Skip to main content
. 2020 Feb 6;10:1952. doi: 10.1038/s41598-020-58722-z

Table 1.

Baseline characteristics of the training and validation cohorts.

Characteristic Training cohort External validation cohort p-value
n = 1,022 (%) n = 298 (%)
Age (years) Median (range) 66 (33–86) 66 (25–85) 0.387
Gender Male 666 (65.2) 195 (65.4) 0.931
Female 356 (34.8) 103 (34.6)
Smoking history Never 461 (45.7) 132 (45.2) 0.884
Former/current 548 (54.3) 160 (54.8)
ECOG 0 503 (49.2) 157 (52.7) 0.292
performance status 1 519 (50.8) 141 (47.3)
CEA ng/mL 2.1 (1.0–230.1) 1.9 (1.0–1070.9) 0.428
WBC 106/L 7,399 ± 3.8 7,611 ± 3.3 0.381
Neutrophil % 59.7 ± 31.1 60.0 ± 12.4 0.722
lymphocyte % 29.9 ± 18.7 28.5 ± 10.9 0.227
Haemoglobin g/dL 13.2 ± 3.8 13.0 ± 1.7 0.518
Platelet 109/L 237 ± 79 239 ± 83 0.610
C-reactive protein mg/dL 0.14 (0.0–34.1) 0.12 (0.02–23.6) 0.450
Pulmonary function FEV1 (L) 2.4 (0.08–352.0) 2.37 (0.96–139.0) 0.915
DLCo (%) 85 (8–173) 83 (9–159) 0.327
Histology Adenocarcinoma 651 (63.7) 2003 (67.1) 0.442
Squamous 303 (29.6) 77 (25.8)
others 68 (6.7) 21 (7.0)
Tumour size cm 2.5 (0.4–13.0) 2.5 (0.3–13.0) 0.226
No. of LN positivity 0 (0–23) 0 (0-31) 0.847
T stage T1 474 (46.5) 160 (53.7) 0.064
T2 433 (42.5) 114 (38.3)
T3/4 113 (11.1) 24 (8.1)
N stage N0 757 (74.9) 239 (80.5) 0.136
N1 135 (13.4) 30 (10.1)
N2 119 (11.8) 28 (9.4)
TNM stage I 669 (65.7) 211 (71.0) 0.218
II 200 (19.6) 50 (16.8)
III 150 (14.7) 36 (12.1)
Tumor differentiation Well 197 (19.7) 75 (25.6) 0.054
Moderately 603 (60.4) 156 (53.2)
Poorly 199 (19.9) 62 (21.1)
Vascular invasion Yes 143 (14.0) 35 (11.9) 0.489
Lymphatic invasion Yes 353 (34.6) 95 (32.0) 0.560
Perineural invasion Yes 59 (5.8) 13 (4.4) 0.642
Resection status* R0 980 (97.5) 289 (98.0) 0.849
R1 19 (1.9) 5 (1.7)
R2 6 (0.6) 1 (0.3)
Neoadjuvant treatment Yes 50 (4.9) 14 (4.7) 0.888
Adjuvant treatment Yes 333 (33.1) 86 (29.3) 0.214
Recurrence Yes 272 (26.6) 76 (25.2) 0.618

ECOG, Eastern Cooperative Oncology Group; CEA, carcinoembryonic antigen; WBC, white blood cell; FEV1, forced expiratory volume in the first second; DLCo, diffusing capacity of the lung for carbon monoxide; LN, lymph node.

*R0, number of cancer cells seen microscopically at the resection margin; R1, microscopic positive margin; R2, macroscopic positive margin.