Olfactory function as an independent prognostic factor in glioblastoma

Objective To determine the role of olfactory function in patients with glioblastoma multiforme (GBM) as a prognostic clinical measure. Methods In a prospective case-control study, olfactory testing was performed in 73 patients with primary GBM at baseline during first-line treatment and at later follow-ups. An age-matched control cohort consisted of 49 patients with neurologic diseases, excluding those known to affect olfactory function per se. Depending on the olfactory testing score, patients were allotted to a hyposmia group (HG) or normosmia group (NG). MRI analysis was performed to assess whether tumor location affects olfactory pathways. Results Patients with GBM had olfactory dysfunction significantly more often compared to the control cohort (p = 0.003). Tumor location could not explain this finding since no relevant difference in MRI-based olfactory pathway involvement was found between HG and NG (p = 0.131). Patients with olfactory dysfunction had significantly worse overall survival (OS) and progression-free survival (PFS) compared to those without dysfunction (median OS 20.9 vs 40.6 months, p = 0.035; median PFS, 9 vs 19 months, p = 0.022). Multivariate analysis in patients without MRI-based involvement of olfactory pathways confirmed olfaction is an independent prognostic factor for OS (hazard ratio [HR] 0.43; p = 0.042) and PFS (HR 0.51; p = 0.049). Conclusion This pilot study provides the first indication that olfactory dysfunction is frequently observed in GBM and may be associated with worse survival outcome in GBM. However, validation of these results in an independent cohort is needed.