Background Adult brainstem gliomas (BSG) are uncommon and poorly understood with respect to prognostic factors. We retrospectively evaluated the clinical, radiographic, histologic, and treatment features from 101 adults with presumed or biopsy proven BSG to determine prognostic factors. Patients and Methods We reviewed the records of patients diagnosed from 1987–2005. We used Cox proportional hazard models to determine prognostic factors. Results These 50 male and 51 female patients ranged in age from 18 to 79 years at diagnosis (median 36 years) with follow-ups from 1 to 261 months (median 47 months). The overall survival for all patients at 5 and 10 years was 58% and 41%, respectively, with a median survival of 85 months (range 1–228). Out of 24 candidate prognosis factors, we selected seven covariates for proportional hazards model by Lasso procedure: age of diagnosis, ethnicity, need for corticosteroids, tumor grade, dysphagia, tumor location, and karnofsky performance status (KPS). Univariate analysis showed that these seven factors are significantly associated with survival. Multivariate analysis showed that four covariates significantly increased hazard for survival: ethnicity, tumor location, age of diagnosis, and tumor grade. Conclusions In this study, we identified four prognostic factors that were significantly associated with survival in adults with BSGs. Overall, these patients have a better prognosis than children with BSGs reported in the literature. These results call for larger prospective studies to fully assess the importance of these factors in the clinical setting and to help stratify patients in future clinical studies.
Brainstem glioma Natural history Chemotherapy Radiation Surgery Prognostic factors
This is a preview of subscription content, log in to check access.
We gratefully acknowledge the support of our patients and their families who have allowed us to perform this study.
Figure S1:(TIF 2188 kb) Lasso yields more accurate model than stepwise algorithm in our data. Figure shows cross-validated prediction errors according to different L1 norms: lower value corresponds to better performance. The blue horizontal line indicates the prediction error from traditional stepwise algorithm. The figure clearly demonstrates that Lasso performs better than stepwise procedure in our data and is the rationale for using this technique for this dataset.
Figure S2A:(TIF 5104 kb)Univariate analysis of survival. Kaplan-Meier survival curves and score (log-rank) tests of age and eight factors that were found to be significant are shown. 2A: age, race, KPS, tumor grade, steroid use, and dysphagia. 2B MRI-hydrocephalus, motor weakness, and tumor location. Two continuous variables, KPS and the age of diagnosis, were divided into patients with values higher than third quartile and those with values less than first quartile.
Figure S3A:(TIF 1102 kb) Imaging of BSGs. Examples of MRI of clinical-imaging subgroups. 4A. a, b- Exophytic/cystic: axial FLAIR (a) and T1-post gad (b) sequences showing an enhancing exophytic mass with mass effect on brainstem; c,d- axial FLAIR (c) and T1-post gad (d) sequences showing an cystic branstem glioma. 4B. Diffuse pontine: axial FLAIR (a) and T1-post gad (b) of a diffuse BSGs in a patient who had a prior PNET as a child. 4C. Tectal: axial T2 weighted (a) and T1-post gad (b) sequences of a patient with a tectal glioma with some enhancement (arrow) 4D. CMJxn: sagital T1-post gad sequence showing an cervicomedullary junction mass (arrow). 4E. Multifocal /gliomatosis: a, b- coronal (a) and axial T1-post gad (b) sequences showing a multifocal BSG (arrows); c, d- axial FLAIR weighted sequences showing diffuse involvement of brainstem, temporal lobes and thalami.
Laws ER Jr, Taylor WF, Clifton MB et al (1984) Neurosurgical management of low-grade astrocytoma of the cerebral hemispheres. J Neurosurg 61:665–673PubMedGoogle Scholar
Cohen ME, Duffner PK, Heffner RR et al (1986) Prognostic factors in brainstem gliomas. Neurology 36:602–605PubMedGoogle Scholar
Selvapandian S, Rajshekhar V, Chandy MJ (1999) Brainstem glioma: comparative study of clinico-radiological presentation, pathology and outcome in children and adults. Acta Neurochir (Wien) 141:721–726, discussion 726–727Google Scholar
Kansal S, Jindal A, Mahapatra AK (1999) Brain stem glioma–a study of 111 patients. Indian J Cancer 36:99–108PubMedGoogle Scholar
Farmer JP, Montes JL, Freeman CR et al (2001) Brainstem gliomas. A 10-year institutional review. Pediatr Neurosurg 34:206–214PubMedCrossRefGoogle Scholar
Barkovich AJ, Krischer J, Kun LE et al (1990) Brain stem gliomas: a classification system based on magnetic resonance imaging. Pediatric Neurosurg 16:73–83CrossRefGoogle Scholar
Pollack IF, Hoffman HJ, Humphreys RP et al (1993) The long-term outcome after surgical treatment of dorsally exophytic brain-stem gliomas. J Neurosurg 78:859–863PubMedGoogle Scholar
Lee BC, Kneeland JB, Walker RW et al (1985) MR imaging of brainstem tumors. AJNR: Am J Neuroradiol 6:159–163PubMedGoogle Scholar
Packer RJ, Zimmerman RA, Luerssen TG et al (1985) Brainstem gliomas of childhood: magnetic resonance imaging. Neurology 35:397–401PubMedGoogle Scholar
Albright AL, Packer RJ, Zimmerman R et al (1993) Magnetic resonance scans should replace biopsies for the diagnosis of diffuse brain stem gliomas: a report from the Children’s Cancer Group. Neurosurgery 33:1026–1029, discussion 1029–30Google Scholar
Steck J, Friedman WA (1995) Stereotactic biopsy of brainstem mass lesions. Surg Neurol 43:563–567, discussion 567–8Google Scholar
Takasato Y, Arai T, Ohta Y et al (1993) Gross total removal of adult brainstem glioma–two case reports. Neurol Med-Chir 33:625–629CrossRefGoogle Scholar