Recurrent anaplastic gliomas
There is less consensus on how to treat anaplastic gliomas (anaplastic as¬tro¬cytoma, anaplastic oligodendroglioma, and mixed anaplastic oligoastrocytoma) at initial diagnosis and therefore even more variability in how these patients are treated at recurrence. The study that showed the benefit of chemoradiation and adjuvant temozolomide for newly diagnosed GB excluded patients with anaplastic tumors.1 A meta-analysis of clinical trials included adults with high-grade glioma who after initial surgery were treated with radiation plus chemotherapy (most often a nitrosourea) or radiation. The results only suggested that chemotherapy provided an additional survival benefit over radiotherapy alone for both GB and anaplastic patients.
Many cite the above data as the rationale for including temozolomide in the initial treatment of anaplastic tumors.33 Others, however, cite the results of NOA-04, a large study of patients with anaplastic glioma randomized to receive initial therapy with radiation or one of two chemotherapy regimens: procarbazine (Matulane), lomustine, and vincristine (PCV) or temozolomide.34 At tumor progression, patients who had received radiation were treated with either PCV or temozolomide, and those initially treated with either chemotherapy regimen were irradiated. This study demonstrated no difference in time to treatment failure or PFS among the three groups or any significant difference between the two chemotherapy regimens. Patients with anaplastic astrocytomas fared worse than those with anaplastic oligodendrogliomas or mixed tumors, suggesting that the latter groups may do well with initial chemotherapy only and then radiation at recurrence.
For patients with anaplastic tumors who have not previously been treated with temozolomide, studies do suggest that its use at recurrence is beneficial. In one phase II study, temozolomide-naive patients, or those who had previously received a nitrosourea, showed a 35% overall response rate and PFS-6 of 46% when treated with temozolomide at first recurrence.35 A recent study (RESCUE) evaluating a continuous low-dose temozolomide regimen of 50 mg/m2 in recurrent GB and anaplastic tumors demonstrated a PFS-6 of 35.7% for the anaplastic subgroup that contained patients who had previously had a variety of initial therapies, including chemoradiation.25
With the knowledge that the presence of chromosome 1p/19q codeletions in anaplastic tumors is prognostic for better outcomes,36 two ongoing studies will hopefully provide definitive answers for the treatment of recurrent anaplastic gliomas. The Chemoradiation and Adjuvant Temozolomide in Non-deleted Anaplastic Tumors (CATNON) study will randomly assign patients after surgery to receive either chemoradiation or radiation alone. Following this therapy, there is a second randomization to adjuvant temozolomide or observation only. The trial endpoint is overall survival. The phase III intergroup study of radiotherapy versus temozolomide alone versus radiotherapy with concomitant and adjuvant temozolomide for patients with 1p/19q codeleted anaplastic glioma (CODEL) will determine whether these patients with inherently better outcomes may do just as well with less aggressive therapy.
Several studies of bevacizumab have included recurrent anaplastic gliomas. Two studies of the combination of bevacizumab and irinotecan produced response rates of 55%–66% and PFS-6 of 56%–61%, suggesting activity.12,37 Extrapolating from the GB data, it is likely that single-agent bevacizumab would be efficacious and less toxic than this combination.
Emerging strategies
The identification of prognostic and predictive markers is paving the way for individualized treatment planning. In addition to the prognostic value of 1p/19q codeletions in anaplastic gliomas, the presence of MGMT promoter methylation in GB is likely predictive of response to temozolomide, although this is still under debate. There has been recent excitement about the demonstration that the presence of mutated isocitrate dehydrogenase 1 (IDH1) in gliomas is a robust independent factor associated with better outcome.38 For example, in a series of patients with anaplastic glioma, patients with the IDH1 mutation had a median survival four times longer than that of those without the mutation (81.1 months vs 19.4 months).39 This raises the question of the role of mutated IDH1 in glioma biology and makes it a potentially valuable therapeutic target.
Ongoing gene-expression profiling studies are showing that histologically indistinguishable GB can be clustered into distinct molecular subtypes, with widely different outcomes and responses to treatment.40 This likely contributed to the failure of past clinical trials, as the populations under study were, in fact, too diverse; potentially efficacious agents for one or more subtypes may have been overlooked. These studies are also identifying novel cellular targets such as MET, fibroblast growth factor receptor (FGFR), heat shock protein-90 (HSP-90), and hypoxia-inducible factor 1? (HIF1?).41 Other research is focusing on targets involved in glioma migration and invasion such as tenascin, the Src family of nonreceptor tyrosine kinases, the Rho family of small GTPases, and integrins. The role of glioma stem cells in glioma development and resistance to therapy is another emerging area of study and has led to the identification of specific glioma stem cell targets such as Notch and Sonic hedgehog.42