Bioinformatics Methods for Learning Radiation-Induced Lung Inflammation from Heterogeneous Retrospective andProspective Data

[摘要] Radiotherapy outcomes are determined by complex interactions between physical andbiological factors, reflecting both treatment conditions and underlying genetics. Recentadvances in radiotherapy and biotechnology provide new opportunities and challenges forpredicting radiation-induced toxicities, particularly radiation pneumonitis (RP), in lungcancer patients. In this work, we utilize datamining methods based on machine learningto build a predictive model of lung injury by retrospective analysis of treatment planningarchives. In addition, biomarkers for this model are extracted from a prospective clinicaltrial that collects blood serum samples at multiple time points. We utilize a 3-wayproteomics methodology to screen for differentially expressed proteins that arerelated to RP. Our preliminary results demonstrate that kernel methods can capturenonlinear dose-volume interactions, but fail to address missing biological factors. Ourproteomics strategy yielded promising protein candidates, but their role in RP as well astheir interactions with dose-volume metrics remain to be determined.

[学科分类] 基础医学