Determining optimal treatment strategy for diffuse glioma: the emerging role of IDH mutations

Abstract

The isocitrate dehydrogenase 1 and 2 (IDH1 and IDH2) genes mutate frequently in gliomas, and it has become increasingly apparent that IDH mutation status accounts for much of the prognostic information previously rendered by histological grading. Most glioblastomas (90–95%) are IDH-wild type and most lower-grade diffuse gliomas (80%) are IDH-mutant. We examine here how IDH mutation status interacts with treatments known to influence survival (surgery, chemotherapy, and radiotherapy) in patients with gliomas, and the impact of IDH mutations on patients’ survival after such treatments. IDH mutations are associated with more complete surgical resection of enhancing disease, and with a better response to radiotherapy. Additionally, there is increasing clinical evidence that, in certain contexts, IDH mutations predict chemotherapeutic sensitivity. Mutations in IDH and other genes are beginning to drive decisions on therapy for diffuse gliomas, and will likely allow tailoring of treatment by molecular profile in the future.

Introduction

Our knowledge about molecular markers in gliomas has undergone a rapid evolution in the last few years. This new era began with the publication of a landmark genetic sequencing study in 2008, in which isocitrate dehydrogenase 1 (IDH1) was identified as a frequently mutated gene in secondary glioblastoma [1]. Since that time, much work has been carried out to further determine the clinical value of mutations in IDH1 and close family member IDH2 in gliomas. Simultaneously, assessment of IDH mutation status has rapidly become incorporated as a standard component in the neuropathological assessment of brain tumors. Detection methods include immunohistochemistry, and PCR analysis with DNA sequencing.

Numerous further genomic alterations are strongly associated with IDH mutations – overlapping to define distinct subgroups of gliomas, including mutations of the TP53 gene and the alpha thalassemia/mental retardation syndrome X-linked (ATRX) gene in astrocytic gliomas [2], whereas oligodendroglial tumors harbor more frequently mutations in the homologue of the Drosophila capicua gene (CIC), mutations of the far upstream element (FUSE) binding protein 1(FUBP 1), co-deletion of chromosome arms 1p and 19q [3, 4], and mutations in promoter of the telomerase reverse transcriptase gene (TERT) [5, 6].

Neuro-oncologists now face the challenge of incorporating these new markers into the treatment of glioma patients, since many unanswered questions remain about the evidence base for using these markers in clinical decision-making. Whether the favorable prognosis seen in association with IDH mutation in gliomas is due to improved natural history or better response to therapy is unknown. Until now, treatment decision-making has been driven by histological type and malignancy grade. IDH-mutant gliomas comprise 80% of diffuse low grade gliomas (WHO grade II), 60% of anaplastic gliomas (WHO grade III) and 5–10% of glioblastomas (WHO grade IV) [7]. IDH mutant status therefore offers the basis for an alternative method of classification. Here, we will focus on the emerging role of the IDH mutations in gliomas, with regards to how they might influence treatment decisions.

Surgical management

Surgery is the first essential step in the classification and management of gliomas. In general, patients with suspected glioma undergo surgical resection to establish a histological and molecular diagnosis and to alleviate neurological symptoms. Due to the close linkage between surgical specimen size and accurate histologic grading [8], the choice for tumor biopsy instead of resection should be reserved for cases where surgical resection is associated with high risk of morbidity (due to location of the tumor or impaired clinical condition of the patient). If a biopsy is indicated, a sufficient amount of tissue should be obtained for molecular testing [9].

In glioblastomas, which are largely IDH wild-type, maximum safe resection of contrast-enhancing tumor is associated with improved survival [10]. In the management of IDH-mutant gliomas, aggressive surgery targeting all non-enhancing areas of disease remains the standard first-line treatment since, in low grade gliomas (which again are largely IDH mutant), a significant correlation has been demonstrated between maximal tumor resection and overall survival [11]. IDH-mutant high grade gliomas are more amenable to complete resection of enhancing tumor; a fact which may contribute to the better prognosis observed after aggressive surgery in this patient population [12]. Possible explanations for this phenomenon are the frequent location of IDH-mutant gliomas in the frontal lobe, the less functional disturbance of adjacent normal brain, or that they may have a less fibro-vascular tumor consistency [13], facilitating more extensive removal. Nevertheless, the lack of Class I evidence – due to lack of prospective randomized surgical trials with determination of IDH mutations -- prevents certainty in assessing the influence of extent of resection.

Radiation therapy

Radiotherapy (RT) is a mainstay in the initial treatment of patients with de novo high-grade gliomas – which are primarily IDH-wild-type. It is administered in the early post-operative phase and is often combined with chemotherapy. In diffuse (low-grade) gliomas, which are largely IDH mutant, the European Organization for Research and Treatment of Cancers (EORTC) trial 22845 showed a significant benefit of early RT in relation to longer progression-free survival and seizure control [14]. However, the trial was conducted before the detection of IDH mutations as a molecular marker in gliomas became standard. Therefore, prospectively collected data on the impact of IDH mutations with regard to RT are still lacking. On the other hand, an increased radiosensitivity of IDH-mutant diffuse gliomas in patients treated with RT has been reported in a small patient cohort [15]. This effect was clinically translated as improved survival. Furthermore, recent in vitro studies [16] and clinical data from EORTC trial 26951 [17] support the hypothesis that IDH-mutant gliomas are highly sensitive to RT. This sensitivity might be due the fact that wild-type IDH enzymes protect cells against gamma irradiation by maintaining NADPH levels that buffer against irradiation-induced reactive oxygen species, thus protecting cells from apoptosis [18].

Chemotherapy

While chemotherapy (CT) with alkylating agents – e.g., temozolomide or lomustine – improves survival in IDH-wild-type gliomas [19], few studies have investigated the role of CT in IDH-mutant gliomas [17, 20]. In this context, the chromosomal 1p/19q-deletion status should be considered as well, because virtually all 1p/19q-deleted gliomas are also IDH-mutant [21]. Long-term follow-up data of the randomized phase III Radiation Therapy Oncology Group (RTOG) trial 9402 [22] and the EORTC trial 26951 [23] demonstrated an improved outcome – in both progression-free and overall survival – in 1p/19q-codeleted anaplastic oligodendroglial tumors when procarbazine, lomustine and vincristine chemotherapy (PCV) was added to RT – for instance, in the RTOG 9402 trial the median survival of those with 1p/19q co-deleted tumors treated with PCV plus RT was twice that of patients receiving RT (14.7 v 7.3 years). Consequently, 1p/19q co-deletion is now incorporated in treatment decision-making in anaplastic oligodendroglial tumors.

Moreover, both aforementioned trials have retrospectively established IDH mutations as a prognostic marker. In the RTOG 9402 trial survival of anaplastic oligodendroglial and mixed oligoastrocytic tumor patients exhibiting an IDH-mutation treated with PCV and RT was significantly longer compared with patients without an IDH mutation (9.4 v 5.7 years) [20]. Thus, the role of IDH as a predictive marker of response to CT with PCV may be dependent upon its linkage with 1p/19q co-deletion and oligodendroglial histology. Accordingly, IDH status was prognostic but not predictive in the retrospective analysis of the German Glioma Network NOA-04 trial, in which patients with anaplastic gliomas were randomized to receive first either RT or CT (PCV or temozolomide) and upon disease progression to receive the other treatment modality [24]. Nevertheless, a recently published study demonstrated that IDH-mutant glioblastomas have a more pronounced radiographic response to chemoradiation therapy by serial quantitative MR volumetry than do their IDH-wild-type counterparts [25]. The predictive value of IDH mutation, separate from 1p/19q co-deletions, in relation to CT response has not yet been investigated prospectively. The role of adjuvant temozolomide in newly-diagnosed IDH mutant malignant gliomas also remains with a less developed evidence base than that of PCV [26].

