Background
Phase I study (University of Toronto/Princess Margaret Hospital). 41 patients with unresectable Child-Pugh A hepatocellular carcinoma (n=31) or intrahepatic cholangiocarcinoma (n=10) not suitable for standard therapies. Six-fraction SBRT over 2 weeks using an individualized dose prescription based on liver volume irradiated and normal tissue complication probability (NTCP) modeling. Dose escalated from 5% to 10% to 20% liver toxicity risk within three liver volume strata.
Interventions and follow up
Arm A: Individualized 6-fraction SBRT (24–54 Gy; median 36 Gy) based on NTCP liver tolerance model
Arm B: N/A — single-arm phase I dose-escalation study
Primary endpoint: Safety (dose-limiting liver toxicity at 3 months)
mFollow up: Reported at 3-month toxicity assessment; survival follow-up reported separately
Arm B: N/A — single-arm phase I dose-escalation study
Primary endpoint: Safety (dose-limiting liver toxicity at 3 months)
mFollow up: Reported at 3-month toxicity assessment; survival follow-up reported separately
Results
Grade 3 liver enzyme elevation: 12% (5 patients)
Radiation-induced liver disease (RILD): 0% within 3 months
Grade 4/5 toxicity: None
Median survival: 11.7 months (HCC), 15.0 months (iCCA)
Child-Pugh A→B decline (3 months): 7 patients (17%)
Radiation-induced liver disease (RILD): 0% within 3 months
Grade 4/5 toxicity: None
Median survival: 11.7 months (HCC), 15.0 months (iCCA)
Child-Pugh A→B decline (3 months): 7 patients (17%)
Adverse events
Main adverse events: No RILD or grade 4/5 toxicity. Grade 3 liver enzyme elevation 12%. Two iCCA patients had transient biliary obstruction after first fractions. 17% had Child-Pugh class decline within 3 months (acceptable given dose escalation).
Conclusions
Individualized 6-fraction SBRT based on NTCP modeling is safe for unresectable HCC and intrahepatic cholangiocarcinoma, with no RILD and acceptable toxicity. Median survivals of 11.7 months (HCC) and 15.0 months (iCCA) were promising, establishing the feasibility foundation for the subsequent Bujold phase I/II trial.
Key Limitations
Key Limitations: Small phase I trial (41 patients); no efficacy endpoints (local control not reported). Median SBRT dose of 36 Gy is lower than ablative doses used in modern practice. Child-Pugh A only — results not applicable to decompensated cirrhosis. No comparison group. Median tumor size 173 mL is large, limiting generalizability to smaller tumors treated today.
Clinical Context
The Toronto SBRT program established individualized NTCP-based dose prescription as a foundational approach for liver SBRT. This framework underpins modern liver SBRT practice and subsequent trials (Bujold phase I/II, RTOG 1112). The 6-fraction regimen over 2 weeks remains a widely used fractionation scheme for liver SBRT.
References