A retrospective study analyzed data from 11,134 patients with psoriasis treated with secukinumab after 3 months, with 4,593 (41.25%) achieving complete skin clearance (PASI 100). The data was split into a training set (70%) and a test set (30%), feature selection was performed by LASSO, and eight machine learning algorithms were developed. Random Forest (training AUC = 0.879, testing AUC = 0.757) and LightGBM (training AUC = 0.834, testing AUC = 0.761) performed best. LASSO identified 5 key predictors: gender, bIGA, bBSA, bPASI and bDLQI. The SHAP analysis confirmed that gender and basic indicators of disease severity are the most important. Factors such as disease duration, BMI, education or comorbidities were significantly associated with the PASI 100 response (all p < 0.05). The model supports personalized treatment in clinical practice.