Gliomas typically exhibit features of astrocytes, or oligodendrocytes, or both (mixed glioma) (Fig. 52-1). Microscopically, gliomas appear as diffusely infiltrating cancers of three types: astrocytic, oligodendroglial, and oligoastrocytic (combining the morphologic features of both oligodendroglioma and astrocytoma).

Figure 52-1 Gliomas.

The World Health Organization uses a three-tiered classification system based on histologic criteria that divides these tumors into low-grade glioma, anaplastic glioma, and glioblastoma multiforme (Table 52-1). Low-grade tumors may contain a high density of almost normal-appearing cells. Here the percentage of cells that are dividing (as determined by mib-1 or KI-67 staining) is often 2% or less. Anaplastic gliomas exhibit more atypical cells, with pleomorphic nuclei having growth rates in the 5–10% range but no evidence of necrosis. Gliomas with high growth rates (>10% mitotic figures) and necrosis are classified as glioblastoma multiforme (GBM). The less common pilocytic astrocytomas are a separate category of glioma that are histologically characterized by Rosenthal’s fibers, usually occur in children, and often have a good prognosis if surgical resection can be achieved. Tumor grade is the most reliable predictor of prognosis. Even if the lesion cannot be safely excised, a needle biopsy is often indicated. Gliomas are not staged as other cancers are because they rarely metastasize outside the CNS. Analysis of tumor samples for genetic abnormalities can help predict response to therapy and will likely lead to a better classification system for gliomas. This classification is valuable prognostically; low-grade gliomas have median survivals of 5–15 years, anaplastic gliomas 2–5 years, and GBM 12–18 months.

Table 52-1 Grades of Gliomas

Glioblastoma

These extremely malignant tumors frequently present with seizures, aphasia, or other focal symptomatology, pointing to the specific areas of pathologic origin. Very infrequently, a glioma may manifest itself more globally, gliomatosis cerebri, wherein there is widespread dissemination of neoplastic cells globally through a hemisphere or even the entire brain per se. These relatively rare patients may present with cognitive or personality changes. On other occasions, even though the patient presents relatively acutely with focal findings, the clinician is surprised to find a diffusely invasive malignant tumor despite the clinical presentation compatible with an acute focal brain pathology. This is the very common, most aggressive, and the least likely of the gliomas to respond to therapy. “Multiforme” refers to the tumor’s gross appearance. Often areas of necrosis, hemorrhage, and fleshy tumor exist within the same tumor focus.

Two types of GBM (Grade IV astrocytomas) are distinguished by molecular features. The classic primary GBM arises relatively suddenly in an older person with no preexisting history. Characteristically, primary GBM have an amplification and overexpression of the epidermal growth factor receptor (EGFR) and ligand (EGF). A mutated form of EGFR, EGFRvIII is another hallmark of primary GBM, present in about 15–20% of cases. EGFRvIII may confer an unfavorable prognosis. p53 mutations are uncommon. Classically secondary GBM arises gradually from a low-grade astrocytoma in a younger adult and harbors a p53 mutation. As it undergoes anaplastic transformation, the secondary GBM accumulates other genetic derangements, most notably, mutation of the Rb gene, deletion of the tumor suppressor gene p16/CDKN2A, and amplification of CDK4.

When clinical behavior and genetic abnormalities of GBM tumors are reviewed, a developmental dichotomy emerges. Younger patients with GBM sometimes have a longer history of symptoms or a history of a lower-grade glioma, suggesting that the tumor developed from a lower-grade precursor, whereas older patients with GBM tend to have relatively sudden symptom onset, suggesting that the malignancy did not evolve from a less aggressive tumor. Genetic analysis of GBM samples from older patients frequently reveals overexpression of the epidermal growth factor receptor and loss of 10q. Tumor samples from younger patients are more likely to show mutations in p53, RB, overexpression of the platelet-derived growth factor receptor, and loss of 19q—changes often seen in lower-grade gliomas.

