Abstract 7314: Technological and economic values of an AI-based pre-screening multi-cancer early detection (MCED) liquid-biopsy test as a companion system for a cancer screening solution

Abstract

Abstract Multi-cancer early detection (MCED) liquid biopsy screening tests provide great promise for both first-time cancer detection, and recurrence monitoring applications. The prominent feature of asymptomatic cancer screening solutions is their demand for ultra-high specificity (e.g., circa 99.99%), while maintaining a reasonable sensitivity. However, the challenges of screening are not limited to these technical specifications. Among other challenges, psychological and economical barriers-to-entry can be mentioned. In this work, we discuss the value of adding a high-sensitivity liquid-biopsy prescreening system to a screening solution. Such a tandem system can retain the high specificity offered by the screening solution, while relieving some of the economical concerns in cancer screening, by providing an opportunity to forgo screening in individuals with negative prescreening results. In order to serve in a general multi-cancer scheme, the prescreening solution must be MCED in nature. While our screening solution is primarily single-cell genomics/proteomics-based, we adopted a predominantly proteomics-based solution for prescreening. A proteomics solution has the potential to be notably more economical due to its simplicity/maturity. This is in contrast to the relatively high cost of the ′omics-based screening methods. Nevertheless, proteomics-based solutions, historically, have suffered from limited sensitivity and low specificity. In order to attain a high sensitivity while maintaining a workable specificity, we have used a complex Artificial Intelligence (AI) based system, comprising machine learning (ML) and Expert Systems (ES). Our MCED blood-based prescreening solution has been successfully tested on a series of cancers, including (among others) the cancers of breast, lung, prostate, ovary, uterus, and cervix, with cancer signal detection sensitivity of ≈92% for early stages (Stage I or II). While a tandem prescreening-screening system is designed primarily for testing asymptomatic individuals for early detection of cancer, a prescreening method has other potential uses. For instance, it can be used for: 1) minimal residual disease (MRD) detection after treatment; 2) recurrence monitoring after achieving remission. In particular, in cases where the original cancer is aggressive and has a high chance of recurrence, e.g., triple-negative breast cancer (TNBC), such a prescreening test can be repeated several times in the course of a year, providing the patient with measurements that are affordable and near real-time. In conclusion, the introduced proteomics/AI-based prescreening system could provide an economical, yet high-sensitivity solution for general adoption, acting primarily as a molecular risk assessment tool, to be performed prior to using a high-specificity screening modality. Citation Format: Bahram G. Kermani. Technological and economic values of an AI-based pre-screening multi-cancer early detection (MCED) liquid-biopsy test as a companion system for a cancer screening solution [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 7314.