LiMAx Prior to Radioembolization for Hepatocellular Carcinoma as an Additional Tool for Patient Selection in Patients with Liver Cirrhosis

Abstract

Background and aims: Radioembolization (RE) has recently demonstrated a non-inferior survival outcome compared to systemic therapy for advanced hepatocellular carcinoma (HCC). Therefore, current guidelines recommend RE for patients with advanced HCC and preserved liver function who are unsuitable for transarterial chemoembolization (TACE) or systemic therapy. However, despite the excellent safety profile of RE, post-therapeutic hepatic decompensation remains a serious complication that is difficult to predicted by standard laboratory liver function parameters or imaging modalities. LiMAx^® is a non-invasive test for liver function assessment, measuring the maximum metabolic capacity for 13C-Methacetin by the liver-specific enzyme CYP 450 1A2. Our study investigates the potential of LiMAx^® for predicting post-interventional decompensation of liver function.

Patients and methods: In total, 50 patients with HCC with or without liver cirrhosis and not amenable to TACE or systemic treatments were included in the study. For patients prospectively enrolled in our study, LiMAx^® was carried out one day before RE (baseline) and 28 and 90 days after RE. Established liver function parameters were assessed at baseline, day 28, and day 90 after RE. The relationship between baseline LiMAx^® and pre-and post-interventional liver function parameters, as well as the ability of LiMAx^® to predict hepatic decompensation, were analyzed.

Results: We observed a strong association between baseline LiMAx^® and bilirubin, albumin, ALBI grade, and MELD score. Patients presenting with Child-Pugh score B 28 days after RE or with a deterioration in Child-Pugh score by at least one point had a significantly lower baseline LiMAx^® compared to those with Child-Pugh score A or with stable Child-Pugh score. The ability of LiMAx^® to predict hepatic decompensation after RE was determined using ROC curve analysis and was compared to MELD score and ALBI grade. LiMAx^® achieved a substantial AUC of 0.8117, comparable to MELD score and ALBI grade.

Conclusion: Patients with lower LiMAx^® values at baseline have a significantly increased risk for hepatic decompensation after RE, despite being categorized as Child-Pugh A. Therefore, LiMAx^® can be used as an additional tool to identify patients at high risk of post-interventional hepatic failure.

Keywords: HCC; LiMAx®; enzymatic liver function test; hepatocellular carcinoma; radioembolization; selective internal radiotherapy.

Figure 1

(A) Baseline LiMAx^® is not different in patients with or without liver cirrhosis (with liver cirrhosis median 279 µg/kg/h, IQR: 220–370.5 µg/kg/h; without liver cirrhosis median 363 µg/kg/h, IQR 291–389.5 µg/kg/h; p = 0.1464); (B) LiMAx^® at d0, d28, and d90 for the entire prospective cohort; (C) LiMAx^® at d0, d28, and d90 only for patients from the prospective cohort with liver cirrhosis. * = p < 0.05, ns = not significant.

Figure 2

Correlation analysis of LiMAx^® at baseline with liver function 28 days after RE in the whole cohort. (A) bilirubin r = −0.55, p < 0.0001; (B) albumin r = 0.33, p = 0.03; (C) ALBI grade r = −0.41, p = 0.005; (D) MELD score r = −0.47, p = 0.0009.

Figure 3

(A) Baseline LiMAx^® was significantly lower in patients presenting Child–Pugh B liver function 28 days after RE (Child–Pugh A median LiMAx 308 µg/kg/h, Child–Pugh B median LiMAx 220 µg/kg/h, p = 0.007); (B) Receiver operating characteristics (ROC) curve to predict the risk of hepatic decompensation in the whole cohort (AUC Meld score 0.803, AUC ALBI grade 0.854, AUC LiMAx 0.818), ** = p < 0.01.

Figure 4

(A): Patients without any deterioration in Child–Pugh Score 28 days after RE presented significantly higher baseline LiMAx values, whole cohort (Δ CPS d28-d0 = 0 median baseline LiMAx 321 µg/kg/h, IQR 261–398 µg/kg/h, Δ CPS d28-d0 ≥ 1 median baseline LiMAx 226.5 µg/kg/h, IQR 200.3–308.3 µg/kg/h, p = 0.004); (B) Patients without any deterioration in Child–Pugh Score 28 days after RE presented significantly higher LiMAx values 28 days after RE, only prospective cohort (Δ CPS d28-d0 = 0 median baseline LiMAx 342 µg/kg/h, IQR 277–420.5 µg/kg/h, Δ CPS d28-d0 ≥ 1 median baseline LiMAx 229 µg/kg/h, IQR 199.5–243.5 µg/kg/h, p = 0.02). * = p < 0.05, ** = p < 0.01.

Figure 5

(A) No significant difference in baseline LiMAx for patients with or without deterioration in Child–Pugh score 90 days after RE. (B) Patients with at least two points decline in Child–Pugh score at d90 presented a significantly lower LiMAx at d28, only prospective cohort (Δ CPS d90-d0 = 0–1 median baseline LiMAx 311 µg/kg/h, IQR 243.5–420.5 µg/kg/h, Δ CPS d90-d0 ≥ 2 median baseline LiMAx 216 µg/kg/h, IQR 200.8–267 µg/kg/h, p = 0.05; (C) Patients without any deterioration in Child–Pugh score showed unchanged LiMAx values at d28. Patients with a deterioration in Child–Pugh score by at least two points at day 90 had slightly worsened LiMAx values, only prospective cohort. * = p < 0.05, ns = not significant.