WHAT THE STUDY ACTUALLY SAYS

Adding ablation to chemoembolisation extends survival in liver cancer

In the phase 3 TORCH trial, following transarterial chemoembolisation with thermal ablation more than doubled median progression-free survival, to 17.7 months from 7.3, in unresectable liver-confined cancer.

Median progression-free survival in TORCHTACE + thermal ablation: 17.7months; TACE alone: 7.3months0months15months30monthsTACE + thermal ablation17.7monthsTACE alone7.3months
Median progression-free survival in TORCH
GroupValue (months)
TACE + thermal ablation17.7 (11.4 to 23.1)
TACE alone7.3 (6.4 to 10.4)
Median progression-free survival in TORCH Liver-confined unresectable hepatocellular carcinoma (BCLC stage B); chemoembolisation followed by thermal ablation versus chemoembolisation alone. Source: JAMA Oncology

Following the standard artery-blocking treatment for inoperable liver cancer with a session of heat-based tumour destruction sharply slowed the disease: in the phase 3 TORCH trial, median progression-free survival was 17.7 months when thermal ablation was added after transarterial chemoembolisation, against 7.3 months with chemoembolisation alone [s1]. Overall survival separated even more widely, and severe side-effects were similar in the two groups [s1].

Hepatocellular carcinoma is the most common primary liver cancer, and many patients are diagnosed when the tumour is confined to the liver but too large or too widespread for surgery or a curative single ablation. For that group — classified as Barcelona Clinic Liver Cancer (BCLC) stage B — the standard treatment is transarterial chemoembolisation, or TACE: a catheter delivers chemotherapy directly into the artery feeding the tumour and then plugs it, cutting off the blood supply [s1]. TACE controls disease for a time but rarely eradicates it, and the trial's premise was that finishing the job with thermal ablation — using radiofrequency heat to kill tumour tissue TACE has softened up — might do better than TACE repeated on its own [s1].

What they did

TORCH was an open-label trial run at two tertiary centres in China, enrolling patients from May 2015 to August 2024 with a data cutoff of 31 October 2025 [s1]. It randomly assigned 241 patients with BCLC stage B, liver-confined disease in a 1:1 ratio: 121 to TACE followed by selective radiofrequency ablation, and 120 to TACE alone [s1]. The two groups were closely matched on tumour burden, scored by the "6-and-12" system that adds a tumour's largest diameter to the number of nodules — most patients in both arms fell into the low or intermediate bands (below 6, or 6 to 12 points) rather than the highest [s1]. The primary endpoint was progression-free survival measured by standard imaging criteria [s1]. The trial was registered as NCT02435953 [s2].

What it showed

At the data cutoff, median progression-free survival was 17.7 months (95% confidence interval, 11.4 to 23.1) with the combination against 7.3 months (6.4 to 10.4) with TACE alone — a hazard ratio of 0.47 (95% confidence interval, 0.34 to 0.65; p<0.001), meaning the combination cut the rate of progression or death by roughly half [s1]. Median overall survival was 88.6 months with TACE-ablation versus 35.1 months with TACE alone (hazard ratio 0.50; 95% confidence interval, 0.34 to 0.73; p<0.001) [s1]. The trial also tracked "untreatable progression" — the point at which the cancer can no longer be treated with these local approaches — which came at a median of 35.1 months with ablation added and 12.3 months without (hazard ratio 0.40) [s1].

The survival figures are unusually long for stage B liver cancer and deserve a reader's caution rather than excitement alone. The trial recruited over nine years at two experienced Chinese centres, the benefit was concentrated in patients with low-to-moderate tumour burden, and the very long overall-survival estimate in the combination group rests on data that will keep maturing [s1]. What TORCH establishes is a large, consistent advantage on the endpoint it was built to measure — time without progression — in the patients it enrolled.

The cost side

Adding an ablation session did not meaningfully raise serious harm. Grade 3 or 4 treatment-related adverse events occurred in 23.2% of the TACE-ablation group and 18.3% of the TACE-alone group [s1]. That both intensive, catheter-and-needle procedures carry similar rates of severe complications is part of the case for the combination: the extra step bought a large gain in disease control without a matching rise in toxicity, at least in the hands of the centres that ran the trial [s1].

What to watch

TORCH is a single-country, two-centre trial, and its striking numbers will need confirmation in other health systems and by other operators before "TACE then ablate" becomes a universal standard for stage B disease. The result speaks only to liver-confined cancer of low-to-moderate burden, not to the larger, multifocal tumours that fell outside its strongest benefit, and not to disease that has spread beyond the liver, where systemic drugs remain the mainstay. For most people, the larger gains in liver cancer are still upstream of any of this — in preventing and treating the chronic viral hepatitis and fatty liver disease that cause most cases in the first place.

Sources

  • [s1] Transarterial Chemoembolization Plus Thermal Ablation in Unresectable Hepatocellular Carcinoma: The Phase 3 TORCH Randomized Clinical Trial. JAMA Oncology. 2026;12(9):980-989. doi:10.1001/jamaoncol.2026.2366.
  • [s2] ClinicalTrials.gov. Transarterial Chemoembolization Combined With Radiofrequency Ablation for Hepatocellular Carcinoma (TORCH, NCT02435953). U.S. National Library of Medicine.

Sources

  1. Transarterial Chemoembolization Plus Thermal Ablation in Unresectable Hepatocellular Carcinoma: The Phase 3 TORCH Randomized Clinical Trial — JAMA Oncology , July 30, 2026
  2. Transarterial Chemoembolization Combined With Radiofrequency Ablation for Hepatocellular Carcinoma (TORCH, NCT02435953) — ClinicalTrials.gov, U.S. National Library of Medicine , August 4, 2026

More on

Related coverage