Fig. 4: Development and application of BC-mqmsPCR for BC diagnosis.
a Comparison between the BC-mqmsPCR and uniplex qMSP assays. The BC-mqmsPCR assay produced a ΔCT values that were higher than cg11754974, cg16652347, cg13828440, and cg18637238 by 2.22, 5.95, 0.71, and 1.76, respectively. FAM represents the DNA methylation signal, and VIC represents the internal reference control signal. The value of ΔRn represents the amount of degradation of the probe during the PCR process, which is the amount of PCR product. ΔRn=R + n - R-n, where R + n represents the fluorescence intensity measured at each point, and R-n represents the fluorescence baseline intensity. b Assessment of analytical sensitivity of BC-mqmsPCR assay. The BC-mqmsPCR assay detected PBMCs DNA signals with as little as 0.01% of 50 ng total PBMCs DNA diluted with pure water. c, d Methylation levels of PBMCs DNA as quantified by BC-mqmsPCR assay in the training set (130 BC patients and 115 normal controls) and validation set (76 BC patients and 55 normal controls). The y-axis represents methylation levels (ΔCT = CTreference – CTbiomarker), in which a higher value represents a higher methylation level. P values were determined by a two-sided Mann–Whitney U-test. ****p ≤ 0.0001. e, f The AUC of BC-mqmsPCR was 0.925 (0.89–0.96) for the training set (e) and 0.918 (0.86–0.98) for the validation set (f). g, h Confusion tables of the binary results of BC-mqmsPCR in the training set (g) and validation set (h). i, j Supervised hierarchical clustering of the differentially methylated BC-mqmsPCR signature between BC patients and normal controls in the training set (i, n = 245) and validation set (j, n = 131). k Distribution of BC-mqmsPCR value in BC (n = 206), LC (n = 42), GC (n = 41), CRC (n = 42) and normal controls (n = 170). The red dashed line represents the cut-off value (3.223). LC lung cancer, GC gastric cancer, CRC colorectal cancer. l The AUC for distinguishing BC from non-BC was 0.913. The boxes in c, d are bounded by the first and third quartile with a horizontal line at the median; minima is the smallest data greater than or equal to the first quartile – 1.5 × interquartile range (IQR); maxima is the largest data point less than or equal quartile + 1.5 × IQR. Source data are provided as a Source Data file.
