This study aimed to identify independent prognostic factors for patients with concurrent hepatitis B virus (HBV)-related hepatocellular carcinoma (HCC) and cancer-related anemia (CRA), and to investigate the impact of traditional Chinese medicine (TCM) on patient outcomes. Furthermore, the objective was to determine the patient subset most likely to benefit from TCM intervention for guiding clinical practice.

A total of 532 patients diagnosed with coexisting HBV-related hepatocellular carcinoma and cancer-related anemia at Beijing Ditan Hospital between June 2008 and June 2017 were included in this retrospective analysis. The cohort comprised 318 patients who received TCM treatment and 214 patients who underwent exclusive modern medical therapy. To evaluate the association between TCM intervention and patient survival, multivariable Cox regression analysis was employed to estimate hazard ratios (HRs) while adjusting for potential confounding factors. Concurrently, Kaplan-Meier survival curves were generated to visualize survival distributions, and dendrograms were utilized to explore hierarchical relationships among prognostic factors and stratify mortality risks in this patient population. Building upon the observed synergistic interaction between C-reactive protein (CRP) and alpha-fetoprotein (AFP) in predicting clinical outcomes, this study further developed a decision tree model. This model systematically compares prognostic outcomes across subgroups defined by different thresholds of these biomarkers, providing a practical tool for risk stratification and treatment decision-making in clinical practice.

Cox regression analysis revealed that Traditional Chinese Medicine served as an independent protective factor for 1-year survival in this patient cohort. Patients who received TCM treatment for more than six months demonstrated a significantly higher survival rate compared to those treated with TCM for less than six months. Furthermore, AFP and CRP were identified as independent risk factors for 1-year survival within the TCM treatment group. The subgroup characterized by low levels of both AFP (≤350 ng/mL) and CRP (≤5 mg/L)—designated as the LALC subgroup—exhibited the most favorable survival outcomes. A decision tree model constructed based on CRP, AFP, platelet count (PLT), and hemoglobin (HGB) provided an intuitive tool for prognostic prediction. Among these variables, CRP level demonstrated a particularly strong association with patient outcomes: patients with CRP ≤5 mg/L achieved a 1-year survival rate of 87%, constituting 47% of the study population. Additional indicators positively influencing prognosis in patients with HCC complicated by anemia included AFP ≤350 ng/mL, PLT ≤100×10⁹/L, and HGB <90 g/L. Patients meeting all these criteria collectively achieved a 1-year survival rate of 60%, representing 11% of the study cohort. Radar chart analysis illustrated that across subgroups with varying durations of TCM treatment, the proportion of the LALC subgroup increased progressively with extended TCM use. This pattern suggests that the mechanism by which TCM improves patient prognosis may be associated with its ability to reduce AFP and CRP levels.

TCM demonstrated protective effects in patients with HBV-related HCC and CRA, and survival was positively associated with the duration of TCM use. Patients with low CRP and AFP levels represent the optimal population for TCM treatment to enhance prognosis. Observing the levels of CRP, AFP, platelets, and hemoglobin can help predict the prognosis of patients.