Table 3. The observational studies on the disease risk of dietary isoflavones by using level in biological samples as biomarker.

Outcome	Subjects	Type of specimen	Biomarkers	Association1)	Study name2)/design	Ref.
CVD	303 adults	24 h urine	ISO	↔	CARDIAC/CS	Yamori et al. [44]
Ischemic stroke	1,422 case; 1,422 control	Spot urine	GEN, DAI, GLY, O-DMA, DHGE, DHDE, equol	↔	SWHS/NCC	Yu et al. [45]
Type 2 diabetes	1,111 case; 1,111 control	Spot urine	ISO, GEN, DAI, O-DMA, DHGE, DHDE	↓ (only DAI)	NHS/NCC	Ding et al. [46]
Prostate cancer	46 case; 54 control	Plasma	GEN	↓	CC	Wu et al. [47]
Liver cancer	18,628 (34% men)	Plasma	GEN, DAI, GLY, equol	↔	JPHC/NCC	Michikawa et al. [48]
	90 case; 175 control (69% men)

CVD, cardiovascular disease; ISO, total isoflavone; GEN, genistein; DAI, daidzein; GLY, glycitein; CS, cross-sectional; FFQ, food frequency questionnaire; Q, quartile or quintile; NCC, nested case-control; HR, hazard ratio; CI, confidence interval; O-DMA, O-desmethylangolensin; DHGE, dihydrogenistein; DHDE, dihydrodaidzein; WHO, World Health Organization; CARDIAC, Cardiovascular Disease and Alimentary Comparison; SWHS, The Shanghai Women's Health Study; NHS, Nurses' Health Study; JPHC, The Japan Public Health Center-based prospective.

¹⁾↑ positive association, ↓ inversely association, ↔ no association; ²⁾WHO-coordinated CARDIAC, SWHS, NHS, JPHC.