Novel urine-based DNA methylation biomarkers for urothelial bladder carcinoma detection in patients with hematuria

Figures & data

Figure 1. Study flowchart.

Table 1. Patients’ demographics.

Median age in years (range)	58 (34-99)
Sex n (%)
• Male	201 (81.7)
• Female	45 (18.3)
Hematuria type n (%)
•Macroscopic	204 (82.9)
•Microscopic	42 (17.1)
Diabetes mellitus n (%)
•No	184 (74.8)
•Yes	62 (25.2)
Hypertension n (%)
•No	174 (70.7)
•Yes	72 (29.3)
Concomitant dysuria/suprapubic pain/urgency n (%)
•No	184 (74.8)
•Yes	62 (25.2)
History of urological problems n (%)
•No	194 (78.9)
•Yes	52 (21.1)
○ Urolithiasis	16 (6.5)
○ Bladder outlet obstruction	16 (6.5)
○ Urinary tract infections	20 (8.1)
Concomitant indwelling bladder foreign body n (%)
•No	238 (96.7)
•Yes	8 (3.3)
○ Catheter	3 (1.3)
○ Bladder stone	5 (2)
Smoking history n (%)
•Never	90 (36.6)
•Former	75 (30.5)
•Current	81(32.9)
Cytology result n (%)
•Negative	168 (68.3)
•Positive	78 (31.7)
Cystoscopy findings n (%)
•Free	100 (40.7)
•bladder lesion	146 (59.3)
Urine methylation assay target n (%)
•GATA4
Methylated	127 (51.6)
Unmethylated	119 (48.4)
•P16
Methylated	52 (21.1)
Unmethylated	194 (78.9)
•P14
Methylated	117 (47.6)
Unmethylated	129 (52.4)
•APC
Methylated	106 (43.1)
Unmethylated	140 (56.9)
•CDH1
Methylated	90 (36.6)
Unmethylated	156 (63.4)
•CD99
Methylated	71 (28.9)
Unmethylated	175 (71.1)
Biopsy findings in positive cystoscopy (Urothelial carcinoma) n (%)
Stage (TNM)
○ Ta, Tis, T1	31 (12.6)
○ ≥ T2	115 (46.7)
Grade (WHO/ISUP 2004)
○ Low grade	7 (2.8)
○ High grade	139 (56.5)

Table 2. Cross tables of different urine assay methylation targets and urine cytology in comparison to cystoscopy/biopsy results in study participants.

	Urothelial bladder carcinoma by cystoscopy and biopsy n %		Total	Diagnostic characteristics % (95%CI)
	Positive	Negative
GATA4 Methylated Unmethylated	122 (83.6%) 5 (5%)	24 (16.4%) 95 (95%)	127 119	SN 83.6% (95%CI: 79–87) SP 95% (95%CI: 92–98) PPV 96.1% (95%CI: 92–99) NPV 79.8% (95%CI: 75–84)
Total Count	146	100	246
P16 Methylated Unmethylated	52 (35.6%) 94 (64.4%)	0 100 (100%)	52 194	SN 35.6% (95%CI: 31–39) SP 100% PPV 100% NPV 51.5% (95%CI: 46–55)
Total Count	146	100	246
P14 Methylated Unmethylated	113 (77.4%) 96 (96%)	4 (4%) 96 (96%)	117 129	SN 77.4% (95%CI: 72–83) SP 96% (95%CI: 92–99) PPV 96.6% (95%CI: 93–99) NPV 74.4% (95%CI: 71–78)
Total Count	146	100	246
APC Methylated Unmethylated	106 (72.6%) 40 (27.4%)	0 100 (100%)	106 140	SN 72.6% (95%CI: 67–79) SP 100% PPV 100% NPV 71.4% (95%CI: 66–75)
Total Count	146	100	246
CDH1 Methylated Unmethylated	90 (61.6%) 56 (38.4%)	0 100 (100%)	90 156	SN 61.6% (95%CI: 57–66) SP 100% PPV 100% NPV 64.1% (95%CI: 60–68)
Total Count	146	100	246
CD99 Methylated Unmethylated	68 (46.6%) 78 (53.4%)	3 (3%) 97 (97%)	71 175	SN 46.6% (95%CI: 42–50) SP 97% (95%CI: 93–99) PPV 95.8% (95%CI: 91–99) NPV 55.4% (95%CI: 51–59)
Total Count	146	100	246
Urine cytology Positive Count Negative Count	57 (39%) 89 (61%)	21 (21%) 79 (79%)	78 168	SN 39% (95%CI: 36–42) SP 79% (95%CI: 76–84) PPV 84.5% (95%CI: 81–88) NPV 61% (95%CI: 57–65)
Total Count	146	100	246

