Transformer-based AI technology improves early ovarian cancer diagnosis using cfDNA methylation markers
Epithelial ovarian cancer (EOC) is the deadliest women’s cancer and has a poor prognosis. Early detection is the key for improving survival (a 5-year survival rate in stage I/II is over 70% compared to that of 25% in stage III/IV) and can be achieved through methylation markers from circulating cell...
Saved in:
Published in | Cell reports. Medicine Vol. 5; no. 8; p. 101666 |
---|---|
Main Authors | , , , , , , , , , , , , , , , , , , , , , , , |
Format | Journal Article |
Language | English |
Published |
United States
Elsevier Inc
20.08.2024
Elsevier |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | Epithelial ovarian cancer (EOC) is the deadliest women’s cancer and has a poor prognosis. Early detection is the key for improving survival (a 5-year survival rate in stage I/II is over 70% compared to that of 25% in stage III/IV) and can be achieved through methylation markers from circulating cell-free DNA (cfDNA) using a liquid biopsy. In this study, we first identify top 500 EOC markers differentiating EOC from healthy female controls from 3.3 million methylome-wide CpG sites and validated them in 1,800 independent cfDNA samples. We then utilize a pretrained AI transformer system called MethylBERT to develop an EOC diagnostic model which achieves 80% sensitivity and 95% specificity in early-stage EOC diagnosis. We next develop a simple digital droplet PCR (ddPCR) assay which archives good performance, facilitating early EOC detection.
[Display omitted]
•Transformer-based MethylBERT learned associations among CpG sites from 110,000 cases•EOC cfDNA methylation markers were screened from 3.3 million CpG sites in 3,000 samples•MethylBERT-based EOC diagnostic model outperforms a LASSO-based model•EOC diagnostic assay was developed by digital PCR
Li et al. utilized transformer-based AI technology to learn the knowledge of methylome-wide features among different CpG sites from 110,000 cancer samples and then applied the learned knowledge to analyze large numbers of cfDNA methylation markers in 754 EOC patients and 1,118 healthy females and developed an accurate EOC diagnostic model. |
---|---|
Bibliography: | ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 These authors contributed equally Lead contact |
ISSN: | 2666-3791 2666-3791 |
DOI: | 10.1016/j.xcrm.2024.101666 |