A Study on Predictive Models and Clinical Outcome of Radiation Pneumonitis

1. Collect clinical information, CT images, and peripheral blood of the lung cancer patients treated with thoracic radiotherapy in Tongji Hospital, Hubei Cancer Hospital, and Jingjiang People's Hospital. 2. Follow up the enrolled patients. All patients enrolled in this study are examined during and one month after radiotherapy. Then, the patients are followed every three months for the first year and every six months thereafter. At each follow-up visits, all patients are asked to undergo a chest CT, and information including survival status, symptoms, CT images, and treatment is collected. Radiation pneumonitis and other toxicities induced by radiotherapy are graded by two radiation oncologists according to the Common Terminology Criteria for Adverse Events 4.0 (CTCAE4.0). 3. Detect serum proteins, gene expression profile, single-nucleotide polymorphisms, and epigenetic modifications that may be associated with radiation pneumonitis. 4. Screen biomarkers that are associated with radiation pneumonitis via univariate and multivariate Cox regression analysis. 5. Construct a predictive model of radiation pneumonitis based on clinical information, radiomics, and biomarkers via machine learning or Least absolute shrinkage and selection operator. 6. Use Kaplan-Meier and Cox model to analyze the association of radiation pneumonitis with survival and efficacy of antitumor treatment. 7. Identify biomarkers and predictors of other toxicities induced by radiotherapy including radiation esophagitis, cardiotoxicity and radiodermatitis.

• Other injuries induced by radiotherapy [Срок оценки: 5 years after radiotherapy]