Fig. 1

Feature selection and construction of the immunomarker classifier using LASSO logistic regression, and the predictive performance of the classifier. A Tuning parameter (λ) selection in the LASSO model via ten-fold cross-validation based on minimum criteria. The binomial deviance was plotted versus log (λ). The dotted vertical line was set with minimum criteria, and the optimal λ value of 0.094 with log (λ) of -2.368 was selected. B LASSO coefficient profiles of the 13 immune markers. A vertical line was drawn at the value selected by 10-fold cross-validation, where optimal λ resulted in 2 features with nonzero coefficients. C ROC curve and Youden’s index of the Immunomarker Classifier. D The immune score boxplots between complete and partial remission groups. E Comparison of the immediate responses between the low-score group and the high-score group. Significance: * p value < 0.05. **p < 0.01