# Evaluation of hyperprogressive disease with atezolizumab plus bevacizumab for hepatocellular carcinoma: A secondary analysis of the IMbrave150 trial

**Authors:** Yuan Gao, Ann‐Lii Cheng, Lee X. Li, Natalie Parent, Ganessan Kichenadasse, Christos S. Karapetis, Andrew Rowland, Ashley M. Hopkins, Michael J. Sorich

PMC · DOI: 10.1002/ijc.35407 · International Journal of Cancer · 2025-03-13

## TL;DR

This study finds that combining atezolizumab and bevacizumab for liver cancer is unlikely to cause rapid tumor growth, unlike some other treatments.

## Contribution

The study shows that the atezolizumab plus bevacizumab combination does not significantly increase hyperprogressive disease risk in liver cancer patients.

## Key findings

- The combination treatment had lower odds of hyperprogressive disease compared to sorafenib.
- Hyperprogressive disease in the combination group may reflect the natural disease course rather than treatment effect.
- High alpha-fetoprotein and neutrophil-to-lymphocyte ratio were linked to increased hyperprogressive disease risk.

## Abstract

The use of Immune checkpoint inhibitors (ICIs) as monotherapy for patients with hepatocellular carcinoma (HCC) has been associated with an increased risk of hyperprogressive disease (HPD), the occurrence of which carries a poor prognosis. However, it is unknown whether contemporary frontline treatment with the combination of atezolizumab and bevacizumab causes significant HPD. This study conducted a secondary analysis of patient‐level data from the IMbrave150 randomized controlled trial of atezolizumab plus bevacizumab versus sorafenib for frontline treatment of HCC. Multiple established definitions of early progression and treatment failure applicable to clinical trials were evaluated, including Response Evaluation Criteria in Solid Tumours (RECIST) HPD, HPD based on percent change of sum of longest diameter (SLD HPD), treatment failure HPD (TF HPD), and fast progression (FP). The incidence of these measures was compared between arms. The risk factors for and prognosis of TF HPD were evaluated. The risk of RECIST HPD and TF HPD was significantly lower with atezolizumab plus bevacizumab treatment than with sorafenib treatment—odds ratio for RECIST HPD: 0.29 (95% CI 0.01 to 0.82), TF HPD: 0.30 (0.17, 0.54). TF HPD was similarly associated with poor prognosis, irrespective of treatment arm. High blood alpha‐fetoprotein and neutrophil‐to‐lymphocyte ratio were both associated with an increased risk of TF HPD. For all definitions of early progression/treatment failure, the risk was either significantly lower with atezolizumab plus bevacizumab than with sorafenib, or there were no differences. Atezolizumab plus bevacizumab treatment is unlikely to cause significant HPD.

What's new?

Hyperprogressive disease (HPD) is defined by unusually rapid tumor growth following treatment initiation. In hepatocellular carcinoma (HCC), HPD is suspected of occurring more commonly with immune checkpoint inhibitors (ICIs), particularly when used as monotherapy, than with other treatment strategies. Here, combination treatment using the ICI atezolizumab plus the antiangiogenic agent bevacizumab was investigated for associations with HPD in HCC patients. Clinical data show that the combination is unlikely to increase HPD risk over sorafenib, a previous standard‐of‐care treatment for HCC. HPD detected in patients administered the combination treatment may be part of the HCC disease course.

## Linked entities

- **Chemicals:** sorafenib (PubChem CID 216239)
- **Diseases:** hepatocellular carcinoma (MONDO:0007256)

## Full-text entities

- **Genes:** AFP (alpha fetoprotein) [NCBI Gene 174] {aka AFPD, FETA, HPAFP}
- **Diseases:** Solid Tumours (MESH:D009369), HPD (MESH:D004194), HCC (MESH:D006528)
- **Species:** Homo sapiens (human, species) [taxon 9606]

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12079626/full.md

## Figures

1 figure with captions in the complete paper: https://tomesphere.com/paper/PMC12079626/full.md

## References

41 references — full list in the complete paper: https://tomesphere.com/paper/PMC12079626/full.md

---
Source: https://tomesphere.com/paper/PMC12079626