| Highlights |
• Single-agent immune checkpoint inhibitors (ICIs) produced objective responses in about 15% of patients with advanced hepatocellular carcinoma in phase II and III trials. • Immunotherapy combinations have been developed with the aim of improving activity and expanding the population of patients who could benefit. • Immunotherapy combinations tested in phase III trials include ICIs and bevacizumab, ICIs and tyrosine kinase inhibitors, and the combination of two ICIs. • Atezolizumab-bevacizumab or durvalumab-tremelimumab (STRIDE regimen) are currently the first-line standard of care. • Future directions include defining treatment sequences, as well as identifying biomarkers, optimal combinations (including with locoregional therapies), and new agents. |
The successful development of systemic therapy for hepatocellular carcinoma (HCC) followed two major advances in our understanding of the disease: a) an appreciation of the importance of patient selection in clinical trials and b) the development of molecular therapies with activity against HCC.
In this review, the authors provide the evidence for combination immunotherapy in advanced HCC, including scientific rationale and clinical data. The goal is to put available data into perspective to better understand the rapidly changing treatment landscape and identify future directions that need to be addressed.
| Fundamental reason |
Cancer progression requires evading immune surveillance. Checkpoint molecules participate in fine-tuning the immune response to both infectious agents and cancer cells. Inhibitory checkpoints include PD-1, CTLA-4, and lymphocyte activation gene 3 (LAG-3), among others.
ICIs targeting PD-1 and its main ligand PD-L1 have an unquestionable antitumor effect in some patients with advanced HCC. PD-1 and PD-L1 inhibitors, including nivolumab, pembrolizumab, camrelizumab, tislelizumab, durvalumab, and atezolizumab, consistently produce objective tumor responses in about 15% of patients in prospective phase II and III trials.
Responses are consistently associated with prolonged survival, but favorable outcomes are also seen among some nonresponders, including those with long-term stable disease and even patients with tumors that initially progress and then stabilize or respond. However, no less than 30% of HCC tumors at this stage show intrinsic resistance to PD-1 or PD-L1 inhibitors and, in some patients, treatment can even increase the rate of tumor growth.
The antitumor activity of PD-1 and PD-L1 inhibitors used as single agents was not broad or potent enough to significantly improve overall survival (OS) vs sorafenib in new patients in phase III trials. Therefore, combining them with other agents that could provide additive or synergistic activity was a rational option.
Angiogenesis is a key player in cancer immune evasion. Proangiogenic factors such as VEGF inhibit cytokine-induced endothelial cell adhesiveness, inducing a state of endothelial cell anergy that tumors take advantage of to evade immune infiltration. On the other hand, these factors also promote T cell exhaustion by upregulating immune checkpoint molecules and directly inhibiting T cell proliferation and cytotoxic activity.
By further enhancing their immunosuppressive activity, they also inhibit the maturation of dendritic cells and increase tumor infiltration by Tregs and myeloid-derived suppressor cells (MDSCs). Similar effects have been demonstrated for other proangiogenic factors, such as angiopoietins, hepatocyte growth factor, and PDGF (platelet-derived growth factor).
Bevacizumab counteracts these immunosuppressive effects and may help increase the number and activation of dendritic cells and cytotoxic T cells, and reverse VEGF-induced T cell exhaustion. On the other hand, VEGF inhibition generally results in a more hypoxic tumor microenvironment due to decreased tumor perfusion and this in turn can activate a number of immunosuppressive mechanisms.
Tumor hypoxia attracts Tregs, regulates the maturation and function of MDSCs, attracts tumor-associated macrophages (TAMs) and differentiates them towards an M2 phenotype (which
participates in tumor progression by suppressing antitumor immunity) and has a very potent negative effect on the function of activated T cells through the accumulation of adenosine. However, such a significant effect seems unlikely given the observed antitumor activity of the combined selective VEGF inhibitors.
| Transforming HCC management: immunotherapy in the clinic |
> Single agent immunotherapies
As with almost all drugs in HCC, immunotherapies are usually studied first in other malignancies, due to concerns about underlying liver disease in patients with HCC. The inability of single-agent ICIs to improve OS has led to the development of combination approaches.
Tremelimumab was first studied as a single agent in patients with HCV-related liver cancer; showed an early signal of efficacy, but there were still concerns regarding the safety of this class in patients with underlying liver disease.
Many other single-agent studies have been performed with other PD-1/PDL1 inhibitors, including pembrolizumab in the second-line setting, and durvalumab and tislelizumab in the first-line setting, and all yielded an objective response rate (ORR). similar in the range of 15-20% and similar well-tolerated side effect profiles. In the first-line setting, these agents have met non-inferiority survival endpoints compared with sorafenib, and in the second-line setting, pembrolizumab has met its OS vs placebo endpoint in some studies but not others.
