BMC Cancer (Dec 2023)

Prognostic implications of alpha-fetoprotein and C-reactive protein elevation in hepatocellular carcinoma following resection (PACE): a large cohort study of 2770 patients

  • Kong-Ying Lin,
  • Qing-Jing Chen,
  • Shi-Chuan Tang,
  • Zhi-Wen Lin,
  • Jian-Xi Zhang,
  • Si-Ming Zheng,
  • Yun-Tong Li,
  • Xian-Ming Wang,
  • Qiang Lu,
  • Jun Fu,
  • Luo-Bin Guo,
  • Li-Fang Zheng,
  • Peng-Hui You,
  • Meng-Meng Wu,
  • Ke-Can Lin,
  • Wei-Ping Zhou,
  • Tian Yang,
  • Yong-Yi Zeng

DOI
https://doi.org/10.1186/s12885-023-11693-6
Journal volume & issue
Vol. 23, no. 1
pp. 1 – 12

Abstract

Read online

Abstract Background Routine clinical staging for hepatocellular carcinoma (HCC) incorporates liver function, general health, and tumor morphology. Further refinement of prognostic assessments and treatment decisions may benefit from the inclusion of tumor biological marker alpha-fetoprotein (AFP) and systemic inflammation indicator C-reactive protein (CRP). Methods Data from a multicenter cohort of 2770 HCC patients undergoing hepatectomy were analyzed. We developed the PACE risk score (Prognostic implications of AFP and CRP Elevation) after initially assessing preoperative AFP and CRP’s prognostic value. Subgroup analyzes were performed in BCLC cohorts A and B using multivariable Cox analysis to evaluate the prognostic stratification ability of the PACE risk score and its complementary utility for BCLC staging. Results Preoperative AFP ≥ 400ng/mL and CRP ≥ 10 mg/L emerged as independent predictors of poorer prognosis in HCC patients who underwent hepatectomy, leading to the creation of the PACE risk score. PACE risk score stratified patients into low, intermediate, and high-risk groups with cumulative 5-year overall (OS) and recurrence-free survival (RFS) rates of 59.6%/44.9%, 43.9%/38.4%, and 20.6%/18.0% respectively (all P < 0.001). Increased PACE risk scores correlated significantly with early recurrence and extrahepatic metastases frequency (all P < 0.001). The multivariable analysis identified intermediate and high-risk PACE scores as independently correlating with poor postoperative OS and RFS. Furthermore, the PACE risk score proficiently stratified the prognosis of BCLC stages A and B patients, with multivariable analyses demonstrating it as an independent prognostic determinant for both stages. Conclusion The PACE risk score serves as an effective tool for postoperative risk stratification, potentially supplementing the BCLC staging system.

Keywords