Rituximab, bevacizumab and CHOP (RA-CHOP) in untreated diffuse large B-cell lymphoma: safety, biomarker and pharmacokinetic analysis.

Bevacizumab is a humanized monoclonal antibody directed against vascular endothelial growth factor (VEGF-A). Non-Hodgkin's lymphoma patients with high serum VEGF levels have an inferior survival compared to patients with low VEGF levels. Bevacizumab was administered through a central line at 15 mg kg(-1) IV on day 1 followed by rituximab (R) and CHOP on day 2 for cycle 1 and day 1 for cycles 2 - 8. Serum levels of bevacizumab and R were measured at specified time points to assess pharmacokinetics (PK). Plasma and urine samples were also analysed for VEGF. Tumor samples were stained for VEGF, CD31 and factor VIII by immunohistochemistry. Thirteen patients with newly-diagnosed DLBCL received a total of 88 cycles (range 2 - 8, median 7). Best response included five CR, six PR, one SD and one PD with an overall response rate of 85% and complete response rate of 38%. The 12-month PFS is 77% and a median follow-up of 16.9 months for the surviving patients. All tumor samples stained strongly positive for VEGF and there was a marginal association between baseline plasma VEGF and response (p = 0.04). Patients with higher plasma VEGF levels were generally younger and had bulky disease. Micro-vessel density did not correlate with presenting disease characteristics, VEGF expression or response. The PK of bevacizumab and rituximab were not influenced by combined treatment. In this patient population, treatment with RA-CHOP did not result in any episodes of grade 3 or 4 proteinuria, heart failure or hemorrhage. The RA-CHOP combination was generally well tolerated and safe.