Abstract Background: Colorectal cancer (CRC) is a major cause of cancer mortality, yet early detection markedly improves outcomes. Current screening tools like colonoscopy and fecal immunochemical test (FIT) face limitations in accessibility, participation, and sensitivity, particularly for advanced adenomas (AA). Emerging evidence indicates that tumor-derived systemic stress can induce genomic and epigenetic abnormalities in bone marrow, thereby remotely reprogramming hematopoiesis. Consistent with this phenomenon, our previous work revealed that colorectal tumors remotely disrupt the genomic integrity of hematopoietic stem and progenitor cells, and that these alterations presist through erythroid differentiation to generate distinct DNA signatures in mature red blood cells (rbcDNA). Leveraging these rbcDNA signatures, we developed a CRC-rbcDNA based classifier for early detection of CRC and AA. Following multi-center validation, this study presents the first prospective clinical evaluation of the classifier and a head-to-head comparison with quantitative FIT (qFIT), aiming to assess its performance in detecting both early CRC and AA. Methods: We conduct a prospective cohort clinical study (NCT05875584) designed to validate the locked rbcDNA classifier and to benchmark its performance against qFIT. A total of 598 individuals were enrolled, of which 585 samples were available for analysis. These comprised 299 non-CRC controls (non-neoplastic findings, non-advanced adenomas, and limited non-CRC malignancies), 206 AAs (high-grade dysplasia, villous features, or lesions ≥10 mm), and 80 CRCs. All samples underwent rbcDNA isolation, purification, and low-coverage whole-genome sequencing to generate rbcDNA profiles. For each participant, the locked classifier generated an rbcDNA-based predictive score, and qFIT results were collected in parallel. Results: Using the predefined threshold, the rbcDNA assay reached 90% sensitivity for CRC, including over 90% of stage I-II tumors, and detected 60% of AA cases. Among 299 controls, specificity was 90% and remained consistent across clinical subgroups. Performance in the prospective cohort closely matched that of the multi-center validation set and remained consistent across demographic, clinical, and pathological characteristics. Compared with qFIT, the rbcDNA assay achieved similar CRC sensitivity (90% vs. 88%) but substantially improved AA sensitivity (60% vs. 18%). Overall, the rbcDNA assay showed substantially higher sensitivity than qFIT for early-stage CRC and AA, which typically present with low tumor burden, while maintaining similar specificity. Conclusions: This prospective clinical study demonstrated rbcDNA can be a highly sensitive and non-invasive method for early detection of colorectal neoplasia, supporting its promise for integration into clinical screening practice. Citation Format: Chengcheng Liu, Xingyun Yao, Haobo Sun, Yurong Jiao, Xiangxing Kong, Jie Jin, Kefeng Ding, Jun Li, Xiaofei Gao. Non-invasive detection of early colorectal neoplasia using red blood cell DNA profiling: a prospective clinical validation abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 7623.
Liu et al. (Fri,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: