Extracellular Vesicles as Liquid Biopsy Biomarkers Predicting Immune Checkpoint Therapy Resistance in Hepatocellular Carcinoma

Highlight

This study identifies programmed death-1 (PD-1), programmed death-ligand 1 (PD-L1), and cytotoxic T-lymphocyte associated protein 4 (CTLA-4) expressed on circulating extracellular vesicles (EVs) as robust liquid biopsy biomarkers in hepatocellular carcinoma (HCC). EV-immune checkpoint protein (IC) levels at baseline and early during therapy inform responses to anti-PD-L1-based immune checkpoint inhibitor (ICI) treatment and predict progression-free survival (PFS) and overall survival (OS). Notably, dynamic changes in EV-ICs preceded clinical progression by imaging by 36–42 weeks, enabling earlier identification of acquired resistance to immunotherapy. These markers were specific to immunotherapy response, with no predictive value in tyrosine-kinase inhibitor (TKI) treatments.

Study Background

Hepatocellular carcinoma represents a major global health burden with limited systemic therapy options and generally poor prognosis. The advent of immune checkpoint inhibitors, particularly those targeting the PD-1/PD-L1 axis, has improved outcomes for some patients with advanced disease. However, clinical benefit is seen in only approximately 30% of treated patients, underscoring the urgent need for predictive biomarkers to guide personalized therapy. Routine tumor tissue acquisition in HCC is challenging due to anatomical and clinical constraints, limiting biomarker development from tumor samples. Liquid biopsy-based approaches, especially those analyzing circulating extracellular vesicles, offer minimally invasive methods to assess tumor and immune status dynamically.

Study Design

This comprehensive, multi-cohort study evaluated circulating EVs carrying immune checkpoint molecules in HCC patients. Three distinct cohorts were included: (1) an HCC explorer cohort (n=40) for initial characterization of membrane-bound immune checkpoint proteins on EVs; (2) an early-stage HCC cohort (n=37) with paired tissue and blood samples to study EV and tissue IC expression correlations; and (3) a treatment cohort (n=202) encompassing patients treated with anti-PD-L1-based ICI therapy and tyrosine-kinase inhibitors (TKIs). The treatment cohort was subdivided into training (n=79; 402 serial samples) and validation (n=82; 146 samples) groups for generation and confirmation of predictive signatures. EV-IC protein levels (PD-1, PD-L1, CTLA-4) were quantified using multiplex immunoassays on serial blood samples. Clinical endpoints included best treatment response, PFS, and OS, while also assessing temporal changes indicative of resistance development.

Key Findings

EV Immune Checkpoint Protein Enrichment and Baseline Associations: PD-1, PD-L1, and CTLA-4 were significantly enriched in EV fractions relative to EV-depleted serum, confirming their EV-associated membrane-bound nature. Baseline EV-IC levels differed markedly between patients who subsequently responded to ICI therapy and non-responders, with responders demonstrating distinct signature profiles.

Predictive Capacity for Treatment Response and Survival: Baseline and early on-treatment (early dynamics) EV-IC measurements robustly discriminated responders from non-responders in both the training and independent validation cohorts. These markers predicted PFS and OS, with higher EV-IC expression correlating with improved outcomes. Conversely, in the TKI-treated group, EV-IC levels bore no predictive value, supporting specificity to ICI therapy.

Early Detection of Acquired Resistance: Serial monitoring revealed that dynamic changes in EV-IC levels during ongoing therapy preceded radiographic progression by approximately 36 to 42 weeks. Notably, upward trends in EV-ICs after initial response indicated emerging resistance well before clinical or imaging-based relapse, offering a critical window for therapeutic interventions.

Biological Relevance and Clinical Applicability: The study demonstrated interplay between tissue and circulating EV immune checkpoint profiles. This supports the biological plausibility of EV-ICs as surrogates reflecting tumor-immune microenvironment activity, bridging the gap created by the lack of routine tissue acquisition in HCC patients.

Expert Commentary

This landmark investigation offers compelling evidence that circulating EVs carrying PD-1, PD-L1, and CTLA-4 serve as informative, non-invasive biomarkers predictive of response and resistance to immune checkpoint blockade in advanced HCC. The ability to detect treatment failure months before imaging could reshape clinical decision-making, enabling earlier therapy adjustment or combination strategies. Notwithstanding, further prospective validation in larger, diverse populations and integration into clinical workflow are necessary before widespread adoption. Additionally, the biological mechanisms underlying EV cargo modulation during therapy warrant further exploration.

Conclusion

The profiling of immune checkpoint proteins on circulating extracellular vesicles represents a promising liquid biopsy approach to guide immunotherapy in hepatocellular carcinoma. These biomarkers provide actionable insights into treatment efficacy and emerging resistance with a lead time that surpasses conventional imaging. Incorporation of EV-IC monitoring could significantly enhance personalized management of HCC, optimizing the use of expensive and potentially toxic immune checkpoint inhibitors.

Funding and Clinical Trial Registration

The study was supported by institutional and governmental cancer research funds. Specific grant information was detailed in the original publication. Clinical trial registration details are available on PubMed entry PMID: 42562419.

References

Gorgulho J, Masood R, Buescher G, et al. Circulating extracellular vesicles carrying PD-1, PD-L1 and CTLA-4 inform resistance to anti-PD-L1-based therapy in HCC. Gut. 2026 Aug 6. doi:10.1136/gutjnl-2025-XXX
Additional context and supportive literature:
– El-Khoueiry AB, Sangro B, Yau T, et al. Nivolumab in patients with advanced hepatocellular carcinoma (CheckMate 040): an open-label, non-comparative, phase 1/2 dose escalation and expansion trial. Lancet. 2017;389(10088):2492-2502.
– Greten TF, Lai CW, Li G, Staveley-O’Carroll KF. Targeted and immune-based therapies for hepatocellular carcinoma. Gastroenterology. 2019;156(2):510-524.
– Kalluri R, LeBleu VS. The biology, function, and biomedical applications of exosomes. Science. 2020;367(6478):eaau6977.