This study focuses on the computational analysis of solubility of an anti-cancer drug namely Temozolomide in SC-CO2, utilizing a dataset consisting of pressure and temperature as input parameters and Temozolomide solubility as the output. Temperature range between 308–338 K, and pressure between 120–400 bar were taken into account for building the models. Three regression models—K Nearest Neighbors (KNN) regression, Huber regression, and Support Vector Machine (SVM) regression—were employed to predict solubility, with hyper-parameter tuning accomplished using the Cuckoo Search (CS) optimization algorithm. The updated results underscored the efficacy of these models in predicting Temozolomide solubility. Notably, the SVM regression model displayed impressive performance, yielding an R-squared score of approximately 0.976, indicating a high degree of accuracy in predicting solubility. The RMSE for SVM regression was approximately 0.000152, signifying minimal prediction error, and the Mean Absolute Error (MAE) was approximately 0.000124.