Univariate and multivariate Cox regression analyses of potential correlations between different clinicopathological parameters and survival of patients with CRC
| Variable | HR (95% CI) | p Value |
|---|---|---|
| Univariate Cox regression analysis | ||
| Age | 1.02 (0.99 to 1.04) | 0.2200 |
| Sex | ||
| Male (n=92) | 0.64 (0.35 to 1.19) | 0.1600 |
| Female (n=56) | 1.00 | |
| Tumour localisation | ||
| Right colon (n=28) | 1.03 (0.46 to 2.33) | 0.9400 |
| Left colon or rectum (n=120) | 1.00 | |
| Differentiation | ||
| Low (n=4) | 0.73 (0.10 to 5.32) | 0.7600 |
| Medium (n=139) | 1.00 | |
| High (n=5) | 3.40 (1.04 to 11.06) | 0.0400 |
| TNM | ||
| 1 (n=12) | 0.04 (0.006 to 0.33) | 0.0024 |
| 2 (n=61) | 0.08 (0.04 to 0.20) | 0.0000 |
| 3 (n=55) | 0.16 (0.07 to 0.32) | 0.0000 |
| 4 (n=20) | 1.00 | |
| Hypermutation | ||
| Yes (n=14) | 0.53 (0.13 to 2.21) | 0.39 |
| No (n=134) | 1.00 | |
| MSI | ||
| Negative (n=111) | 1.00 | |
| Low (n=18) | 0.95 (0.4 to 2.26) | 0.9000 |
| High (n=19) | 0.34 (0.08 to 1.41) | 0.1400 |
| KRAS G12/13X | ||
| Negative (n=103) | 1.00 | |
| Positive (n=45) | 0.87 (0.44 to 1.74) | 0.6920 |
| Prognostic signature mutation | ||
| Yes (n=40) | 0.21 (0.064 to 0.67) | 0.0091 |
| No (n=108) | 1.00 | |
| Multivariate Cox regression analysis | ||
| Differentiation | ||
| Low (n=4) | 1.55 (0.20 to 12.0) | 0.68 |
| Medium (n=146) | 1 | |
| High (n=5) | 0.99 (0.29 to 3.40) | 0.99 |
| TNM | ||
| 1 (n=12) | 0.05 (0.006 to 0.36) | 0.0033 |
| 2 (n=66) | 0.09 (0.04 to 0.21) | 0.0000 |
| 3 (n=57) | 0.16 (0.08 to 0.34) | 0.0000 |
| 4 (n=20) | 1 | |
| Prognostic signature mutation | ||
| Yes (n=38) | 0.27 (0.083 to 0.89) | 0.031 |
| No (n=117) | 1 | |
p Values <0.05 were bolded.
Low TNM staging and mutation(s) in a five-gene signature composed of CDH10, COL6A3, SMAD4, TMEM132D and VCAN conferred significantly lower hazard ratios in both analyses. Patients with undermined MSI status were excluded from univariate analysis but included in multivariate analysis. Patients with missing survival data were excluded from both analyses.
MSI, microsatellite instability.