PROGNOSTIC ACCURACY OF CHILD-PUGH VERSUS ALBI SCORE IN PATIENTS WITH HEPATOCELLULAR CARCINOMA TREATED WITH SYSTEMIC TREATMENT

Liver function assessmentis key for managing hepatocellular carcinoma (HCC), because most HCC patients also have a hepatopathy. Child-Pugh score (CP) is widely used and addresses ascites, hepatic encephalopathy, albumin, bilirubin and prothrombin time to classify patients in three groups. However, it relies on non-standardized parameters (ascites, encephalopathy) and extremes of laboratorial variables are not properly stratified. The albumin-bilirubin score (ALBI) was developed as a nomogram using only albumin and bilirubin as continuous variables to assess liver function in HCC patients. This study aims to evaluate prognostic accuracy of ALBI score in patients with HCC under systemic treatment, comparing to CP score. This is a retrospective analysis with data collected from medical registries of HCC patients under systemic therapy. Patients were classified according to ALBI and CP classes. Overall survival (OS) was estimated by Kaplan-Meier method and compared by log-rank. Cox regression model was used to define the prognostic role of each classification. A Harell’s C index compared the prognostic accuracy of both models. We included 373 patients: 75.6% were male, median age 62.2 years-old, 54.4% had viral etiology, and 2.7% had no hepatopathy. A total of 310 (83.1%) were CPA and 63 (16.8%) were CP B, with a median OS of 11.2 and 3.8 months, respectively (HR 3.17 [95% CI 2.37 to 4.24; p=0.000]). 137 (36.7%) were classified as ALBI I, 209 (56%) as ALBI II and 27 (7.24%) as ALBI III with a median OS of 12.6 (95% CI 10.6 to 18.3), 8.7 (95% CI 7.2 to 10) and 6.3 (95% CI 3.2 to 10) months respectively. Comparing ALBI II to ALBI I, we found a HR 1.46 [95% CI 1.16 to 1.84; p=0.001], and between ALBI III and ALBI I, HR was 2.20 [95% CI 1.42 to 3.41; p=0.000]. There was no difference between the prognostic accuracy of ALBI vs CP score with difference between C coefficients = 0.01 [95% CI -0.01 to 0.03; p=0.339]. ALBI score was able to discriminate two prognostic subgroups within CPA patients (ALBI I 12.8 months [95% CI 10.6 to 18.3] and ALBI II 10.2 months [95% CI 9.3 to 11.6]), with a HR 1.27 [95% CI 1.03 to 1.56; p=0.02]. In conclusion, ALBI score shows comparable prognostic accuracy to CP score in HCC patients under systemic treatment, using objective parameters to assess liver function. Moreover, ALBI score was able to stratify different prognostic groups within CPA. Accurate assessment of hepatopathy is key for trials interpretation and clinical decisions.