Increased Efficacy and Toxicity with Combination Anti-Vegf Therapy Using Sorafenib and Bevacizumab.

3004 Background: The multikinase inhibitor, sorafenib (sor), targets raf-kinase and vascular endothelial growth factor receptor 2 (VEGFR2). Bevacizumab (bev) is a monoclonal antibody against VEGF. We theorized that combining the two agents would have synergistic therapeutic effects. Methods: Eligible pts had advanced solid tumors, good end organ function, and PS ≤ 1. A phase I dose-escalation study (cohort 1) used sor 200 mg po BID, and bev 5 mg/kg q2 wks IV (dose level, DL1) or 10 mg/kg (DL2). After DLT was reached, DL4 was added to cohort 1 which tested sor 200 mg po BID on days 1–5 q wk with bev IV 5 mg/kg q 2 wks in an attempt to minimize toxicity. Cohort 2 is an expansion phase at MTD (DL1: sor 200 mg po BID daily with bev IV 5 mg/kg), randomized to start with single drug for 4 wks. Cohort 2 has extensive translational endpoints, including imaging. Results: 34 pts have enrolled, 16 in cohort 1 and 18 in cohort 2. Tumor types include ovarian cancer (13), melanoma (4), and renal cell carcinoma (3). The median age was 58 yrs (30–76); pts received a median of 6.5 (1–15) prior regimens. In cohort 1, 3/16 pts had PR (19%; duration 4, 5+, 12+ mos), all with ovarian cancer, and 11/16 had stable disease (SD; 69%; 3+ - 13+ mos). 10/13 assessable cohort 2 pts had SD (77%; 3+ - 6.5+ mos). The most common DL1 AEs Gr ≥ 2 were hypertension (HTN; 6/6) and hand-foot syndrome (4/6). Other Gr 3 AEs in DL1 were leukopenia and infection (2; 2). DLT was reached at DL2 when 2/6 patients developed Gr 3 proteinuria (4, 5 gm/24 hr) and 3/6 developed uncontrolled Gr 3 HTN. Dose reduction on daily sor (DL1 and DL2) occurred for 11/12 pts at a median of 2 cycles, but is delayed on the intermittent schedule (DL4). In all pts, common Gr 1–2 AEs were elevated AST/ALT (18), rhinorrhea (12), fatigue (10), and anorexia (4). 4 pts developed Gr 2–3 diarrhea and 4 Gr 2–3 nausea/vomiting. Gastrointestinal fistulae occurred in two pts with PR, occurring within 4 wk of initiation of therapy, both in areas of tumor regression. Serial PET and DCE-MRI results and proteomic signal pathway analysis will be reported. Conclusions: The combination of sorafenib and bevacizumab appeared to increase both clinical effect and toxicity at below recognized single agent doses. Sor 200 mg po BID d1–5 qwk and bev of 5 mg/kg q2 wks is suggested for further study, pending further dose escalation. No significant financial relationships to disclose.