Treatment options for recurrent glioblastoma are rare, with their response uncertain. This study aimed to determine the response of chemotherapy including bevacizumab in combination with vincristine and carboplatin for glioblastoma at first recurrence in a single-institution cohort.Clinical data of patients who received chemotherapy including bevacizumab, vincristine, and low-dose carboplatin for recurrent glioblastoma between 2008 and 2014 were analyzed. Differences between those who received combination chemotherapy (chemotherapy-positive) and those who did not (chemotherapy-negative) were estimated by Fisher exact test or Wilcoxon rank-sum test, as appropriate. Survival curves were estimated using the Kaplan-Meier method, and differences between survival curves were estimated by the log-rank test. Univariate analysis of treatment response for all recurrent glioblastoma patients and secondary recurrence patients under different conditions were evaluated using Wilcoxon rank-sum test or the Kruskal-Wallis test.Although mortality rates were similar between the chemotherapy-negative and chemotherapy-positive groups (26.7% vs 28.6%), median overall survival was significantly longer in the chemotherapy-positive group than the chemotherapy-negative group (P = .006). There were no chemotherapy-related serious complications such as gastrointestinal perforation, serious bleeding, or new-onset seizure during chemotherapy, whereas others side effects including proteinuria and hypertension were more common albeit well controlled by medication.This study revealed combination regimen of bevacizumab, vincristine, and low-dose carboplatin as a potentially effective therapeutic approach in recurrent glioblastoma. More in-depth understanding of the mechanism underlying this combination treatment and potential contribution of alternative genetic therapeutic in recurrent glioblastoma is necessary.