Patient-derived organoid models of metastatic colorectal cancer to study response to PIK3CA inhibitors

Abstract

Abstract Background More than 70% of patients diagnosed with colorectal cancer with liver metastasis have unresectable disease, so the available treatment is systemic chemotherapy that has limited efficacy and causes harmful side-effects. Thus novel and more effective drugs are needed to improve patients’ survival and quality of life. Recent studies have identified several genomic alterations in CRLM that can theoretically be targeted by novel cancer-specific drugs. Among those is the gain-of-function mutations in PIK3CA – the gene encoding the p110a catalytic subunit of PI3K – that are present in 18% of metastatic colorectal tumours. Following the recent FDA-approval of Alpelisib (a PI3Ka-specific inhibitor) HR-positive breast cancer, there is increasing interest to test this drug efficacy to treat CRLM patients with PIK3CA mutations. Although widely used to test cancer cell response, immortalized cell lines fail to represent tumour genomic alterations and heterogeneity. Our goal is to develop patient-derived primary cell models that can be used to reliably test patients’ drug response. Methods Fresh colorectal cancer liver metastasis surgical specimens were dissociated and digested into single cells and seeded in matrigel with appropriate culture media. Originated organoid cultures were then submitted to a minimum of 5 passages followed by a freeze-thaw cycle before deemed established culture. To validate organoid tumour origin, we then performed whole-genome sequencing of paired organoid and respective fresh tumour. Results The established patient-derived organoids from colorectal liver metastasis add an overall mutation match of > 80% when compared to with the original tumour. We have found so far two CRLM organoid lines with mutations in the helical domain of PIK3CA (E545A and E542K). Conclusions We have demonstrated patient-derived organoids recapitulate the genome of their tumour of origin. Libraries of these primary cell models should in the near future be used to more accurately determine drug effectiveness prior to clinical trials. Ongoing testing of Alpelisib alone or in combination with other PI3K pathway target genes will determine whether these patients could benefit from these drugs. Legal entity responsible for the study Barros-Silva, J.D. Funding Cancer Research UK. Disclosure All authors have declared no conflicts of interest.