the proteome may reflect the biological foundations underlying variation in disease behaviour among and within sarcoma subtypes. Overall, this thesis demonstrates how prognosis can be determined through examination of the proteome and highlights the capacity of quantitative proteomics to offer novel insights into sarcoma.