Clinical Significance of a Gene Signature Generated from Tumor Budding Grade in Colon Cancer
| Autor: | Hitoshi Tsuda, Hideki Ueno, Eiji Shinto, Takehiro Shiraishi, Kazuo Hase, Koichi Okamoto, Yuichiro Yoshida, Masato Yamadera, Ken Nagata, Satsuki Mochizuki, Yoshiki Kajiwara, Yoji Kishi |
|---|---|
| Rok vydání: | 2020 |
| Předmět: |
Oncology
Candidate gene medicine.medical_specialty animal structures Colorectal cancer Anion Transport Proteins Antiporters Disease-Free Survival 03 medical and health sciences 0302 clinical medicine Tumor budding Surgical oncology Internal medicine Humans Medicine Clinical significance Neoplasm Staging business.industry Microarray analysis techniques Hazard ratio Prognosis medicine.disease Mesothelin 030220 oncology & carcinogenesis Colonic Neoplasms 030211 gastroenterology & hepatology Surgery Neoplasm Recurrence Local DNA microarray business |
| Zdroj: | Annals of Surgical Oncology. 27:4044-4054 |
| ISSN: | 1534-4681 1068-9265 |
| DOI: | 10.1245/s10434-020-08498-3 |
| Popis: | Tumor budding, a microscopic finding of dedifferentiation at the invasive margin, has been reported as a definite prognostic marker in colon cancer (CC). Herein, we aimed to generate a molecular budding signature (MBS) based on DNA microarray data and to examine its prognostic significance. Frozen tissue samples from 85 patients with stage II/III CC were used for DNA microarray analyses. First, we selected candidate genes that were differentially expressed (twofold change) between the invasive frontal regions and corresponding tumor centers of three extremely high-grade budding tumors. Subsequently, using microarray data from whole-tissue sections of the 85 patients, we selected MBS-constituent genes from the candidates based on correlation to the pathological budding grade. The MBS score was calculated using the sum of the logarithm of the expression of each gene. We selected seven MBS-constituent genes: MSLN, SLC4A11, WNT11, SCEL, RUNX2, MGAT3, FOXC1. A comparison of relapse-free survival (RFS) rates revealed a significant impact of the MBS score [5-year RFS, 77.4% (score-high) vs. 95.1% (score-low); P = 0.044]. Analyses of public databases revealed that low MBS score patients exhibited better prognosis than those with high-score cancers (GSE14333: 5-year RFS, 83.1% vs. 66.6%, P = 0.028; GSE39582: 5-year disease-free survival, 72.2% vs. 58.1%, P = 0.0005). Multivariate analysis revealed that the MBS score is an independent prognostic indicator in GSE39582 (hazard ratio, 1.611; P = 0.013). We developed a new gene classification method, i.e., MBS, and demonstrated its clinical relevance as an indicator of high recurrence risk of CC. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink