Predictive value of tumor recurrence using urinary vascular endothelial factor levels in patients receiving radiation therapy for Glioblastoma Multiforme (GBM).

BACKGROUND: Glioblastoma Multiforme (GBM) is the most common primary malignant tumor of the central nervous system. Standard of care includes maximal resection followed by chemoradiotherapy. Tumors need adequate perfusion and neovascularization to maintain oxygenation and for removal of wastes. Vascular endothelial growth factor (VEGF) is a well characterized pro-angiogenic factor. We hypothesized that the increases in urinary VEGF levels would occur early in the course of tumor recurrence or progression. We examine the feasibility of collecting and analyzing urinary VEGF levels in a prospective, multi-institutional trial (Radiation Therapy Oncology Group, RTOG, 0611) as well as the role of VEGF as a marker of tumor recurrence. METHOD: We evaluated VEGF levels in urine specimens collected post-operatively, at the conclusion of radiation therapy (RT) and one month following RT. Urinary VEGF levels were correlated with tumor progression at one year. VEGF levels were measured by enzyme-linked immunosorbant assay in urine specimens and normalized to urinary creatinine levels. Sample size was determined based on a 50% 1-year recurrence rate. With a sensitivity and specificity of 80%, the expected 95% confidence interval was (0.69, 0.91) with 100 patients. A failure was defined as documented disease progression, recurrence or death before one year. RESULTS: 202 patients were enrolled between February-2006 and October-2007. Four patients were ineligible as they did not receive RT. Of the remaining 198 patients, 128 had all three samples collected. In this group, 35 patients (27.3%) did not progress, 89 (69.5%) had progression and 4 (3.1%) died without evidence of progression. Median VEGF levels at baseline were 52.9 pg/mg Cr (range 0.2- 15,034.4); on the last day of RT, 56.6 (range 0-2,377.1); and at one month follow-up, 70.0 (range 0.1-1813.2). In patients without progression at 1-year, both baseline VEGF level and end of RT VEGF level were lower than those of patients who progressed: 40.3 (range 0.2-350.8) vs. 59.7 (range 1.3-15,034.4) and 41.8 (range 0-356.8) vs. 69.7 (range 0-2,377.1), respectively. This did not reach statistical significance. Comparison of the change in VEGF levels between the end of RT and one month following RT, demonstrated no significant difference in the proportions of progressors or non-progressors at 1-year for either the VEGF increased or VEGF decreased groups. CONCLUSION: Urine can be collected and analyzed in a prospective, multi-institutional trial. In this study of patients with GBM a change in urinary VEGF levels between the last day of RT and the one month following RT did not predict for tumor progression by one year.