High Serum Vascular Endothelial Growth Factor (VEGF) Level Is An Adverse Prognostic Factor In High Risk Diffuse Large B-Cell Lymphoma (DLBCL) Patients Treated with Dose-Dense Chemoimmunotherapy and Systemic CNS Prophylaxis. Results From a Nordic Phase II Study

VEGF is one of the most specific and critical regulators of angiogenesis. It regulates normal endothelial proliferation, permeability, and survival, but is also an angiogenic mediator in tumours, and has been implicated in the pathogenesis and progression of cancer. Elevated serum VEGF (S-VEGF) levels have been associated with unfavourable outcome in different hematologic malignancies including DLBCL. However, to our knowledge there is no data about the prognostic relevance of S-VEGF levels in the rituximab era. The aim of this study was to determine if pretreatment S-VEGF levels and VEGF gene expression in the DLBCL tissue correlate with survival of patients treated with chemoimmunotherapy.

Study population consisted of DLBCL and follicular lymphoma (FL) grade 3 patients, who were less than 65 years old and had high-risk (aaIPI 2–3) disease. All patients were treated in the Nordic phase II protocol with six courses of R-CHOEP14 followed by systemic CNS prophylaxis with one course of high-dose methotrexate and one course of high-dose cytarabine. Pretreatment serum was available for VEGF Enzyme-Linked Immuno-Sorbent Assay (ELISA) from 112 trial patients. VEGF gene expression could be analyzed from the exon-based microarrays of 41 pretreatment biopsies. In the present report of the 112 patients with a median follow-up of 37 months, (range 17–64 months), 3-year relapse free survival (RFS) was 76% and overall survival (OS) 82%.

S-VEGF levels at diagnosis varied from undetectable to 2000 pg/ml, median concentration being 530 pg/ml. 10 patients had values higher than the highest standard, 2000 pg/ml. 28 patients (25%) had S-VEGF level higher than 925 pg/ml (highest quartile). According to Kaplan-Meier analyses, RFS at 3 years was poorer among the patients with high (the highest quartile) than low S-VEGF-levels (59% vs 82%, p=0.005) The risk of relapse was 2.9 fold among the patients with high S-VEGF level (95% CI 1.34–6.26, p=0.007). In contrast to serum data, VEGF gene expression in the lymphoma tissue of the same patients did not associate with outcome. Furthermore, no correlation between S-VEGF levels and VEGF gene expression was found. To validate the gene expression results we used microarray data from an independent set of 233 DLBCL patients treated with a combination of rituximab and CHOP-like chemotherapy (Lenz et al., 2008). Again, no correlation with VEGF gene expression and outcome could be observed.

High pretreatment S-VEGF level is an adverse prognostic factor for young high-risk DLBCL/FL grade 3 patients treated with chemoimmunotherapy, whereas VEGF gene expression in the lymphoma tissue correlates neither with the S-VEGF levels nor the prognosis. The results suggest that S-VEGF does not necessarily arise from the lymphoma tissue but may instead reflect host response to lymphoma.

No relevant conflicts of interest to declare.