Introduction: Treatment failure and/or frequent reoccurrence are the major clinical problems of bladder cancer (BLCa), suggesting that a more personal approach is mandatory. The genomic has a limited role in the identification of therapeutic options, therefore patient-derived organoids (PDOs) have been used for functional assays for predicting drug response in different cancer types. Our research aims to analyze PDOs to determine drug response profiles of BLCa patients to personalize therapy accordingly. We aim to correlate PDOs’ drug sensitivity to their molecular and genomic profiles in association with the parental tumor (PT) as well as to the clinical course of the patient.
Methods: To derive PDOs, tumor samples were dissociated into single cells and cultured in ultra-low attachment plates. The drug response was evaluated by viability assay after 48 hours of treatment. Whole exome sequencing was performed on genomic DNA from PDOs, PT, and blood (germline control). Histological evaluation of PDOs and PT morphology was performed.
Results: We successfully obtained PDOs from 11 NMIBC and 6 MIBC tissues. Genomic analysis of PDOs and PT showed good similarity of mutational burden with preservation of sub-clonal populations. Important BLCa mutations were found in PDOs and matched PT (e.g. loss of p53 for MIBC and amplification of FGFR3 for NMIBC). Additionally, conservation of the tumor phenotype in PDOs was confirmed. A panel of BLCa standard-of-care and FDA-approved drugs, selected based on clinical importance and the possibility of drug repurposing, were tested on PDOs. Surprisingly, a significant response to SOC (cisplatin alone or in combination with gemcitabine) was observed in 8 out of 17 analyzed samples and the non-SOC drugs were effective in few samples. Investigation of drug sensitivity showed partial association with phenotypic features, mutational profiles, and clinical outcomes. For example, Docetaxel sensitivity was associated to copy number gain in TUBB1 and MIBC organoids were significantly more sensitive to a low dose of cisplatin than NMIBC organoids. Moreover, in a case study, alteration caused by treatment was captured in organoids derived from chronologically different tumors.
Conclusions: To conclude, the genetic and phenotypic profiles of PDOs and parental tumor were very similar, supporting that PDOs are preserving tumor heterogeneity and are therefore a good model of BLCa. This indicates that PDOs can represent a tool to predict drug sensitivity in a personalized manner.
Source of Funding: Swiss National Science Foundation (SNF), grant number REF-18-434. Novartis, grant number REF-39-781.
.jpg)
.jpg)
