Introduction

Gliomas account for the majority of primary CNS tumours and are most commonly high-grade, characterised by aggressive growth and poor prognosis. Treatment is determined by a number of different prognostic factors including age of the patient, performance status, tumour location and histological grade. There have been a number of advances in the diagnosis and management of HGG which are reviewed here.

Longer-term data on chemo-radiotherapy for GBM

The gold standard management of patients with newly diagnosed GBM under the age of 70 years is now maximal safe resection with concomitant chemoradiation using temozolomide, an oral alkylating agent, followed by six cycles of adjuvant temozolomide. This is based on the pivotal EORTC trial reported in 2004, demonstrating a hazard ratio for death of 0.63 (95 % CI 0.52–0.75) using this regimen compared with radiotherapy alone, with median survival increasing from 12.1 to 14.6 months and 2-year survival rates from 10.4 to 26.5 % [1]. The 5-year follow-up of this cohort was published recently by Stupp et al. [2] and showed that the survival advantage was maintained in the chemoradiation group (5 year overall survival 9.8 %compared to 1.9 %). The fact that 10 % of patients with GBM are now surviving 5 years has not previously been reported in any prospective study and raises the expectation of important improvements in survival with other new agents.

There is increasing awareness of the role of epigenetic modifications in determining response to treatment. In GBM, one of the most important prognostic factors for improved survival is reduced availability of the DNA repair protein methyl guanyl methyl transferase (MGMT), whose gene expression is prevented by methylation of a promoter region upstream from the coding region. MGMT promoter methylation has been shown to be associated with improved survival in patients treated with chemoradiation. Important data were reported by Stupp et al. [2] in their 2009 paper on 206 cases where MGMT promoter methylation status was known. Their analysis provided long-term information on the effect of MGMT status, confirming that it is a strong prognostic factor in GBM, with an overall hazard ratio of 0.49 (0.32–0.76) for 5-year survival. Additionally, within the limits of small numbers, the analysis also confirmed that MGMT status was predictive of TMZ response, at least in terms of progression-free survival. Finally, this report confirmed long-term efficacy of combination treatment in the 60–70 years age group (HR 0.7, 95 % CI 0.5–0.97).

First line therapy in older patients with high grade glioma

The seminal study that prompted the widespread use of chemo-radiotherapy in GBM in 2004 also reported age greater than 70 years to be a poor prognostic factor and suggested that TMZ/RT may be less efficacious and poorly tolerated in the older age group [1]. In general, patients greater than 70 with GBM are treated with best supportive care or biopsy/limited debulking followed by a short course of radiotherapy. Two recent studies suggest that single agent temozolomide may be an alternative to radiotherapy in older patients, who traditionally experience more treatment toxicity from brain irradiation.

The Nordic phase-3 trial randomized nearly 300 glioblastoma patients to one of three treatments: TMZ 200 mg/m2 in a standard 5 days per month schedule, standard RT (60 Gy), or hypofractionated RT (34 Gy in 10 fractions) [3]. Overall, there was longer survival with TMZ compared to standard RT (9.3 vs. 6.0 months) and a trend towards longer survival with hypofractionated RT compared to standard RT (7.7 vs. 6.0 months). Interestingly, this effect was demonstrated to arise from the subgroup of patients >70 years only, in which both TMZ (p < 0.001) and hypofractionated RT (p = 0.02) were superior to standard RT. Furthermore, there was a trend in the >70 years group for superiority of TMZ over hypofractionated RT (p = 0.09). MGMT promoter methylation had no effect on response to RT, but predicted better response to TMZ (HR 0.56, p = 0.02).

The NOA-08 trial randomly assigned nearly 400 patients over the age of 65 years with anaplastic astrocytoma or glioblastoma to either TMZ (100 mg/m2 in a dense 7 days on/7 days off schedule) or RT (60 Gy) [4]. The results showed that TMZ monotherapy was non-inferior to radiotherapy in terms of overall survival (OS) or progression-free survival (PFS). Additionally, it was shown that MGMT promoter methylation significantly predicted the effect of TMZ on progression free survival (not overall survival). Compared to the RT group as a reference, patients with MGMT promoter methylation responded better to TMZ than radiotherapy (HR 0.53) while those without promoter methylation tumours did worse on TMZ (HR 1.95).

In both trials, quality of life was equivalent or superior in the TMZ groups but there was, as expected, an excess of haematological adverse events. The treatment was well tolerated. These trials indicate that TMZ monotherapy is equivalent and possibly superior to standard RT in patients over 65 years. Importantly, MGMT promoter status seems to strongly predict better outcomes and greater benefit from TMZ, and should be considered when deciding on treatment. This oral alkylating agent is administered orally and can be commenced rapidly once the treatment decision is made. This has to be balanced against the longer duration of treatment and the risk of haematological adverse events. The brief (2-week) hypofractionated RT reported in the Nordic trial is also a valid option, based on their data.

Bevacizumab in recurrent glioblastoma multiforme

Bevacizumab (Avastin) is a monoclonal antibody directed against vascular endothelial growth factor (VEGF). VEGF is a crucial factor for angiogenesis [5] and bevacizumab is currently used in a number of cancers. Improved overall survival has been shown in non-small cell lung cancer [6] and advanced colorectal cancer [7]. Progression-free survival is improved in other cancers, including metastatic renal tumours [8] and advanced breast carcinoma [9]. GBM is a highly angiogenic tumour, and one of its histo-pathological hallmarks is microvascular proliferation [10]. Based on the results of a Phase II study of irinotecan and bevacizumab in recurrent GBM, which showed an improvement in 6-month progression-free survival, the FDA approved its use in 2009 [11]. This decision has prejudiced further studies to evaluate its role in recurrent GBM.

There are a number of published case series and prospective studies that provide data on the use of bevacizumab in patients with recurrent glioblastoma, both as monontherapy and in combination with other chemotherapy. While a small number of retrospective reports suggest improved survival [12–14], most reports and all prospective data seem to agree that the drug improves progression-free survival and quality of life in approximately two-thirds of recurrent glioblastoma patients, but has no significant effect on overall survival [11, 15–35]. This is further confirmed in two large (c. 500 cases) meta-analyses [36, 37]. Also consistently reported is that the drug is well tolerated with the main concerns being bleeding or, less commonly, thrombosis. There is evidence for other effects on tumour behaviour, with many patients experiencing out-of-radiotherapy-field progression [38].

In the absence of large randomized control trials, these data are our best estimate of the effect of bevacizumab in recurrent glioblastoma. Bevacizumab is expensive (in the region of £10,000 GBP per month) and given the lack of an effect on overall survival it is highly unlikely that it will become available for UK patients with recurrent GBM, based on current NICE criteria. If patients more likely to respond could be identified, the potentially important effects on quality of life and delayed progression in a proportion of patients may justify its use. In terms of newly-diagnosed GBM, interim analyses of trials examining chemoradiation plus or minus bevacizumab have also failed to demonstrate any improvement in overall survival in patients treated with bevacizumab at diagnosis, and possibly worse QoL, although there is again an effect on progression-free survival [39, 40].

Adjuvant therapy in grade III (anaplastic) oligodendroglioma

Oligodendrogliomas are chemosensitive tumours, characterised by the presence of co-deletion of chromosomes 1p/19q, which has been known to be associated with a prolonged and durable response to PCV chemotherapy since the 1990s [41]. Two randomised controlled trials of PCV in anaplastic oligodendroglioma were published in 2006; the EORTC trial of adjuvant PCV [42] and the North American trial of neo-adjuvant PCV [43]. While both showed improved PFS, neither could not demonstrate improved overall survival from the addition of PCV chemotherapy. Both trials confirmed the strong prognostic value of 1p/19q codeletion but neither found a predictive role.

The 12-year follow-up data has recently been analysed for both cohorts. The European group shows a significant overall survival benefit in patients treated with PCV (median 42.3 vs. 30.6 months), particularly in those with 1p/19q co-deletion (median not reached vs. 112 months, HR 0.56). There was no substantial evidence for a benefit from PCV in patients without 1p/19q co-deletion [44]. The 12-year data from the North American study again validates 1p/19q co-deletion as a positive prognostic factor. In addition, while there is no benefit of treatment with PCV in the non-deleted group of patients, those with co-deletion are demonstrated to have statistically significantly better overall survival (median 14.7 compared to 7.3 years, HR 0.59) and PFS [45].

These data now inform us that standard of care for patients with anaplastic oligodendroglioma who carry the 1p/19q co-deletion is post-surgical radiotherapy followed by PCV chemotherapy. It is unclear, whether temozolomide may provide a similar benefit in this group.

Conclusions

The management of CNS gliomas continues to evolve. This brief review describes recent developments that impact on patient management, and highlights the dynamic nature of this progressive field. The speed at which optimal patient management can change supports the importance of the multidisciplinary team model in the assessment and treatment of glioma patients.