BPG is committed to discovery and dissemination of knowledge
Retrospective Study
Copyright ©The Author(s) 2025.
World J Gastroenterol. Sep 28, 2025; 31(36): 111293
Published online Sep 28, 2025. doi: 10.3748/wjg.v31.i36.111293
Table 1 Demographic and clinical factors in pathological complete response and non-pathological complete response groups

All (n = 70)
pCR (n = 15)
Non-pCR (n = 55)
P value
Age, median (range)66 (45-78)64 (52-76)66 (45-78)0.97
Gender
Male/female60/1012/348/70.48
Tumor site
Cervical/upper/middle/lower5/14/33/181/5/6/34/9/27/150.60
TNM staging
cT1/2/3/40/2/19/490/0/3/120/2/16/370.56
cN0/1/2/31/17/29/230/1/5/91/16/24/140.06
cStage I/II/III/IVA0/0/10/600/0/0/150/0/10/450.075
Total radiation dose
40 Gy/up to 60 Gy59/1111/448/70.19
Table 2 Predictive performance and receiver operating characteristic curve analyses of the five radiomics features with the highest area under the curve values and the best performing artificial intelligence-based machine-learning model for pathological complete response

AUC
Sensitivity
Specificity
Accuracy
Cut off
95%CI
P value
ADC skewness0.770.670.820.790.370.61-0.880.005
GLCM entropy (b = 1000 second/mm²)0.760.870.620.679.280.64-0.850.002
GLCM autocorrelation (b = 0 second/mm²)0.760.530.930.845276.10.60-0.870.009
Skewness (b = 0 second/mm²)0.730.870.420.640.720.61-0.840.006
Kurtosis (b = 0 second/mm²)0.720.730.640.665.050.55-0.820.005
Machine learning radiomics model0.850.800.850.81NA0.73-0.93< 0.001
Table 3 Univariate and multivariate Cox regression analyses for relapse-free survival
Univariate
Multivariate
HR
95%CI
P value
HR
95%CI
P value
Radiomics features
ADC skewness0.420.16-1.090.076
GLCM entropy (b = 1000 second/mm²)0.250.11-0.560.0010.320.14-0.750.009
GLCM autocorrelation (b = 0 second/mm²)2.000.61-6.570.25
Skewness (b = 0 second/mm²)0.600.29-1.220.16
Kurtosis (b = 0 second/mm²)0.760.37-1.540.45
Pathological features
pT3 vs pT0-21.080.54-2.180.82
pN + vs pN -3.141.48-6.630.0032.201.00-4.800.042
Grade 3 or 2 vs grade 10.500.25-1.010.054