Introduction:

Enfortumab vedotin (EV) is an antibody-drug conjugate (ADC) targeting NECTIN4, a surface protein highly expressed in bladder tumors, and is currently approved for metastatic urothelial carcinoma (mUC). Based on its demonstrated benefit, EV is currently being evaluated in earlier UC settings, including intravesical administration for patients with non-muscle invasive bladder cancer (NMIBC). Because ADCs are associated with primary or secondary resistance, it is crucial to understand mechanisms of EV resistance. Previous studies suggest that loss of NECTIN4 expression and upregulation of the drug efflux pump P-glycoprotein may lead to EV resistance. However, other mechanisms have not been explored. Therefore, we sought to develop a preclinical bladder cancer model of EV resistance with which we could investigate mechanisms of and strategies to overcome resistance.

Methods:

The RT112 bladder cancer cell line, which expresses high levels NECTIN4, was used to develop a model of resistance. RT112 cells underwent “cycles” of treatment with 5-7 days EV with subsequent recovery and passaging of surviving cells. Each “cycle” used escalating doses of EV. Parental RT112 cells were cultured in parallel. Cells were profiled with flow cytometry and western blotting.

Second-generation chimeric antigen receptor (CAR) T-cells were engineered with a single chain fragment variable (scFv) designed with the variable regions of the heavy (VH) and light chains (VL) of enfortumab, the human monoclonal antibody specific for NECTIN4.

Drug dose-response and CAR T-cell killing assays were performed with RT112 parental and EV-exposed cells. After 24h, drug or CAR T-cells were added at indicated concentrations or effector-to-target (E:T) ratios, and cell count and growth monitored using an IncuCyte S3.

Results:

RT112 cells undergoing treatment “cycles” with escalating concentrations of EV starting at 0.5 µg/ml and culminating at 30 µg/ml yielded a generation of cells that exhibited a 4-5-fold increase in IC50 for EV. The “EV resistant” RT112 cells had comparable surface NECTIN4 levels but increased surface TROP2 levels compared to parental cells. The EV resistant cells were less sensitive to MMAE, the payload of EV. EV resistant RT112 cells were more sensitive to sacituzumab govitecan (SG), a Trop-2-directed antibody conjugated to the topoisomerase inhibitor SN-38. The EV resistant RT112 cells were equally susceptible to killing by NECTIN4-directed CAR T cells compared to parental RT112 cells.

Conclusion:

In conclusion, we developed a preclinical in vitro model of bladder cancer resistant to EV. Resistance to EV was largely due to resistance to the payload MMAE, and not due to downregulation of the surface target NECTIN4. The EV resistant bladder cancer cells remained sensitive NECTIN4-directed CAR T-cells, suggesting that NECTIN4 remains a relevant target even after cells develop EV resistance, and non-overlapping mechanisms of resistance to ADCs versus CAR T therapy. Finally, treatment using SG represents an alternative strategy after EV resistance.

Funding: UCSF Inquiry Yearlong Research Fellowship