作者: Helena Liljedahl , Anna Karlsson , Gudrun N. Oskarsdottir , Annette Salomonsson , Hans Brunnström
DOI: 10.1002/IJC.33242
关键词:
摘要: Disease recurrence in surgically treated lung adenocarcinoma (AC) remains high. New approaches for risk stratification beyond tumor stage are needed. Gene expression-based AC subtypes such as the Cancer Genome Atlas Network (TCGA) terminal-respiratory unit (TRU), proximal-inflammatory (PI) and proximal-proliferative (PP) have been associated with prognosis, but show methodological limitations robust clinical use. We aimed to derive a platform independent single sample predictor (SSP) molecular subtype assignment that could function setting. Two-class (TRU/nonTRU=SSP2) three-class (TRU/PP/PI=SSP3) SSPs using AIMS algorithm were trained 1655 ACs (n = 9659 genes) from public repositories vs TCGA centroid subtypes. Validation survival analysis performed 977 patients overall (OS) distant metastasis-free (DMFS) endpoints. In validation cohort, SSP2 SSP3 showed accuracies of 0.85 0.81, respectively. captured relevant biology previously prognosis. analysis, OS DMFS cases discordantly classified between favored classification. resected Stage I patients, identified TRU-cases better (hazard ratio [HR] 0.30; 95% confidence interval [CI] 0.18-0.49) (TRU HR 0.52; CI 0.33-0.83) age, IA/IB gender. was transformed into NanoString nCounter assay tested 44 RNA formalin-fixed tissue, providing prognostic (relapse-free interval, 3.2; 1.2-8.8). conclusion, gene can provide information early-stage ACs. may overcome critical applicability signatures cancer.