Virtual Library
Start Your Search
T. Fujiwara
Author of
-
+
P3.04 - Poster Session/ Biology, Pathology, and Molecular Testing (ID 235)
- Event: WCLC 2015
- Type: Poster
- Track: Biology, Pathology, and Molecular Testing
- Presentations: 1
- Moderators:
- Coordinates: 9/09/2015, 09:30 - 17:00, Exhibit Hall (Hall B+C)
-
+
P3.04-024 - Lung Adenocarcinoma with Neuroendocrine Feature Revealed by Transcriptome Profiling (ID 881)
09:30 - 09:30 | Author(s): T. Fujiwara
- Abstract
Background:
Although our previous transcriptomic analyses revealed that a subgroup in lung adenocarcinoma with a neuroendocrine feature exhibits poor prognosis, the link of the phenotype with patient outcomes has been limited only to two populations derived from a Japanese and a US institution. Here we performed additional transcriptomic profiling analyses to elucidate whether our method was useful also to other populations.
Methods:
Seven independent web-based datasets of lung adenocarcinoma, either on expression microarrays or on an RNA-seq platform, were examined. The expression level of the ASCL1 geneset (100 probes closely correlated with ASCL1 expression) was used to define the neuroendocrine character based on the method we previously reported. Subtyping was performed by consensus clustering with non-negative matrix factorization. Correlation of overall survival was analyzed with the Kaplan-Meier method.
Results:
The neuroendocrine subtype was identified from each of seven independent cohorts with 45, 90, 117, 183,196, 443 and 548 lung adenocarcinoma samples. Among them, three datasets showed statistically significant association with patient survival (p<0.05). The neuroendocrine subtype was inversely correlated with expression of ubiquitination genes. Somatic mutations identified in the neuroendocrine subtype with the TCGA data were common ones such as TP53, STK11 and KRAS.
Conclusion:
Transcriptomic profiling partially reproduced the neuroendocrine subtype in lung adenocarcinoma samples derived from the independent datasets.
