Abstract Background: Delayed diagnosis of cancer lead to advanced-stage presentation, contributing to increased patient anxiety, healthcare costs, and mortality. Early detection enables timely intervention, improving outcomes. OncoSeek is an AI-powered multi-cancer early detection (MCED) test with a reagent cost of 25. It integrates seven protein tumor markers (PTMs) and clinical data and has previously demonstrated 58. 4% sensitivity, 92. 0% specificity, and 70. 6% tissue-of-origin accuracy in studies involving over 15, 000 individuals. This study investigates its novel application in aiding cancer diagnosis among outpatients presenting with tissue masses. Methods: We prospectively enrolled 732 outpatients with tissue masses referred for cancer diagnosis and treatment, and recruited 355 non-cancer individuals as controls. Plasma/serum PTM levels were measured, and probability of cancer (POC) indexes were calculated using a machine learning algorithm to distinguish malignancies from benign tumors. To enhance sensitivity and reduce false negatives in this symptomatic cohort, a lower cutoff was applied, targeting 80% specificity. Results: Among patients with tissue mass, OncoSeek correctly identified 445 of 682 confirmed cancers, with a cancer detection rate of 65. 2%, comparable to the 66. 3% sensitivity reported in the SYMPLIFY study, a prospective study in symptomatic patients referred from primary care. The assay detected 15 cancer types, including bile duct, breast, cervix, colorectum, endometrium, esophagus, gallbladder, head and neck, liver, lung, lymphoma, ovary, pancreas, prostate, and stomach. Sensitivity increased with cancer stage: 40. 4% in stage I, 61. 2% in stage II, 72. 7% in stage III, and 86. 0% in stage IV. Notably, POC index levels were significantly higher in cancer patients compared to those with benign tumor (n=50; p 0. 001), while no significant difference was observed between benign tumor patients and non-cancer individuals (p 0. 05), supporting the test’s discriminatory capability. Conclusion: OncoSeek demonstrated strong clinical utility in evaluating patients with tissue masses, achieving 65. 2% sensitivity and robust discrimination between malignant and benign lesions. This supports its potential role in expediting cancer diagnosis and mitigating diagnostic delays. Citation Format: Mao Mao, Shen Yong, Chang yin. yin, Wu Wei, Li Shi. yong, Xing ping. Pingping, Ding chen. yu, Zhu dan. dan, Xu qing. xia. Low-cost blood test to assist cancer diagnosis in patients with tissue masses 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 C126.
Mao et al. (2025) studied this question.