Introduction:

Neoadjuvant chemotherapy prior to radical cystectomy (RC) for muscle-invasive urothelial carcinoma (UCCx) is the standard of care, though the absolute survival benefit is small, and some patients progress during chemotherapy. While progress has been made in the prediction of sensitivity to platinum-based chemotherapies, providing more accurate, personalized, and clinically-relevant chemotherapy response prediction is an unmet need. We present our early and ongoing experience with rapid, organoid-based drug-screening, which can be performed within 5-10 days of transurethral resection of bladder tumor (TURBT) or RC, and only requires only a small amount of tissue (~0.3 grams),. This novel platform allows for the testing of a large number of drugs or drug combinations and is potentially informative for the prediction of neoadjuvant and adjuvant chemotherapy response.

Methods:

After informed consent was obtained, approximately 1 gram of tumor was procured from patients undergoing TURBT or RC and divided between DNA/RNA sequencing, organoid drug-screening, and patient-derived xenograft model development. Briefly, tissue was dissociated, filtered, and resuspended in organoid media for serial passage and drug screening. Drugs were tested at a single “Cmax” concentration, which is the maximum plasma concentration in human trials so as to provide physiologic relevance. Testing was performed in duplicate and normalized to DMSO controls. For each drug tested at Cmax concentration, organoid viability was assessed, where 0 = complete response, and 100 = no response relative to DMSO controls. This was plotted on a red-green scale for ease of viewing in the Table where green indicates drug sensitivity, and red indicates drug resistance. The number of drugs tested per sample was dependent upon quanitity of cells available upon tissue dissociation and subject to variability over the pilot period as the protocol was refined. Wherever possible, FDA-approved bladder cancer drugs and drug combinations were utilized including FGFR inhibitors. In addition, exploratory analyses with immuno-oncology agents were performed, given the innate advantages of retained tumor heterogeneity and possibility of retained immune cells within the organoid system. Anti-PD1, PDL1, and CTLA4 agents were tested alone and in combination. 

Results:

Drug response analyses were available 5-10 days following procedure. The first patient underwent RC for a locally-advanced pT3 UCCx with 75% squamous differentiation and dose response curves presented in the Figure, show a non-chemo-sensitive tumor. The second (TURBT for cT3N1 pure UCCx, sensitive to FGFR inhibition), third (RC for pure squamous cancer, sensitive only to doxorubicin), and fourth (TURBT for T4 UCCx, exquisitely sensitive to MVAC) patient tumor drug-response results are presented in the Table. The fifth patient underwent RC for pT2N1 UCCx after neoadjuvant gemcitabine-cisplatin. Interestingly, his tumor was MVAC-sensitive, while it was insensitive to gemcitabine-cisplatin. The sixth and seventh patients (RC for pT4, and pT2b, respectively UCCx with focal squamous) showed sensivity to MVAC, and doxorubicin, respectively. Cell counts were lower for patients seven and eight necessitating the testing of fewer drugs. The eighth patient underwent TURBT for a locally advanced small cell cancer of urothelial origin (with adjacent CIS), and as expected, was cisplatin-sensitive. All organoids showed sensitivity to doxorubicin. DNA and RNA sequencing and PDX models are ongoing for all patients, as are clinical determinations of response to neoadjuvant and adjuvant chemotherapy, and immunotherapy.

Conclusion:

This platform allows for the rapid determination of neoadjuvant chemotherapy response within 5-10 days of TURBT, and may further guide selection of therapeutic agents in patients with locally advanced bladder cancer. Correlation with clinical response to systemic therapies is ongoing as are comparisons with tumor FGFR status, PDL1 status, tumor mutational burden, and presence of DNA damage repair pathway mutations. Organoid-based drug testing has the potential for truly personalized therapeutics in bladder cancer, and may also facilitate future drug development and clinical trial efficiency.

Funding: N/A