Skip to main content
. 2022 Dec 15;12(12):5668–5683.

Table 2.

Multivariate cox regression analysis of test and validation cohorts

Cohort Variable Hazard Ratio (95% CI) P Value
A. Glasgow training cohort Lymph Node Metastases (Positive) 1.92 (1.17-3.15) 0.01
Tumour size (> 30 mm) 1.48 (1.06-2.06) 0.021
Tumor Grade (High) 1.77 (1.25-2.52) 0.001
Resection Margin Involvement (< 1 mm) 1.79 (1.18-2.72) 0.006
Adjuvant chemotherapy (≥ 3 cycles) 0.54 (0.38-0.77) 0.001
miR-21 expression (High, HS > 45) 2.11 (1.51-2.96) < 0.0001
B. Pisa validation cohort Tumor Stage (T3/T4) 4.47 (1.06-18.9) 0.042
miR-21 expression (High, HS > 45) 2.03 (1.21-3.39) 0.007
C. Sydney validation cohort Tumour size (> 30 mm) 2.10 (1.42-3.12) < 0.0001
Venous Invasion (Positive) 1.30 (0.98-1.75) 0.070
Resection Margin Status (Involved, < 1 mm) 1.65 (1.23-2.20) < 0.0001
Adjuvant chemotherapy (≥ 3 cycles) 0.59 (0.42-0.86) 0.006
miR-21 expression (High, HS > 45) 2.59 (1.89-3.53) < 0.0001
D. ICGC validation cohort Tumour size (> 30 mm) 2.01 (1.22-3.31) 0.006
Gender 1.63 (1.04-2.56) 0.032
Tumour Grade (High) 2.40 (1.41-4.06) 0.001
Adjuvant chemotherapy (≥ 3 cycles) 0.41 (0.23-0.72) 0.002
miR-21 expression (High, HS > 45) 2.16 (1.32-3.51) 0.002