Multiomics analysis of immune correlatives in hepatocellular carcinoma patients treated with tremelimumab plus durvalumab

Treg-CD8+ T-cell interactions as key immune correlates of tremelimumab/durvalumab responses in HCC

Combination immunotherapy regimens with checkpoint therapy are now widely used as first-line systemic therapy of hepatocellular carcinoma (HCC). However, objective responses are limited to a fraction of the patients—such as in the STRIDE (single tremelimumab [anti-CTLA4] regular interval durvalumab [anti-PD-L1]) treatment arm of the HIMALAYA trial, where an objective response rate of 20.1% was observed (DOI: 10.1056/evidoa2100070). Efforts to identify immune correlatives of treatment response are crucial for better tailoring of therapies and identification of resistance mechanisms. Immune features in the tumor microenvironment are particularly promising as predictive biomarkers and previous work has highlighted a likely predictive role for CD8⁺ T cells and NK cells using spatial profiling approaches in patient cohorts with checkpoint therapy regimen (DOI: 10.1038/s41591-022-01868-2; DOI: 10.1136/gutjnl-2024-332837; DOI: 10.1038/s41586-025-08668-x). In the current publication in Gut by Myojin et al., the team led by Prof. T. Greten used a multiomic approach to identify immune correlatives of response in patients treated with tremelimumab and durvalumab as well as locoregional therapies in a phase 2 prospective trial with longitudinal sampling. Patients with a progression-free survival (PFS) longer than 6 months were assessed as responders (13/28, 46%). Using bulk transcriptomics, the authors first identified enhanced interferon signaling as a surrogate of immune activity in responders. Single-cell transcriptomic analysis of tumor samples found enhanced T-cell exhaustion signatures at baseline in non-responders, as well as some increase in cytokine production post-therapy. Highly multiplexed imaging analysis using the CODEX platform was then used to understand the spatial cellular interactions. Interestingly, this approach showed an increase in tumor immune cell infiltration independent of the response. However, in non-responders, a stronger interaction between regulatory T cells (Tregs) and CD8⁺ T cells was observed, suggesting that Treg activity limits immune invigoration. In line with this, pathways associated with production of inhibitory and immunosuppressive cytokines (e.g., IL-10, TGF-β) were overrepresented in non-responders. Analysis of corresponding blood samples also identified elevated Treg frequencies in non-responders, while activated CD8⁺ T cells dominated in responders. In the trial cohort studied, the anti-CTLA4 treatment dose was split over 4 different applications, while in the approved STRIDE regimen, this dose is combined and applied at the beginning of therapy. Using a mouse model mimicking these treatment strategies, the authors showed that a STRIDE-like approach led to stronger CD8⁺ T-cell activation and reduced Treg and myeloid cell infiltration compared to the split dosing approach. 
The study highlights that both immune features in the tumor immune microenvironment and the peripheral blood can serve as surrogates of immunotherapy response in HCC. The findings further highlight that differences in dosing and application strategies of checkpoint therapy antibodies may crucially influence the strength of T-cell reinvigoration. Pharmacodynamic monitoring of the T-cell response as a surrogate of checkpoint therapy efficacy was first implemented in melanoma treatment and helped to understand different mechanisms of immunotherapy (DOI: 10.1038/nature22079; DOI: 10.1038/s41590-024-02027-0). The combination of CTLA4 blockade with PD-1 blockade in particular changes CD4⁺ T-cell activation in lymphatic tissues including regulatory T-cell function, which aligns with the data of this study in HCC. Crosstalk between CD4⁺ and CD8⁺ T cells emerges as a critical mechanism underlying successful immunotherapy in HCC. Including immunomonitoring in prospective clinical trials will be crucial for improving HCC immunotherapies.