Phase II Trial of Standard Dose Cyclophosphamide, Doxorubicin, Vincristine, Prednisone (CHOP) and Rituximab (R-CHOP) Plus Bevacizumab for Advanced Stage Diffuse Large B-Cell (DLBCL) NHL: Southwest Oncology Group Study S0515

VEGF and VEGF receptors (VEGFR) are frequently over-expressed in DLBCL. Elevated levels of circulating VEGF have also been associated with a worse prognosis and resistance to chemotherapy. This trial was initiated to determine if the addition of anti-VEGF targeted therapy to standard R-CHOP would improve PFS without adding significant toxicity in previously untreated patients with stage III, IV or bulky stage II DLBCL.

64 eligible patients were treated with standard dose R-CHOP plus bevacizumab administered at 15mg/kg on day1. Patients received therapy every 21 days for a maximum of 8 cycles. Both biologics were administered prior to chemotherapy. Pretreatment tumor biopsies from 40 patients were submitted for construction of tissue microarrays for VEGF/VEGFR immunohistochemical analysis. Two 1 mm tissue cores were assessed for each DLBCL specimen and expression of VEGF and VEGFR scored on a scale of 0 (no expression) to +3 (strong expression) by 2 pathologists. Both lymphoma and endothelial cell (blood vessel) staining was assessed with the higher score being recorded.

73 patients were enrolled with 9 patients found ineligible, the majority (n = 7) secondary to incorrect histology. Patient characteristics for the 64 eligible patients included: median age 68 years with 77% of the patients over 60 years (age range: 22–85), 44 males, IPI scores: 3 low, 22 low-intermediate, 29 high-intermediate, and 8 high. Median follow-up is 2.0 years. Based on prior studies, and adjusted for the observed IPI factor frequencies seen in this study, the projected 1 year historical PFS rate in this population would be 71%, while the observed 1 year PFS estimate was 77% (95% CI: 66 – 87%). The PFS estimate at 2-years was 69% (95% CI: 57 – 80%). The 1 and 2 year overall survival estimates were 86% and 79%, respectively. Among all patients (including ineligibles), a total of 30 patients experienced 40 serious adverse events (SAE) while on study. The majority of SAE were hematologic, but also included 2 patients with sudden death, 5 patients with GI perforations (4 occurring after cycle 1), and 5 patients with grade 3 or 4 thrombotic events. Additional adverse events of interest included 7 patients who developed hypertension (4 patients with grade 3), and 13 patients who developed grade 2 or 3 left ventricular dysfunction. VEGF expression was observed in 98% (IHC scoring: +1 in 18%, +2 in 38%, and +3 in 43%) of lymphoma cells and VEGFR was expressed in 95% (IHC scoring:+1 in 28%, +2 in 40%, and +3 in 28%) of blood vessels by immunohistochemistry analysis.

Despite high protein expression of VEGF and VEGFR in newly diagnosed DLBCL specimens, the addition of bevacizumab to standard R-CHOP chemotherapy did not significantly improve PFS from expected results with R-CHOP alone. Although the observed 1-year PFS estimate trended higher than the historical estimate, a target PFS of 81% was pre-specified as warranting further investigation. Significant toxicities were associated with the addition of bevacizumab, including an increased incidence of cardiac dysfunction and GI perforation. Our results do not support additional studies of this regimen in this patient population.

Stopeck:Genentech: Consultancy, Honoraria, Speakers Bureau. Off Label Use: Bevacizumab therapy in NHL. Fisher:Roche: Consultancy.