UC Davis

Background

The overarching goal of this project is to establish a patient-derived bladder cancer xenograft (PDX) platform, annotated with deep sequencing and patient clinical information, to accelerate the development of new treatment options for bladder cancer patients. Herein, we describe the creation, initial characterization and use of the platform for this purpose.

Methods and findings

Twenty-two PDXs with annotated clinical information were established from uncultured unselected clinical bladder cancer specimens in immunodeficient NSG mice. The morphological fidelity was maintained in PDXs. Whole exome sequencing revealed that PDXs and parental patient cancers shared 92-97% of genetic aberrations, including multiple druggable targets. For drug repurposing, an EGFR/HER2 dual inhibitor lapatinib was effective in PDX BL0440 (progression-free survival or PFS of 25.4 days versus 18.4 days in the control, p = 0.007), but not in PDX BL0269 (12 days versus 13 days in the control, p = 0.16) although both expressed HER2. To screen for the most effective MTT, we evaluated three drugs (lapatinib, ponatinib, and BEZ235) matched with aberrations in PDX BL0269; but only a PIK3CA inhibitor BEZ235 was effective (p<0.0001). To study the mechanisms of secondary resistance, a fibroblast growth factor receptor 3 inhibitor BGJ398 prolonged PFS of PDX BL0293 from 9.5 days of the control to 18.5 days (p<0.0001), and serial biopsies revealed that the MAPK/ERK and PIK3CA-AKT pathways were activated upon resistance. Inhibition of these pathways significantly prolonged PFS from 12 day of the control to 22 days (p = 0.001). To screen for effective chemotherapeutic drugs, four of the first six PDXs were sensitive to the cisplatin/gemcitabine combination, and chemoresistance to one drug could be overcome by the other drug.

Conclusion

The PDX models described here show good correlation with the patient at the genomic level and known patient response to treatment. This supports further evaluation of the PDXs for their ability to accurately predict a patient's response to new targeted and combination strategies for bladder cancer.