Machine Learning for Predicting and Managing Quality of Life in Lung Cancer Immunotherapy Patients

Lung cancer is the leading cause of cancer incidence and mortality in China, and it holds the same position in the United States. Non-small cell lung cancer (NSCLC) is the most common histological type, accounting for approximately 85% of lung cancer cases. Treatment strategies based on pathology, molecular subtyping, and clinical staging include surgery, radiotherapy, chemotherapy, targeted therapy, and immunotherapy. In recent years, immunotherapy has been extensively researched and applied in lung cancer treatment. It works by blocking the binding of PD-L1 on tumor cells to PD-1 on T cells, thereby releasing the inhibition of T cell function and killing the tumor cells. Immunotherapy has become the standard treatment for advanced NSCLC without driver mutations, and it covers the entire spectrum of non-surgical locally advanced NSCLC consolidation therapy, perioperative neoadjuvant, and adjuvant therapy for early-stage NSCLC. However, not all patients benefit from immunotherapy, with only a small subset experiencing clinical benefit. Therefore, identifying resistance mechanisms, selecting populations that benefit from treatment, and overcoming therapy resistance are complex and challenging clinical issues that require collaboration among basic, translational, and clinical oncology research teams.

In 1993, the World Health Organization (WHO) introduced the concept of Quality of Life (QoL), which refers to an individual's perception of their position in life within their cultural and value system, relating to their goals, expectations, standards, and concerns. Few studies focus on cancer patients' QoL, particularly those using patient-reported outcomes (PRO) as a primary endpoint. Most clinical trials for cancer drugs use PROs as secondary or exploratory endpoints. There is limited research that considers PROs as the primary endpoint. Therefore, it is essential to further investigate the relationship between cancer patients' health-related quality of life and prognosis, as well as its relevance to immunotherapy. This would facilitate better early identification of immune-related adverse events and systematic management, improving treatment adherence, QoL, and ensuring optimal treatment outcomes.

This project aims to develop a risk warning model for health-related quality of life in lung cancer patients receiving immunotherapy based on machine learning. By using cluster analysis, the study will clean, validate, and analyze the health-related quality of life data from the QLQ-C30 and QLQ-LC13 questionnaires from clinical trials available on the Vivli Global Clinical Research Data Sharing and Analysis Platform. The goal is to identify the distribution characteristics of these data and explore whether patient-reported outcomes can predict the efficacy of immunotherapy, thus serving as biomarkers to identify potential beneficiaries of immunotherapy. Furthermore, based on a risk warning and stratified management approach, the project aims to design appropriate symptom intervention strategies for different PRO types in immunotherapy patients, ultimately helping healthcare providers better understand the symptom burden that lung cancer patients may experience during immunotherapy and offering practical guidance for symptom management.

• ORR (Objective Response Rate) [Time frame: Two weeks after the intervention]
• PFS (Progression-Free Survival) [Time frame: Two weeks after the intervention]