Table 2.
Univariate analysis of recurrence predictors.
| All patients (Cohort A) | LN-negative patients (without MAC, MIC, ITC) (Cohort B) | ||||||
|---|---|---|---|---|---|---|---|
| Predictora | Total N (relapse/event) | HR (95% CI) | P valued | Total N (relapse/event) | HR (95% CI) | P valued | |
| Age (years) | 379 (43) | 1.00(0.98; 1.03) | 0.725 | 320 (23) | 0.98 (0.95; 1.02) | 0.345 | |
| Stage pT | 1a | 75 (3) | Ref. | 75 (3) | Ref. | ||
| 1b1–1b2 | 249 (28) | 3.39 (1.03; 11.16) | 0.045 | 209 (15) | 2.13 (0.61; 7.35) | 0.234 | |
| >1b2 | 55 (12) | 5.99 (1.69; 21.24) | 0.006 | 36 (5) | 3.67 (0.88; 15.35) | 0.075 | |
| Tumour type | Squamous | 280 (26) | Ref. | 241 (13) | Ref. | ||
| Adenocarcinoma | 76 (10) | 1.53 (0.74; 3.17) | 0.255 | 63 (7) | 2.35 (0.94; 5.90) | 0.069 | |
| Adenosquamous | 18 (7) | 7.29 (3.12; 17.01) | <0.001 | 12 (3) | 8.32 (2.32; 29.86) | 0.001 | |
| Grade | 1 | 37 (6) | ref. | 36 (6) | ref. | ||
| 2 | 144 (13) | 0.57 (0.22; 1.51) | 0.257 | 122 (6) | 0.31 (0.10; 0.95) | 0.040 | |
| 3 | 138 (23) | 1.10 (0.45; 2.69) | 0.841 | 103 (11) | 0.66 (0.24; 1.77) | 0.406 | |
| LVSI | No | 235 (18) | Ref. | 223 (17) | Ref. | ||
| Yes | 144 (25) | 2.48 (1.35; 4.54) | 0.003 | 97 (6) | 0.84 (0.33; 2.12) | 0.706 | |
| Number of positive LN | 379 (43) | 1.18 (1.11; 1.26) | <0.001 | 320 (23) | – | – | |
| LN positivity | No (any type) | 319 (23) | Ref. | 320 (23) | – | – | |
| Yes | 60 (20) | 5.46 (2.99; 9.95) | <0.001 | 0 (0) | – | – | |
| LN positivity | ITC, negative | 329 (25) | ref. | 320 (23) | – | – | |
| MAC, MIC | 50 (18) | 5.40 (2.94; 9.90) | <0.001 | 0 (0) | – | – | |
| Largest tumour size (P) | 378 (43) | 1.02 (1.00; 1.03) | 0.007 | 319 (23) | 1.00 (0.98; 1.02) | 0.909 | |
| Largest tumour size (P) binarisedb | ≤Cut-offc | 263 (18) | Ref. | 195 (12) | Ref. | ||
| >Cut-offc | 115 (25) | 3.69 (2.01; 6.78) | <0.001 | 124 (11) | 1.54 (0.68; 3.50) | 0.299 | |
| Depth of stromal invasion (P) | 378 (27) | 1.06 (1.00; 1.12) | 0.035 | 187 (12) | 1.04 (0.95; 1.14) | 0.379 | |
| Largest tumour size (I) | 379 (43) | 1.03 (1.01; 1.05) | <0.001 | 320 (23) | 1.01 (0.99; 1.04) | 0.265 | |
| Parametrial invasion (P) | No | 323 (33) | Ref. | 281 (19) | Ref. | ||
| Yes | 56 (10) | 1.70 (0.84; 3.45) | 0.141 | 39 (4) | 1.47 (0.50; 4.31) | 0.488 | |
| Minimal TFD (I) | 321 (33) | 0.84 (0.76; 0.93) | 0.001 | 278 (18) | 0.89 (0.79; 0.99) | 0.030 | |
| TFD binarised (I)b | >3.5 | 146 (4) | Ref. | 140 (4) | Ref. | ||
| ≤3.5 | 175 (29) | 7.16 (2.52; 20.39) | <0.001 | 134 (14) | 4.37 (1.44; 13.27) | 0.009 | |
| Tumour volume (mm3) | 361 (43) | 1.00 (1.00; 1.00) | 0.021 | 302 (23) | 1.00 (1.00; 1.00) | 0.831 | |
CI confidence interval, ITC isolated tumour cells, LN lymph node, MAC macrometastasis, MIC micrometastasis, TFD tumour-free distance, (I) assessed by imaging, (P) assessed by pathology.
aHazard ratios are computed using Cox proportional hazards model; bcut-off determined by ROC analysis, the criterion was the highest value of the sum of sensitivity and specificity; ccut-off for Cohort A: 32.5, Cohort B: 25.5; dlevel of significance P < 0.05.