Lung cancer is one of the leading causes of cancer related deaths in the United States. A novel volatile analysis platform is needed to complement current diagnostic techniques and better elucidate chemical signatures of lung cancer and subsequent treatments. A systems biology bottom-up approach using cell culture volatilomics was employed to identify pathological volatile fingerprints of lung cancer in real time. An advanced secondary electrospray ionization (SESI) source, named SuperSESI was used in this study and directly attached to a Thermo Q-Exactive high-resolution mass spectrometer (HRMS). We performed a series of experiments to determine if our optimized SESI-HRMS platform can distinguish between cancer types by sampling their in vitro volatilome profiles. We detected 60 significant volatile organic compound (VOC) features in positive mode that were deemed of cancer cell origin. The cell derived features were used for subsequent analyses to distinguish between our two studied lung cancer types, non-small cell lung cancer (NSCLC) and small cell lung cancer (SCLC). Partial least squares-discriminant analysis (PLS-DA) model revealed a good separation of the two cancer types, suggesting unique chemical composition of their headspace profiles. A receiver operating characteristic (ROC) curve using 10 prominent features was used to predict disease type, with an area under the curve (AUC) of 0.811. Cultures were also treated with cisplatin to determine the feasibility of classifying drug treatment from expelled gases. A PLS-DA model revealed independent clustering based on their headspace profiles. An ROC curve using the top features driving separation of PLS-DA model suggested good accuracy with an AUC of 1. It is thus possible to benefit from the advantages of this platform to distinguish the unique volatile fingerprints of cancers to uncover potential biomarkers for cancer type differentiation and treatment monitoring.