Frontiers in Neurology (Nov 2021)

Epidemiologic Features, Survival, and Prognostic Factors Among Patients With Different Histologic Variants of Glioblastoma: Analysis of a Nationwide Database

  • Li-Tsun Shieh,
  • Chung-Han Ho,
  • Chung-Han Ho,
  • How-Ran Guo,
  • How-Ran Guo,
  • Chien-Cheng Huang,
  • Chien-Cheng Huang,
  • Yi-Chia Ho,
  • Sheng-Yow Ho,
  • Sheng-Yow Ho,
  • Sheng-Yow Ho

DOI
https://doi.org/10.3389/fneur.2021.659921
Journal volume & issue
Vol. 12

Abstract

Read online

Background: Glioblastoma (GBM) is the most common primary intracranial malignancy. Previous studies found incidence of GBM varies substantially by age, sex, race and ethnicity, and survival also varies by country, ethnicity, and treatment. Gliosarcoma (GSM) and giant cell glioblastoma (GC-GBM) are different histologic variants of GBM with distinct clinico-pathologic entities. We conducted a study to compare epidemiology, survival, and prognostic factors among the three.Methods: We identified GBM patients diagnosed between 2000 and 2016 using the Taiwan Cancer Registry and followed them using the death registry. Survival was compared among conventional GBM and two histologic variants. The potential confounding factors evaluated in this study included registered year, age, sex, and treatment modality (resection, radiotherapy, and chemotherapy).Results: We enrolled 3,895 patients, including 3,732 (95.8%) with conventional GBM, 102 (2.6%) with GSM, and 61 (1.6%) with GC-GBM. GC-GBM patients had younger mean age at diagnosis (49.5 years) than conventional GBM patients (58.7 years) and GSM patients (61.3 years) (p < 0.01). The three groups had similar sex distributions (p = 0.29). GC-GBM had a longer median survival [18.5, 95% confidence interval (CI): 15.8–25.3 months] than conventional GBM (12.5, 95%CI: 12.0–13.0 months) and GSM (12.8, 95%CI: 9.2–16.2 months), and the differences in overall survival did not attain statistical significance (p = 0.08, log-rank test). In univariate analysis, GC-GBM had better survival than conventional GBM, but the hazard ratio (0.91) did not reach statistical significance (95%CI: 0.69–1.20) in the multivariate analysis. Young ages (≤ 40 years), female sex, resection, radiotherapy, and chemotherapy were factors associated with better survival in overall GBMs. In subtype analyses, these factors remained statistically significant for conventional GBM, as well as radiotherapy for GSM.Conclusion: Our analysis found conventional GBM and its variants shared similar poor survival. Factors with age ≤ 40 years, female sex, resection, radiotherapy, and chemotherapy were associated with better prognosis in conventional GBM patients.

Keywords