Prognostic Factors and Treatment Patterns in the Management of Giant Cell Glioblastoma.

BACKGROUND

There is a lack of literature guiding treatment of giant cell glioblastoma (gcGBM), a rare subtype of glioblastoma (GBM). We used a national hospital-based registry to explore treatment patterns and outcomes associated with gcGBM.

METHODS

Adult patients (age ≥18 years) diagnosed with gcGBM or GBM between 2004 and 2014 were identified from the National Cancer Database. χ analysis and Wilcoxon rank sum testing were used to compare characteristics between the gcGBM and GBM cohorts. Kaplan-Meier statistics, univariable and multivariable Cox regression, and propensity score matching were used to evaluate association between patient, tumor and treatment factors, and survival outcomes. Correlation analysis was used to evaluate historical trends in the treatment of gcGBM. Landmark analysis allowed for accounting of immortal time.

RESULTS

In total, 683 patients with gcGBM were identified. Patients with gcGBM had improved survival compared with patients with GBM (15.5 months from landmark vs. 11.7; P < 0.001). Increased age (P < 0.001) was associated with worse survival whereas being of female sex (P = 0.023) and having a median income >$63,000 (P = 0.004) predisposed patients to improved outcomes. Patients receiving trimodal therapy (biopsy and/or surgery, radiotherapy, and chemotherapy) experienced better outcomes compared with those receiving either biopsy and/or surgery only or biopsy and/or surgery and radiotherapy without systemic therapy (median survival, 17.55 months vs. 6.68 months; P < 0.001).

CONCLUSIONS

gcGBM has favorable prognosis compared with GBM and should be aggressively managed with trimodal therapy. Prospective studies of gcGBM are warranted to better characterize gcGBM treatment outcomes.