Dose-toxicity analysis for NET patients treated with 166Horadioembolization following PRRT

Abstract

Aim/Introduction: Patients with neuroendocrine tumor liver metastases can beneft from additional holmium-166 (166Ho)-radioembolization after peptide receptor radionuclide therapy (PRRT). However, following radioembolization, hepatotoxicity is feared and has a large variation in its severity. The aim of this study is to assess the hepatotoxicity and the corresponding dose to the healthy liver following 166Ho-radioembolization. Materials and Methods: Twenty-eight patients included in the prospective HEPAR PLuS study were considered. Hepatotoxicity was assessed at 3 months follow-up (3M FU) according to a fve point scale, and a score ≥3 was considered signifcant [1]. Dose to the healthy liver was computed based on manual and automatic segmentation. Manual segmentation was performed by an experienced physician on the contrast enhanced CT acquired up to weeks before radioembolization and manually registered to the corresponding SPECT. Automatic healthy liver segmentation was obtained by thresholding the technetium-99m (99mTc) reconstruction simultaneously acquired after 166Ho-radioembolization, according to the 166Ho-99mTc dual-isotope protocol. T-test was used to assess the statistically signifcant diference between dose using the manual and the automatic segmentations. Results: Among the considered patients (20 males, 8 females) median (IQR) age was 66 (56-71) years. Tumor origin was pancreas, small intestine, colorectal, lung and unknown for 8, 7, 4, 3 and 6 patients, respectively. WHO grade was 1 for 12 patients and 2 for the remaining ones, while 16 patients presented an ECOG status of 0 and 12 an ECOG status of 1. Median (IQR) tumor burden was 6.4% (3.1-22.8%). A total of 29 treatments (15 whole liver and 14 partial) were considered. Median (IQR) administered activity was 6.5GBq (4.8-8.2 GBq). At 3M FU, hepatotoxicity was graded 0 in 2 subjects, 1 in 21, and 2 in 5 subjects. One patient presented a grade 5 hepatotoxicity Median (IQR) dose to the healthy liver (manually delineated) was 23Gy (2), 17Gy (7) and 28Gy (5) among patients with hepatotoxicity graded 0, 1 and 2 respectively. The patient with the highest toxicity (graded 5) received 30Gy to the healthy liver. No statistically signifcant diference was found between dose computed on the manual and automatic segmentation of the healthy liver (P>0.1). Conclusion: Signifcant hepatotoxicity was encountered in only one patient (out of 28) after sequential treatment with PRRT and 166Ho-radioembolization. For all the procedures assessed, dose to the healthy liver was below 30Gy. Dose to the healthy liver can be estimated using manual or automatic healthy liver segmentations. References: [1] Braat MNGJA et al., Eur.J.Gastroenterol.Hepatol. (2017)