Treating patients with advanced hepatocellular carcinoma and impaired liver function: Broadening the reach of anti-cancer therapy

Abstract

According to the Barcelona Clinic Liver Cancer (BCLC) classification, patients with very early (BCLC-0) and early (BCLC-A) disease can undergo a potentially curative approach, such as surgical resection, local ablative therapies and orthotopic liver transplantation (OLT), which is the only curative approach for both HCC and liver cirrhosis. 7 Also, some initial real-life experiences seem to confirm the feasibility and safety of treating CP-B patients with lenvatinib, despite a poor OS compared to CP-A patients. 8 Focusing on second-line options, cabozantinib achieved a significant improvement in OS and progression-free survival (PFS) also in those patients whose liver function deteriorated to CP-B by week 8 of treatment within the CELESTIAL trial. 9 Also, these patients had a similar rate of dose reductions and toxicity-related treatment discontinuations compared to the rest of the population. According to a large retrospective study, CP was confirmed to be a negative prognostic factor; in fact, CP-B patients treated with nivolumab achieved lower OS (7.3 months vs 16.3 months, HR = 1.95, 95% CI 1.35-2.81, P < .001) compared to CP-A patients, but PFS, ORR and TRAEs rate did not appear to be influenced by CP. 14 Among factors defining CP-B class, diuretic-refractory ascites have been reported to carry worse prognostic significance compared to other parameters in CP-B patients receiving nivolumab or pembrolizumab 15; furthermore, albumin–bilirubin grade has been described to better predict survival compared to CP in patients with moderate to severe liver impairment receiving ICI monotherapy. 16 Data about the safety and efficacy of ICI combination therapies in CP-B are currently unavailable.

Hepatocellular carcinoma (HCC) usually arises on the background of liver cirrhosis, which carries an intrinsic risk of death because of liver failure and poses a major treatment challenge. Patient stratification is therefore important to avoid unnecessary treatment for those who are more likely to die from underlying liver disease rather than from cancer. In clinical practice, the severity of liver disfunction is commonly graded with the Child–Pugh (CP) score, which divides patients into three categories (A–B–C), with worsening liver function according to clinical and laboratory criteria.¹ For CP-C patients, exclusive supportive care is recommended. CP-B patients, which represent a high percentage of HCC cases, are a heterogeneous category including patients presenting with both mild and moderate–severe liver function. Treatment options for these patients are limited, and they are usually excluded from clinical trials as their impaired liver function is thought to act as a strong confounder and a competitive cause of death.

According to the Barcelona Clinic Liver Cancer (BCLC) classification, patients with very early (BCLC-0) and early (BCLC-A) disease can undergo a potentially curative approach, such as surgical resection, local ablative therapies and orthotopic liver transplantation (OLT), which is the only curative approach for both HCC and liver cirrhosis. In particular, according to the EASL and the AASLD guidelines, surgical resection is indicated only in patients with good liver function (CP-A) because of the risk of post-surgical liver failure.^2,3 Concerning local ablative strategies (radiofrequency ablation, microwave ablation and percutaneous ethanol injection), the treatment choice should be personalised, since there is no a priori contraindication for CP-B patients, but these patients often suffer from ascites and coagulation disorders, which are known contraindications for local ablation.⁴ In cases of BCLC B HCC, the recommended treatment is trans-arterial chemoembolization. AASLD guidelines restrict this approach to CP-A patients and only limited CP-B patients,³ whereas EASL guidelines include also asymptomatic patients with CP B7, with particular precautions because of the risk of post-treatment liver failure and ischaemic necrosis.²

