Postoperative standard chemoradiotherapy benefits primary glioblastoma patients of all ages

Abstract

Background: Glioblastoma is the most malignant tumor of the central nervous system. Several prediction models have been produced to aid in prognosis assessment. Age, a primary decision factor for prognosis, is associated with increased genomic alterations, however the exact link between increased age and poor prognosis is unknown.

Objective: In this study, we aimed to reveal the underlying cause of poor prognosis in elderly patients.

Methods: This study included data on 616 primary GBM tumor samples from the CGGA and TCGA databases and 41 nontumor brain tissue samples obtained from GSE53890. Hallmarks and clinicopathological characteristics were evaluated in both tumor and nontumor brain tissues. The association between choice of treatment regimen and age was measured using ANOVA and Student's t test.

Results: Age was a robust predictor of poor prognosis in glioma. No age-related hallmarks of cancer were detected, including pathological characteristics or mutations. However, treatment choice was strongly significantly different between old and young patients. Combined chemo-radiation therapy could benefit old and young GBM patients, however, old patients are currently less likely to choose it.

Conclusion: The vast divergence in prognosis between young and old GBM patients is largely caused by choice of treatment rather than age-related tumor genomic characteristics. Postoperative standard radio- and chemotherapy provide strong benefits to primary glioblastoma patients of all ages.

Keywords: adjuvant therapy; age; genomic alteration; glioblastoma; prognosis.

Figure 1

Survival difference between old and young GBM patients. (A and B) Statistical difference in survival between the old and young group. Older patients have a poorer OS and PFS prognosis. (C and D) P‐value between prognosis of patients using arbitrary ages as the cut‐off. Points below the black dotted line represented statistically significant differences in prognosis. The number at risk and the cumulative number of events are important parameters of Kaplan‐Meier survival analysis. The number at risk represents the number of subjects at risk at the corresponding time. Cumulative number of events represents the total number of deaths at the corresponding time. The log‐rank test was used to assess the statistical significance of stratified survival groups

Figure 2

Correlation between biofunction, genomic characteristics, and age. A, Age distribution of hallmarks of cancer in normal and GBM samples. Both normal and tumor tissue have hallmarks positively or negatively correlated with age. R and P values were obtained from the pearson correlation analysis between age and hallmark scores. A significant positive correlation was represented by red dots. A significant negative correlation was represented by blue dots. Gray dots represented no significant correlation. The calculation method of the hallmark scores is detailed in the methods section. B and C, Age distribution of GBM transcriptome subtypes. The yellow dots represent the classical subtype. The red dots represent the mesenchymal subtype. The green dots represent the neural subtype. The blue dots represent the proneural subtype. No significant difference was observed among the 4 subtypes. One‐way ANOVA analysis was performed to assess the statistical significance of variance. D, Overall genomic characteristics distribution ranked by age. No statistical difference was observed as age increased. Pearson correlation analysis was performed to assess the statistical significance between age and genome characteristics

Figure 3

Age distribution in EGFR amplification status in primary GBM. A and D, No significant difference was observed across age groups between amplification and wildtype EGFR status in CGGA and TCGA databases. An unpaired t test was used in the differential analysis. B, C, E and F, Kaplan‐Meier survival analysis for EGFR amplification status in GBM. The log‐rank test was used in survival analysis

Figure 4

Age distribution in TP53 mutation status in primary GBM. A and D, No significant difference across age groups was observed between TP53 mutation and wild‐type GBMs. An unpaired t test was used in the differential analysis. B, C, E and F, Kaplan‐Meier survival analysis of TP53 mutation status. Statistically different PFS was observed in the TCGA database between TP53 mutation and wild‐type group (P = .01). The log‐rank test was used in survival analysis

Figure 5

Treatment options in young and elder GBM patients. A and D, Age distribution of different treatment options. Younger patients tended to receive standardized combination therapy more often in CGGA and TCGA databases. The unpaired t test was used in the differential analysis. B, C, E, and F, Kaplan‐Meier survival analysis for different treatment options. Patients who received standardized postoperative adjuvant therapy had significantly longer overall and PFS survival time. RT + TMZ represents patients undergoing postoperative adjuvant radiotherapy and TMZ chemotherapy combined therapy. RT/TMZ/Observe means patients only receive RT or TMZ or Observe after surgery. The log‐rank test was used in survival analysis

Figure 6

Postoperative standard adjuvant therapy benefited all age grades of GBM patients. A and B, Younger patients who received combination therapy had a better prognosis than those did not (P < .0001 and P = .00017 in OS and PFS survival analysis respectively). C and D, Results were verified by the TCGA database (OS, P < .0001 and PFS, P = .019). E and F, Older patients could also benefit from combination therapy (OS, P < .0001 and PFS, P = .0002)