ANGI-05 PATHOGENESIS OF RESISTANCE (MIMICRY AND CO-OPTION) TO ANTI-ANGIOGENIC TREATMENT FOR GLIOBLASTOMA

PURPOSE: Vessel co-option and vascular mimicry are important resistant factors with ant-angiogenic treatment for glioblastoma, but those precise evaluation is not clear. We had three types of glioblastoma surgically removed specimens treated with / without bevacizumab (Bev). Using these samples, pathogenesis of co-option and mimicry was morphometrically clarified. MATERIALS / METHODS: Three types of glioblastoma specimens were analyzed; 1) Bev naive (N group, n 14), 2) Bev effective that was treated preoperative neoadjuvant Bev (E group, n 5), 3) Bev refractory that recurred with continuous Bev treatment for paired E group (R group, n 5). Vascular density was defined as a number of type IV collagen covered lumen. Vascular mimicry was measured as a ratio of CD34 negative / type IV collagen positive lumen. Vessel co-option was graded to 3 degrees (-), (+), (++) at tumor margin. RESULTS: (1)Vascular density was significantly lower with E group (p<0.01) and R gr up (p<0.02) compared to N group. (2)Mimicry was significantly higher with R group compared to N and E group (p<0.01). Between paired samples, refractory case was constantly higher than effective sample. (3) Co-option was increases with R group compared to N group. DISCUSSION/CONCLUSION: The effect of Bev for glioblastoma was investigated on three points (vascular density, vascular mimicry and vessel co-option) and two pathogeneses were clarified. In Bev refractory case, density was decreased, but mimicry and co-option were increased compared to Bev naive case. In Bev effective case, density was decreased, but mimicry and co-option were unchanged. Anti-angiogenic treatment for initial and Bev refractory glioblastoma should consider targeting co-option and mimicry in addition to Bev.