Circulating cell-free DNA-based methylation patterns for breast cancer diagnosis
Xianyu Zhang,Dezhi Zhao,Yanling Yin,Ting Yang,Zilong You,Dalin Li,Yanbo Chen,Yongdong Jiang,Shouping Xu,Jingshu Geng,Yashuang Zhao,Jun Wang,Hui Li,Jinsheng Tao,Shan Lei,Zeyu Jiang,Zhiwei Chen,Shihui Yu,Jian-Bing Fan,Da Pang
DOI: https://doi.org/10.1038/s41523-021-00316-7
2021-08-16
npj Breast Cancer
Abstract:Abstract Mammography is used to detect breast cancer (BC), but its sensitivity is limited, especially for dense breasts. Circulating cell-free DNA (cfDNA) methylation tests is expected to compensate for the deficiency of mammography. We derived a specific panel of markers based on computational analysis of the DNA methylation profiles from The Cancer Genome Atlas (TCGA). Through training ( n = 160) and validation set ( n = 69), we developed a diagnostic prediction model with 26 markers, which yielded a sensitivity of 89.37% and a specificity of 100% for differentiating malignant disease from normal lesions [AUROC = 0.9816 (95% CI: 96.09-100%), and AUPRC = 0.9704 (95% CI: 94.54–99.46%)]. A simplified 4-marker model including cg23035715 , cg16304215 , cg20072171 , and cg21501525 had a similar diagnostic power [AUROC = 0.9796 (95% CI: 95.56–100%), and AUPRC = 0.9220 (95% CI: 91.02–94.37%)]. We found that a single cfDNA methylation marker, cg23035715 , has a high diagnostic power [AUROC = 0.9395 (95% CI: 89.72–99.27%), and AUPRC = 0.9111 (95% CI: 88.45–93.76%)], with a sensitivity of 84.90% and a specificity of 93.88%. In an independent testing dataset ( n = 104), the obtained diagnostic prediction model discriminated BC patients from normal controls with high accuracy [AUROC = 0.9449 (95% CI: 90.07–98.91%), and AUPRC = 0.8640 (95% CI: 82.82–89.98%)]. We compared the diagnostic power of cfDNA methylation and mammography. Our model yielded a sensitivity of 94.79% (95% CI: 78.72–97.87%) and a specificity of 98.70% (95% CI: 86.36–100%) for differentiating malignant disease from normal lesions [AUROC = 0.9815 (95% CI: 96.75–99.55%), and AUPRC = 0.9800 (95% CI: 96.6–99.4%)], with better diagnostic power and had better diagnostic power than that of using mammography [AUROC = 0.9315 (95% CI: 89.95–96.34%), and AUPRC = 0.9490 (95% CI: 91.7–98.1%)]. In addition, hypermethylation profiling provided insights into lymph node metastasis stratifications ( p < 0.05). In conclusion, we developed and tested a cfDNA methylation model for BC diagnosis with better performance than mammography.
oncology