Radioembolisation
Radioembolisation delivers tiny radioactive beads directly into the blood vessels feeding a liver tumour through a catheter, so the radiation stays concentrated in the liver instead of spreading through the whole body.
Overview
Transarterial radioembolisation (TARE), also called selective internal radiation therapy (SIRT), delivers yttrium-90-loaded microspheres (glass, e.g. TheraSphere, or resin, e.g. SIR-Spheres) through a hepatic artery catheter to treat unresectable hepatocellular carcinoma, and is also used for liver-dominant metastases from colorectal cancer and neuroendocrine tumours. The pivotal SARAH and SIRveNIB trials tested Y-90 TARE against sorafenib in advanced HCC and found radioembolisation was not superior to systemic therapy on survival, which tempered early enthusiasm for TARE as a stand-alone replacement for systemic treatment in advanced disease.
The practice-changing development since has been personalised dosimetry. The randomised phase 2 DOSISPHERE-01 trial compared a standard single-compartment dosing approach (target 120±20 Gy to the perfused liver lobe) against a personalised multi-compartment approach targeting at least 205 Gy to the tumour in locally advanced HCC, and found median overall survival of 26.6 months with personalised dosing versus 10.7 months with standard dosing (HR 0.42, 95% CI 0.22-0.83, p=0.0096), with objective response 71.4% versus 35.7% and 36% of personalised-arm patients downstaged to surgery versus 4% in the standard arm (Garin et al., Lancet Gastroenterol Hepatol 2021). This has shifted practice guidance toward partition-model dosimetry using pre-treatment MAA SPECT rather than fixed body-surface-area dosing.
What is still unsettled: TARE's role relative to newer systemic options (immunotherapy combinations, TACE) in intermediate-stage disease is being re-tested in ongoing trials combining radioembolisation with checkpoint inhibitors (e.g. durvalumab and bevacizumab), and while personalised dosimetry clearly improves outcomes in the DOSISPHERE-01 population, routine adoption of partition-model dosimetry across centres worldwide is uneven, similar to the dosimetry debate in radioligand therapy.
Notes
top- SARAH and SIRveNIB both found Y-90 TARE was not superior to sorafenib on survival in advanced HCC, so TARE has not replaced systemic therapy in that setting.
- DOSISPHERE-01 is the clearest randomised evidence that personalised, partition-model dosimetry meaningfully changes outcomes in radioembolisation (median OS 26.6 vs 10.7 months), but routine uptake of this dosimetry approach still varies by centre.
- Trials combining TARE with checkpoint inhibitors and anti-angiogenics (e.g. durvalumab/bevacizumab, NCT06040099) are testing whether locoregional and systemic immunotherapy combine safely and effectively; results are pending.