Also unclear is the value of IDH mutation for predicting response to chemotherapy in diffuse low grade tumors. Contradictory results have been obtained in small retrospective studies. A larger prospective trial, the phase III RTOG 9802 trial, was conducted to compare the addition of PCV to RT versus treatment with RT alone for “high risk” WHO grade II diffuse glioma patients. This “high-risk” cohort included patients with subtotal tumor resection or > 40 years of age. The long-term follow-up data demonstrate that PCV + RT prolongs both progression-free and overall survival compared with RT alone [27]. However, in this trial, multivariable models that incorporate 1p/19q co-deletion and IDH mutational analyses have not yet been reported. Hence, no statements can be made about the predictive implication of molecular markers in this patient group. A further limitation is that all patients in RTOG 9802 received RT, precluding determination of the value of CT alone in this patient group. And lastly, no conclusions can be drawn on the effect of treatment with temozolomide, as patients in RTOG 9802 were treated with PCV.

On the other hand, there is increasing evidence that IDH mutant gliomas might follow a common mechanism of resistance to chemotherapy. In a recently published study, the exomes of 23 IDH-mutant low-grade gliomas and their subsequent recurrent tumors were sequenced [28]. In 6 of 10 patients treated with temozolomide, these tumors were hyper-mutated at recurrence and harbored driver mutations as a consequence of temozolomide-induced mutagenesis, indicative of an alternative evolutionary path to high-grade gliomas. Thus, a connection between temozolomide treatment, driver mutations in oncogenic signaling pathways, and malignant progression was suggested. A further study aimed to investigate the possible causative relationship between IDH mutation and response to temozolomide in vitro [29]. In this study, IDH mutations introduced into immortalized untransformed human astrocytes initiated expression alterations in DNA repair genes that facilitated repair of temozolomide-induced DNA damage, resulting in temozolomide.

The results of the ongoing CODEL (NCT00887146) and CATNON (NCT00626990), multicenter phase III trials may well clear these uncertainties. The CODEL trial addresses newly diagnosed anaplastic gliomas with 1p/19q codeletion – mainly oligodendroglial tumors -- and the CATNON trial is designed for patients with newly diagnosed anaplastic gliomas without 1p/19q codeletion – primarily astrocytic tumors. Once available, the results from these clinical studies can hopefully determine the influence of molecular stratification by IDH mutation on resistance to temozolomide.

Finally, novel methods directly targeting the IDH mutation itself hold significant promise for improving treatment of these gliomas. Inhibitors of the mutant enzyme can slow the growth of an IDH-mutant glioma in mice [30], and a vaccintion strategy targeting the mutant IDH1 epitope has been shown to induce anti-tumor immunity in a humanized murine model [31]. Clinical trials of these experimental therapies are underway.

Conclusion

IDH mutation status is emerging as a useful guide for treatment decision-making in gliomas. To optimize the application of existing therapies, we would propose the following conclusions based on the available evidence: 1) Enhancing high-grade IDH wild-type gliomas should be managed with an aim towards complete surgical resection of enhancing disease, and combined/concurrent chemoradiation therapy with temozolomide according to the Stupp glioblastoma regimen, whereas in non-enhancing or diffuse lower-grade IDH wild-type gliomas, surgical resection followed by radiotherapy alone is the current standard of care, awaiting the results of the ongoing EORTC/RTOG phase III trials 2) IDH mutant 1p/19q-codeleted gliomas should be managed with maximal safe surgical resection of both enhancing and non-enhancing disease, followed by sequential radiation therapy and PCV chemotherapy, 3) IDH mutant 1p/19q-non-deleted glioma management can be controversial, given the conflicting available evidence. These gliomas should be managed with maximal safe surgical resection of enhancing and non-enhancing disease, which can be followed by radiation alone, or sequential radiation and chemotherapy, or enrollment in clinical trials of chemotherapy or IDH-mutant directed agents.