Preoperative Inverse Albumin-to-Globulin Ratio Predicts Worse Oncologic Prognosis Following R0 Resection for Hilar Cholangiocarcinoma

Zhi-Peng Liu,Zheng-Jun Cheng,Hai-Su Dai,Wei-Yue Chen,Xing-Chao Liu,Hai-Ning Fan,Yu Pan,Jie Bai,Yan Jiang,Yan-Qi Zhang,Shi-Quan Deng,Zhi-Yu Chen
DOI: https://doi.org/10.21203/rs.3.rs-1327409/v1
2022-01-01
Abstract:Abstract Background & Aims: Inverse albumin-to-globulin ratio (IAGR) is defined as an AGR in serum of less than 1, which commonly indicates poor protein synthesis and severe inflammation. This study aimed to evaluate whether preoperative IAGR is associated with worse oncologic survival after R0 resection for hilar cholangiocarcinoma (HCCA).Methods: Patients who underwent R0 resection for newly diagnosed HCCA from January 2011 to December 2018 were enrolled from a multicenter database. Based on their preoperative AGR levels, all patients were divided into the IAGR and NAGR groups. We compared their clinical characteristics and long-term oncologic prognosis. Univariate and multivariate Cox regression analyses were performed to identify risk factors for overall survival (OS) and recurrence-free survival (RFS). The time-dependent area under the curve (AUC) was compared to determine the predictive ability of AGR and other liver function indicators for long-term oncologic prognosis.Results: Overall, 323 patients with R0 resection of HCCA were included; 58 (18.0%) were in the IAGR group. The 5-year OS and RFS rates in the IAGR group were 6.3% and 2.5%, which were significantly worse than those in the NAGR group (24.5% and 21.2%, P < 0.05). Multivariate Cox regression analyses identified IAGR an independent risk factor for predicting worse OS (HR: 1.619, 95% CI: 1.161-2.258, P = 0.005) and RFS (HR: 1.773, 95% CI: 1.289-2.438, P < 0.001). The time-dependent AUCs of AGR for OS and RFS were superior to those of other liver function indicators.Conclusion: Preoperative IAGR is valuable for predicting worse OS and RFS in patients who underwent R0 resection for HCCA. AGR has ideal predictive ability for long-term oncological prognosis.
What problem does this paper attempt to address?