1. Combination Assay of Methylated HOXA1 with Tumor Markers Shows High Sensitivity for Detection of Early-Stage Hepatocellular Carcinoma.
- Author
-
Kunimune, Yuki, Suehiro, Yutaka, Saeki, Issei, Yamauchi, Yurika, Tanabe, Norikazu, Matsumoto, Toshihiko, Higaki, Shingo, Fujii, Ikuei, Suzuki, Chieko, Okayama, Naoko, Nishioka, Mitsuaki, Ichihara, Kiyoshi, Nagano, Hiroaki, Sakaida, Isao, Takami, Taro, and Yamasaki, Takahiro
- Subjects
LOGISTIC regression analysis ,TUMOR markers ,HEPATOCELLULAR carcinoma ,LIVER diseases ,ALPHA fetoproteins - Abstract
Introduction: Patients with hepatitis virus-related hepatocellular carcinoma (viral HCC) are decreasing as hepatitis control improves, but those with non-viral-related HCC (non-viral HCC) are increasing in Japan. No established surveillance system exists for patients with non-viral HCC, so they are often diagnosed at an advanced stage. To address this, we performed this study. Methods: We collected serum samples from 516 participants (154 healthy subjects, 93 chronic liver disease [CLD] patients without HCC, and 269 HCC patients). Participants were divided into a control group comprising healthy subjects and patients with CLD and an HCC group. We evaluated serum methylated HOXA1 (m-HOXA1) copy numbers using modified combined restriction digital PCR (CORD) assay (1-step CORD assay). We assessed diagnostic performance of m-HOXA1 compared to HCC tumor markers alpha-fetoprotein (AFP) and des-gamma-carboxy prothrombin (DCP) and created a novel index to improve HCC prediction. Results: Serum m-HOXA1 level was significantly higher in each HCC stage group versus the control group. Its sensitivity was 69.1% and specificity was 78.5% for diagnosing HCC. The area under the curve (AUC) of m-HOXA1 was superior to that of AFP and equal to that of DCP. Multivariate logistic regression analysis revealed independent contributions of m-HOXA1, DCP, and AFP, in that order of strength, to diagnose HCC after adjustment for age and sex. We designated the predictive probability of HCC based on the regression model as the ASDAm-H1 (Age, Sex, DCP, AFP, and m-HOXA1) index. Its diagnostic accuracy was 0.96 by AUC with a sensitivity of 86.2% and specificity of 93.9%. Sensitivity was identical for viral and non-viral HCCs. When limited to early-stage HCC, sensitivity of the ASDAm-H1 index was 76.3%. Conclusions: We showed distinguished performance of the ASDAm-H1 index to detect viral and non-viral HCC, even at an early stage. This index might have potential as a non-viral HCC surveillance system. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF