This article is mentioned in:

Introduction

Malignant primary brain tumors are a fairly uncommon malignancy, representing 2–3% of all adult tumors in Saudi Arabia. Glioblastoma multiforme (GBM) is the most common and most malignant primary tumor of the brain. It arises from astrocytes, and is characterized by rapid growth and short time to progression. Consequently, it is associated with one of the worst 5 year survival rates amongst all human cancer types (1–4).

The standard of care in patients with GBM includes maximal surgical resection, followed by radiotherapy (60 Gy in 30 fractions) with concomitant and adjuvant temozolomide (TMZ). The addition of TMZ to radiation therapy has increased both the median survival (12.1 to 14.6 months) and the 2 year survival duration (10 to 26%) (5). Despite recent advances in the understanding of the molecular mechanism of tumorigenesis, the outcome of patients with cancer remains poor and therefore there is an urgent requirement for more effective initial treatments for this intractable disease (6–10). Recent therapies under investigation include immunotherapy, chemotherapy, targeted molecular therapy, antiangiogenic therapy, gene therapy, radiation-enhancement and drugs for overcoming resistance (11).

Although prognosis is extremely poor, a limited number of patients with GBM do survive past 36 months. However, a limited understanding of the predictors for survival amongst patients with GBM exists. Additionally, no studies have as yet examined the outcome of GBM in Saudi Arabia. Therefore the aim of the present study was to assess the real world outcome of patients with GBM in this region and to determine the important clinical, pathological and molecular prognostic factors correlated with patient outcomes in this population.

Patients and methods

Study design

The present single-center retrospective cohort study was performed in the Comprehensive Cancer Center at King Fahad Medical city (Riyadh, Saudi Arabia). The present study was approved by the local Ethics Committee of the Comprehensive Cancer Center (no. 14-163).

Patients

All adult patients (>18-years-old), who were diagnosed with histologically proven GBM between January 2008 and December 2013 were included in the present study.

Intervention

Patients underwent either a biopsy or a resection, following which they received standard or hypo-fractionated radiotherapy (XRT), with or without concurrent temozolmide (TMZ), as the current standard of care. Following completion of XRT, a proportion of patients received further cycles of TMZ.

Data collection

The data were obtained from electronic medical records, where a data collection form was developed to collect patient demographics, pathology, XRT and chemotherapy details, and progression and survival outcomes. The patient and tumor characteristics are categorized, as shown in Tables I and II.

Table I. Patient characteristics (n=90).

Table I.

Patient characteristics (n=90).

Characteristic	No. patients, n (%)
Age, years (range)
Median	49 (18–81)
≥65	18 (20)
Gender
Male	67 (74)
Female	23 (26)
Eastern Cooperative Oncology Group
≤2	54 (60)
>2	36 (40)
Co-morbidities
Yes	36 (40)
No	54 (60)
Year of diagnosis
2008	14 (15)
2009	17 (19)
2010	10 (11)
2011	16 (18)
2012	15 (17)
2013	18 (20)

Table II. Tumor characteristics (n=90).

Table II.

Tumor characteristics (n=90).

Characteristic	No. patients, n (%)
Tumor site
Frontal	22 (24)
Temporal	21 (23)
Parietal	13 (14)
Other	34 (38)
Hemisphere
Right	44 (49)
Left	32 (36)
Bilateral	14 (16)
Pathology
GBM	78 (87)
GBM variant	12 (13)
Oligodendroglioma	7 (58)
Primitive neuroectodermal tumor	2 (17)
Gliosarcoma	2 (17)
Giant cell	1 (8)
MGMT status
Methylated	2 (2)
Unmethylated	5 (6)
Unknown	83 (92)

[i] GBM, glioblastoma multiforme; MGMT, O⁶-methylguanin-DNA-methyltransferase

Statistical analysis

The OS and progression-free survival (PFS) were estimated using Kaplan-Meier methodology. Univariate analyses were performed using the log-rank test and multivariate analyses using the Cox proportional hazards model. P<0.05 was considered to indicate a statistically significant difference. All analyses were performed using SPSS version 22 (IBM SPSS, Armonk, NY, USA).

Results

A total of 129 patients with newly diagnosed GBM were identified for inclusion in the present study (Fig. 1). Of these patients, 39 had missing data and were excluded from the analysis. Therefore, the data from 90 patients was included for analysis in the final study population: 67 (74.4%) males and 23 (25.6%) females, with a median age of 49.0 (Table I).

