Drug repurposing, an important aspect of this process, entails discovering new therapeutic uses for existing medications, leveraging their known safety profiles and pharmacokinetics to accelerate the development of effective treatments. Nowadays, the field is evolving with the integration of machine learning, which enhances drug repurposing by detecting patterns and relationships within biochemical and pharmacological data, optimizing drug formulations, predicting efficacy and safety from existing datasets, and discovering new medications for diseases such as cystic fibrosis. Quantitative Structure–Activity Relationship (QSAR) modeling is one of the key approaches that employ statistical and machine learning methods to predict a compound's biological activity based on its chemical structure. This use of machine learning in drug repurposing shows the prospect of producing more effective treatments for CF and other genetic disorders. Machine Learning Algorithms applied for this prediction are Random Forest, Linear Regression, SVR, CNN and MLP; they have R2 scores of 0.98, 0.99, 0.95, 0.96, and 0.97, respectively for the most commonly occurring mutation of CFTR gene, i.e. F508del.