Development of biomarker assays to dissect subtype-specific heterogeneity and anti-EGFR treatment response in colorectal cancer

Abstract

Colorectal cancer (CRC) is the third most frequently diagnosed type of cancer and the second leading cause of cancer deaths worldwide. Although extensive biological heterogeneity at multiple "-omics" levels has been demonstrated, a very limited number of biomarkers are currently taken into account to guide treatment decisions in the clinic. The reason for this is partially due to the lack of validated assays suitable for routine clinical application. Building on gene expression CRC subtypes previously identified (CRCAssigner: enterocyte, goblet-like, transit-amplifying, stem-like, inflammatory; Consensus Molecular Subtypes (CMS): CMs1, CMS2, CMS3, CMS4) as potential biomarkers of clinical interest, the aims of the work presented in this thesis were: - Develop and validate assays able to stratify CRC patients into clinically meaningful subgroups based on the subtypes; - Demonstrate the potential value of the assays as companion diagnostic tools for prediction of benefit from epidermal growth factor receptor (EGFR) targeted drugs cetuximab and panitumumab. A total of 825 clinically annotated CRC samples were analysed. A gene expression assay for nCounter platform (NanoString Technologies) was developed to classify CRC samples into CMS. A second assay was developed to integrate the first CMS assay into a previously validated assay for CRCAssigner classification. The integrated assays enabled the simultaneous classification of samples into CRCAssigner and CMS subtypes. The accuracy of the assays was assessed in both fresh-frozen and formalin-fixed paraffin embedded samples. Orthogonal validation of the results was performed using matched RNAseq data and by confirming known subtype-specific associations with clinico-pathological features. Assay detection of similar biological features in multiple Caucasian and Asian populations was demonstrated. Finally, the potential of the assays as companion diagnostic for anti-EGFR benefit prediction was demonstrated in a retrospective cohort; the results were validated in samples prospectively collected from patients who received anti-EGFR monotherapy within an international phase III clinical trial.