Figure 1. Description of study cohort.

Tumor characteristics were described in Table II. The majority of patients (87%) were diagnosed with a GBM, whilst the remaining 13% were diagnosed with GBM variants, most commonly oligodendroglioma (n=7/12, 58%). Frontal (24%) and temporal (23%) tumors were the most common tumor sites, with almost half of the tumors (49%) located in the right hemisphere. In general, the methylation status of the tumor could not be determined.

In the study cohort, 81% of patients underwent resection, whilst the remaining 19% underwent biopsy only (Fig. 2). Of those patients who had their tumors resected, the majority (81%) were deemed to have residual disease following surgery. The majority (98%) of patients received XRT; most received standard XRT (n=67/88, 76%) whilst the remaining 24% received a hypo-fractionated dosage (Fig. 2). The median dose of radiotherapy was 59.4 Gy. Concurrent TMZ was received by 72% of patients. Notably, whilst nearly all patients (n=65/67, 97%) receiving standard XRT also received TMZ, only a quarter (n=5/21, 24%) of those receiving hypo-fractionated dose also received TMZ. Following completion of XRT, the majority of patients (n=55/90, 61%) received further cycles of TMZ, although half were unable to complete their chemotherapy regimen.

Figure 2. Treatment exposure of patients.

The median PFS was 5.3 months [95% confidence intervals (CI), 4.47–6.62], with 43% PFS at 6 months (Fig. 3A). The median OS was 13.7 months (95% CI, 10.1–17.5), with 53% OS at 1 year and median follow-up 12.5 months (1.5–70.2 months; Fig. 3B). The PFS and OS were also examined in the subset of patients who received standard XRT with concurrent TMZ (n=65). In these patients, the median PFS was 6.7 months (95% CI, 4.7–11), with a 6 month PFS of 55% (Fig. 4A). The median OS was 19.7 months (95% CI, 11.9–27.4), with a 1 year OS of 65% (Fig. 4B).

Figure 3. (A) PFS and (B) OS for all patients (n=90). PFS, Progression-free survival; OS, overall survival.

Figure 4. (A) PFS and (B) OS probability over time for patients who received combined modality treatment (n=65). PFS, Progression-free survival; OS, overall survival.

To determine the factors, which may be associated with PFS and OS, univariate and multivariate analyses were performed. In univariate analysis, age, Eastern Cooperative Oncology Group (ECOG) performance scale, surgery type, presence of residual disease, type of radiotherapy and receipt of chemotherapy were all predictors of both PFS and OS (Tables III and IV). However, in multivariate analysis, only ECOG ≤2 [odds ratio (OR), 0.3; P<0.001], absence of residual disease (OR, 0.3; P=0.02) and receipt of adjuvant TMZ (OR, 0.5; P=0.05) were significant predictors of survival (Tables III and IV).

Table III. Univariate and multivariate analysis for progression-free survival.

Table III.

Univariate and multivariate analysis for progression-free survival.

	Univariate			Multivariate
Characteristic	OR	CI	P-value	OR	CI	P-value
Age
≥65 (n=18)
<65 (n=72)	0.5	0.3–0.8	0.007	1.0	0.5–1.9	0.7
Eastern Cooperative Oncology Group
>2 (n=36)
≤2 (n=54)	0.2	0.1–0.3	<0.001	0.3	0.2–0.5	<0.001
Surgery
Biopsy (n=17)
Any resection (n=73)	0.5	0.3–0.8	0.006	0.9	0.5–1.7	0.8
Residual
Yes (n=76)
No (n=14)	0.3	0.1–0.5	<0.001	0.4	0.1–0.8	0.01
Radiotherapy
Hypo-fractionated (n=21)
Standard (n=67)	0.2	0.1–0.3	<0.001	0.7	0.3–1.5	0.3
Chemotherapy
No (n=35)
Yes (n=55)	0.3	0.2–0.4	<0.001	0.4	0.2–0.7	0.004

[i] OR, odds ratio; CI, confidence intervals.

Table IV. Univariate and multivariate analysis for overall survival.

Table IV.

Univariate and multivariate analysis for overall survival.

