Neutrophil–lymphocyte ratio (NLR) is a marker of systemic inflammatory response and its elevation has recently been shown to be a poor prognostic factor in many malignancies including colon, prostate and bladder cancer. The primary aim of this study was to assess the prognostic impact of NLR in a clinically annotated cohort of patients with glioblastoma multiforme (GBM). We hypothesised that elevated NLR would be associated with worse prognosis. Between 2004 and 2009, 137 patients had surgery for GBM and were assessed for consideration of adjuvant therapy at our institution. Of these, 84 patients with an evaluable pre-corticosteroid full blood count result were identified and included in the final analysis. Median overall survival was 9.3 months (range 0.7–82.1). On univariate analysis, age >65 years, gender, ECOG performance status ≥2, frontal tumour, extent of surgical resection, completion of adjuvant chemoradiation protocol and NLR > 4 were significantly correlated with overall survival. Patients with NLR > 4, had a worse median overall survival at 7.5 months versus 11.2 months in patients with NLR ≤ 4 (hazard ratio 1.6, 95 % CI 1.00–2.52, p = 0.048). On multivariate analysis NLR > 4 remained an independent prognostic indicator for poor outcome. These data are an important reminder of the potential relevance of host immunity in GBM. In our cohort, NLR > 4 conferred a worse prognosis independent of other well established prognostic factors. If validated in other cohorts NLR may prove to be a useful addition in predicting prognosis in GBM patients. The demonstration that host immunity plays a role in GBM biology suggests that investigation of emerging therapies which modulate host immune response are warranted in this disease.