Yttrium-90 radiation segmentectomy requires tumor-specific dosimetry beyond prescribed segment dose

Evidence is strongest in hepatocellular carcinoma, where higher tumor doses correlate with pathological necrosis, while non-HCC dose thresholds remain uncertain.

KEY POINTS

  • This narrative review examined yttrium-90 radiation segmentectomy (RS) for hepatocellular carcinoma (HCC), intrahepatic cholangiocarcinoma, and liver metastases from colorectal, neuroendocrine, and other malignancies. Literature published from 1995 through September 2026 was evaluated, with emphasis on original clinical studies, explant pathology, absorbed-dose measurements, and treatment outcomes. RS involves superselective transarterial delivery of radioactive microspheres to tumor-bearing hepatic segments, with the intention of delivering an ablative dose while preserving uninvolved liver.
  • Prescribed segmental dose and actual tumor-absorbed dose are not necessarily equivalent. In single-compartment MIRD dosimetry, dose is assigned uniformly across the perfused volume, making calculated segment and tumor doses identical by model construction. However, partition and voxel-based methods account for preferential tumor uptake and heterogeneous microsphere distribution. In the original HCC series, median segmental dose was 521 Gy versus estimated tumor dose of 1,214 Gy; the prospective RASER cohort reported corresponding values of 584 versus 1,004.6 Gy. Consequently, dose thresholds cannot be transferred directly between dosimetric models.
  • Explant studies demonstrate a dose-response relationship for complete pathological necrosis in HCC. In a multicenter series of 303 patients with 364 tumors treated with glass microspheres before transplantation, complete pathological necrosis occurred in 68% overall and 76% (184/241) of tumors receiving more than 400 Gy. An optimized combination of dose, microsphere-specific activity, particle density, and angiosome-to-tumor volume ratio was associated with 91% complete necrosis (57/63 tumors). Smaller explant studies also demonstrated increasing necrosis with higher doses, although thresholds depended on the microsphere product and calculation method.
  • Prospective and large-cohort HCC studies support substantial antitumor activity in selected patients. The prospective, single-arm RASER study treated 29 patients with solitary HCC ≤3 cm, achieving 83% complete response and 96% two-year overall survival. The multicenter LEGACY study included 162 patients with solitary HCC ≤8 cm, reporting 88.3% objective response and 86.6% three-year overall survival. However, LEGACY included different selective infusion strategies, and only 95/162 patients were explicitly categorized as receiving radiation segmentectomy, limiting attribution of all cohort outcomes specifically to segmental treatment.
  • Microsphere type affects the interpretation of absorbed-dose thresholds. Glass and resin microspheres differ in specific activity, particle number, and dose distribution. In a glass-microsphere explant series, complete pathological necrosis increased from 53% at doses >190 Gy to 75% at >500 Gy. In a resin-microsphere study, post-treatment PET/CT tumor D50 ≥300 Gy was associated with a lower two-year progression rate than D50 below 300 Gy (17% vs 61%; p=0.047). These values are device- and model-specific, not interchangeable universal prescriptions.
  • Evidence for non-HCC malignancies remains substantially less mature. In 36 patients with 57 colorectal liver metastases, mean tumor dose ≥400 Gy combined with mean absorbed dose ≥350 Gy to a 5-mm peritumoral margin was associated with 83.3% one-year local tumor progression-free survival, but this margin-dose hypothesis derives from a single retrospective study. A small intrahepatic cholangiocarcinoma series involving 10 patients and 13 tumors reported 100% objective response after segmental doses above 400 Gy. Among 18 patients with 23 neuroendocrine liver metastases, response was 83% by RECIST 1.1, with 100% response by mRECIST among 12 hypervascular lesions evaluated using enhancement criteria.
  • Treatment safety depends on the volume of functional liver exposed and baseline hepatic reserve. In a retrospective HCC analysis, treatment of approximately 14.5% or more of liver volume was associated with increased risk of biochemical hepatic deterioration among patients with ALBI grade 2 or Child–Pugh B function, but no comparable threshold was identified for patients with better preserved function. The review also emphasizes substantial between-study heterogeneity in patient selection, response criteria, dosimetric models, and segmental versus lobar delivery. No prospective study has established a universal relationship between prescribed segmental dose and achieved tumor dose across different histologies.

CLINICAL TAKEAWAY

Yttrium-90 radiation segmentectomy is a potentially curative locoregional option for selected patients with early-stage hepatocellular carcinoma, supported by pathological dose-response correlations and prospective single-arm evidence. However, treatment planning must distinguish segmental from tumor-absorbed dose and account for microsphere type, tumor vascularity, and hepatic reserve, while histology-specific thresholds for non-HCC tumors require further prospective validation.

SOURCE

Cancers

Browse more research Suggest a correction