Abstract 2946: The prevalence of Globo H in different tumor types: Breast, pancreatic, lung, gastric, colorectal, liver, and esophageal cancers

Abstract

Abstract Background: Globo H (GH) is a hexassacharide glycosphingolipid overexpressed on tumor epithelial cells in several cancer types. Ongoing clinical trials of GH-targeting drugs include therapeutic vaccines, a monoclonal antibody, and an antibody drug conjugate. Consequently, the level of GH expression may provide a critical biomarker for appropriate patient selection. We present the prevalence of GH expression in various human cancers using a validated immunohistochemistry (IHC) assay. Methods: A validated IHC assay was used to analyze a total of 562 specimens across 7 cancer types, including breast (131), lung (77), colorectal (75), gastric (73), pancreatic (72), liver (70), and esophageal cancer (64) specimens. Anti-GH monoclonal antibody (VK9) was used for IHC staining. GH expression level was assessed by certified pathologists, and results are presented using an H-score system (0 to 300). The H-score was calculated as: H-score = (% of weak intensity x 1) + (% of moderate intensity x 2) + (% of strong intensity x 3). Results: Among the specimens tested, pancreatic cancer had the highest median H-score of 90.0 (range 0-300) with a 76.4% prevalence using cutoff of H-score ≥ 1, while liver cancer had the lowest median H-score of 0.0 (range 0 to 80) with a 17.1% prevalence. Using an H-score of 100 as a cutoff [approved in the Investigational Device Exemption (IDE) application as a pre-screening criterion of GH monoclonal antibody in the Phase 2 clinical trial OBI-888-001 (NCT03573544)], the GH prevalence of the pancreatic, esophageal, gastric, breast, lung, colorectal, and liver cancers are 50%, 17.2%, 24.7%, 13%, 10.4%, 16%, and 0%, respectively. In addition, subtypes of each cancer differ in expression profile. While immune cells are typically GH negative, some GH-positive immune cells were observed in the intra- or peri-tumor region, suggesting the shedding of GH to tumor-surrounding or infiltrated immune cells. Conclusions: Overexpression of GH in human pancreatic, gastric, lung, colorectal, esophageal, and breast tumors renders GH a potential therapeutic target for these cancers. Furthermore, the presence of GH-positive immune cells in the intra- or peri-tumor region lends support to the proposed mechanism of cancerous cells shedding GH-ceramide to suppress normal immune functions. The heterogeneous expression of GH among different molecular subtypes may provide a biomarker in the selection of patients for GH-directed therapies. The GH IHC assay has now been validated for use and the IDE approved by the FDA for patient screening. Citation Format: I-Ju Chen, Ming-Chen Yang, Yu-Jung Chen. The prevalence of Globo H in different tumor types: Breast, pancreatic, lung, gastric, colorectal, liver, and esophageal cancers [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 2946.