DCYTB is a predictor of outcome in breast cancer that functions via iron-independent mechanisms

Breast Cancer Research

Table 2 Univariate and multivariate regression analysis of combined cohorts 1 and 2

	Univariate			Multivariate^ac (LN status excluded)			Multivariate^bc (LN status included)
Covariates	Cox P value ^*	HR	95% CI	Cox P value ^*	HR	95% CI	Cox P value^*	HR	95% CI
DCYTB	1.11E-07	0.79	0.73–0.87	0.03	0.84	0.71–0.98	0.25	0.90	0.75–1.08
Size	0.41	1.00	0.99–1.01	0.02	1.01	1.00–1.02	0.05	1.01	0.99–1.02
Grade 2	1.20E-05	2.34	1.60–3.42	3.90E-03	2.07	1.26–3.38	0.01	1.97	1.17–3.34
Grade 3	2.53E-11	3.64	2.50–5.32	7.51E-04	2.53	1.48–4.35	0.01	2.29	1.27–4.15
ER status	4.14E-07	0.56	0.45–0.69	0.59	0.89	0.57–1.37	0.36	0.79	0.47–1.32
LN status	7.14E-05	1.53	1.24–1.89	-	-	-	0.01	1.55	1.09–2.21

LN lymph node, HR hazard ratio, CI confidence interval, DCYTB duodenal cytochrome b, ER estrogen receptor,
*Likelihood ratio test P value
^a612 patients had complete clinical annotation for size, grade, age and ER status
^b464 patients had complete clinical annotation for size, grade, ER and LN status
^cER status was not significant in the multivariate analysis, possibly due to collinearity with grade in this dataset or to criteria used in assembling cohort #2 (in particular, the requirement for a relatively high (>20%) event rate, which may have enriched for studies with fewer ER+/LumA tumors)

ISSN: 1465-542X