S1106 Early Detection of Hepatocellular Carcinoma Recurrence in the Post-Transplant Population: A Comparison of RETREAT and Cleveland Clinic Florida Scoring System

Abstract

INTRODUCTION: Liver transplantation (LT) for hepatocellular carcinoma (HCC) is curative in most cases, however, recurrence is observed in some patients. The Risk Estimation of Tumor Recurrence After Transplant (RETREAT) score is an externally-validated scoring system for prediction of post-LT HCC recurrence. The Cleveland Clinic Florida Scoring System (CCFSS) is a potential new scoring system for prediction of HCC recurrence. Our study aimed to compare the RETREAT and CCFSS. METHODS: We conducted a retrospective cohort study of adult patients with HCC who underwent LT. The RETREAT score was classified as follows: low risk (scores 0-3), moderate risk (score 4), and high risk (score 5 or higher). The CCFSS is based on explant pathology and categorized as: low (score 1), moderate (score 2), and high risk (score 3). Mantel- Haenzsel chi-square analyses were conducted to compare the RETREAT and CCFSS classifications for detecting HCC recurrence. RESULTS: A total of 52 patients with HCC underwent LT. The median follow-up period was 37 months. The median age of the patient population was 65.5 years (range 30-74 years) and 75% were male. The most common underlying liver disease was hepatitis C (n = 29, 55.8%), followed by fatty liver disease (n = 13, 25%). On explant pathology, 41 patients (78.8%) had viable tumor size <5 cm, two patients (3.8%) had ≥5 cm tumor size and the rest had a complete pathological response. Sixteen patients (31%) showed microvascular invasion. In 14 patients (31%), the tumors were poorly differentiated. Thirty-four patients (65.4%) had only one or no HCC lesions on explant pathology. Four patients had post-LT HCC recurrence, all within 2 years of LT (Table 1). Both risk scores had a sensitivity of 75%; the specificity of RETREAT score was 95.8% while the specificity of CCFSS was 60.4%. The RETREAT score was better able to detect low, moderate and high levels of risk (P < 0.001), compared to the CCFSS score (P = 0.480). AFP levels at the time of LT were associated with HCC recurrences (P = 0.014). Patient age, gender, ethnicity, liver disease etiology, MELD scores, HCC lesions at the time of diagnosis, and number/type of locoregional therapy (LRT), were not predictive of HCC recurrence. CONCLUSION: This is the first study to evaluate the CCFSS as a potential new scoring system to predict HCC recurrence after LT. The RETREAT score is more specific than CCFSS. The incorporation of AFP levels at the time of LT improves the estimation of HCC recurrence in the post-LT period.Table 1