Table 1.
Baseline characteristics of training and validation cohorts.
| Variables | Training Cohortan = 25,350 (%) | Validation Cohortan = 10,860 (%) | P value |
|---|---|---|---|
| Age, years | 0.963 | ||
| Median (range) | 60 (18-72) | 60 (20-72) | |
| Gender | 0.607 | ||
| Male | 13275 (52.4) | 5655 (52.1) | |
| Female | 12075 (47.6) | 5205 (47.9) | |
| Histological grade | 0.857 | ||
| Grade I | 2266 (8.9) | 999 (9.2) | |
| Grade II | 18404 (72.6) | 7856 (72.3) | |
| Grade III | 4273 (16.9) | 1815 (16.7) | |
| Grade IV | 407 (1.6) | 190 (1.7) | |
| AJCC 8th T stage | 0.281 | ||
| T1 | 2640 (10.4) | 1077 (9.9) | |
| T2 | 4190 (16.5) | 1782 (16.4) | |
| T3 | 15486 (61.1) | 6695 (61.6) | |
| T4a | 1751 (6.9) | 763 (7.0) | |
| T4b | 1283 (5.1) | 543 (5.0) | |
| AJCC 8th N stage | 0.258 | ||
| N0 | 14762 (58.2) | 6293 (57.9) | |
| N1a | 3240 (12.8) | 1399 (12.9) | |
| N1b | 3352 (13.2) | 1398 (12.9) | |
| N2a | 2233 (8.8) | 940 (8.7) | |
| N2b | 1763 (7.0) | 830 (7.6) | |
| AJCC 8th TNM stage | 0.383 | ||
| I | 5493 (21.7) | 2321 (21.4) | |
| IIA | 8011 (31.6) | 3456 (31.8) | |
| IIB | 667 (2.6) | 260 (2.4) | |
| IIC | 591 (2.3) | 256 (2.4) | |
| IIIA | 1150 (4.5) | 455 (4.2) | |
| IIIB | 6949 (27.4) | 2974 (27.4) | |
| IIIC | 2489 (9.8) | 1138 (10.5) | |
| AJCC 6th TNM stage | 0.358 | ||
| I | 5476 (21.6) | 2318 (21.3) | |
| IIA | 7971 (31.4) | 3445 (31.7) | |
| IIB | 1251 (4.9) | 516 (4.8) | |
| IIIA | 1134 (4.5) | 448 (4.1) | |
| IIIB | 5577 (22.0) | 2381 (21.9) | |
| IIIC | 3941 (15.5) | 1752 (16.1) | |
| Tumor size, mm | 0.810 | ||
| Median (range) | 42 (1-150) | 41.5 (1-150) | |
| Positive lymph nodes | 0.158 | ||
| Mean (range) | 1.6 (0–59) | 1.7 (0–54) | |
| Examined lymph nodes | 0.315 | ||
| Median (range) | 16 (1-89) | 16 (1-89) |
AJCC, American Joint Committee on Cancer.
a
Ratio of training to validation cohorts is 7:3 by randomized number using R software.