Vascular endothelial growth factor (VEGF) plays an important role in angiogenesis and progression of cancer, and blood level of VEGF has been known to predict of outcome in several types of cancer. However, the impact of blood VEGF levels on prognosis of diffuse large B-cell lymphoma (DLBCL) has not been fully elucidated. Here we investigated the prognostic implication of circulating VEGF levels in patients with DLBCL.

Patients and Methods

The study involved 127 DLBCL patients treated by rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone (R-CHOP) at National Cancer Center, Korea. Both serum and plasma of the patients were obtained at diagnosis. VEGF were measured using ELISA kit (R&D systems) according to manufacturer’s guidance. We investigated the correlation between clinical parameters and blood VEGF levels. Survival rates and hazard ratios (HRs) in terms of risk for overall survival were determined using Kaplan-Meier method and Cox proportional hazard regression analysis.


The median patient age was 56 years, and 75 (59%) patients were men. Clinical characteristics and international prognostic index (IPI), including age, performance, lactate dehydrogenase (LDH), Ann Arbor stage and extranodal involvement were evaluated. The mean (±SD) value of serum and plasma VEGFs were 713 (±599.8) and 107 (±164.4) pg/mL, and they were correlated as r=0.61 (p<0.001). Both serum and plasma VEGF showed the significant (p<0.01) correlation with serum LDH level, Ann Arbor stage and multiple extranodal involvement, but not with age, gender nor performance. IPI showed strong prediction for prognosis in our data set (HR 3.80, 95%CI 1.64-8.81, p=0.02), and the VEGF levels of high/high-intermediate IPI group were significantly higher than those of low/low-intermediate group [serum VEGF 1088(±838.0) and 557(±377.2), p<0.001 and plasma VEGF 79(±131.6) and 176(±212.2), p=0.002]. With the median follow-up of 44 months, high serum VEGF levels (higher than the median) were significantly associated with short survival (HR 2.74, 95%CI 1.13-6.60, p=0.025), though the plasma VEGF levels did not show the association similar to the serum samples (HR 1.40, 95%CI 0.63-3.12, p=0.414). The patients with higher serum VEGF than the median value showed significantly lower survival rate compared to the low group (3-year survival rates, 68.6% vs. 85.6%, p=0.019).


These findings suggest that high serum VEGF levels can predict poor clinical outcome in patients with DLBCL. This study also represents the different type of specimen for VEGF measurement affected the results. We might have to select adequate specimen type for VEGF measurement, although further validation in large cohort and mechanism studies for this data are warranted.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.