3:18 PM Abstract No. 239 Feasibility of radiation segmentectomy with resin microspheres prescribed using medical internal radiation dosimetry model

Abstract

Radiation segmentectomy (RS) defined as radioembolization of ≤ 2 hepatic segments has only been described with glass microspheres. This study evaluates the feasibility and safety of transarterial yttrium-90 resin microsphere RS prescribed using Medical Internal Radiation Dosimetry (MIRD) model. IRB-approved, retrospective review of all patients undergoing RS with resin microspheres prescribed using MIRD model dosimetry (December 2015 to December 2018). Types of tumor include hepatocellular carcinoma (HCC) =16, cholangiocarcinoma (CC) = 12, carcinoma of unknown primary = 2, and others = 4. Mean tumor size was 5.7 ± 2.6 cm (range, 1.3 – 10.7). Coil embolization was performed in 6 (16%) procedures to prevent non-target embolization. Technical success was defined as ≥90% of prescribed activity delivered. Blood flow in the treated artery at end of the procedure and postprocedural grade 3 adverse events by Common Terminology Criteria for Adverse Events (CTCAE) v5.0 was assessed. Student’s t-test was used for statistical analysis. 34 patients (Age: 70 ± 9 years, 74% male) underwent 37 resin microsphere RS prescribed using MIRD model. Mean target absorbed dose for MIRD prescription was 145 ± 27 Gy (range, 120-200). Mean target area volume was 366 ± 292 cm3 (range, 75-1678). Target absorbed dose ≥150 Gy was prescribed in 20/37 (54%) procedures. Mean prescribed activity MIRD model was 1.2 ± 0.9 GBq (range, 0.3–4.5) compared to 0.4 ± 0.3 GBq (range, 0.1-1.4) if body surface area model had been used (P = <0.0001). Mean delivered activity was 1.2 ± 0.9 Gbq (range, 0.2 – 4.4). Mean percentage of prescribed activity delivered was 100 ± 8% (81-109%). Mean lung shunt fraction was 7% (3-34%). Technical success was achieved in 34/37 (92%). Blood flow at the end of treatment was brisk 17/37 (46%), slow 11/37 (30%) or sluggish 7/ (19%). Stasis only occurred in 2/37 (5%) procedures. Grade 3 adverse events included abdominal pain (3/37, 8%), radiation pneumonitis (2/37, 5%), and nausea and vomiting (1/37, 3%). The MIRD model can be safely used to perform dosimetry for radiation segmentectomy using resin microspheres with a high technical success rate. Target dose >190Gy can be achieved in selected cases