Serum discrimination of early-stage lung cancer patients using electrospray-ionization mass spectrometry

Abstract

The goal of this study was to evaluate the usefulness of electrospray ionization-mass spectrometry (ESI-MS) technology to distinguish sera of early-stage lung cancer patients from control individuals. ESI-MS m/z (mass divided by charge) data were generated from sera of 43 non-small cell lung cancer patients (pathological stages I and II) and 21 control individuals. Identifications of m/z peak area significances between cancer and control ESI-MS sera spectra were performed using t-tests. A "leave one out" cross validation procedure, which mimics blinded sera analysis and corrects for "over-fitting" of data, yielded discriminatory cancer versus control distribution p value and ROC curve area value of <0.001 and 0.87, respectively. Analysis without the "leave one out" cross validation procedure yielded a ROC curve area of 0.99 for discrimination of sera from lung cancer patients versus control individuals. Predictive value measurements revealed overall test efficiency and sensitivity for distinguishing sera from lung cancer patients from controls (using "leave one out" cross validation) of 80% and 84%, respectively. ESI-MS serum analysis between control individuals and lung cancer patients who smoked or did not smoke had p values in ranges indicating that smoking effects are not pronounced in our analysis. These studies indicate that ESI-MS analyses of sera from early stage non-small cell lung cancer patients were helpful in distinguishing these patients from control individuals.