Estimation of the Survival of Patients With Lung Squamous Cell Carcinoma Using Genomic Copy Number Aberrations

Abstract

BackgroundEstimation of the survival of patients with lung squamous cell carcinoma (SCC) on the basis of histopathology is inadequate. The aim of this study was to identify genomic regions with potential value for estimating the prognosis of these patients. Patients and MethodsDepending on their survival time, 100 patients with primary lung SCC were separated into high- or low-risk prognostic groups, and their copy number aberrations (CNAs) were analyzed using array-comparative genomic hybridization (array-CGH). ResultsWe identified 123 CNA regions that were significantly associated with survival. Among these regions, some have been reported previously (eg, amplifications of 8p12, 3q27.1, and loss of 9p21.3 and 13q34) but others have never been reported. For example, gains of 3q27.1, 5p13.2, and 5p13.3 were found to be associated with a favorable prognosis, but patients harboring gains of 11q23.3, 11q13.1, and 14q32.3, and deletions of 3p21.3 and 9p21.3 tended to have poor survival. Among the 123 CNA regions, 41 were further selected to construct a survival estimation model that could effectively separate SCC patients into high- or low-risk groups with an accuracy of 92%, sensitivity of 90%, and specificity of 94%. The results of the array-CGH were further validated in an independent cohort of 45 formalin-fixed, paraffin-embedded specimens using real-time polymerase chain reaction. ConclusionA number of CNA regions were found to be associated with the survival of SCC patients, and we were able to construct a model to estimate prognosis on the basis of these regions. Assessment of these CNAs could potentially assist in clinical decision-making regarding adjuvant therapy after surgery.