Prediagnostic Circulating Anti-Müllerian Hormone Concentrations Are Not Associated with Prostate Cancer Risk

Martha M. Sklavos; Cindy Ke Zhou; Ligia A. Pinto; Michael B. Cook

doi:10.1158/1055-9965.EPI-14-0803

DOI: 10.1158/1055-9965.EPI-14-0803 Published November 2014

Article Figures & Data

Tables

Additional Files

Table 1.

Baseline characteristics by case status

		Cases (n = 998)
Baseline characteristics	Controls (n = 999)	Total	P	Aggressive^a	P	Nonaggressive^b	P
AMH quartiles (ng/mL)			0.599		0.283		0.525
Q1 (0.04–2.64)	248 (24.8%)	222 (22.2%)		78 (21.3%)		134 (22.0%)
Q2 (2.65–4.21)	249 (24.9%)	260 (26.1%)		107 (29.2%)		150 (24.6%)
Q3 (4.22–6.74)	252 (25.2%)	260 (26.1%)		97 (26.4%)		159 (26.1%)
Q4 (6.75–25.0)	250 (25.0%)	256 (25.7%)		85 (23.2%)		167 (27.4%)
Age at randomization, y			1.000		0.180		0.462
≤59	151 (15.1%)	151 (15.1%)		40 (10.9%)		109 (17.9%)
60–64	345 (34.5%)	346 (34.7%)		127 (34.6%)		211 (34.6%)
65–69	327 (32.7%)	325 (32.6%)		124 (33.8%)		193 (31.6%)
≥70	176 (17.6%)	176 (17.6%)		76 (20.7%)		97 (15.9%)
Marital status			0.098		0.562		0.052
Married or living as married	861 (86.2%)	882 (88.4%)		319 (86.9%)		545 (89.3%)
Single^c	138 (13.8%)	113 (11.3%)		46 (12.5%)		64 (10.5%)
First-degree relative(s) with prostate cancer			<0.001		<0.001		<0.001
No	924 (92.5%)	864 (86.6%)		320 (87.2%)		524 (85.9%)
Yes	52 (5.2%)	114 (11.4%)		39 (10.6%)		74 (12.1%)
BMI at baseline (kg/m²)			0.411		0.538		0.399
<25	275 (27.5%)	265 (26.6%)		104 (28.3%)		152 (24.9%)
25–30	505 (50.6%)	530 (53.1%)		192 (52.3%)		327 (53.6%)
≥30	210 (21.0%)	190 (19.0%)		67 (18.3%)		122 (20.0%)
History of diabetes			0.015		0.086		0.054
No	909 (91.0%)	937 (93.9%)		345 (94.0%)		571 (93.6%)
Yes	85 (8.5%)	57 (5.7%)		21 (5.7%)		36 (5.9%)
Cigarette smoking status			0.005		0.001		0.124
Never	394 (39.4%)	453 (45.4%)		181 (49.3%)		265 (43.4%)
Current	96 (9.6%)	66 (6.6%)		21 (5.7%)		44 (7.2%)
Former	509 (51.0%)	478 (47.9%)		164 (44.7%)		301 (49.3%)
Food energy from diet (kcal/d)			0.429		0.604		0.487
Q1 (≤1707)	243 (24.3%)	217 (21.7%)		76 (20.7%)		135 (22.1%)
Q2 (1718–2212)	244 (24.4%)	268 (26.9%)		97 (26.4%)		168 (27.5%)
Q3 (2213–2804)	243 (24.3%)	235 (23.5%)		89 (24.3%)		142 (23.3%)
Q4 (≥2805)	244 (24.4%)	249 (24.9%)		88 (24.0%)		153 (25.1%)
Enlarged prostate			0.044		0.677		0.020
No	738 (73.9%)	696 (69.7%)		267 (72.8%)		417 (68.4%)
Yes	261 (26.1%)	301 (30.2%)		100 (27.2%)		192 (31.5%)
Biopsy of prostate			<0.001		0.008		<0.001
No	922 (92.3%)	864 (86.6%)		318 (86.6%)		529 (86.7%)
Yes	56 (5.6%)	105 (10.5%)		35 (9.5%)		67 (11.0%)

NOTE: Column percentages may not sum to 100% due to missings. P values were estimated by χ² tests using nonmissing values. There were 21 prostate cancer cases that could not be classified to a subtype due to missing tumor characteristics.

