Current Standards of Care in Glioblastoma Therapy

Introduction

With an incidence of 3 to 5 per 100,000, glioblastoma fulfills the criteria of a rare cancer. Despite this, glioblastoma is the most common and most aggressive primary brain tumor and accounts for 12% to 15% of all intracranial neoplasms and 45% to 50% of all gliomas. Patients of any age may be affected, but it is most commonly observed in individuals over the age of 50 years. The suspicion of the presence of a glioblastoma is typically raised if a patient presents with neurologic symptoms and an imaging study shows the presence of a suspect lesion. On computed tomography (CT), the lesion typically appears as a contrast-enhancing lesion with peritumoral edema. However, smaller lesions may be missed. MRI is significantly more sensitive than CT and represents the modality of choice. Glioblastoma appears as a heterogeneous enhancing lesion with or without necrotic core. The margins are usually diffuse as a reflection of the infiltrative nature of the lesion. There is typically prominent peritumoral edema.

At present, despite aggressive treatment, the prognosis of glioblastoma remains dismal, with overall an estimated median life expectancy of only 14-16 months following diagnosis. Two-year and 5-year survival rates remain around 30% and 10% respectively. With these outcomes, it is essential to tailor the treatments for each individual patient to offer the best possible outcome with the optimal quality of life.

The initiation of therapy for glioblastoma therefore depends on many factors. These factors include, but are not limited to, the patient’s preoperative level of function, performance status, age, and the resources available to the patient and the treating physicians. This chapter describes the standard of care for newly diagnosed glioblastoma and provides a brief overview of the evidence that supports these treatment paradigms.

Surgical intervention

With radiographic and clinical findings suggestive of glioblastoma, referral to a neurosurgeon and specialized neuro-oncology team for evaluation is warranted. In those patients deemed capable of tolerating a neurosurgical intervention, safe maximal resection is typically preferred rather than biopsy. The decision to proceed with resection accomplishes several goals. The first is to provide an accurate initial diagnosis with a specific tumor grading, which functions as an important step in the initiation of a chemoradiation regimen. The analysis of molecular markers also provides essential molecular information about the tumor, which will assist with possible prognostic and treatment implications. Moreover, in the presence of significant neurologic deficits linked to an important mass that compresses surrounding tissue, a neurosurgical resection may also offer the possibility to improve the symptoms of the patient.

A biopsy should be performed in patients for whom an extensive resection cannot be performed safely. The main goal of the biopsy is to confirm the diagnosis of glioblastoma. The amount of tissue collected should also be sufficient to be able to perform some of the essential molecular marker analyses, especially the determination of the status of methylation of O6-methylguanine-methyl transferase (MGMT) gene promoter. This information is essential because these patients often present with poor performance status and treatment with temozolomide or radiation therapy alone might be important options to minimize further clinical deterioration (discussed later). However, note that the diagnosis established on biopsy might not represent the most aggressive part of the tumor: In 2001, Jackson and colleagues examined a consecutive series of 81 patients with imaging findings suggestive of glioma who underwent stereotactic biopsy followed by resection, and found that diagnosis based on biopsy or resection in the same patient differed in 30% of the cases reviewed. Woodworth and colleagues reviewed the histology of 21 stereotactic biopsies and found that although stereotactic biopsy samples correctly represented glioma in 91% of cases, 14% of cases were unable to be sufficiently graded for more definitive classification.

Surgical resection has been suggested to be an independent favorable prognostic factor for increases in Karnofsky Performance Scale (KPS), overall survival, and progression-free survival (PFS).

Beyond simply the decision to undergo surgical resection as opposed to biopsy, Sanai and colleagues showed that the extent of tumor resection and its correlation with survival could be estimated, based on retrospective data. They examined 500 consecutive newly diagnosed patients with supratentorial glioblastomas who underwent resection followed by chemoradiation. A significant survival advantage was imparted in those patients who had 78% extent of resection or greater ( Fig. 6.1 ). The strongest evidence to suggest that the extent of resection can determine the outcome of patients stems from the work of Stummer and colleagues, who evaluated the potential role of 5 aminolevulinic acid (5-ALA), an orally administered amino acid that allows neurosurgeons to visualize the tumor intraoperatively when illuminated with fluorescent blue light and to maximize surgical resection. This prospective randomized phase III trial included only patients for whom the neurosurgeon thought that a (near) complete resection could be achieved. Those with no residual tumor after the surgery showed a statistically significant improvement in overall survival compared with patients with poorer resections ( Fig. 6.2 ).

Carmustine wafers (Gliadel Wafer, Arbor Pharmaceuticals, LLC, Atlanta, GA) were first approved as adjunctive therapy by the US Food and Drug Administration (FDA) for recurrent glioblastoma in 1996, and then subsequently for de novo glioblastoma in 2003 based on trials conducted before the availability of temozolomide. Despite its approval by the FDA, carmustine wafers are not readily used as first-line treatment at many centers, partially because of cost and availability, but conflicting evidence from 2 recent meta-analyses has also questioned the modest survival benefit: both studies confirmed a statistically significant but modest survival advantage (16.2 months in the group in which carmustine wafers were added to adjuvant chemoradiotherapy vs 14 months in the standard treatment group) but also highlighted the complication profile (cited as high as 42.7%) for the use of carmustine wafers as routine treatment in de novo glioblastoma. Moreover, because of the difficulty in interpretation of postoperative MRI in the presence of carmustine wafers, their use is a contraindication to inclusion in most clinical trials. Unfortunately, carmustine wafers were never formally tested and compared within or against temozolomide-containing regimens.

Chemotherapy and radiation

Because glioblastomas are very infiltrative, neurosurgical resection cannot be considered as curative and additional treatments must be considered. These treatment decisions should ideally be defined by a multidisciplinary tumor board involving representatives from neurosurgery, radiation therapy, oncology, neuropathology, neuroradiology, and neurology. This approach should allow optimizing the management of every patient, based on the clinical situation, performance status, molecular markers, and other considerations. Management consists of a combination of radiation therapy to the site of tumor resection and residual tumor, usually with a safety margin of 2 cm, for a total of 58 to 60 Gy in fractions of 1.8 to 2.0 Gy. During this course of radiation therapy, the oral alkylating agent temozolomide is given daily at a dosage of 75 mg/m ²/d to increase the radiosensitivity of the tumor.

After a break of 4 weeks, a new MRI is performed and temozolomide maintenance therapy is started for a maximum of 6 cycles (150 mg/m ²on days 1–5 every 28 days in the first cycle, increased to 200 mg/m ²in subsequent cycles in the absence of significant bone marrow and liver toxicity; Fig. 6.3 ).