Stratification of resectable lung adenocarcinoma by molecular and pathological risk estimators
Background: Mortality in early stage, resectable lung cancer is sufficiently high to warrant consideration of post-surgical treatment. Novel markers to stratify resectable lung cancer patients may help with the selection of treatment to improve outcome.
Methods: Primary tumour tissue from 485 patients, surgically treated for stage I-II lung adenocarcinoma, was analysed for the RNA expression of 31 cell cycle progression (CCP) genes by quantitative polymerase chain reaction (PCR). The expression average, the CCP score, was combined with pathological stage into a prognostic score (PS). Cox proportional hazards regression assessed prediction of 5-year lung cancer mortality above clinical variables. The PS threshold was tested for risk discrimination by the Mantel-Cox log-rank test.
Results: The CCP score added significant information above clinical markers (all patients, P = 0.0029; stage I patients, P = 0.013). The prognostic score was a superior predictor of outcome compared to pathological stage alone (PS, P = 0.00084; stage, P = 0.24). Five-year lung cancer mortality was significantly different between the low-risk (90%, 95% confidence interval (CI) 81-95%), and high-risk groups (65%, 95% CI 57-72%), P = 4.2 x 10(-6)).
Conclusions: The CCP score is an independent prognostic marker in early stage lung adenocarcinoma. The prognostic score provides superior risk estimates than stage alone. The threefold higher risk in the high-risk group defines a subset of patients that should consider therapeutic choices to improve outcome.