| Abstract: |
Objectives: We aim to leverage more comprehensive phenotypic and genotypic clinical data to enhance the treatment response predictions. Materials and Methods: The study cohort includes 213 NSCLC patients who underwent ICI therapy. Patients were categorized based on treatment outcomes: those with complete or partial responses were considered responders, while those exhibiting stable or progressive disease were deemed non-responders. Comprehensive phenotypic and genomic features were selected for prediction. We developed 9 machine learning models. The model demonstrating the highest area under the receiver operating characteristic curve (AUROC) performance was further analyzed using Shapley additive explanation values to interpret the predictive factors. Results: There were 72 patients who responded to the treatment, while 141 patients were considered non-responders. In total, 57 features were included, encompassing demographics, tumor status, treatment information, pre-treatment information, serum CBC, serum chemistry, and vital signs. The KNN model excelled among the models, achieving an AUROC score of 0.862 and outperforming the conventional PD-L1 biomarker’s AUROC of 0.619. The top features influencing ICI treatment response include the ECOG performance status of 0, lower red cell distribution width, higher mean platelet volume, etc. Discussion: The significance of functional status, inflammatory biomarkers, and PD-L1 expression are revealed. This research underscores the potential of using a more nuanced combination of biochemical markers and clinical data to enhance the precision of immunotherapy efficacy predictions, compared with single prognostic biomarkers such as PD-L1. Conclusion: Our findings emphasize the complex interplay among various risk factors that influence the effectiveness of ICI. Lung cancer remains the most prevalent form of cancer in the United States, and non-small cell lung cancer (NSCLC) is its most common type. While traditional treatments often offer limited success, immune checkpoint inhibitors (ICIs) have revolutionized cancer care by effectively harnessing the body’s immune system to fight cancer. However, responses to these therapies can vary among patients. This research explores the potential of machine learning to predict which NSCLC patients will respond to ICIs by analyzing both genetic and clinical data before treatment begins. Conducted with data from the Georgetown-Lombardi Comprehensive Cancer Center, we developed a model that significantly outperforms traditional methods like PD-L1, demonstrating potential as a valuable supplementary prognostic tool. Our model integrates a variety of patient data to predict responses to immunotherapy, ranging from genetic markers to simple clinical indicators like blood pressure and blood test results. This study also identifies key factors that could guide more personalized treatment plans. The result underscores the power of machine learning in enhancing the prediction and effectiveness of cancer treatments and creates more opportunities for more targeted therapy options that could improve survival rates and quality of life for cancer patients. © The Author(s) 2025. |
