Search for a command to run...
Glioblastoma has a poor prognosis with median survival of 12–14 months. Long-term survivors (LTS) defined as those alive at least 2 years from diagnosis, comprise 13% of this population. This study aims to provide a clinical profile of LTS at two institutions in Melbourne, Australia. Histological diagnosis of glioblastoma from 1st January 2006 to 31st December 2012 were identified from pathology and oncology databases. Demographic, treatment and survival characteristics were recorded, with follow-up to 31st December 2015. Independent sample t-test, chi-squared and fisher exact were used to identify differences between LTS and those surviving less than 2 years. Survival estimated by Kaplan-Meier. 776 patients were identified with 154 (20%) surviving >2 years. Compared with patients surviving < 2 years, LTS were more likely to be younger (median age 56 vs 65yrs, p<.001), have ECOG 0–2 (95% vs 64%, p<.001), unilateral tumours (88% vs 78%, p.004), gross tumour resection (90% vs 61%, p<.001), and receive chemoradiotherapy (69% vs 40%, p<.001). The most common presenting symptoms amongst LTS were headache (42%), seizure (28%) and speech disturbance (16%), with 79% prescribed anticonvulsants. Of LTS, 111 patients (72%) progressed at a median of 20.1 months from diagnosis, with 46% of these undergoing a second craniotomy. Temozolomide was the most common non-surgical second line treatment (41%), followed by radiotherapy (12%). One third of LTS received three or more lines of treatment, and 10% underwent three or more craniotomies. The median overall survival for LTS was 38.3 months, compared with 11.0 months (entire cohort) (p<.001). LTS of glioblastoma (20%) are more likely to be younger, have unilateral tumours, good performance status and undergo multimodality treatment, and multiple lines of therapy. These data may assist in predicting LTS at diagnosis and understanding their clinical journey to facilitate planning of treatment and supportive care.