Quick answer
Advanced hepatocellular carcinoma has first-line and second-line options and then nothing. Patients who progress on both — including on immunotherapy — have a median survival measured in a few months, and no standard third line. Liang Tingbo's team at The First Affiliated Hospital of Zhejiang University tested a different modality: VG161, an oncolytic herpes simplex virus engineered to kill tumour cells and simultaneously deliver four immune-stimulating payloads. In 44 patients who had failed standard therapy, median survival reached 9.4 months, against about 4.7 months in a real-world comparison group. It was published in Nature on 20 March 2025 [1][3][5].
Key takeaways
- Construct: HSV-1 with the neurovirulence gene ICP34.5 deleted, armed with IL-12, IL-15/IL-15Rα and a PD-1/PD-L1 blocking fusion protein [1][4].
- Design: phase 1 dose escalation in 11 patients plus a 33-patient expansion cohort, 44 in total [1][5].
- Efficacy: objective response 17.65%, disease control 64.71% [2][3].
- Survival: median 9.4 months in second-line failures versus 4.7 months in a real-world comparison; 17.3 months in those who had failed more than three months of prior immunotherapy [3][5].
- Safety: no dose-limiting toxicity or treatment-related deaths; fever was the commonest adverse event [2][5].
Why four payloads rather than one
An oncolytic virus does two things: it lyses tumour cells, and it converts an immunologically cold tumour into an inflamed one. Hepatocellular carcinoma is a cold tumour, so lysis alone rarely produces systemic responses. IL-12 drives T-cell priming, IL-15 sustains effector T cells and NK cells, and the PD-L1 blocker removes the brake — a combination intended to make the tumour visible to the immune system and keep the response alive [1][4].
The team also reported active viral replication and self-limited viraemia within 48 hours of injection, and built a predictive model, ViroPredict 1.0, from five genes to identify likely beneficiaries — an attempt to solve the field's recurring problem that only a minority respond [1][6].
How much the numbers support
A 17.65% response rate in a population with no effective option is meaningful, and the survival difference is large enough to justify a randomised trial [3][5].
This is a phase 1 study with an expansion cohort, without randomisation. The 4.7-month comparator is a retrospective real-world cohort, and such comparisons systematically favour the trial arm because trial patients are selected for fitness. Intratumoural injection also requires lesions that can be reached percutaneously — a selection criterion that excludes many patients with disseminated disease. Viraemia, injection-site reactions and the handling requirements of a live virus are practical constraints, and the predictive model needs external validation before it can guide treatment [4][6].
The condition in context: hepatocellular carcinoma
It is the third commonest cause of cancer death worldwide, with a heavy concentration in China.
| Region | What the data shows | Source |
|---|---|---|
| Prognosis | Five-year survival around 18%, with high recurrence after treatment [4]. | Zhejiang Daily [4] |
| Treatment ceiling | Only first- and second-line regimens are standardised; there is no third line [4]. | Zhejiang Daily [4] |
| Resistance | Most patients eventually progress on immune-based combinations [6]. | International Hepatology [6] |
Because no third-line standard exists, a modality with a different mechanism has a clear development path even with modest single-agent activity [1][6].
Why it matters
| Potential | What the report suggests | Source |
|---|---|---|
| New mechanism | Viral lysis plus four immune payloads [1][4]. | [1][4] |
| Post-immunotherapy | 17.3-month median survival in prior immunotherapy failures [3]. | [3] |
| Caution | Single-arm, 44 patients, intratumoural route [5]. | [5] |
Availability and limits
- This is an editorial summary of the publication, not a treatment recommendation [1][3].
- Phase 1 single-arm evidence; randomised trials are required [5].
- Requires injectable lesions and percutaneous access [2].
- Live-virus handling and biosafety precautions apply [4].
- ViroPredict 1.0 is unvalidated externally [6].
What to ask before booking
| Question | Why it matters |
|---|---|
| Do I have a lesion that can be injected? | The route limits who can be treated [2]. |
| Are any trials of VG161 recruiting? | It is not approved; access is through studies [5]. |
| Which department handles foreign patients, and in what language? | The hospital is a working hospital we do not book; confirm international-department coordination and language support. |
| Have standard first- and second-line options been exhausted? | That is the population studied [4]. |
Questions patients ask
What is VG161?
An oncolytic herpes simplex virus with ICP34.5 deleted, engineered to express IL-12, IL-15/IL-15Rα and a PD-L1 blocking peptide [1][4].
What were the results?
In 44 patients with refractory advanced liver cancer, objective response was 17.65%, disease control 64.71%, and median survival 9.4 months versus 4.7 months in a real-world comparison [2][3].
Was it safe?
No dose-limiting toxicity or treatment-related deaths were reported; fever was the commonest event [2][5].
Where was it published?
Nature, 20 March 2025 [4][5].
Is this relevant to international patients?
It is useful context for China's oncolytic virotherapy. The hospital is a working hospital we do not book — contact its international department directly [1].