DOI: 10.1007/s00259-018-4064-6Pages: 2110-2121

Radioembolization with 90Y glass microspheres for hepatocellular carcinoma: significance of pretreatment 11C-acetate and 18F-FDG PET/CT and posttreatment 90Y PET/CT in individualized dose prescription

1. Hong Kong Sanatorium & Hospital, Department of Nuclear Medicine & PET

2. Hong Kong Sanatorium & Hospital, Medical Physics & Research Department

3. Hong Kong Sanatorium & Hospital, Comprehensive Oncology Center

Correspondence to:
Chi Lai Ho
Tel: (+852) 2835-8848




The aim of this study was to establish an algorithm for the prescription of 90Y glass microsphere radioembolization (90Y-GMRE) of HCC in individual patients based on the relationship between tumour dose (TD) and response validated by 90Y PET/CT dosimetry and dual-tracer PET/CT metabolic parameters.


The study group comprised 62 HCC patients prospectively recruited for 90Y-GMRE who underwent pretreatment dual-tracer (11C-acetate and 18F-FDG) PET/CT as surrogate markers of HCC cellular differentiation. Pretreatment tumour-to-nontumour ratio on 99mTc-MAA SPECT/CT (T/NTMAA) was correlated with posttreatment 90Y PET/CT T/NT90Y after quantification validation. The TD–response relationship for HCC of different tracer groups was assessed on follow-up PET/CT 2 months after treatment.


90Y PET/CT was accurate in the measurement of recovery of injected 90Y activity (81.9–99.9%, median 94.8%). Pretreatment SPECT/CT T/NTMAA was strongly correlated with posttreatment 90Y PET/CT T/NT90Y (5.6 ± 3.2 versus 5.9 ± 3.5, T/NT90Y 1.01 × T/NTMAA + 0.161, r = 0.918, P < 0.05). The response rates were 72.4% (21/29), 70.6% (12/17) and 25% (4/16) for well, moderately and poorly differentiated HCC, respectively. The cut-off TD for a good response was significantly different between poorly differentiated and well/moderately differentiated HCC (262 Gy versus 152/174 Gy) with 89.2% sensitivity and 88% specificity. At a limiting tolerated liver dose of 70 Gy, the T/NTMAA thresholds for predicting a good response in poorly differentiated and well/moderately differentiated HCC were 3.5 and 2.0/2.3. Disregarding HCC cellular differentiation, the cut-off TD became 170 Gy, with lower sensitivity (70.3%) and specificity (76%).


90Y PET/CT can provide accurate dosimetry for 90Y-GMRE. Pretreatment T/NTMAA predicts posttreatment T/NT90Y. The TD thresholds for a good response are tracer-dependent, with a strong correlation between HCC radiosensitivity and cellular differentiation and other PET-based parameters. These cytokinetic factors improve treatment efficacy while minimizing organ damage for the prescription of personalized 90Y-GMRE.

To access the full text, please Sign in

If you have institutional access, please click here

  • Accepted: May 27, 2018
  • Online: Jun 11, 2018

Article Tools