Abstract 2888: Utilizing PDX models to better understand factors that predict response to SGN-CD228A, an antibody drug-conjugate (ADC) for solid tumors

Patient-derived xenograft (PDX) models are widely recognized as a powerful preclinical tool. In contrast to cell line-derived xenograft (CDX) models, patient tumors are excised and implanted directly into mice and as such, retain tumor characteristics that are often lost in cell lines. Multiple studies have shown that PDX models have similar histology, genetic mutations, and gene expression patterns to actual patient samples. Additionally, they reflect diversity of both disease stage and treatment history, enabling an assessment of how these factors may impact therapeutic response to a given treatment. SGN-CD228A, which targets melanotransferrin (CD228, p97, MFI2, MELTF), recently entered Phase I clinical trials. CD228 has broad expression in solid tumors and we are testing SGN-CD228A in a basket trial which includes melanoma, mesothelioma, Her2- breast cancer, pancreatic cancer, colorectal cancer, and non-small cell lung cancer (NSCLC). Within these six indications, CD228 is detectable by an IHC assay in 50-80% of patients, depending on the tumor type. In order to better understand the relationship between CD228 expression and SGN-CD228A anti-tumor activity, we have tested 64 PDX models spanning triple-negative breast cancer (TNBC), mesothelioma, and NSCLC. A single 3 mg/kg dose of SGN-CD228A caused measurable tumor growth inhibition (%TGI) in 89% of all models tested, 29% of which had %TGI values of >100%. Similarly, when using the best tumor volume response as a metric, treatment with SGN-CD228A resulted in >30% tumor volume reduction in 36% of the models, with several models achieving a complete response. Of the three tumor types tested, TNBC appeared to be the most sensitive, with 59% of the models achieving >30% tumor volume reduction. In this study, we included models that have a wide range of CD228 expression as assessed by RNA seq, IHC, mass spectrometry, and western blotting. We are attempting to establish how expression of both CD228 RNA and protein relate to response, as well as define a minimum level of CD228 expression required for anti-tumor activity. Additionally, we assessed factors in the tumor microenvironment (TME), such as the vasculature and immune cell infiltrate, to determine how these variables might influence the anti-tumor activity of SGN-CD228A. Finally, we are utilizing RNAseq data to define a gene signature that may predict response to SGN-CD228A. In summary, we have found that SGN-CD228A is active in a diversity of solid tumor PDX models and biomarker data collected in these studies may guide interpretation of clinical data. Citation Format: Sharsti Sandall, Rebecca Mazahreh, Marsha Mason, John Gosink, Christopher Hale, Sean Allred, Rachael Hu, Abbie Wong, David Ortiz, Robert Lawrence, Kelly Hensley, Christine O9Day, Phillip Garfin, Scott Peterson, Timothy Lewis. Utilizing PDX models to better understand factors that predict response to SGN-CD228A, an antibody drug-conjugate (ADC) for solid tumors [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 2888.