Skip to main content
. 2025 Aug 27;16:1646038. doi: 10.3389/fimmu.2025.1646038

Figure 4.

Panel A shows a plot of AUC versus Log Lambda with data points and error bars. Panel B displays a series of coefficient trajectories against Log Lambda. Panel C is a ROC curve plot with an AUC of 0.70, showing sensitivity against 1-specificity.

Results of the LASSO regression analysis and ROC curves for risk factors associated with SJS/TEN related to anticancer therapies. (A) Optimal regularization parameter λ in the LASSO regression model; (B) Evolutionary trajectories of variable coefficients across the λ-spectrum; (C) AUC of the multifactorial logistic regression analysis. LASSO, least absolute shrinkage and selection operator; ROC, receiver operating characteristic; AUC, area under curve; SJS/TEN, Stevens-Johnson syndrome and/or toxic epidermal necrolysis.