Caren van Roekel

169 Dose-response relationship in 166 Ho-radioembolization on the average change in TLG of all hepatic tumors. Patient-level analysis was performed both including and excluding tumors that formed after baseline (which were labeled as progressive disease). All other analyses ignored the formation of new lesions at follow-up, as they were not targeted by the treatment. Linear mixed-effect models were used to assess the relation between tumor absorbed dose and response and to account for correlation of tumors within patients. Dose was used as dependent variable and log- transformed to fulfill model assumptions. Nested models were compared using Akaike’s Information Criterion. The dose-effect relationship was best explained using a random intercept per patient without random slopes. A geometric mean of the tumor absorbed dose per response category was estimated. On a patient level, response categories CR and PR were merged in the analysis due to otherwise too limited numbers per category. To test the hypothesis of an ordered relationship across response categories, a trend test was performed with response as a continuous variable in the model. Overall survival was defined as the interval between treatment and death from any cause, with censoring of patients who were still alive at their last known follow-up date. The survival curve was estimated by the Kaplan-Meier method. A log-rank test was used to compare median overall survival between patients with and without a metabolic liver response. Baseline characteristics of these groups, consisting of primary tumor type, gender, age, previous treatments, WHO performance score, presence of extra hepatic disease, number of tumors and tumor load, were scrutinized for differences that could have biased the survival analysis. Analyses were performed using R statistical software, version 3.4.0. A two-sided p-value <0.05 was considered statistically significant. RESULTS Thirty-six patients with a total of 98 tumors were included in this study. Baseline characteristics are listed in Table 1. Eleven patients of the HEPAR I study were excluded because of absence of post-treatment 166 Ho-SPECT/CT (n=4) or due to unavailability of the corresponding low-dose CT with the 166 Ho-SPECT (n=7). Five patients of the HEPAR II study were excluded because of absence of post- treatment 166 Ho-SPECT/CT (n=2), absence of baseline FDG PET/CT (n=1), absence of follow-up FDG PET/CT (n=1) and no FDG-uptake in the tumor (n=1). 6