Atezolizumab plus bevacizumab versus lenvatinib or sorafenib in non-viral unresectable hepatocellular carcinoma: an international propensity score matching analysis

BACKGROUND

A growing body of evidence suggests that non-viral hepatocellular carcinoma (HCC) might benefit less from immunotherapy.

MATERIALS AND METHODS

We carried out a retrospective analysis of prospectively collected data from consecutive patients with non-viral advanced HCC, treated with atezolizumab plus bevacizumab, lenvatinib, or sorafenib, in 36 centers in 4 countries (Italy, Japan, Republic of Korea, and UK). The primary endpoint was overall survival (OS) with atezolizumab plus bevacizumab versus lenvatinib. Secondary endpoints were progression-free survival (PFS) with atezolizumab plus bevacizumab versus lenvatinib, and OS and PFS with atezolizumab plus bevacizumab versus sorafenib. For the primary and secondary endpoints, we carried out the analysis on the whole population first, and then we divided the cohort into two groups: non-alcoholic fatty liver disease (NAFLD)/non-alcoholic steatohepatitis (NASH) population and non-NAFLD/NASH population.

RESULTS

One hundred and ninety patients received atezolizumab plus bevacizumab, 569 patients received lenvatinib, and 210 patients received sorafenib. In the whole population, multivariate analysis showed that treatment with lenvatinib was associated with a longer OS [hazard ratio (HR) 0.65; 95% confidence interval (CI) 0.44-0.95; P = 0.0268] and PFS (HR 0.67; 95% CI 0.51-0.86; P = 0.002) compared to atezolizumab plus bevacizumab. In the NAFLD/NASH population, multivariate analysis confirmed that lenvatinib treatment was associated with a longer OS (HR 0.46; 95% CI 0.26-0.84; P = 0.0110) and PFS (HR 0.55; 95% CI 0.38-0.82; P = 0.031) compared to atezolizumab plus bevacizumab. In the subgroup of non-NAFLD/NASH patients, no difference in OS or PFS was observed between patients treated with lenvatinib and those treated with atezolizumab plus bevacizumab. All these results were confirmed following propensity score matching analysis. By comparing patients receiving atezolizumab plus bevacizumab versus sorafenib, no statistically significant difference in survival was observed.

CONCLUSIONS

The present analysis conducted on a large number of advanced non-viral HCC patients showed for the first time that treatment with lenvatinib is associated with a significant survival benefit compared to atezolizumab plus bevacizumab, in particular in patients with NAFLD/NASH-related HCC.

Identification of lenvatinib prognostic index via recursive partitioning analysis in advanced hepatocellular carcinoma

BACKGROUND

After the advent of new treatment options for advanced hepatocellular carcinoma (HCC), the identification of prognostic factors is crucial for the selection of the most appropriate therapy for each patient.

PATIENTS AND METHODS

With the aim to fill this gap, we applied recursive partitioning analysis (RPA) to a cohort of 404 patients treated with lenvatinib.

RESULTS

The application of RPA resulted in a classification based on five variables that originated a new prognostic score, the lenvatinib prognostic index (LEP) index, identifying three groups: low risk [patients with prognostic nutritional index (PNI) >43.3 and previous trans-arterial chemoembolization (TACE)]; medium risk [patients with PNI >43.3 but without previous TACE and patients with PNI <43.3, albumin-bilirubin (ALBI) grade 1 and Barcelona Clinic Liver Cancer stage B (BCLC-B)]; high risk [patients with PNI <43.3 and ALBI grade 2 and patients with PNI <43.3, albumin-bilirubin (ALBI) grade 1 and Barcelona Clinic Liver Cancer stage C (BCLC-C)]. Median overall survival was 29.8 months [95% confidence interval (CI) 22.8-29.8 months] in low risk patients (n = 128), 17.0 months (95% CI 15.0-24.0 months) in medium risk (n = 162) and 8.9 months (95% CI 8.0-10.7 months) in high risk (n = 114); low risk hazard ratio (HR) 1 (reference group), medium risk HR 1.95 (95% CI 1.38-2.74), high risk HR 4.84 (95% CI 3.16-7.43); P < 0.0001. The LEP index was validated in a cohort of 127 Italian patients treated with lenvatinib. While the same classification did not show a prognostic value in a cohort of 311 patients treated with sorafenib, we also show a possible predictive role in favor of lenvatinib in the low risk group.

CONCLUSIONS

LEP index is a promising, easy-to-use tool that may be used to stratify patients undergoing systemic treatment of advanced HCC.

