Identification of candidate molecular markers predicting chemotherapy resistance in non-small cell lung cancer

Empirical chemotherapy for patients with non-small cell lung cancer (NSCLC) is recommended, even without knowledge of the tumor's specific biological characteristics, and many patients may not benefit. The goal of this study was to identify potential serum biomarkers that influence resistance to chemotherapy, and to build a model that could be used to predict resistance to chemotherapy of patients with advanced NSCLC.

A total of 97 NSCLC patients were classified as stage IIIB and stage IV. The chemotherapy regimen was cisplatin plus docetaxel. All patients received two cycles of chemotherapy. Serum protein profiles were detected using surface-enhanced laser desorption/ionization time-of-flight mass spectrometry (SELDI-TOF MS) and the spectra were analyzed with a support vector machine (SVM).

For the 93 eligible patients, 22 patients had a partial response (23.7%); 27 patients had stable disease (29.0%) and 44 (47.3%) had progressive disease. One hundred and twenty-eight mass peaks were detected from the chemotherapy sensitive group and the chemotherapy resistant group by receiver operator characteristic curve. The top 10 peaks with the highest area under the curve values were selected, randomly combined, and fed into the SVM. The SVM model with the highest accuracy was used as the diagnostic model. The model constructed using five protein peaks with mass/charge ratios of 3955 Da, 6207 Da, 7992 Da, 9214 Da, and 15,086 Da separated the chemotherapy resistant group from the chemotherapy sensitive group with a sensitivity of 83.3% and specificity of 85.7%.

SELDI-TOF MS may provide a useful means in the search for serum biomarkers for predicting chemotherapy resistance in patients with advanced NSCLC.