Table 2 Associations between heritable DNA methylation marks (associated with breast cancer in multiple-case families) and risk of breast cancer in the general population (Melbourne Collaborative Cohort Study)

Site	Chr.	Position	Gene name	OR^a	95% CI	p
cg06536614	5	135416381	VTRNA2-1 (MIR886)	0.95	0.83–1.10	0.497
cg10306192	11	102576374	MMP27	1.09	0.94–1.27	0.235
cg18110333	6	292329	DUSP22	0.96	0.83–1.11	0.588
cg00124993	5	135416412	VTRNA2-1 (MIR886)	0.97	0.84–1.12	0.667
cg26328633	5	135416394	VTRNA2-1 (MIR886)	0.98	0.85–1.13	0.761
cg25340688	5	135416398	VTRNA2-1 (MIR886)	0.95	0.82–1.09	0.441
cg18514595	22	49579968	unannotated	1.14	0.99–1.31	0.077
cg26896946	5	135416405	VTRNA2-1 (MIR886)	0.94	0.82–1.09	0.426
cg11035303	3	43465503	ANO10	1.01	0.88–1.16	0.894
cg23012654	14	97493395	unannotated	0.95	0.83–1.10	0.503
cg26773954	13	111969980	unannotated	1.02	0.88–1.17	0.813
cg22901919	4	141317067	CLGN	0.91	0.78–1.06	0.224
cg04417708	17	4043867	ZZEF1	1.00	0.87–1.15	0.989
cg18584561	2	11682017	*GREB1*	1.18 ^b	1.03–1.36	0.015
cg11608150	5	135415948	unannotated	0.93	0.80–1.07	0.311
cg01741999	2	219137824	*PNKD*	1.26	1.03–1.54	0.027
cg01074083	16	17516862	XYLT1	0.98	0.84–1.13	0.749
cg02096220	4	129212177	unannotated	1.02	0.89–1.18	0.743
cg03916490	7	1080558	*C7orf50*	0.83	0.72–0.96	0.012
cg27639199	15	81666528	*TMC3*	1.19	1.03–1.36	0.018
cg25188166	3	119420208	unannotated	0.96	0.83–1.10	0.551
cg05865327	14	102274741	PPP2R5C	1.04	0.90–1.20	0.589
cg23947138	13	114782778	RASA3	0.89	0.78–1.02	0.091
cg05187003	21	34641507	IL10RB	1.00	0.86–1.15	0.950

^a Odds ratio from conditional logistic regression of the risk of breast cancer on M-values (per 1 s.d.), adjusting for body mass index, tobacco smoking, alcohol drinking, time between blood collection and cancer diagnosis, and sample type (dried blood spots, peripheral blood mononuclear cells, and buffy coats). Cases and controls were individually matched on year of birth, year of blood draw, country of birth, and sample type for the vast majority of them (97%))
^b Results are presented here using the methylation values as continuous, although the association was not linear. A better model fit was obtained by categorising into hypo/hemi/hypermethylated groups (i.e., peaks). Bold text indicates statistically significant associations

Search