> Improving survival with combinations
Early promising results from single-arm phase I/II studies with ICI were tempered by phase III readouts. Although durable single-agent responses were confirmed that were significantly greater than those reported with sorafenib, or even placebo, single-agent ICIs initially did not meet their endpoints in phase III trials. To improve these results, efforts to identify a biomarker have not yielded easily translatable results.
Furthermore, combining ICIs with other agents has been a successful strategy to improve OS and reestablish the standard of care for patients with advanced liver cancer.
The approval of the PD-L1 antibody atezolizumab and the VEGF antibody bevacizumab marked a change in practice. For the first time since the approval of sorafenib in 2007, a first-line regimen demonstrated superiority for OS in HCC. Based on the observation that HCC is a hypervascular tumor, anti-VEGF/VEGFR therapies have been extensively studied in HCC and, until recently, the only approved drugs in this space were related to this pathway.
Fundamental understanding of the biological effects of targeting the VEGF pathway has evolved over time, from one focused on “normalizing” tumor vasculature and affecting tumor oxygenation and nutrition, to effects on the immune microenvironment that change it from being immunosuppressive. an immune activator.
The IMbrave150 study was a global open-label phase III study that evaluated the combination of atezolizumab and bevacizumab vs sorafenib. The results of this study have supported the global approval of this regimen.
> ICI and VEGF multikinase inhibitors
Combining ICIs with multikinase inhibitors is another VEGF-based approach to increase therapeutic activity. The difference between TKIs and mAbs is that TKIs not only affect VEGFR but also other kinases that may play a role in modulating the activity of ICIs.
> Combined ICI
Co-targeting the immune priming phase with CTLA-4 inhibition in combination with the immune effector phase with PD-1/PD-L1 inhibition has become an attractive approach in cancer medicine. The first combination that was tested, based on this approach, in HCC was that of ipilimumab and nivolumab in the second line after previous sorafenib.
| Future directions |
The rapid transformation of the therapeutic arsenal for patients has left important knowledge gaps and, therefore, areas for future research. On the one hand, now that ICI-based therapy has moved into the first-line setting, optimal management in progression is not defined. For patients who do not benefit from one ICI combination, is there a role for another? Additionally, with several first-line combinations available, it would be valuable to develop more biomarkers that would help identify patients most likely to benefit from one combination over the other.
More insights can be gained by implementing radiomics and radiogenomics, a new area of research that seeks to correlate imaging features with genetic profiles. Hopefully, analyzes of tumors and circulating biomarkers included in ongoing immunotherapy studies, along with non-invasive image-based metrics, will provide the foundation for precision medicine in HCC.
Recent preclinical and clinical data suggest that non-viral HCC, and particularly NASH (non-alcoholic steatohepatitis)/NAFLD (non-alcoholic fatty liver disease)-related HCC, may be less responsive to immunotherapy. However, this observation has not been confirmed by other trials and redefinition of stratification factors beyond viral/nonviral etiology to consider NASH/NAFLD as an independent factor should be encouraged.
Another important challenge concerns patients with Child-Pugh B cirrhosis. Further evaluation of immunotherapy in this population should be encouraged, as nivolumab, tested in this setting, showed reassuring safety and efficacy. Recently, a real-world collaborative effort showed preliminary evidence of the safety and efficacy of atezolizumab plus bevacizumab in patients with Child-Pugh B cirrhosis, with tolerability similar to that of those with Child-Pugh A cirrhosis.
Another future direction is represented by the combination of ICI and locoregional treatments in early stages of HCC and ongoing trials may address unmet needs, such as increasing the percentage of patients eligible for curative therapies, including liver transplantation in the setting of a extreme migration stage, and reduce relapse rates.
Finally, other immune checkpoints may regulate T cell function and play important roles in tumor immune escape. Trials of ICIs targeting LAG-3, TIM-3 (T-cell immunoglobulin and mucin domain 3), or TIGIT (T-cell immunoreceptor with Ig and ITIM domains) in combination with PD-1/PD-L1 or CTLA inhibitors -4.
| Conclusions |
The prognosis of patients with advanced HCC has changed significantly with the development of combination regimens that are based on single-agent immunotherapy data. While we are seeing significant improvements in survival, with more patients achieving durable responses and favorable side effect profiles, there are still unmet needs for our patients.
Active research is seeking to better understand the mechanisms of intrinsic and acquired resistance to these regimens which will then direct the next generation of studies. There are already numerous early-stage studies evaluating new “triplet” regimens in the first line, as well as new molecular targets in the second line, in an effort to reverse ICI resistance. Results from ongoing phase III studies incorporating immune-based combinations into earlier lines of therapy are awaited and, if positive, will be practice-changing.
In summary, the addition of immunotherapy combinations to the HCC landscape is changing the natural history of the disease and will be the backbone of future drug development.
