- September 11, 2018
- Category: Scientific Publications
Tumor Targeting and Three-Dimensional Voxel-Based Dosimetry to Predict Tumor Response, Toxicity, and Survival after Yttrium-90 Resin Microsphere Radioembolization in Hepatocellular Carcinoma
Carole Allimant1, MD, Marilyne Kafrouni2, MSc, Julien Delicque1, MD, Diana Ilonca2, MD, Christophe Cassinotto1, MD, PhD, Eric Assenat3, MD, PhD, Jose Ursic-Bedoya4, MD, MSc, Georges-Philippe Pageaux4, MD, PhD, Denis Mariano-Goulart5, MD, PhD, Serge Aho6, MD, PhD, and Boris Guiu7, MD, PhD
1 Department of Radiology, Saint-Eloi University Hospital, 80 avenue Augustin Fliche, Montpellier 34295, France.
2 Department of Nuclear Medicine, Guy de Chauliac University Hospital, Montpellier, France.
3 Department of Digestive Oncology, Saint-Eloi University Hospital, 80 avenue Augustin Fliche, Montpellier 34295, France
4 Department of Hepatology, Saint-Eloi University Hospital, 80 avenue Augustin Fliche, Montpellier 34295, France
5 Department of Nuclear Medicine, Guy de Chauliac University Hospital, Montpellier, France; PhyMedExp, INSERM U1046, CNRS UMR9214, University of Montpellier, Montpellier, France.
6 Department of Statistics and Epidemiology, University Hospital of Dijon, Dijon, France.
7 Department of Radiology, Saint-Eloi University Hospital, 80 avenue Augustin Fliche, Montpellier 34295, France; Montpellier Cancer Research Institute, INSERM U1194, Montpellier, France. Electronic address: Bfirstname.lastname@example.org.
Purpose: To identify predictive factors of tumor response, progression-free survival (PFS), overall survival (OS), and toxicity using three-dimensional (3D) voxel-based dosimetry in patients with intermediate and advanced stage hepatocellular carcinoma (HCC) treated by yttrium-90 (90Y) resin microspheres radioembolization (RE).
Materials and Methods: From February 2012 to December 2015, 45 90Y resin microspheres RE procedures were performed for HCC (Barcelona Clinic Liver Cancer stage B/C; n ¼ 15/30). Area under the dose-volume histograms (AUDVHs) were calculated from 3D voxel-based dosimetry to measure 90Y dose deposition. Factors associated with tumor control (ie, complete/partial response or stable disease on Modified Response Evaluation Criteria in Solid Tumors) at 6 months were investigated. PFS and OS analyses were performed (Kaplan-Meier). Toxicity was assessed by occurrence of radioembolization-induced liver disease (REILD).
Results: Tumor control rate was 40.5% (17/42). Complete tumor targeting (odds ratio ¼ 36.97; 95% confidence interval, 1.83–747; P < .001) and AUDVHtumor (odds ratio ¼ 1.027; 95% confidence interval, 1.002–1.071; P ¼ .033) independently predicted tumor control. AUDVHtumor 61 Gy predicted tumor control with 76.5% sensitivity and 75% specificity. PFS and OS in patients with incomplete tumor targeting were significantly shorter than in patients with complete tumor targeting (median PFS, 2.7 months [range, 0.8–4.6 months] vs 7.9 months [range, 2.1–39.5 months], P < .001; median OS, 4.5 months [range, 1.4–23 months] vs 19.2 months [range, 2.1–46.9 months], P <.001). Patients with incomplete tumor targeting and AUDVHtumor < 61 Gy, incomplete tumor targeting and AUDVHtumor > 61 Gy, complete tumor targeting and AUDVHtumor < 61 Gy, and AUDVHtumor > 61 Gy had median PFS of 2.7, 1.8, 6.3, and 12.1 months (P < .001). REILD (n ¼ 4; 9.5%) was associated with higher dose delivered to normal liver (P ¼ .04).
Conclusion: Complete tumor targeting and 90Y dose to tumor are independent factors associatedwith tumor control and clinical outcomes.