Table 3.
Univariate analysis | Multivariate analysis | ||||
---|---|---|---|---|---|
Variable | 5-year GCCS | Log rank χ2 test | P | HR(95%CI) | P |
Sex | 3.059 | 0.08 | |||
Male | 30.4% | ||||
Female | 28.5% | ||||
Age | 23.923 | <0.001 | <0.001 | ||
<60 | 32.0% | Reference | |||
≥60 | 28.4% | 1.295(1.221-1.373) | |||
Race | 51.463 | <0.001 | <0.001 | ||
White | 27.6% | Reference | |||
African American | 28.2% | 1.042(0.960-1.131) | |||
Others | 37.5% | 0.809(0.751-0.871) | |||
Location | 72.114 | <0.001 | 0.0251 | ||
Cardia, NOS | 26.1% | Reference | |||
Fundus of stomach | 31.4% | 0.827(0.700-0.978) | |||
Body of stomach | 30.5% | 0.769(0.687-0.861) | |||
Gastric antrum | 32.6% | 0.778(0.718-0.842) | |||
Pylorus | 29.7% | 0.857(0.748-0.982) | |||
Lesser curvature of stomach NOS | 37.1% | 0.693(0.625-0.768) | |||
Greater curvature of stomach NOS | 31.2% | 0.812(0.707-0.931) | |||
Overlapping lesion of stomach | 23.3% | 0.860(0.774-0.955) | |||
Stomach, NOS | 26.6% | 0.891(0.801-0.992) | |||
Pathological grading | 98.930 | <0.001 | <0.001 | ||
High/ Moderate | 37.9% | Reference | |||
Poor/ Anaplastic | 27.4% | 1.156(1.077-1.241) | |||
Unknown | 21.7% | 1.186(1.021-1.377) | |||
Histotype | 45.168 | <0.001 | 0.060 | ||
Adenocarcinoma | 32.1% | Reference | |||
Mucinous/signet ring cell | 24.0% | 1.061(0.998-1.129) | |||
pT Stage | 731.610 | <0.001 | <0.001 | ||
T1 | 60.2% | Reference | |||
T2 | 51.1% | 1.156(0.969-1.380) | |||
T3 | 31.7% | 1.835(1.584-2.126) | |||
T4 | 17.9% | 2.604(2.246-3.019) | |||
Tx | 4.2% | 4.453(3.216-6.164) | |||
pN Stage | 620.680 | <0.001 | <0.001 | ||
N1 | 42.9% | Reference | |||
N2 | 32.2% | 1.371(1.270-1.480) | |||
N3a | 19.7% | 2.186(2.010-2.378) | |||
N3b | 9.2% | 3.524(3.169-2.919) | |||
No. of LNs | 135.822 | <0.001 | <0.001 | ||
≤6 | 20.3% | Reference | |||
7-14 | 29.0% | 0.586(0.536-0.640) | |||
≥15 | 32.6% | 0.390(0.356-0.427) |
NI: not included in the multivariate survival analysis.
P values were adjusted for age, race, location, pathological grading, histotype stage, tumor stage and No. of LNs as covariates.