Background: Acute-on-chronic liver failure (ACLF) is a disease characterized by systemic inflammatory response with high mortality. Nowadays, there is no prediction model for its long-term prognosis. We aimed to establish and validate a prognostic prediction model incorporating inflammation indexes to forecast the long-term prognosis of patients with hepatitis B-related ACLF (HBV-ACLF). Methods: Retrospective analysis of 986 patients’ clinical data with HBV-ACLF from Third Affiliated Hospital of Sun Yat-sen University between January 2014 and December 2018 were conducted. The patients were randomly divided into the training cohort (690 cases) and the validation cohort (296 cases) according to the ratio of 7:3. LASSO and Cox regression analysis were used to determine the independent risk factors for long-term mortality. Results: The following variables were identified: age, cirrhosis, hepatic encephalopathy, total bilirubin (TBIL), international normalized ratio (INR), Monocyte to lymphocyte ratio (MLR), and Neutrophil to platelet ratio (NPR), and a new nomogram was constructed to predict the survival rate of 1 -month, 3-month, and 12-month by weighting the scores of each variable. The C-index was 0.777 (95%CI 0.752-0.802), and the AUC was 0.829 (95%CI 0.798-0.859) in the training cohort. The predictive value of the nomogram demonstrated a superior ability to predict long-term survival rate compared to MELD score (0.767, 95% CI: 0.730-0.804, P<0.001), and COSSH-ACLF II score (0.807, 95%CI: 0.774-0.840, P=0.028). Evaluation using calibration curves and decision curve analysis (DCA) suggested its practical utility. Conclusions: The novel inflammation scoring system, including MLR and NPR, can well predict long-term mortality in HBV- ACLF patients.