	Univariate			Multivariate
Characteristic	OR	CI	P-value	OR	CI	P-value
Age
≥65 (n=18)
<65 (n=72)	0.5	0.3–0.8	0.009	0.6	0.3–1.2	0.7
Eastern Cooperative Oncology Group
>2 (n=36)
≤2 (n=54)	0.2	0.1–0.3	<0.001	0.3	0.1–0.5	<0.001
Surgery
Biopsy (n=17)
Any resection (n=73)	0.4	0.2–0.8	0.004	0.9	0.5–1.7	0.8
Residual
Yes (n=76)
No (n=14)	0.2	0.1–0.5	<0.001	0.3	0.1–0.8	0.02
Radiotherapy
Hypo-fractionated (n=21)
Standard (n=67)	0.2	0.1–0.4	<0.001	0.7	0.3–1.5	0.3
Chemotherapy
No (n=35)
Yes (n=55)	0.2	0.1–0.4	<0.001	0.5	0.2–1.0	0.05

[i] OR, odds ratio; CI, confidence intervals.

Discussion

The present retrospective analysis is the first study, to the best of our knowledge, to examine the outcome of patients with malignant GBM in Saudi Arabia, and to determine the important clinical and pathological prognostic factors that are correlated with the outcome in this region.

A median PFS of 5.3 months and a median OS of 13.7 months was demonstrated. These results are consistent with previous studies in patients with GBM (5,12,13). A 6 month PFS of 43% and a 1 year OS of 53% was observed, which was also in line with previous studies. Notably, the PFS and OS were improved in the subgroup analysis of patients receiving combined modality treatment, reflecting recent studies which demonstrated that addition of concomitant chemotherapy to XRT resulted in improved outcomes (14,15).

In the present multivariate analysis, only ECOG, residual disease and receipt of chemotherapy were significant predictors of survival. Previous studies have reported that age, tumor grade (anaplastic glioma vs. GBM), Karnofsky performance status, the number of molecular alterations and the extent of initial surgical resection are all prognostic factors for outcome in patients with GBM. Therefore in general, the present results are in line with those reported elsewhere. Although no randomized controlled trials have been performed to establish the benefit of maximal surgical resection over a more limited resection, numerous previous studies have suggested that maximal resection, particularly gross total resection, does improve survival (16–24). However, evidence remains conflicting, with further studies failing to show a benefit with more extensive surgical resection (either subtotal resection vs. biopsy, or complete vs. subtotal resection) (25–27). The association of chemotherapy with survival is supportive of recent trials demonstrating superior outcomes for those patients treated with combined modality therapy compared with radiotherapy alone (5). Age was not predictive of XRT, which is in contrast to previous studies reporting that age is a reliable predictor for outcome and treatment response in elderly patients with glioma (28–31). Another previous study revealed that age was only prognostic in patients undergoing biopsy, however, not in those undergoing resection (32). As a result of the small number of biopsy patients within the present study, age was not determined as a prognostic for biopsy patients within the present population.

Certain limitations to the present study exist. Due to the retrospective nature of the present study, certain data was missing and therefore a third of the original population were excluded from the analysis due to this. Although the present study attempted to control for potential confounders through multivariate analyses, a randomized controlled trial may provide a more robust environment to control for factors which may influence the outcome. Additionally, other factors which could not be collected, including tumor size and extent of resection, may have impacted on patient outcome. Another limitation was that treatment conditions for patients in the present study were heterogeneous since administration of adjuvant chemotherapy was based on personal preferences and experience. In addition, since the present study was a single centre study, it is possible that results are not fully generalizable across Saudi Arabia. Furthermore, evaluation of pseudo-progression may not have been systematic as the common response criteria were unclear. Molecular profiling data for O6-methylguanin-DNA-methyltransferase-status was not part of routine assessment during the majority of the study time-frame and is therefore not complete for the majority of patients. Previous studies have suggested that symptom type and histological factors, including necrosis, endothelial abnormalities and degree of anaplasia are important prognostic factors and these were not included in the present analysis (33–35).

In conclusion, the present study found that outcomes for patients in Saudi Arabia with GBM are in line with those reported in previous studies worldwide. Good performance status at diagnosis, absence of residual disease and chemotherapy were independently associated with an improved outcome. Future studies must examine these factors further, whilst clinicians must focus on optimizing these components within their treatment of patients with GBM, in order to improve outcomes.

References