Genetic profiling using plasma-derived cell-free DNA in therapy-naïve hepatocellular carcinoma patients: a pilot study

Background

Hepatocellular carcinomas (HCCs) are not routinely biopsied, resulting in a lack of tumor materials for molecular profiling. Here we sought to determine whether plasma-derived cell-free DNA (cfDNA) captures the genetic alterations of HCC in patients who have not undergone systemic therapy.

Patients and methods

Frozen biopsies from the primary tumor and plasma were synchronously collected from 30 prospectively recruited, systemic treatment-naïve HCC patients. Deep sequencing of the DNA from the biopsies, plasma-derived cfDNA and matched germline was carried out using a panel targeting 46 coding and non-coding genes frequently altered in HCCs.

Results

In 26/30 patients, at least one somatic mutation was detected in biopsy and/or cfDNA. Somatic mutations in HCC-associated genes were present in the cfDNA of 63% (19/30) of the patients and could be detected 'de novo' without prior knowledge of the mutations present in the biopsy in 27% (8/30) of the patients. Mutational load and the variant allele fraction of the mutations detected in the cfDNA positively correlated with tumor size and Edmondson grade. Crucially, among the seven patients in whom the largest tumor was ≥5 cm or was associated with metastasis, at least one mutation was detected 'de novo' in the cfDNA of 86% (6/7) of the cases. In these patients, cfDNA and tumor DNA captured 87% (80/92) and 95% (87/92) of the mutations, suggesting that cfDNA and tumor DNA captured similar proportions of somatic mutations.

Conclusion

In patients with high disease burden, the use of cfDNA for genetic profiling when biopsy is unavailable may be feasible. Our results support further investigations into the clinical utility of cfDNA in a larger cohort of patients.

Safety and efficacy of subcutaneous hepatitis B immunoglobulin after liver transplantation: an open single-arm prospective study

Life-long hepatitis B immunoglobulin (HBIG) administration is a main component of prophylactic strategy to prevent hepatitis B virus (HBV) reinfection after liver transplantation (LT). Long-term effects of HBIG treatment are known only for intravenous (IV) and intramuscular formulations. To evaluate safety and efficacy of self-administered SC HBIG, 135 LT patients receiving a 48-week treatment were analyzed. The dose of HBIG was 500 IU or 1000 IU if body weight was <75 kg or ≥75 kg, respectively. Patients were switched from the monthly IV HBIG treatment to weekly SC HBIG 2-3 weeks after the last IV dosage. All patients were able to SC self-injection after a single training. The treatment was effective in maintaining trough anti-HBs levels >100 IU/L. No severe drug-related side effects occurred. Fifteen injection-site small hematomas and four cases of mild itch occurred. At the end of the study, anti-HBs median titer was 232 IU/L (115-566 IU/L) and 97.8% of patients had an anti-HBs level >150 IU/L. Due to high mean level of anti-HBs titers observed during this study, individualized treatment schedules should be further investigated. In conclusion, SC HBIG for long-term prophylaxis of post-LT HBV reinfection resulted safe, well accepted, and effective in maintaining adequate anti-HBs levels.

Elevated CD4+/CD25+ T-cell frequency and function during hepatitis C virus recurrence after liver transplantation

BACKGROUND/AIM

Factors involved in hepatitis C virus (HCV) recurrence versus acute cellular rejection are not fully understood. The aim of the present study was to investigate whether patients with recurrence after liver transplantation (OLT) showed similar CD4(+)/CD25(+) cell frequency and function as those who became chronically infected.

PATIENTS AND METHODS

After written informed consent, we enrolled 20 patients (group A) who underwent OLT with HCV recurrence within 6 months. HCV-RNA and hypertransaminasemia were used to assess the reactivation of viral hepatitis. CD4(+)/CD25(+) T cells were enumerated using a flow cytometry assay, gated on CD3 cells, stained for FoxP3. After immunomagnetic sorting (Dynal, Oslo, NW), Treg suppressor activity was measured, as the ability to inhibit proliferation of autologous CD4(+)/CD25(-) T cells (anti-CD3/CD28 stimulation-1:2, 1:20 ratio). Eight patients with acute hepatitis C who evolved to a chronic infection after 6 months (group B) were used as positive controls, while 10 healthy individuals were negative controls (group C).

RESULTS

We did not observe any difference in CD4(+)/CD25(+) frequency or function among group A compared with group B (CD4(+)/CD25(+) = 14% +/- 2% versus CD4(+)/CD25(+) = 16% +/- 3%), although both groups were significantly increased with respect to group A (CD4(+)/CD25(+) = 6% +/- 3%; Mann-Whitney U test, P < .01).

