Table 3. Univariate and multivariate analysis for progression-free survival and overall survival from the second-line of chemotherapy.

	Progression-Free Survival						Overall Survival
	Univariate			Multivariate			Univariate			Multivariate
	HR	(95% CI)	P	HR	(95% CI)	P	HR	(95% CI)	P	HR	(95% CI)	P
Age (years)
≤ 60 vs > 60R	0.85	(0.57-1.27)	0.43				1.18	(0.76-1.82)	0.44
Gender
Male vs female R	0.65	(0.40-1.05)	0.08				0.62	(0.37-1.03)	0.06
ECOG performance status
0-1 vs ≥ 2 R	0.52	(0.32-0.86)	0.009	0.56	(0.34-0.91)	0.02	0.35	(0.21-0.59)	<.0001	0.34	(0.20-0.57)	<.0001
Primary tumor site
Junction vs stomach R	0.76	(0.50-1.13)	0.18				0.70	(0.45-1.09)	0.11
Histological type
Diffuse/mixed vs intestinal R	1	(0.53-1.90)	0.99				0.82	(0.41-1.63)	0.57
Tumor grade
Well/moderate vs poor R	0.69	(0.46-1.05)	0.09				0.57	(0.35-0.93)	0.06
Measurable disease
Yes vs No R	0.36	(0.19-0.67)	0.001	0.42	(0.21-0.85)	0.01	0.36	(0.19-0.67)	0.001	0.39	(0.19-0.79)	0.009
Number of metastatic sites
0-1 vs ≥ 2 R	0.63	(0.40-1.0)	0.049	0.70	(0.43-1.13)	0.14	0.61	(0.37-0.99)	0.048	0.59	(0.34-1.02)	0.06
Second-line chemotherapy
With vs without trastuzumab R	0.51	(0.33-0.79)	0.002	0.56	(0.35-0.89)	0.01	0.44	(0.27-0.73)	0.002	0.47	(0.28-0.79)	0.004

Abbreviations: R, reference; HR, hazard ratio; ECOG, Eastern Cooperative Oncology Group; PFS, progression-free survival; OS, overall survival.

Multivariate analysis was performed on variables potentially predictive of the risk of disease progression or death in univariate analysis (threshold, 5%).

For PFS and OS multivariate analysis, 7 patients were excluded because they had at least one missing data among the variables selected on univariate analysis; finally, 97 patients were selected for multivariate analysis.

A P value < 0.05 was considered statistically significant.