No abstract

Background: Glioblastoma (GB) is the most common malignant primary central nervous system tumor in adults. Standard-of-care therapy includes surgical resection, radiotherapy and temozolomide, but nearly all patients experience disease progression. The purpose of this study was to describe 2 cohorts of patients with recurrent GB submitted to second-line treatment with procarbazine/lomustine/vincristine (PCV) or bevacizumab/irinotecan (BI). Material and Methods: Retrospective analysis of GB patients treated in our center with PCV or BI, after progression with temozolomide, between 2004 and 2012. Results: Among 60 patients, 41 were treated with BI and 19 with PCV. According to the Macdonald criteria, the overall response rate in the BI group was 66% (n = 27) while it was 11% (n = 2) in the PCV group. The median progression-free survival was 5 and 3 months in the BI and PCV group, respectively. The median overall survival (OS) since second-line chemotherapy was 9 months in the BI group and 5 months in the PCV group. The latter group had a worse toxicity profile (grade 3-4: 52.6% vs. 22.0%; grade 1-2: 89.5% vs. 68.3%). Conclusions: The BI cohort had higher response rates, almost twice the OS and a lower degree of toxicity in contrast to the PCV group. The small number of patients and historical cohorts limits these comparisons.

Dismal prognosis of glioblastoma (GBM) prompts for the identification of response predictors and therapeutic resistance mechanisms of current therapies. The authors investigated the impact of c-Met, HGF, VEGFR2 expression and microvessel density (MVD) in GBM patients submitted to second-line chemotherapy with bevacizumab. Immunohistochemical expression of c-Met, HGF, VEGFR2, and MVD was assessed in tumor specimens of GBM patients treated with bevacizumab, after progression under temozolomide. Survival analysis was evaluated according to the expression of the aforementioned biomarkers. c-Met overexpression was associated with a time-to-progression (TTP) after bevacizumab of 3 months (95% CI, 1.5–4.5) compared with a TTP of 7 months (95% CI, 4.6–9.4) in patients with low or no expression of c-Met (p = 0.05). VEGFR2 expression was associated with a TTP after bevacizumab of 3 months (95% CI, 1.8–4.2) compared with a TTP of 7 months (95% CI, 5.7–8.3) in patients with no tumoral expression of VEGFR2 (p = 0.009). Concomitant c-Met/VEGFR2 overexpression was associated with worse overall survival (13 months) compared with concomitant c-Met/VEGFR2 negative expression (19 months; p = 0.025). Our data support the hypothesis that c-Met and VEGFR2 overexpression have a role in the development of glioblastoma early resistance and might predict poorer responses to anti-angiogenic therapies.