Note: SN: Sensitivity, SP: specificity, PPV: Positive predictive value, NPV: Negative predictive value.

Figure 2. Receiver operator characteristic (ROC) curve for urine-based methylation biomarkers and urine cytology for urothelial bladder carcinoma detection in the study cohort.

Table 3. Univariate and multivariate analyses for predictors of positive cystoscopy/biopsy for urothelial bladder carcinoma (UBC).

Patient and tumor characteristics	UBC by cystoscopy and biopsy NoYes N (%) N (%)	Univariate analysis HR (95%CI) p value	Multivariate analysis HR (95%CI) p value
Median age in years (Range)	56 (36–75) 60 (34–99)	1.01 (0.9–1.03) 0.47
Sex n (%) Male Female	80 (39.8) 121 (60.2) 20 (44.4) 25 (55.6)	1.2 (0.9–1.5) 0.56
Hematuria n (%) Macroscopic Microscopic	82 (40.2) 122 (59.8) 18 (42.9) 24 (57.1)	1.4 (0.95–1.85) 0.75
Concomitant dysuria/suprapubic pain/urgency n (%) No Yes	73 (39.7) 111 (60.3) 27 (43.5) 35 (56.5)	1.2 (0.89–1.5) 0.59
History of urological problems n (%) No Yes	81 (41.8) 113 (58.2) 19 (36.5) 33(63.5)	1.12 (0.8–1.4) 0.5
Indwelling bladder foreign body n (%) No Yes	99 (41.6) 139 (58.4) 1 (12.5) 7 (87.5)	1.15 (0.88–1.34) 0.1
Smoking n (%) Never Former Current	43 (47.8) 47 (52.2) 25 (33.3) 50 (66.7) 32 (39.5) 49 (60.5)	0.92 (0.87–1.23) 0.17
Cytology result n (%) Negative Positive	83 (61) 53 (39) 17 (15.5) 93 (84.5)	1.17 (0.74–0.1.64) 0.07
Urine methylation assay target n (%) GATA4 Methylated Unmethylated P16 Methylated Unmethylated P14 Methylated Unmethylated APC Methylated Unmethylated CDH1 Methylated Unmethylated CD99 Methylated Unmethylated	5 (3.9) 122 (96.1) 95 (79.8) 24 (20.2) 0 52 (100) 100 (51.5) 94 (48.5) 4 (3.4) 113 (96.6) 96 (74.4) 33 (25.6) 0 106 (100) 100 (71.4) 40 (28.6) 0 90 (100) 100 (64.1) 56 (35.9) 3 (4.2) 68 (95.8) 97(55.4) 78 (44.6)	2.58 (1.94–2.94) <0.001 3.42 (2.9–3.8) <0.001 2.72 (2.34–2.98) <0.001 4.9 (4.3–5.42) <0.001 3.35 (2.97–3.86) <0.001 2.67 (2.32–2.91) <0.001	4.2 (2.78–8.24) 0.001 2.1 (1.72–3.12) 0.003 1.4 (1.1–1.8) 0.007 2.1 (1.8–2.5) 0.006 2.32 (1.92–2.73) 0.002 2.12 (1.83–2.53) 0.003

Figure 3. Boxplots of urine-based DNA methylation biomarkers GATA4, P16, P14, APC, CDH1 and CD99 for detection of different urothelial bladder carcinoma stages (3A) and grade (3B).