458 Background: Transarterial chemoembolization (TACE) and selective internal radiation therapy (SIRT) are commonly used for treatment of liver tumors. The use of TACE, which is macroembolic, prior to SIRT may cause hemodynamic changes in tumor vasculature that impair yttrium-90 (90Y) microsphere delivery to the targeted lesions. This work aims to quantify dosimetric tumor coverage using positron emission tomography (PET) dosimetry after SIRT alone compared to TACE followed by SIRT. Methods: A total of 40 consecutive hepatocellular carcinoma (HCC) SIRT patients who had a post-SIRT PET scan were evaluated. Yttrium-90 PET scan was performed within 2 hours after completion of SIRT. 3D dose distribution of each patient was then reconstructed from the PET images. To quantify the quality of SIRT microsphere delivery, we evaluated the selectivity index (SI) defined as the ratio of the average dose inside the treated lesion(s) and the average dose of the normal liver. The SI values of patients were compared based on whether TACE was previously used. Results: Thirty patients had complete data for analysis. The average dose for a total of 125 targeted lesions was 106 Gy (range 0 Gy – 495 Gy). The average SI of the entire patient group was 3.0, indicating that the targeted lesion dose was three times higher than normal liver dose. The average SI was 1.7 for the 13 patients who had prior TACE and 3.8 for the 17 patients who did not have prior TACE (p = 0.01). Nearly 80% of the patients with prior TACE demonstrated poor 90Ymicrosphere delivery (SI < 2) while none demonstrated excellent microsphere delivery (SI > 4). On the other hand, the incidence of SI < 2 and SI > 4 among patients with no prior TACE was 24% and 40%, respectively. Conclusions: 3D dose evaluation using post-SIRT PET suggests that 90Y microsphere delivery to liver tumors is impaired among patients who received prior TACE compared to those who receive SIRT alone. These data warrant evaluation of clinical outcomes in SIRT patients based on prior use of TACE and may provide a rationale for the use of SIRT prior to TACE.