Clinical Summary:

• Design/Population: This multicenter, open-label, phase 2 study evaluated transarterial tirapazamine embolization (TATE) combined with nivolumab in patients with advanced hepatocellular carcinoma who had progressed on at least one prior immunotherapy-containing regimen. Eligible patients had liver-dominant BCLC stage C disease, Child-Pugh A to B7 liver function, and measurable hepatic lesions.
• Key Outcomes: The combination demonstrated encouraging intrahepatic and extrahepatic response rates with durable disease control in a heavily pretreated population. The regimen showed promising clinical activity despite prior exposure to multiple systemic and locoregional therapies, supporting further evaluation in a randomized trial.
• Clinical Relevance: These findings suggest that combining hypoxia-activated transarterial therapy with immune checkpoint inhibition may represent a novel treatment strategy for patients with immunotherapy-refractory hepatocellular carcinoma.

Nadine Abi-Jaoudeh, MD, University of California, Irvine, California, discusses results from a phase 2 study evaluating transarterial tirapazamine embolization (TATE) combined with nivolumab in patients with advanced hepatocellular carcinoma that had progressed following prior immunotherapy. The study explored whether locoregional induction of tumor hypoxia could enhance antitumor immune responses when paired with PD-1 blockade.

The combination demonstrated encouraging response rates and durable disease control in a heavily pretreated population, including activity outside the liver despite treatment being directed only to hepatic lesions. These findings support further investigation of TATE plus nivolumab as a potential second-line treatment strategy for advanced hepatocellular carcinoma.

These results were presented at the American Society of Clinical Oncology (ASCO) Gastrointestinal Cancers Symposium in San Francisco, California.

Transcript: 

My name is Nadine Abi-Jaoudeh, I'm the section chief of IR and director of clinical research in the department of radiology at the University of California, Irvine. I will be discussing a poster presented at ASCO GI 2026 titled, “A phase 2 open-label single-arm multicenter study combining transarterial tirapazamine embolization (TATE) with nivolumab in immune-refractory advanced HCC patients.” 

As we all know, there are several first-line therapies available for patients with advanced HCC, including atezo/bev, durva/tremi, or ipi/nivo. However, once patients fail or progress on first-line immunotherapy, there's really no established second-line therapy for HCC patients and this is the population that was targeted with this trial. They had to be patients with BCLC- stage C, they had to have progressed on at least 1 line of systemic therapy that had to be immunotherapy, they had to have a Child-Pugh between 5 and 7, ECOG 0 to 2, they must have at least 1 liver lesion measuring a minimum of 2 cm but it could be up to 15 cm and, in fact, up to 50% of the liver could be replaced with tumor so, a pretty advanced population. 

We assessed the efficacy as determined by overall response rate by RECIST and mRECIST, as well as safety of the combination of nivolumab plus transarterial tirapazamine embolization. Patients were started on nivolumab every 3 weeks. After the first infusion, specifically 1 week after the initiation of nivolumab, patients underwent TATE. This consists of catheterizing the arteries, supplying the tumor and going as super-selective as possible, and then we would infuse TPZ followed by embolization. TPZ (tirapazamine) is a hypoxia-activated agent so really, we're relying on the synergistic effect between embolization, which causes hypoxia, and the hypoxia-activated agent. The rationale is we would infuse the TPZ and then immediately after embolize the tumor with bland embolization. This bland embolization induces hypoxia, which activates the TPZ already in the tumor. This results in necrosis of the tumor cells and release of DAMP as well as antigens. The rationale was that this inflammation and immune activation that resulted from the TATE, with the primed immune system by nivolumab, would result in an immune reaction.  

We had 18 patients that were treated on this trial that were evaluable. Of those, we had a 55% response rate by RECIST and 19% by mRECIST in the liver. This is much higher than even some of the response rates that were seen in first-line advanced HCC trials, with dual checkpoint inhibitors, it was around 25 to 30%. Five of the patients had extrahepatic disease. In immune-refractory population, since TATE is only done in the liver, we had no reason to suspect that there would be a response in those patients in the extrahepatic disease, but we did see that, in fact, 60% of patients had a response in their extrahepatic disease, including complete response by RECIST and 2 partial responses of 63% and 65%, respectively, by RECIST. These responses were durable, and in fact, we had a disease control rate of 77.8% by RECIST. 

This population, as I mentioned, was very advanced. A lot of them had received an average 1.9 lines of systemic therapy, the majority was atezo/bev, but there was durva/tremi, ipi/nivo, and some of them had received nivo either in the combination or as a single agent. A lot of them had received several locoregional therapies ahead of the transarterial tirapazamine embolization, including transarterial radioembolization, transarterial chemoembolization, ablation, and surgery. We still had very encouraging results in this population, and we hope to move to the randomized control trial so that we can in the future establish a second-line therapy for HCC patients who have failed first-line immunotherapy. 

Source: 

Abi-Jaoudeh N, Valerin JB, Fernando D, et al. A phase 2, open-label single-arm, multi-center study of trans-arterial tirapazamine embolization (TATE) combined with nivolumab (nivo) in patients with advanced immunotherapy-refractory hepatocellular carcinoma (HCC). Presented at ASCO Gastrointestinal Cancers Symposium. January 8 - 12, 2026; San Francisco, California. Abstract 538