CONCLUSION

Patients developing HCV recurrence after OLT have the same immunoregulatory network as patients with acute hepatitis C evolving to persistent infection, likely suggesting that CD4(+)/CD25(+) numbers may be a marker to predict recurrence of HCV after OLT.

Efficacy of high dose of recombinant alpha 2b interferon on long term response in chronic hepatitis C and cirrhosis: prospective randomized multicentre study

BACKGROUND/AIMS

The long-term response to alpha-Interferon in HCV-related chronic liver diseases is disappointing. A randomized controlled trial was conducted to investigate: 1) if doubling the standard regimen of 3 MU recombinant alpha 2b-interferon thrice weekly for one year could improve the long-term response, and 2) the efficacy of these two schedules in cirrhotic patients.

PATIENTS AND METHODS

A series of 80 anti-HCV positive patients with biopsy proven liver disease (52 chronic hepatitis and 28 cirrhosis) were randomized to receive either 3 MU or 6 MU alpha 2b-interferon.

RESULTS

Based on "intention-to-treat analysis", 38% in the 3 MU group and 53% in the 6 MU group had end-of-treatment response. After 24 months, 18% had long-term response: 5% in 3 MU group and 30% in 6 MU group (p < 0.008). HCV genotype had no influence on the response rate. Thirty-eight percent of the cirrhotics treated with 6 MU had long-term response, while none of those treated with 3 MU had long-term response (difference 38%; 95% confidence internal 10%-67%; p = 0.03). At the end of treatment, 38% of patients lost HCV-RNA. After 24 months only 19% remained HCV-RNA negative: 12 patients (31%) in the 6 MU group and 2 (6%) in the 3 MU group (p < 0.05).

CONCLUSIONS

6 MU of alpha 2b-interferon thrice weekly for 12 months is significantly better than 3 MU in inducing a long-term response and permanent loss of HCV-RNA. This result is particularly striking in the subgroup of cirrhotics.

Resistance to alpha interferon therapy in HCV chronic liver disease: role of hepatic fibrosis

The response rate to interferon in HCV chronic liver disease is insufficient to date and the causes of this failure are not fully understood. Hepatic fibrosis hinders the blood-hepatocyte exchange of substances and we hypothesized that this process may also reduce the efficacy of interferon. Serum levels of connective tissue metabolites are related, to some extent, to the amount of extracellular matrix in the liver. Therefore, the usefulness was evaluated of serum tests of connective tissue metabolism compared to standard biochemical and histological parameters in predicting the probability of primary response to interferon. Sixty-eight patients with HCV chronic liver disease were treated with alpha-interferon for 1 year. At multivariate analysis time 0, the serum level of the P1 fragment of laminin was found to be the only factor independently associated with the response to treatment. As is well known, higher serum concentrations of the P1 fragment of laminin are associated with active basement membrane turnover and derangement of the hepatic structure. Therefore, this process seems to reduce the probability of response to interferon and, if confirmed, evaluation of serum the P1 fragment of laminin may be a useful test to predict the response to interferon and to define the therapeutic strategy, especially as far as the dose of interferon is concerned.

Prevalence of anti-HCV among spouses and offspring of anti-HCV positive subjects: an Italian multicentre study

The role of familial environment in the spreading of hepatitis C virus (HCV) infection is not well established. We studied 1670 family members for 578 anti-HCV+ subjects enrolled in 8 centres distributed throughout Italy. The prevalence of anti-HCV positivity was significantly higher in spouses than in offspring (15.6% and 2.1% respectively; p < 0.01), with no difference between northern and central-southern regions of Italy. Anti-HCV positivity was found almost exclusively in adults; among offspring, during the first two decades of life, the prevalence of anti-HCV positivity was significantly lower than in subjects over 20 years old (0.6% vs 3.1%, respectively).

[Prevalence of antibody to hepatitis C virus in hospital personnel]

Prevalence of antibody to hepatitis C virus (anti-HCV) has been widely investigated in many categories; however no data are available on hospital personnel. The aim of our study was to investigate whether hospital personnel are at risk for HCV infection.

METHODS

sera collected during a prospective study on HBV infection in hospital workers done in our institution in 1985 were analyzed for the ELISA test for anti-HCV from Ortho Diagnostic System. Sera were stored at -20 degrees C and were never defrosted until tested. A population of a consecutive series of healthy volunteer blood donors was used as a control group.

RESULTS

the anti-HCV prevalence was higher in hospital personnel, than in blood donors (4.5 versus 1.1, p less than 0.001, Odds Ratio 4.5, Confidence Limits 2.9-7.2).

CONCLUSION

although anti-HCV is not an "ideal" test for epidemiological purposes, our study suggests that hospital personnel is at high risk for HCV infection.