Safety of Radioembolization in the Setting of Angiographically Apparent Arterioportal Shunting

Purpose: To retrospectively analyze adverse events (AE) in patients with hepatocellular carcinoma (HCC) treated with yttrium-90 radioembolization in the setting of angiographically apparent arterioportal shunts (APSs). Materials and Methods: Thirty-two patients with HCC underwent radioembolization with APSs from January 2011 to September 2016, totaling 34 administrations using resin (6) and glass (28) microspheres. APSs were graded angiographically as segmental (9), ipsilobar (15), contralobar (7), or main portal (2), according to portal perfusion. Tumors were categorized as solitary (9), multifocal (7), or infiltrative (16). Both unilobar (25) and bilobar (7) disease was treated. Child Pugh Score was A (22), B (10), or C (2), with a median Model for End-Stage Liver Disease (MELD)/Na-MELD of 8/8.5. Median procedure dose was 132.6 Gy. AEs were graded using Combined Terminology Criteria for Adverse Events (CTCAE) version 4.0. Tumor response was assessed using the modified Response Evaluation Criteria in Solid Tumors (mRECIST). Results: CTCAE grade ≥3 AEs were observed in 22% of patients. Barcelona Clinic Liver Cancer (BCLC) C patients with nonsegmental shunts who received lobar administrations had a grade ≥3 AE rate of 38% compared with the remaining cohort, which was 12% (P =.076). No events were reported in patients with segmental shunts (P =.023). Imaging analysis revealed mRECIST complete response (17), partial response (13), stable disease (3), and progressive disease (1). Overall survival at 6 months and 12 months was 72% and 57%, respectively. Conclusions: Radioembolization in the setting of APS may have a higher AE profile than reported literature when BCLC-C patients with nonsegmental shunts receive lobar administrations. Segmental shunts are generally well tolerated.