Surgical resection only remains the standard choice to treatment for early stage I non-small lung cancer (NSCLC) patients. Preliminary studies suggest the application of adjuvant chemotherapy with surgery (ACT) for more severe NSCLC patients is associated with higher prognosis compared to those that underwent surgical resection only. However, at an individual level, not all patients may benefit from chemotherapy. Given the immense personal and financial costs associated with ACT, finding those patients that are most likely to benefit from ACT is paramount. Thus, the purpose of this research is to utilize gene expression and clinical data from lung cancer patients to develop a statistical decision support system to identify subgroups of early stage lung cancer patients that are most likely to benefit from ACT treatment or surgical resection only. Cox regression models are trained using data from a randomized control trial from the National Cancer Institute utilizing explicit treatment interaction terms. To handle high dimensions inherent in gene expression data, LASSO and ridge penalties are applied to find the most significant interacting covariates. Then the risk scores constructed from the models may be used to stratify patients according to a high risk or low risk group respective to ACT treatment. After applying the model to an independent set, our methods show that patients that underwent the treatment according to their risk group exhibited slightly higher survival than those that did not.