Hepatocellular carcinoma (HCC) is the most common primary liver cancer, representing 85-90% of all primary liver cancers1. Worldwide, liver cancer is the sixth most frequent cancer and the fourth most frequent cause of cancer related death, representing a major global health problem2. The pathogenesis of HCC is an incompletely understood complex process that occurs in the context of chronic inflammatory damage arising from hepatocyte necrosis, regeneration and fibrosis deposition3.

There is a growing literature about the role of the tumour microenvironment (TME) and the cross-talk between tumour cells and their surrounding microenvironments in the pathogenesis and progression of HCC, tumour invasion and metastasis4. The TME is the cellular milieu in which the HCC tumour grows5. It is immunosuppressed through various mechanisms, facilitating tumour progression. Chronic inflammation, for example, due to chronic hepatitis B or C infection, creates an immunosuppressive microenvironment that permits HCC tumourgenesis and progression6. During HCC progression, as the microenvironment components interact with each other and with the tumour cells, they acquire an abnormal phenotype and modulate the biological behaviour of the tumour, facilitating cancer progression and metastasis7. By understanding components of the TME and their interaction with one another, the underlying mechanisms of hepatocarcinogenesis can be better defined, allowing the development of targeted therapies.

In very early and early stage HCC (Barcelona Clinic Liver Cancer 0 and A stages), liver resection and radiofrequency ablation are appropriate first line modalities8. Resection can also be considered in select Barcelona Clinic Liver Cancer (BCLC) B patients with preserved liver function (Child Pugh A)9, 10. Following resection for HCC, tumour recurrence complicates 70% of cases at five years11. Tumour recurrence can be broadly classified as early or late. Early recurrence generally occurs within two years and is predicted by adverse histological features including large tumour size, presence of multifocality, microvascular invasion, poor histological differentiation, satellite lesions. Late recurrence however, usually represents ‘de novo’ lesions arising from the carcinogenic effect of the underlying chronic liver disease12, 13.

Currently there are no effective therapies to reduce this risk of recurrence, and unfortunately Sorafenib failed to prevent tumour recurrence after hepatic resection or ablation14. To address this unmet clinical need in HCC, immune checkpoint inhibitor (ICIs) therapies are being evaluated in the neoadjuvant and adjuvant settings.

Immune Checkpoint Molecules
Currently most studies investigating immune evasion have focused on the programmed death receptor 1 (PD-1)/programmed death ligand 1 (PD-L1) pathway. PD-1 is an immune-inhibitory receptor involved in the regulation of T-cell activation15. It is expressed on activated T and B cells and natural killer T cells. Its ligand, PD-L1 is expressed on hematopoietic cells and non-hematopoietic cells, such as T cells, B cells, endothelial cells, epithelial cells, cancer cells and normal tissue16. Programmed death ligand 2 (PD-L2) expression on the other hand, is largely limited to antigen presenting cells including dendritic cells, macrophages and mast cells16. Wang et al showed that the expression of PD-L1 and PD-L2 is de¬tectable on Kupffer cells and liver sinusoidal endothelial cells in viral and non-viral hepatitis15.

Interaction of PD-1 with its ligand leads to immune suppression and peripheral tolerance by promoting regulatory T-cell proliferation, exhaustion of effector T cells17 and suppression of the activation of immune cells18. Tumour cells hijack this pathway by overexpressing PD-L1, enabling immune evasion18. Immune checkpoint inhibitors (ICIs) inhibit this interaction, causing resurrection of T-cell mediated anti-tumour effect19. Since the approval of ipilimumab in 2011 for malignant melanoma, several ICIs have been developed and undergoing Phase 3 trials in HCC.

The prognostic role of PD-L1 expression in HCC remains inconsistent and is controversial. Some studies16, 20 and a meta-analysis21 have shown that the expression of PD-L1 is correlated with a poor prognosis 18 following curative hepatectomy for HCC, whereas a recent metanalysis found no significant prognostic role of PD-L1 in this setting15. Wu et al22 showed that PD-L1 expression on Kupffer cells was increased in tumour tissue compared with non-tumour tissue in patients with HCC and that this correlated with poorer survival. Furthermore, they showed that PD-1+CD8+ T cells in HCC had reduced proliferative ability and effector function, as demonstrated by reduced granule and cytokine function, when compared with PD-1- T cells, indicating that the quality and function of CD8+ T cells is altered in HCC but can be restored when this interaction is blocked22. However, responses to ICIs in clinical trials have been suboptimal. In advanced HCC, the response rate to nivolumab (anti-PD-1 monoclonal antibody) ranged between 15%–20%, with responses occurring irrespective of tumour PD‐L1 expression23.

We therefore aim to interrogate the immune landscape of HCC in patients undergoing curative resection or ablation and correlate this with the risk of tumour recurrence and response to ICIs in a prospective cohort of patients with HCC.