No evidence of histology subtype-specific prognostic signatures among lung adenocarcinoma and squamous cell carcinoma patients at early stages
Background Non-small cell lung cancer (NSCLC) is the predominant histological type of lung cancer, accounting for up to 85% of cases. Disease stage is commonly used to determine adjuvant treatment eligibility of NSCLC patients, however, it is an imprecise predictor of the prognosis of an individual patient. Currently, many researchers resort to microarray technology for identifying relevant genetic prognostic markers, with particular attention on trimming or extending a Cox regression model. Among NSCLC, adenocarcinoma (AC) and squamous cell carcinoma (SCC) are two major histology subtypes. It has been demonstrated that there exist fundamental differences in the underlying mechanisms between them, which motivated us to postulate there might exist specific genes relevant to prognosis of each histology subtype. Results In this article, we propose a simple filterer feature selection algorithm with a Cox regression model as the base. Applying this method to a real-world microarray data, no evidence has been found to support the existence of histology-specific prognostic gene signature. Nevertheless, a 31-gene prognostic gene signature for the early-stage AC and SCC samples is obtained, which provides comparable performance when compared with other relevant signatures. Conclusions Our proposal is conceptually simple and straightforward to implement. Therefore, it is expected that other researchers, especially those with less statistical knowledge and experience, can adapt this method readily to test their own research hypotheses.