Back to Search
Start Over
Abstract 5000: A stromal liver gene signature predictive of HCC risk across all liver disease etiologies
- Source :
- Cancer Research. 77:5000-5000
- Publication Year :
- 2017
- Publisher :
- American Association for Cancer Research (AACR), 2017.
-
Abstract
- Background & Aims: HCC risk-predictive biomarkers will enable identification of patients who most need close monitoring and cancer preventive intervention. We previously reported a stromal liver gene signature predictive of HCC risk in multiple independent patient cohorts mainly affected with HCV (Hoshida 2008, Hoshida 2013, King 2014). We aimed to determine whether the gene signature is similarly prognostic in HBV, alcohol abuse, and NAFLD/NASH patients. Methods: The liver gene signature was analyzed using an FDA-approved diagnostic platform (NanoString), and an independent cohort of curatively resected early-stage HCC patients (n=223, median follow up 4.5 years, HCV [n=67, 30%], HBV [n=39, 17%], alcohol [n=51, 23%], NAFLD [n=66, 30%], and NASH [n=52, 23%, a subset of 66 NAFLD patients]) were analyzed. Complete surgical resection was radiologically and histologically confirmed, and HCC recurrence in the cohort was confirmed to be mainly late recurrence, i.e., de novo carcinogenesis, based on a recurrence hazard plot. Prognostic prediction was performed by using a previously developed model without making any modification, and evaluated for association with de novo HCC recurrence in multivariable modeling and subgroup analyses. Results: The gene signature assay classified the patients into high- (n=49, 22%), intermediate- (n=127, 57%), and low- (n=47, 21%) risk groups. The high-risk prediction was associated with higher HCC recurrence (HR=4.8, p&lt0.001) as well as overall death (HR=4.8, p&lt0.001) in multivariable modeling. The association with HCC risk remained significant in subgroups of patients with HCV (HR=3.0, p=0.028), HBV (HR=6.1, p&lt0.001), alcohol (HR=6.1, p=0.023), and NASH (HR=10.9, p=0.03) in multivariable models including cirrhosis and AJCC stage 2/3. The association was diminished in NAFLD patients (HR=3.02, p=0.16), suggesting that presence of active inflammation and/or fibrosis is critical for the signature to predict HCC risk in NAFLD. Established cirrhosis was less frequent in NAFLD/NASH (23%/21%) compared to other etiologies (50%). Conclusions: The prognostic liver gene signature was implemented in a clinically applicable assay, and successfully validated for de novo HCC recurrence after curative resection of early-stage HCC in all major etiologies of chronic liver disease and HCC. The assay will enable prioritization of patients for HCC surveillance and enrichment of high risk patients for more cost-effective clinical trials of new HCC chemopreventive therapies. Citation Format: Shigeki Nakagawa, Naoki Umesaki, Takanobu Yamao, Yuki Kitano, Kensuke Yamamura, Kota Arima, Tatsunori Miyata, Takayoshi Kaida, Yujin Hoshida, Katsunori Imai, Daisuke Hashimoto, Yo-ichi Yamashita, Akira Chikamoto, Hideo Baba. A stromal liver gene signature predictive of HCC risk across all liver disease etiologies [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5000. doi:10.1158/1538-7445.AM2017-5000
Details
- ISSN :
- 15387445 and 00085472
- Volume :
- 77
- Database :
- OpenAIRE
- Journal :
- Cancer Research
- Accession number :
- edsair.doi...........eb2f252c2159b8b7257e99e4d14329f8
- Full Text :
- https://doi.org/10.1158/1538-7445.am2017-5000