<p>Early accurate breast cancer diagnosis is a cornerstone of medical diagnostics and successful prognosis. This study investigates the use of advanced metaheuristic optimization algorithms for integrated breast cancer prediction in this novel study. Three benchmark databases the Wisconsin Diagnostic Breast Cancer (WDBC), Wisconsin Prognostic Breast Cancer (WPBC) and Coimbra datasets has been considered in the work. As part of this study, the impact of six state-of-the-art metaheuristic algorithms, namely, Multiverse Optimization Algorithm (MVO), Gravitational Search Algorithm (GSA), Teaching-Learning Based Optimization (TLBO), Whale Optimization Algorithm (WOA), Cuckoo Search Optimization Algorithm (CSOA) and Grey Wolf Optimization (GWO) on the feature selection of factors affecting breast cancer classification is ascertained. The optimized feature subsets are then used to train SVM classifiers, and their performance is evaluated using metrics such as accuracy, sensitivity, specificity, and F1-score. An end-to-end pipeline consisting of data preprocessing, split train-test-validation, and performance tracking was constructed to maintain the quality and generalizability of the results. Experimental results demonstrate that MVO attained the highest accuracy with least features for all three databases. These results demonstrate that not only quantitative measures of predictive power can be drastically improved through feature optimization, but also that model complexity is minimized to the point where pathways are viable for development into novel and rapid diagnostic tools. This Paper demonstrates the power of combining metaheuristic optimization algorithms for machine learning for solving critical healthcare problems. The work done shines a light of innovation, and opens the door to more precise, trustworthy and accessible diagnostic systems in the fight against breast cancer.</p>