BPG is committed to discovery and dissemination of knowledge
Retrospective Study
Copyright: ©Author(s) 2026.
World J Gastrointest Oncol. Jun 15, 2026; 18(6): 117851
Published online Jun 15, 2026. doi: 10.4251/wjgo.v18.i6.117851
Table 1 Clinical characteristic of esophageal patients, median (interquartile range)/mean (SD)
Variables
Whole cohort (n = 206)
Training cohort (n = 144)
Validation cohort (n = 62)
P value
Age 64.1 (7.5)64.5 (7.5)63.1 (7.4)0.218
Gender0.570
Female45 (21.8%)33 (22.9%)12 (19.4%)
Male161 (78.2%)111 (77.1%)50 (80.6%)
Body mass index23.1 (2.65)23.1 (2.7)23.1 (2.5)0.965
Smoking and drinking0.267
None86 (41.7%)60 (41.7%)26 (41.9%)
Smoking alone40 (19.4%)29 (20.1%)11 (17.7%)
Drinking alone13 (6.3%)6 (4.2%)7 (11.3%)
Both67 (32.5%)49 (34%)18 (29%)
Squamous cell carcinoma level1 (0.6, 1.4)1 (0.6, 1.4)1.1 (0.8, 1.3)0.520
Cytokeratin 19 fragment antigen 21-1 level1.4 (0.9, 1.8)1.4 (0.9, 1.9)1.3 (0.8, 1.8)0.290
Carcinoembryonic antigen level3.4 (2.6, 4.3)3.4 (2.6, 4.4)3.5 (2.6, 4.2)0.924
Carbohydrate antigen 19922.3 (14.9, 29.7)22.5 (14.8, 30)22.2 (16.8, 27.2)0.828
Tumor location0.480
Upper63 (30.6%)44 (30.6%)19 (30.6%)
Middle130 (63.1%)89 (61.8%)41 (66.1%)
Lower13 (6.3%)11 (7.6%)2 (3.2%)
Tumor length3 (2, 4.5)3.4 (2, 4.5)3 (2, 4)0.082
Computed tomography-reported nodal status0.407
Negative148 (71.8%)101 (70.1%)47 (75.8%)
Positive58 (28.2%)43 (29.9%)15 (24.2%)
Tumor grade0.057
G116 (7.8%)7 (4.9%)9 (14.5%)
G2150 (72.8%)109 (75.7%)41 (66.1%)
G340 (19.4%)28 (19.4%)12 (19.4%)
Depth of tumor invasion0.912
cT147 (22.8%)33 (22.9%)14 (22.6%)
cT258 (28.2%)40 (27.8%)18 (29%)
cT391 (44.2%)63 (43.8%)28 (45.2%)
cT410 (4.9%)8 (5.6%)2 (3.2%)
Node stage0.506
pN0127 (61.7%)89 (61.8%)38 (61.3%)
pN146 (22.3%)35 (24.3%)11 (17.7%)
pN219 (9.2%)12 (8.3%)7 (11.3%)
pN314 (6.8%)8 (5.6%)6 (9.7%)
Lymph node metastasis status0.944
Negative127 (61.7%)89 (61.8%)38 (61.3%)
Positive79 (38.3%)55 (38.2%)24 (38.7%)
Table 2 Performances of computed tomography to evaluate lymph node metastasis status in patients with esophageal cancer
StatisticValue (95%CI)
Whole cohort (n = 206)
Training set (n = 144)
Validation set (n = 62)
Overall lymph node status evaluation
Sensitivity42.11% (30.86%-53.98%)42.59% (29.23%-56.79%)40.91% (20.71%-63.65%)
Specificity80% (72.08%-86.50%)77.78% (67.79%-85.87%)85% (70.16%-94.29%)
PPV55.17% (44.38%-65.50%)53.49% (41.21%-65.36%)60% (38.06%-78.55%)
NPV70.27% (65.70%-74.47%)69.31% (63.64%-74.45%)72.34% (64.34%-79.13%)
Accuracy66.02% (59.11%-72.46%)64.58% (56.19%-72.37%)69.35% (56.35%-80.44%)
Thoracic lymph node status evaluation
Sensitivity47.92% (33.29%-62.81%)51.35% (34.40%-68.08%)36.36% (10.93%-69.21%)
Specificity86.08% (79.68%-91.06%)85.05% (76.86%-91.20%)88.24% (76.13%-95.56%)
PPV51.11% (39.11%-62.99%)54.29% (40.66%-67.30%)40% (18.39%-66.35%)
NPV84.47% (80.46%-87.79%)83.49% (78.25%-87.66%)86.54% (80.26%-91.04%)
Accuracy77.18% (70.84%-82.73%)76.39% (68.60%-83.06%)79.03% (66.82%-88.34%)
Abdominal lymph node status evaluation
Sensitivity28.89% (16.37%-44.31%)26.67% (12.28%-45.89%)33.33% (11.82%-61.62%)
Specificity95.03% (90.44%-97.83%)93.86% (87.76%-97.50%)97.87% (88.71%-99.95%)
PPV61.9% (41.80%-78.62%)53.33% (31.05%-74.36%)83.33% (38.76%-97.53%)
NPV82.7% (79.82%-85.25%)82.95% (79.59%-85.85%)82.14% (76.24%-86.83%)
Accuracy80.58% (74.51%-85.75%)79.86% (72.37%-86.08%)82.26% (70.47%-90.80%)
Table 3 Comparison of two machine-learning models with computed tomography in evaluating lymph node metastasis in patients with esophageal cancer
StatisticTraining set
Validation set
Value
95%CI
Value
95%CI
Computed tomography predicted
Sensitivity42.6%29.25%-56.8%40.9%20.7%-63.7%
Specificity77.8%67.8%-85.8%85.0%70.2%-94.3%
PPV53.5%41.2%-65.4%60.0%38.1%-78.6%
NPV69.3%63.6%-74.4%72.3%64.3%-79.1%
Accuracy64.6%56.2%-72.4%69.4%56.4%-80.4%
Model-1 predicted
Sensitivity85.45%73.34%-93.50%66.67%44.68%-84.37%
Specificity88.76%80.31%-94.48%92.11%78.62%-98.34%
PPV82.46%72.18%-89.49%84.21%63.45%-94.25%
NPV90.80%83.82%-94.95%81.40%71.15%-88.59%
Accuracy87.50%80.97%-92.42%82.26%70.47%-90.80%
Model-2 predicted
Sensitivity89.09%77.75%-95.89%75.00%53.29%-90.23%
Specificity85.39%76.32%-91.99%89.47%75.20%-97.06%
PPV79.03%69.34%-86.27%81.82%63.38%-92.12%
NPV92.68%85.56%-96.44%85.00%73.75%-91.95%
Accuracy86.81%80.16%-91.87%83.87%72.33%-91.98%


Write to the Help Desk