Noninvasive prediction of the clinical benefit of immunotherapy in hepatocellular carcinoma

Abstract

Long-term survival following a diagnosis of hepatocellular carcinoma (HCC) is greatly diminished when transplantation and surgical resection are ruled out. Fortunately, the advent of immune checkpoint inhibitors (ICIs) has revolutionized the treatment of advanced unresectable HCC (uHCC), prolonging median survival by over a year. T lymphocytes normally eliminate neoplastic cells, but some tumors suppress this response by binding to immune checkpoint receptors. Blocking this interaction via ICIs restores immune-mediated targeting of cancer cells. While ICI-based combination immunotherapy is currently recommended as the first-line systemic therapy for uHCC, the objective radiological response rate remains limited to 20-30%, as not all tumors exploit this mechanism. Consequently, strategies are being explored to modulate the immune microenvironment into a "hot" environment more responsive to ICIs by combining local therapies such as transarterial chemoembolization, ablation, and radiation therapy. Therapeutic options have also expanded beyond ICIs, emphasizing the importance of selecting the most appropriate treatment. Therefore, the development of biomarkers capable of predicting the efficacy of immunotherapy is a priority. Direct evaluation of immune cell infiltration through biopsy is currently the most effective method but involves issues such as invasiveness and susceptibility to sampling bias. In this review, we aim to highlight promising non-invasive biomarkers and scoring systems that have the potential to improve treatment outcomes, including blood-based biomarkers such as lymphocyte ratios, cytokines, C-reactive protein, and alpha-fetoprotein; imaging biomarkers such as MRI, ultrasound, and contrast-enhanced CT; and other clinical indicators such as sarcopenia, grip strength, and diversity of the gut microbiome.

Keywords: Biomarker; Hepatocellular carcinoma; Immune checkpoint inhibitor; Immune microenvironment; Prediction.