S1349 Utility of LAD Score for Treatment Response Assessment and Recurrence Monitoring in Patients With Hepatocellular Carcinoma

Abstract

Listen

Translate article

Introduction: The GALAD score, which incorporates the biomarkers AFP, AFP-L3% and DCP, has excellent performance for detection of hepatocellular carcinoma (HCC). However, its accuracy in treatment response assessment and recurrence prediction are unknown. We aimed to evaluate the accuracy of these biomarkers in predicting the presence of post-treatment viable tumor and HCC recurrence. Methods: We identified patients with cirrhosis or chronic hepatitis B infection who underwent curative surgical or locoregional treatment (Y-90, TACE, ablation) for HCC between May 2019 to May 2022, with GALAD labs obtained greater than 1 month post-treatment. Because the study cohort already has known HCC, we removed the demographic factors to calculate a score based solely on the biomarkers (the “LAD” score). The LAD score is calculated from the following formula: .04×(AFP-L3) +2.34×log(AFP)+1.33×log(DCP). Evidence of viable tumor was determined by imaging, biopsy, or liver explant pathology after transplant. Patients with non-viable tumor on initial post-treatment imaging were further analyzed for tumor recurrence. The sensitivity and specificity of the LAD score in predicting tumor viability were calculated. The optimal receiver operating characteristic (ROC) curve was used to calculate the cutoff for the LAD score. Results: A total of 78 patients (97 treatments) with surveillance imaging and biomarkers were analyzed. The number of treatments per patient ranged between 1 and 3 (mean 1.2). Mean age at time of HCC detection was 67 years, and a third of patients were female. The most common treatment modality was TACE (48.6%), followed by Y-90 (29.9%), ablation (16.5%), and resection (7.2%). About a third of cases (34.0%) had a viable tumor on initial post treatment images. The optimal LAD cutoff was calculated at 2.23, yielding a sensitivity and specificity of 87.4% and 60.9% for detection of viable tumors, respectively. After median follow up of 21.3 months, 11 cases (17.2%) developed recurrent HCC (mean time to recurrence 3.5 months) among cases with post-treatment non-viable tumor. A total of 7 recurrence cases (64%) had a positive LAD score at negative post-treatment surveillance images. Positive LAD score had a borderline association with recurrence (HR 3.17; 95%CI [0.92, 10.92], p = 0.067). Conclusion: LAD score is increased in most HCC patients with post-treatment viable tumors, suggesting its utility for post treatment response assessment. It might provide risk stratification for HCC recurrence.Figure 1.: Incidence of HCC Recurrence in LAD Positive vs Negative Cases