Abstract C118: AI/ML-assisted detection of HMGA2 RNA isoforms in prostate cancer patient tissue

Abstract RNA In Situ Hybridization (RISH) is a powerful tool to analyze RNA expression spatially within cells and tissues. However, tools to quantify this technology is not readily available due to high cost of the software available to under-resourced communities. We therefore partnered with the Center for Equitable Artificial Intelligence and Machine Learning Systems (CEAMLS) to identify and quantify HMGA2, a biomarker associated with cancer metastasis and resistance to therapy. RISH was utilized to examine the RNA expression of HMGA2 full-length (wild-type) and HMGA2 truncated in prostate cancer (PCa) tissue from 81 patients. Artificial Intelligence and Machine Learning (AI/ML) was utilized to train a sample data set before analyzing the entire cohort. RISH demonstrated that HMGA2 wild-type isoform was more abundant and increased with Gleason grade in tissues from men of African descent compared to their European American and Asian counterparts. However, HMGA2 truncated isoform presented no clear pattern and was less abundant. Our research suggests that AI/ML may be feasible methods to analyze RISH data. Additionally, HMGA2 wild-type may be a biomarker associated with PCa disparities. This study underscores the necessity for collaboration between basic and clinical scientists in using AI/ML to better understand biomarkers in cancer progression and disparities. Acknowledgements: The authors acknowledge the use of core facilities supported by the National Institute on Minority Health and Health Disparities through grant number 5U54MD013376 and National Institute of General Medical Sciences through grant number 5UL1GM118973 Citation Format: Bor-Jang Hwang, Oluwatunmise Akinniyi, Sharon Harrison, Dennise Gibbs, Charles G. Waihenya, Andrew Gachii, Precious E. Dike, Camille Ragin, Fahmi Khalifi, Valerie Odero-Marah. AI/ML-assisted detection of HMGA2 RNA isoforms in prostate cancer patient tissue [abstract]. In: Proceedings of the 18th AACR Conference on the Science of Cancer Health Disparities; 2025 Sep 18-21; Baltimore, MD. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2025;34(9 Suppl):Abstract nr C118.