Skip to main content

Advertisement

Table 3 Logistic regression analysis of risk factors for dMMR status

From: Comparison of immunological characteristics between paired mismatch repair-proficient and -deficient colorectal cancer patients

Factors Univariate analysis Multivariate analysis
OR (95% CI) p-value OR (95% CI) p-value
Age at diagnosis (≥ 60) 0.44 (0.28–0.69) < 0.001 0.48 (0.25–0.94) 0.032
Gender (female) 0.77 (0.49–1.19) 0.241   
Primary tumor site < 0.001   0.001
 Left colon 0.55 (0.30–0.99) 0.045 0.43 (0.17–1.09) 0.076
 Right colon 1.98 (1.11–3.53) 0.02 2.28 (0.86–6.07) 0.098
 Rectum (reference)
Stage (7th AJCC) 0.965   
 I 0.79 (0.28–2.21) 0.651   
 II 0.78 (0.31–1.98) 0.603   
 III 0.80 (0.30–2.13) 0.653   
 IV (reference)   
Tumor stage 0.007   0.004
 T1 0.26 (0.07–0.90) 0.034 0.04 (0.01–0.33) 0.002
 T2 0.39 (0.18–0.88) 0.024 0.38 (0.11–1.37) 0.139
 T3 0.33 (0.17–0.63) 0.001 0.20 (0.07–0.59) 0.004
 T4 (reference)
Node stage (positive) 0.819   
Negative lymph nodes (high) 2.09 (1.34–3.24) 0.001 1.10 (0.52–2.30) 0.810
Tumor histological grade 0.002   0.588
 Well-differentiated 0.25 (0.02–2.87) 0.266 7.91 (0.38–163.50) 0.181
 Moderately differentiated 0.35 (0.21–0.60) < 0.001 1.12 (0.48–2.63) 0.788
 Poorly differentiated 0.50 (0.25–1.00) 0.051 0.93 (0.30–2.89) 0.896
 Mucinous (reference)  
Nerve invasion (no) 2.22 (1.07–4.58) 0.032 1.80 (0.59–5.45) 0.299
Vascular invasion (no) 2.32 (1.10–4.91) 0.028 3.66 (1.04–12.98) 0.044
MHC class I expression (low) 15.16 (7.81–29.42) < 0.001 24.66 (8.74–69.64) <0.001
CD3 expression (high) 0.90 (0.57–1.43) 0.662   
CD4 expression (high) 0.43 (0.27–0.68) < 0.001 0.34 (0.14–0.83) 0.017
CD8 expression (high) 1.89 (1.17–3.06) 0.01 2.97 (1.30–6.78) 0.01
CD56 expression (high) 0.37 (0.23–0.59) < 0.001 0.92 (0.78–1.08) 0.294
PD-1 expression (high) 1.19 (0.73–1.93) 0.483   
PD-L1 expression (positive) 0.58 (0.38–0.90) 0.015 0.55 (0.27–1.13) 0.103
  1. OR: odds ratio; 95% CI: 95% confidence interval
  2. Only the meaningful factors (p < 0.05) in univariate analysis were brought into the multivariate analysis