Baseline Cytokines May Predict Response to Immunotherapy and Anti-VEGF Therapy in Advanced Hepatocellular Carcinoma

Clinical Summary:

• Design/Population: This retrospective biomarker study analyzed serial blood samples from 56 patients with advanced hepatocellular carcinoma enrolled in the Johns Hopkins Liver Biobank who received first-line immune checkpoint inhibitor therapy with or without bevacizumab. Baseline and on-treatment cytokine profiles were evaluated as potential predictors of treatment outcomes.
• Key Outcomes: Elevated baseline inflammatory cytokines were associated with shorter progression-free survival, while selected cytokine profiles identified patients who appeared to derive greater benefit from the addition of bevacizumab. Treatment with bevacizumab also attenuated increases in several cytokines associated with myeloid-mediated immunosuppression.
• Clinical Relevance: These findings suggest that circulating cytokines may serve as biomarkers for treatment selection and provide insight into the immunomodulatory effects of anti-VEGF therapy in advanced hepatocellular carcinoma.

Catherine Wilbur, MD, Johns Hopkins University, Baltimore, Maryland, discusses a translational biomarker study evaluating baseline and dynamic cytokine profiles in patients with advanced hepatocellular carcinoma treated with immune checkpoint inhibitors with or without bevacizumab.

The analysis identified inflammatory cytokine signatures associated with clinical outcomes and suggests that bevacizumab may alter the systemic immune microenvironment by suppressing myeloid-associated inflammatory signaling. These hypothesis-generating findings support prospective validation of cytokine-based biomarkers to personalize first-line treatment selection for advanced hepatocellular carcinoma.

Transcript: 

Hi, my name is Catherine Wilbur. I'm an assistant professor of oncology at Johns Hopkins. I'm excited to speak with you today about our recent publication looking at baseline and dynamic cytokines as biomarkers for immune checkpoint inhibitor and anti-VEGF therapy in advanced hepatocellular carcinoma.

We know that the treatment landscape for first-line treatment of hepatocellular carcinoma (HCC) is rapidly changing. Over the past few years, we've seen multiple new regimens become NCCN- and FDA-approved for treatment, including atezolizumab plus bevacizumab and durvalumab plus tremelimumab. In this setting, bevacizumab is the anti-VEGF agent.

What we tried to do was determine whether there were biomarkers that could help predict which patients should receive which regimen in the first-line setting. As part of the Johns Hopkins Liver Biobank, patients receiving immune checkpoint inhibitor therapy as standard of care were enrolled into our biobank protocol. They had blood drawn at baseline and then throughout their treatment course. We retrospectively analyzed 56 patients enrolled in this protocol who had received treatment with either immune checkpoint inhibitor alone or immune checkpoint inhibitor plus bevacizumab.

Among those 56 patients, the median age was about 65 years. The majority were White, were male, had cirrhosis, and most commonly had hepatitis C as the underlying cause. Of those 56 patients, 35 received upfront treatment with immune checkpoint inhibitor alone, and 21 patients received immune checkpoint inhibitor plus bevacizumab.

We then looked at their cytokines, specifically their baseline cytokine levels before starting treatment. What we found was that above-median levels of MIP-1α, MIP-1β, and IL-8 were associated with worse progression-free survival. So patients with a high myeloid inflammatory signature at baseline actually had worse clinical outcomes.

We also looked at baseline cytokine levels to see whether treatment with bevacizumab plus immune checkpoint inhibition compared with immune checkpoint inhibitor alone affected those outcomes. What we found was that elevated baseline levels of IL-6 and IL-12p40, together with below-median levels of RANTES, were associated with longer median progression-free survival in patients treated with bevacizumab plus immune checkpoint inhibition.

Finally, we looked at cytokine levels throughout the course of treatment. What we saw was that bevacizumab actually attenuated, or blunted, the increase in multiple cytokines during treatment. Examples included MIP-1β, MIG, IL-1 receptor alpha, and soluble CD40 ligand.

Overall, when we look at these data, we recognize that this is a highly exploratory analysis. We really performed this study as a hypothesis-generating starting point as we begin thinking more and more about which treatment is right for which patients.

Overall, these findings suggest that bevacizumab modulates systemic immune signaling. Many of the cytokines that appeared to be prognostic in this study are involved in the recruitment and activation of immunosuppressive myeloid cell populations.

As we think about next steps, it's exciting that we may ultimately identify prognostic biomarkers that help us better select treatment for individual patients.

We're excited to move this work into a larger, prospective, randomized study to determine exactly which cytokines are most informative and to better define their prognostic implications.

Source: 

Wilbur HC, Zhao LX, Nakazawa M, et al. Baseline and dynamic cytokines as biomarkers for immune checkpoint and anti-VEGF therapy in advanced hepatocellular carcinoma. ESMO Gastrointest Oncol. Published online: February 4, 2026. doi: 10.1016/j.esmogo.2025.100273