. 2020 Feb 19;10:2976. doi: 10.1038/s41598-020-59923-2

Table 3.

Univariable analysis of variables to predict systemic recurrence in TNBC.

Variables		Hazard ratio (95% CI)	P value
Age		0.978 (0.94, 1.017)	0.267
Neoadjuvant chemotherapy	Yes	1.656 (0.665, 4.121)	0.278
Neoadjuvant chemotherapy	No
Adjuvant chemotherapy	Yes	0.617 (0.248, 1.53)	0.297
Adjuvant chemotherapy	No
Radiotherapy	Yes	0.824 (0.239, 2.84)	0.76
Radiotherapy	No
^*pCR	Yes	0.048 (0.003, 0.926)	<0.001
	Not applicable	0.244 (0.097, 0.613)	0.002
	No
Pathologic invasive cancer size		1.049 (1.021, 1.078)	<0.001
^*Nuclear grade	3	0.934 (0.337, 2.59)	0.892
	Not available	0.288 (0.014, 5.843)	0.317
	2
^*Histologic grade	3	0.113 (0.018, 0.694)	0.055
	2	0.188 (0.029, 1.221)	0.124
	Not available	0.042 (0.001, 1.244)	0.036
	1
Lymphovascular invasion	Yes	13.2 (5.155, 33.804)	<0.001
Lymphovascular invasion	No
Surgery type	Brest conserving surgery	0.167 (0.063, 0.443)	<0.001
Surgery type	Total mastectomy
Number of metastatic axillary lymph nodes after surgery		1.139 (1.084, 1.196)	<0.001
Rad score		23.322 (10.346, 52.575)	<0.001

TNBC = triple-negative breast cancer, CI = confidence interval, *Unbiased hazard ratio was estimated with Firth’s penalized maximum likelihood estimation method because the data was completely separable.