Hepatic decompensation is the major driver of mortality in hepatocellular carcinoma patients treated with atezolizumab plus bevacizumab: The impact of successful antiviral treatment.

BACKGROUNDAIMS: Unlike other malignancies, hepatic functional reserve competes with tumour progression in determining the risk of mortality from hepatocellular carcinoma (HCC). However, the relative contribution of hepatic decompensation over tumour progression in influencing overall survival (OS) has not been assessed in combination immunotherapy recipients. APPROACHRESULTS: From the AB-real observational study(n=898), we accrued 571 patients with advanced/unresectable HCC, Child-Pugh A class treated with frontline atezolizumab+bevacizumab(AB). Hepatic decompensation and tumour progression during follow-up were studied in relationship to patients' OS using time-dependent Cox model. Baseline characteristics were evaluated as predictors of decompensation in competing risks analysis. During a median follow-up of 11.0 months (95%CI 5.1-19.7), 293 patients(51.3%) developed tumour progression without decompensation and 94(16.5%) developed decompensation. In multivariable time-dependent analysis, decompensation(hazard ratio[HR] 19.04, 95%CI 9.75-37.19), HCC progression(HR 9.91, 95%CI 5.85-16.78), albumin-bilirubin(ALBI) grade 2/3(HR 2.16, 95%CI 1.69-2.77) and number of nodules>3(HR 1.63, 95%CI 1.28-2.08) were independently associated with OS. Pre-treatment ALBI grade 2/3(subdistribution HR [sHR] 3.35, 95%CI 1.98-5.67) was independently associated with decompensation, whereas viral aetiology was protective(sHR 0.55, 95%CI 0.34-0.87). Among patients with viral aetiology, effective antiviral treatment was significantly associated with lower risk of decompensation (sHR 0.48, 95%CI 0.25-0.93). CONCLUSIONS: Hepatic decompensation identifies patients with the worst prognosis following AB and is more common in patients with baseline ALBI>1 and non-viral aetiology. Effective antiviral treatment may protect from decompensation, highlighting the prognostic disadvantage of patients with non-viral aetiologies and the importance of multi-disciplinary management to maximise OS.

Characteristics and outcomes of immunotherapy-related liver injury in patients with hepatocellular carcinoma versus other advanced solid tumours.

BACKGROUND & AIMS: Immune-related liver injury (irLI) is commonly observed in patients with cancer treated with immune checkpoint inhibitors (ICIs). We aimed to compare the incidence, clinical characteristics, and outcomes of irLI between patients receiving ICIs for hepatocellular carcinoma (HCC) vs. other solid tumours. METHODS: Two separate cohorts were included: 375 patients with advanced/unresectable HCC, Child-Pugh A class treated with first-line atezolizumab+bevacizumab from the AB-real study, and a non-HCC cohort including 459 patients treated with first-line ICI therapy from the INVIDIa-2 multicentre study. IrLI was defined as a treatment-related increase of aminotransferase levels after exclusion of alternative aetiologies of liver injury. The incidence of irLI was adjusted for the duration of treatment exposure. RESULTS: In patients with HCC, the incidence of any grade irLI was 11.4% over a median treatment exposure of 4.4 months (95% CI 3.7-5.2) vs. 2.6% in the INVIDIa-2 cohort over a median treatment exposure of 12.4 months (95% CI 11.1-14.0). Exposure-adjusted-incidence of any grade irLI was 22.1 per 100-patient-years in patients with HCC and 2.1 per 100-patient-years in patients with other solid tumours (p <0.001), with median time-to-irLI of 1.4 and 4.7 months, respectively. Among patients who developed irLI, systemic corticosteroids were administered in 16.3% of patients with HCC and 75.0% of those without HCC (p <0.001), and irLI resolution was observed in 72.1% and 58.3%, respectively (p = 0.362). In patients with HCC, rates of hepatic decompensation and treatment discontinuation due to irLI were 7%. Grade 1-2 irLI was associated with improved overall survival only in patients with HCC (hazard ratio 0.53, 95% CI 0.29-0.96). CONCLUSIONS: Despite higher incidence and earlier onset, irLI in patients with HCC is characterised by higher rates of remission and lower requirement for corticosteroid therapy (vs. irLI in other solid tumours), low risk of hepatic decompensation and treatment discontinuation, not negatively affecting oncological outcomes. IMPACT AND IMPLICATIONS: Immune-related liver injury (irLI) is common in patients with cancer receiving immune checkpoint inhibitors (ICIs), but whether irLI is more frequent or it is associated with a worse clinical course in patients with hepatocellular carcinoma (HCC), compared to other tumours, is not known. Herein, we compared characteristics and outcomes of irLI in two prospective cohorts including patients treated with ICIs for HCC or for other oncological indications. irLI is significantly more common and it occurs earlier in patients with HCC, also after adjustment for duration of treatment exposure. However, outcomes of patients with HCC who developed irLI are not negatively affected in terms of requirement for corticosteroid therapy, hepatic decompensation, treatment discontinuation and overall survival.

Immunotherapy vs Best Supportive Care for Patients With Hepatocellular Cancer With Child-Pugh B Dysfunction.

Importance: Whether patients with Child-Pugh class B (CP-B) cancer with unresectable hepatocellular carcinoma (uHCC) benefit from active anticancer treatment vs best supportive care (BSC) is debated. Objective: To evaluate the association of immune checkpoint inhibitor (ICI)-based therapies vs BSC with overall survival (OS) of patients with uHCC and CP-B liver dysfunction. Design, Setting, and Participants: This retrospective, multicenter, international clinical case series examined data of patients with CP-B with uHCC who were receiving first-line ICI-based regimens from September 2017 to December 2022 whose data were extracted from an international consortium and compared with a cohort of patients with CP-B receiving BSC. Patients were treated in tertiary care centers across Europe, US, and Asia in routine clinical practice. After applying the inclusion criteria, 187 and 156 patients were left in the ICI and BSC groups, respectively. The propensity score was calculated for the following variables: age, alpha-fetoprotein levels, Child-Pugh score, extrahepatic spread, portal vein tumor thrombosis, cirrhosis, ascites, and baseline Eastern Cooperative Oncology Group performance status. Exposures: Patients in the ICI group received first-line systemic therapy with either atezolizumab plus bevacizumab (A+B) (n = 141) or nivolumab (n = 46). Main Outcomes and Measures: OS in the inverse probability of treatment weighting (IPTW) populations was the main outcome, and it was estimated with Kaplan-Meier method; univariable Cox regression test was used to make comparisons between the 2 groups. Results: The median age was 66 (IQR, 61-72) and 73 (IQR, 66-81) years in the ICI (33 women [18%]) and BSC groups (41 women [26%]), respectively. In the IPTW populations, median OS was significantly longer in the ICI group (7.50 months; 95% CI, 5.62-11.15) compared with BSC (4.04 months; 95% CI, 3.03-5.03; hazard ratio, 0.59; 95% CI, 0.43-0.80; P < .001). Multivariable analysis confirmed that ICI exposure was associated with a reduction of approximately 50% in the risk of death (hazard ratio, 0.55; 95% CI, 0.35-0.86; P < .001), and the presence of portal vein tumor thrombosis, an Eastern Cooperative Oncology Group performance score of greater than 1, and alpha-fetoprotein levels of 400 ng/mL or greater were associated with increased risk of death. Conclusions and Relevance: The results of this case series provide comparative evidence of improved survival in association with ICI treatment compared with BSC in patients with uHCC with CP-B liver dysfunction.