Dose-response relationship after yttrium-90-radioembolization with glass microspheres in patients with neuroendocrine tumor liver metastases

In radioembolization, response is achieved through the irradiation and damaging of tumor DNA. For hepatic metastases of neuroendocrine tumors, a dose-response relationship has not been established yet. This study assesses whether increasing tumor-absorbed doses lead to increased response rates. We included all patients who underwent yttrium-90 (${}^{90}$Y) glass microspheres radioembolization in our center if both pre- and post-treatment contrast-enhanced CT and post-injection PET/CT were available. Up to five hepatic tumors and the healthy hepatic tissue were delineated, and absorbed dose was quantified using post-injection PET/CT. Response was measured according to RECIST 1.1 on patient and tumor level. Linear mixed models were used to study the relationship between absorbed dose and response on tumor level. Logistic regression analysis was used on patient level to study dose-response and hepatic dose-toxicity relationships. A total of 128 tumors in 26 patients (31 procedures) were included in the response analysis. While correcting for confounding by tumor volume, a significant effect of response on dose was found (p = 0.0465). Geometric mean of absorbed dose for responding tumors was 170 Gy, for stable disease 101 Gy, and for progressive disease 67 Gy. No significant dose-toxicity relationship could be identified. In patients with neuroendocrine tumor liver metastases, treated with ${}^{90}$Y-radioembolization, a clear dose-response relationship was found. We propose to perform ${}^{90}$Y-radioembolization with an absolute minimum planned tumor-absorbed dose of 150 Gy.