Table 3 Multivariate logistic regression model of the treatment response for selected variables.
Variable | OR (95% CI) | P | AUC (95% CI) |
|---|---|---|---|
Model 1: Prediction with clinical features | 0.846 (0.724–0.968) | ||
HR status (negative vs. positive) | 5.52 (1.15–36.40) | 0.046 | |
LN status (negative vs. positive) | 4.09 (0.65–36.40) | 0.148 | |
Clinical stage (III vs. I–II) | 0.28 (0.04–1.53) | 0.156 | |
HER2 (3+ vs. 2+/FISH+) | 4.03 (0.31–120.5) | 0.324 | |
TILs (intermediate/high vs. low) | 5.10 (0.99–33.52) | 0.063 | |
Model 2: Prediction with genetic biomarkers | 0.842 (0.728–0.957) | ||
PIK3CA (mutant vs. WT) | 0.08 (0.01–0.34) | 0.0015 | |
MYC (amplification vs. neutral) | 0.10 (0.01–0.63) | 0.024 | |
Model 3: Prediction with combined predictors | 0.912 (0.827–0.997) | ||
HR status (negative vs. positive) | 2.65 (0.32–26.65) | 0.367 | |
LN status (negative vs. positive) | 4.42 (0.49–59.21) | 0.212 | |
Clinical stage (III vs. I–II) | 0.45 (0.04–4.68) | 0.502 | |
TILs (intermediate/high vs. low) | 6.33 (0.71–91.11) | 0.120 | |
HER2 (3+ vs. 2+/FISH+) | 10.54 (0.41–648.7) | 0.200 | |
PIK3CA (mutant vs. WT) | 0.07 (0.01–0.41) | 0.007 | |
MYC (amplification vs. neutral) | 0.22 (0.02–2.27) | 0.211 |
