Research Article

Development and External Validation of Machine Learning-Based Models for Predicting Lung Metastasis in Kidney Cancer: A Large Population-Based Study

Table 2

Characteristics in the training, internal, and external test cohorts.

VariablesTraining cohortInternal test cohortExternal test cohort
Non-LMLMNon-LMLMNon-LMLM
TotalN = 34,775N = 1,780N = 14,853N = 814N = 468N = 24

Age
Mean (SD)59.6 (12.4)61.3 (11.1)59.8 (12.3)61.1 (10.8)59.5 (10.8)60.0 (9.07)

Sex
Female13,622 (37.3%)514 (28.9%)5,524 (37.2%)232 (28.5%)173 (37.0%)4 (16.7%)
Male22,933 (62.7%)1,266 (71.1%)9,329 (62.8%)582 (71.5%)295 (63.0%)20 (83.3%)

Laterality
Left17,982 (49.2%)889 (49.9%)7,302 (49.2%)429 (52.7%)248 (53.0%)12 (50.0%)
Right18,573 (50.8%)891 (50.1%)7,551 (50.8%)385 (47.3%)220 (47.0%)12 (50.0%)

Race
Asian2,163 (5.9%)131 (7.4%)908 (6.1%)62 (7.6%)468 (100%)24 (100%)
Black3,986 (10.9%)134 (7.5%)1,671 (11.3%)44 (5.4%)0 (0%)0 (0%)
Others720 (2.0%)27 (1.5%)299 (2.0%)12 (1.5%)0 (0%)0 (0%)
White29,686 (81.2%)1,488 (83.6%)11,975 (80.6%)696 (85.5%)0 (0%)0 (0%)

T_stage
T124,358 (66.6%)190 (10.7%)10,281 (69.2%)84 (10.3%)393 (84.0%)8 (33.3%)
T23,920 (10.7%)310 (17.4%)1,550 (10.4%)140 (17.2%)29 (6.2%)7 (29.2%)
T37,580 (20.7%)1,012 (56.9%)2,855 (19.2%)476 (58.5%)42 (9.0%)6 (25.0%)
T4697 (1.9%)268 (15.1%)167 (1.1%)114 (14.0%)4 (0.9%)3 (12.5%)

N_stage
N034,928 (95.5%)1,171 (65.8%)14,437 (97.2%)557 (68.4%)452 (96.6%)18 (75.0%)
N11,627 (4.5%)609 (34.2%)416 (2.8%)257 (31.6%)16 (3.4%)6 (25.0%)

Tumor_size
Mean (SD)53.6 (37.4)102 (48.9)51.1 (34.4)101 (55.6)42.9 (22.0)68.6 (27.0)

Histology
8120126 (0.3%)24 (1.3%)45 (0.3%)11 (1.4%)0 (0%)0 (0%)
8255926 (2.5%)80 (4.5%)373 (2.5%)45 (5.5%)0 (0%)0 (0%)
82604,202 (11.5%)70 (3.9%)1,777 (12.0%)28 (3.4%)22 (4.7%)0 (0%)
831024,731 (67.7%)1,146 (64.4%)10,164 (68.4%)535 (65.7%)427 (91.2%)21 (87.5%)
83124,060 (11.1%)269 (15.1%)1,544 (10.4%)113 (13.9%)16 (3.4%)3 (12.5%)
83171,550 (4.2%)14 (0.8%)654 (4.4%)6 (0.7%)0 (0%)0 (0%)
Others960 (2.6%)177 (9.9%)296 (2.0%)76 (9.3%)3 (0.6%)0 (0%)

Grade
I4,111 (11.2%)42 (2.4%)1,724 (11.6%)19 (2.3%)69 (14.7%)0 (0%)
II18,545 (50.7%)330 (18.5%)7,799 (52.5%)146 (17.9%)266 (56.8%)5 (20.8%)
III10,779 (29.5%)747 (42.0%)4,260 (28.7%)324 (39.8%)109 (23.3%)12 (50.0%)
IV3,120 (8.5%)661 (37.1%)1,070 (7.2%)325 (39.9%)24 (5.1%)7 (29.2%)

LM, lung metastasis; 8120, transitional cell carcinoma; 8255, adenocarcinoma with mixed subtypes; 8260, papillary adenocarcinoma; 8310, clear cell adenocarcinoma; 8312, renal cell carcinoma; 8317, chromophobe renal carcinoma.