AIM:To propose an appropriate staging system for hepatocellular carcinoma(HCC)classification.METHODS:Here,288 in-patients with HCC were studied and divided into three groups:those with expansive growth,invasive growth(including satellite nodules,nodule fusions and direct tumor invasion of adjacent organs),or disseminative growth(including vascular involvement,regional lymph node metastasis and distant metastasis).A survival analysis was performed using a Kaplan-Meier analysis,and prognostic factors for overall survival were determined by the Cox proportional hazards regression model.RESULTS:The overall survival(OS)of patients with invasive tumor growth was shorter than that of patients with expansive tumor growth(27.796±3.730and 57.398±4.873 mo,respectively,P<0.001).No significant difference in survival was observed between patients with vascular involvement and patients with regional lymph node metastasis(21.667±4.773 and14.619±2.456 mo,respectively,P=0.801).The OS of patients with distant metastasis(6.417±1.395mo)was shorter than that of the other groups(P<0.001).No significant difference in survival was observed between patients with expansive tumor growth and vascular and/or regional lymph node involvement and patients with invasive tumor growth and no vascular and/or lymph node involvement(25.762±7.024,21.200±7.794 and 39.533±5.840 mo,respectively;P=0.871,0.307 and 0.563,respectively).CONCLUSION:These data led to the proposal of a new staging system:the Expansive-Invasive-Disseminative growth staging classification. AIM: To propose an appropriate staging system for hepatocellular carcinoma (HCC) classification. METHODS: Here, 288 in-patients with HCC were studied and divided into three groups: those with expansive growth, invasive growth (including satellite nodules, nodule fusions and direct tumor survival of adjacent organs), or disseminative growth (including vascular involvement, regional lymph node metastasis and distant metastasis). A survival analysis was performed using a Kaplan-Meier analysis, and prognostic factors for overall survival were determined by the Cox proportional hazards regression model .RESULTS: The overall survival (OS) of patients with invasive tumor growth was shorter than that of patients with expansive tumor growth (27.796 ± 3.730 and 57.398 ± 4.873 mo, respectively, P <0.001) .No significant difference in survival was observed between patients with vascular involvement and patients with regional lymph node metastasis (21.667 ± 4.773 and 14.619 ± 2.456 mo, respectively, P = 0.801). The OS of patients wit Significant difference in survival was observed between patients with expansive tumor growth and vascular and / or regional lymph node involvement and patients with invasive tumor (6.417 ± 1.395 mo) was shorter than that of the other groups (P <0.001) growth and no vascular and / or lymph node involvement (25.762 ± 7.024, 21.200 ± 7.794 and 39.533 ± 5.840 mo, respectively; P = 0.871, 0.307 and 0.563, respectively) .CONCLUSION: These data led to the proposal of a new staging system : the Expansive-Invasive-Disseminative growth staging classification.