Abstract 556: Bystander effects of EGFR-targeting 40H3 antibody-drug conjugates in glioblastoma with heterogeneous EGFR expression

Abstract

Abstract An important aspect to determine the efficacy of a targeted macromolecule in treating brain tumors is penetration and activity in tumor across a heterogeneously intact blood-brain barrier. In this study, novel epidermal growth factor receptor (EGFR)-targeted antibody drug conjugates (ADCs) were constructed with a focus on cytotoxicity in glioblastoma (GBM) with an EGFR heterogeneous cell population. Initially, the cytotoxicity of free payloads with multiple mechanisms of cell kill was assessed. These included tesirine (TS, induces DNA cross-links), deruxtecan (Dxd) and SN38 (inhibit topoisomerase I), and an auristatin (MMAE) and maytansinoid (DM1) (inhibit microtubule polymerization). Across four GBM PDXs with varied EGFR expression (GBM6, 39 and 108- EGFRviii amplified; GBM10 - non-amplified), free TS and MMAE were consistently potent with EC50 values ranging from 0.50-45.9 pM and 7.9-229.1 pM, respectively. In contrast, the other free toxins were less potent; EC50 values: Dxd, 0.21-13.1 nM; SN38, 0.87-5.6 nM; DM1, 2.1-19.9 nM. Based on these results, the 40H3 EGFR-specific IgG was used to construct ADCs with TS and MMAE. 40H3-TS had a drug:antibody ratio (DAR) of 2.5 and was potently cytotoxic in GBM6, GBM39 and GBM108, while minimal cytotoxicity was observed in GBM10 or normal astrocyte SVG-A cells. 40H3-MMAE had a DAR of 3 and was similarly effective in GBM6 and GBM39 but less potent in GBM108, GBM10, and SVG-A. Bystander cytotoxicity was evaluated U87 cells expressing eGFP/fLuc2 (U87eGFP/fluc2) or EGFRviii (U87EGFRviii). Using live-cell imaging, U87EGFRviii cells treated with 40H3-TS or 40H3-MMAE had significantly reduced cell confluence relative to control. The same drug treatments in U87eGFP/fLuc2 cells had no effect on growth/confluence. However, in a 1:1 mixed culture, overall confluence was reduced from 89% in control to 35% and 32% after 40H3-TS (p<0.0001) or 40H3-MMAE (p<0.0001) treatment, respectively. This bystander killing of U87eGFP/fLuc2 cells was indicated by a reduction in confluence of green-fluorescent cells from 50% in control versus 29% with 40H3-TS (p<0.0001) and 26% with 40H3-MMAE (p<0.0001) treatment. In GBM39 orthotopic tumors, a single infusion of 10 or 20 µg 40H3-TS via convection enhanced delivery (CED) reduced the bioluminescence signal 7 days post treatment by ~10-fold (p=0.03) as compared to 40H3 control. However, 50% and 80% mortality was observed within a week of infusing 10 and 20 µg 40H3-TS, respectively. Neurotoxicity was associated with neuron loss in treated hemisphere as determined by NeuN staining. In summary, these data highlight the potential for novel EGFR-targeted ADCs to provide potent direct and bystander cytotoxicity to GBM cells. However, further selection and optimization of the conjugated toxins will be required to balance potency and bystander killing with toxicity for EGFR-targeted ADCs. Citation Format: Sonia Jain, Eric Chun Ho, Kendra A. Porath, Antonella Antignani, David J. FitzGerald, Jann N. Sarkaria. Bystander effects of EGFR-targeting 40H3 antibody-drug conjugates in glioblastoma with heterogeneous EGFR expression [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 556.