Background – Antibody-drug conjugates (ADCs) such as the HER2-targeting agent trastuzumab deruxtecan, the TROP-2 targeting agent sacituzumab govitecan, and the NECTIN4 targeting agent enfortumab vedotin-ejfv, have resulted in increased clinical response rates in previously difficult to treat solid tumor indications, but only a minority of patients exhibit such responses. Increased understanding of the molecular underpinnings of sensitivity and resistance to ADCs beyond membranous target expression is needed. To that end, we developed a platform for generating, deeply characterizing, and functionally screening extensive panels of patient-derived organoids. Here, to demonstrate feasibility, we report on a screening panel of 60 PDOs representing 10 solid tumor indications and focus on response to ADCs now widely used in the clinical setting: trastuzumab deruxtecan and sacituzumab govitecan.

Methods – PDOs were cultured in the presence of 8-dose dilutions of either trastuzumab deruxtecan or sacituzumab govitecan as monotherapies. Experiments were performed in 384 well plates with each dose in technical triplicate. Cells were plated and drugs were added after 24 hours. Cells were cultured for an additional 6 days with no media refreshes. The screen included staurosporine as a positive control and vehicle negative controls. On Day 7, CellTiter-Glo (CTG) was added to the plates to terminate the experiments and assess cell viability. To evaluate treatment activity, cell viability measurements were normalized to vehicle controls and dose-response curves were generated to estimate key efficacy metrics including IC50, EC50, and inverse area under the curve (iAUC).

Results – ADC specific responses were observed across the 60 models tested. As anticipated, response to ADC exposure was driven by target expression. This was most obvious in response to enfortumab vedotin-ejfv, and also observable to a lesser extent with sacituzumab govitecan and trastuzumab deruxtecan. To better understand other molecular indicators of response, differential gene expression analysis was performed. This revealed additional potential biomarkers of specific ADC response including VIL1, MYO7B, HNF4A, EFS, EPS8L3, and several of which interestingly regulate cell-cell junction proteins.

Conclusions – The results shown here demonstrate that screening pan indication panels of PDOs can reveal ADC specific differences in efficacy which are rooted in molecular factors that include, but are not limited to target expression. Further analysis is ongoing, including establishing the prevalence of non-ADC target biomarkers in real-world patient populations via the projection of our screen-derived data onto Tempus RWD.

VIEW THE PUBLICATION