Prostate cancer is a major cause of morbidity and mortality in men, ranking behind lung cancer and colorectal cancer. Early cancer detection and informative predictive models are vital to improving patient outcomes and making the best clinical choices. However, clinical and histopathological heterogeneity is a significant obstacle in developing an accurate and reliable predictive model. This study involves examining the prediction of prostate cancer progression, given clinical and histopathological variables, using three variants of Boosting, namely AdaBoost, Gradient Boosting, and XGBoost. Two metrics for measuring the performance of algorithms, Accuracy and AUC, are used as a comparison method. The results show that XGBoost performs better than the other models, obtaining an AUC of 0.94, which demonstrates its performance in helping prostate cancer diagnosis become more accurate and assisting medical decision-making.