Prognostic Prediction of Pemetrexed-Platinum Chemotherapeutic Regimen for NSCLC by Serum Metabolomics.

Jie Wang,Yanhua Tian,Zhijie Wang,Hua Bai,Jianchun Duan
DOI: https://doi.org/10.1200/jco.2017.35.15_suppl.e20603
IF: 45.3
2017-01-01
Journal of Clinical Oncology
Abstract:e20603 Background: Pemetrexed-platinum chemotherapeutic doublet has become a standard schedule as the first-line treatment for advanced non-squamous NSCLC due to its promising efficacy and a favorable toxicity profile. However, chemotherapy resistance inevitably limits the therapeutic effect of this preferred chemotherapeutic regimen, especially for these intrinsically resistant patients. Development of predictive biomarkers to identify those patients who can benefit from this pemetrexed-based regimen has great significance for personalized chemotherapy. Methods: Total 354 patients who were diagnosed with inoperable stage IIIB or IV lung adenocarcinoma (LADC), assigned to receive pemetrexed-platinum doublet chemotherapy as the first-line treatment were enrolled into this prospective study, including 251 cases diagnosed ahead as training set for predictive model establishment and the subsequently diagnosed 103 cases into validation set. Ultraperformance liquid chromatography coupled to mass spectrometry (UPLC-MS) analytical platform were applied to perform the metabolic profiling studies of pre-treatment serum samples. Results: In the training set, eight potential biomarkers, comprising of hypotaurine, taurine, choline, betaine, dimethylglycine, uridine, dodecanoylcarnitine and L-palmitoylcarnitine, were discovered to be differentially expressed between the patients in disease progressive group (PD, n = 53) and disease control group (including CR, PR and SD, n = 198). In the validation set (PD vs non-PD: 23 vs 80), targeted quantitative analysis of these metabolic biomarkers revealed similar levels as that observed in the training set. Logistic regression model was accordingly established based on this eight-metabolic biomarker panel, to discriminate patients who might benefit from this chemotherapy regimen with over 90% accuracy. Conclusions: This prospective study identified blood-based biomarker panel by metabonomics analysis could predict the clinical outcome of pemetrexed-platinum chemotherapy prior to treatment for LADC patients, which needed to be validated by clinical trials of large samples.
What problem does this paper attempt to address?