Research Article

A Recurrence-Specific Gene-Based Prognosis Prediction Model for Lung Adenocarcinoma through Machine Learning Algorithm

Table 1

Clinical characteristics of included patients for survival model construction and validation.

TCGA training cohort (288)TCGA testing cohort (128)External validation cohort (335)

Sex
 Female167 (56.04%)64 (50%)189 (56.42%)
 Male131 (43.96%)64 (50%)146 (43.58%)
Age
 ≥60201 (67.45%)95 (74.22%)234 (69.85%)
 <6088 (29.53%)32 (25%)101 (30.15%)
 Unknown9 (3.02%)1 (0.78%)0 (0%)
Pathologic T
 T1109 (36.58%)41 (32.03%)110 (32.84%)
 T2160 (53.69%)67 (52.34%)202 (60.29%)
 T321 (7.05%)13 (10.16%)16 (4.78%)
 T46 (2.01%)6 (4.69%)5 (1.49%)
 Unknown2 (0.67%)1 (0.78%)2 (0.60%)
Pathologic N
 N0201 (67.45%)80 (62.50%)299 (89.25%)
 N152 (17.45%)26 (20.31%)88 (26.27%)
 N238 (12.75%)17 (13.28%)53 (14.93%)
 N32 (0.67%)0 (0%)0 (0%)
 Unknown5 (1.68%)5 (3.91%)0 (0%)
Pathologic MNA
 M0192 (64.43%)83 (64.84%)0 (0%)
 M112 (4.03%)5 (3.91%)0 (0%)
 Unknown94 (31.54%)40 (31.25%)335 (100%)
Tumor stage
 I171 (57.38%)64 (50.00%)150 (33.86%)
 II69 (23.15%)33 (25.78%)252 (56.88%)
 III43 (14.43%)21 (16.41%)29 (6.55%)
 IV12 (4.03%)6 (4.69%)12 (2.71%)
 Unknown3 (1.01%)4 (3.13%)0 (0%)