Plasma cell-free DNA (cfDNA) as a predictive and prognostic marker in patients with metastatic breast cancer

Breast Cancer Research

Table 4 Logistic regression analysis between stratified clinical parameter (response to treatment) and stratified or normalized continuous biomarkers, showing the p value, odds ratio and the 95% confidence interval for this odds ratio. Univariate and multivariate analysis

Categorical variable to analyse	Variable	p value	Odds ratio	95% CI for OR		Type of analysis	Type of variable
Categorical variable to analyse	Variable	p value	Odds ratio	Lower	Upper	Type of analysis	Type of variable
Response (RECIST 2 categories) vs	CTC threshold	0.030	5.133	1.174	22.450	Univariate	Stratified biomarker
	cfDNA threshold	*Cannot be quantified
	CA15-3 threshold	0.165	0.528	0.214	1.302
	AP threshold	0.056	4.222	0.961	18.542
	CTC threshold	0.265	2.502	0.499	12.534	**Multivariate
	cfDNA threshold	*Cannot be quantified
	CA15-3 threshold	0.012	0.297	0.116	0.766
	AP threshold	0.171	3.070	0.616	15.298
Response (RECIST 2 categories) vs	CTC count	0.405	27.561	0.011	67106.275	Univariate	Normalized continuous biomarker
	cfDNA yield	0.035	15.917	1.223	207.106
	CA15-3 levels	0.722	0.938	0.661	1.331
	AP levels	0.139	1.611	0.856	3.031
	CTC count	0.809	1.445	0.073	28.702	**Multivariate
	cfDNA yield	0.055	12.481	0.946	164.576
	CA15-3 levels	0.290	0.814	0.556	1.192
	AP levels	0.212	1.575	0.771	3.215

*Cannot be quantified as all patients in the group responding to treatment have low levels of cfDNA. No patients with high levels of cfDNA responding to treatment
**Multivariate analysis represent the combination and analysis of all biomarkers (CTC count, cfDNA yield, CA15-3 levels and AP levels) together, as a joint analysis, to explore the possible interactions between them and how this interactions affects their effect on the clinical variables under study

ISSN: 1465-542X