↵^aAggressive prostate cancer: biopsy Gleason ≥ 7, or clinical stage ≥III, or fatal.
↵^bNonaggressive prostate cancer: biopsy Gleason ≤ 6, and clinical stage I or II.
↵^cSingle: widowed/divorced/separated/never married.

Table 2.

Age-adjusted ORs and 95% CIs for AMH quartiles and prostate cancer risk

		Cases
AMH quartiles (ng/mL)	Controls	Total		Aggressive^a		Nonaggressive^b
Q1 (0.04–2.64)	248	222	1.00 (reference)	78	1.00 (reference)	134	1.00 (reference)
Q2 (2.65–4.21)	249	260	1.17 (0.91–1.50)	107	1.39 (0.99–1.95)	150	1.11 (0.83–1.48)
Q3 (4.22–6.74)	252	260	1.16 (0.90–1.49)	97	1.26 (0.89–1.78)	159	1.15 (0.86–1.54)
Q4 (6.75–25.0)	250	256	1.15 (0.89–1.48)	85	1.14 (0.80–1.63)	167	1.22 (0.91–1.63)
P_trend			0.134		0.512		0.074

NOTE: There were 21 prostate cancer cases that could not be classified to a subtype due to missing tumor characteristics.

↵^aAggressive prostate cancer: biopsy Gleason ≥ 7, or clinical stage ≥III, or fatal.
↵^bNonaggressive prostate cancer: biopsy Gleason ≤ 6, and clinical stage I or II.

Table 3.

Age-adjusted ORs and 95% CIs for AMH quartiles and prostate cancer risk restricted to the subpopulation with PSA < 4.0 ng/mL at baseline

		Cases
AMH quartiles (ng/mL)	Controls	Total		Aggressive^a		Nonaggressive^b
Q1 (0.04–2.64)	232	142	1.00 (reference)	56	1.00 (reference)	79	1.00 (reference)
Q2 (2.65–4.21)	233	165	1.15 (0.86–1.54)	75	1.35 (0.91–1.99)	89	1.10 (0.77–1.57)
Q3 (4.22–6.74)	241	163	1.10 (0.82–1.46)	67	1.17 (0.78–1.74)	95	1.13 (0.79–1.60)
Q4 (6.75–25.0)	242	191	1.27 (0.96–1.69)	67	1.19 (0.79–1.77)	121	1.42 (1.01–1.99)
P_trend			0.135		0.713		0.035

NOTE: There were 12 prostate cancer cases that could not be classified to a subtype due to missing tumor characteristics.

↵^aAggressive prostate cancer: biopsy Gleason ≥ 7, or clinical stage ≥III, or fatal.
↵^bNonaggressive prostate cancer: biopsy Gleason ≤ 6, and clinical stage = I or II.

Additional Files

Tables

Supplemental Tables

Files in this Data Supplement:

Supplemental Tables - Supplemental Tables. Supplemental Table 1. Baseline characteristics by AMH quartiles. Supplemental Table 2. Baseline characteristics by AMH quartiles restricted to controls only. Supplemental Table 3. Spearmen correlations between circulating AMH levels and androgens in overall and control men with androgen measurements. Supplemental Table 4. Sensitivity analysis: Age-adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for AMH quartiles and prostate cancer risk sequentially excluding prostate cancer cases who were diagnosed within two, three and then five years after sera collection. Supplemental Table 5: Sensitivity analyses: Age-adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for AMH quartiles and prostate cancer risk. Supplemental Table 6. Sensitivity analysis: Age-adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for AMH quartiles and prostate cancer risk restricted to subpopulation with PSA at T0 <4.0 ng/ml at baseline.
Supplemental Figures - Supplemental Figures. Supplemental Figure 1. AMH distribution for cases and controls. Supplemental Figure 2. AMH levels in all participants stratified by 5 year age groups.