With the growing volume of text documents, finding relevant information online has become increasingly difficult. Automatic Text Summarization (ATS) offers a solution that allows for the processing of extensive collections of documents and the efficient extraction of the most relevant content. Despite significant progress, ATS still faces challenges such as managing long and repetitive sentences, preserving textual coherence, and maintaining semantic alignment. This work introduces an extractive text summarization approach based on topic modeling to address these limitations. The proposed method focuses on producing summaries with highly representative sentences, reduced redundancy, concise content, and strong semantic consistency. Its effectiveness is demonstrated through experimental evaluations on DUC datasets, where it performs better than state-of-the-art techniques.