Systemic therapies are the treatment of choice for patients with advanced (BCLC-C) or intermediate stage (BCLC-B) liver cancer not amenable for loco-regional therapies. While any systemic agent would be contraindicated for CP-C patients, there is a subgroup of CP-B patients which could still be potential candidates for systemic therapy. Data from clinical trials are scattered in this subgroup and most of the evidence comes from real-life experiences and is therefore limited for newly approved drugs. In particular, the only positive large phase III study enrolling patients up to CP-B was the SHARP trial, which led to the historical approval of sorafenib as the first treatment for advanced HCC.⁵ A large meta-analysis, comprising 8678 patients treated with sorafenib within 30 different clinical trials, found that, despite an expected worse survival, CP-B patients achieved a comparable response, safety and tolerability to sorafenib compared to CP-A patients.⁶ Of note, CP-B accounted for 19% of all analysed patients, with a median overall survival (OS) of 4.6 months vs 8.8 months in the CP-A group. Less robust data are available for the other multi-kinase inhibitors (MKIs). In a post hoc analysis conducted on patients enrolled in the REFLECT trial, lenvatinib was shown to be safe and effective even in those whose liver function deteriorated to CP-B during treatment. In this subgroup, lenvatinib reported an overall response rate (ORR) of 28.3% (vs 42.9% in the CP-A group) and an incidence of treatment-related adverse events (TRAEs) grade ≥3 rate of 71.7% (vs 54.7% in the CP-A group).⁷ Also, some initial real-life experiences seem to confirm the feasibility and safety of treating CP-B patients with lenvatinib, despite a poor OS compared to CP-A patients.⁸ Focusing on second-line options, cabozantinib achieved a significant improvement in OS and progression-free survival (PFS) also in those patients whose liver function deteriorated to CP-B by week 8 of treatment within the CELESTIAL trial.⁹ Also, these patients had a similar rate of dose reductions and toxicity-related treatment discontinuations compared to the rest of the population. On the other side, regorafenib achieved a significantly worse performance in terms of survival and toxicity in patients with CP-B compared to CP-A, thus discouraging its use in this subset of patients.¹⁰ Also, ramucirumab did not prove to be effective in CP-B patients with AFP < 400 ng/mL in the initial REACH study, while at the same time causing a higher rate of grade 3-4 adverse events,¹¹ and for this reason, the subsequent phase III REACH-2 study only included patients with CP < 7.¹²

The only immune checkpoint inhibitor (ICI) prospectively tested in patients with impaired liver function is nivolumab, an antibody to PD-1 monoclonal. In particular, in one of the cohorts of the CheckMate-040 trial, nivolumab was administered to 49 CP B7-8 patients, showing a good toxicity profile, with 4.1% discontinuing the treatment because of AE, and good efficacy data.¹³ These findings have recently been confirmed by real-life data, which showed the feasibility of this approach even in the presence of moderate liver disfunction. According to a large retrospective study, CP was confirmed to be a negative prognostic factor; in fact, CP-B patients treated with nivolumab achieved lower OS (7.3 months vs 16.3 months, HR = 1.95, 95% CI 1.35-2.81, P < .001) compared to CP-A patients, but PFS, ORR and TRAEs rate did not appear to be influenced by CP.¹⁴ Among factors defining CP-B class, diuretic-refractory ascites have been reported to carry worse prognostic significance compared to other parameters in CP-B patients receiving nivolumab or pembrolizumab¹⁵; furthermore, albumin–bilirubin grade has been described to better predict survival compared to CP in patients with moderate to severe liver impairment receiving ICI monotherapy.¹⁶ Data about the safety and efficacy of ICI combination therapies in CP-B are currently unavailable. Early real-life clinical experiences have just been published for the combination of atezolizumab plus bevacizumab; however, the number of CP-B patients included in the analyses was too low to derive conclusions.¹⁷

In conclusion, liver disfunction represents one of the most common barriers to the effective treatment of liver cancer and adequate patient selection is pivotal to carefully balance the risk and benefits of therapies. With new promising therapeutic options becoming available, concerted efforts are required to collect more evidence and to offer the best treatment even to this conspicuous group of patients which have historically been underrepresented in clinical trials.

ACKNOWLEDGEMENTS

AD is supported by grant funding from the European Association for the Study of the Liver (Andrew Burroughs Fellowship). The authors are fully responsible for all content and editorial decisions.

CONFLICT OF INTEREST

All authors have declared no conflicts of interest.

AUTHOR CONTRIBUTIONS

All the authors contributed to writing and revising the original draft.