Diagnosis, Treatment, and Prognosis

MRI is the most specific diagnostic modality (Fig. 52-2). On most occasions, one sees focal heterogeneous irregular-margined cystic mass lesions with perilesion edema, gadolinium rim enhancement, and often enough mass effect to produce a transtentorial herniation. In contrast, the occasional patients with gliomatosis cerebri have a characteristic diffusely abnormal MRI picture characterized by multiple areas of subtle white matter enhancement with extension into the cortical mantle, extending far beyond what their clinical presentation usually dictates (Fig. 52-3).

Figure 52-2 Glioma MRI and Pathology.

Figure 52-3 Gliomatosis Cerebri.

Even with early diagnosis, the prognosis remains grim and most patients will fail therapy within 12 months of diagnosis. The first treatment step is to perform as wide a surgical resection as is functionally tolerable. Younger patients with a normal examination who have had a gross total resection have the best prognosis. Postoperative radiation therapy (RT) clearly benefits many patients as those GBM patients who receive RT have a median survival twice that of those who did not.

Combining RT with concomitant and adjuvant chemotherapy is now the standard of care for patients with GBM. RT plus temozolomide leads to a modest benefit in overall survival (14.6 vs. 12.1 months). However, more importantly, there is a significant increase in the percentage of those surviving 2 or more years (26.5% vs. 10.4%). Bevacizumab, an antagonist of vascular endothelial growth factor, has recently proven safe and effective in patients with recurrent GBM. Recent reports indicate a 6-month progression-free survival of 46%. It is now an urgent priority to determine how best to use this new tool and what agents might work synergistically with it.

When patients have failed these Food and Drug Administration (FDA)–approved treatments, a clinical trial should be considered. Molecular research is defining a number of potential glioma cell targets. These are mostly second messenger molecules involved in pathways that enhance cell proliferation or inhibit programmed cell death. The goal is to treat a selected group of patients whose tumors overexpress the specific target of the treatment drug.

Low-Grade Glioma

Clinical Vignette

This 34-year-old right-handed woman presented with generalized seizures. Several months earlier, she noted episodes of an unusual smell but these did not cause her immediate concern. Brain MRI demonstrated a right temporal lobe lesion, bright on T2 and FLAIR imaging but hypointense on T1, with no evidence of enhancement after gadolinium (Fig. 52-4). The patient was treated with oxcarbazepine and admitted to the hospital. Open biopsy was nondiagnostic but subsequent temporal lobectomy revealed an oligodendroglioma with a Ki-67 index of 3.8%. Postoperatively, the patient was treated with monthly temozolomide for 1 year. She is now receiving no treatment and has been clinically and radiographically stable for 2 years.

Figure 52-4 Oligodendroglioma.

Clinical Presentation/Pathology

Low-grade gliomas (LGGs) are slow growing with a symptom history that can extend from months to years. Although easily defined by MRI (see Fig. 52-4), LGGs often do not enhance with gadolinium. Their course is usually relatively stable for several years before eventually progressing. At time of diagnosis, LGGs have a much better prognosis than GBM. However, eventually LGGs progress to become glioblastomas with their inherent poor prognosis. Histologically, low-grade gliomas are classified as astrocytomas, oligodendrogliomas, or oligoastrocytomas (mixed glioma). A low mitotic index, younger patient age, and a supratentorial nonelegant locus (i.e., not affecting language function) that is amenable to resection predict a longer progression-free survival.

Treatment and Prognosis

The choice of therapeutic modalities is always an issue. Retrospective studies suggest that gross total resection, for gliomas that can be safely removed, provides longer progression-free survival. However, the surgeon can never remove all tumor tissue when dealing with infiltrative gliomas. These lesions always harbor an innate, almost serpiginous invasion of what appears to be grossly normal brain tissue to the surgeon’s eye. At the time of resection, these characteristics prevent appreciation of the full microscopic extent of the entire tumor mass. Therefore, gliomas eventually will demonstrate progression even after what appears initially to be a gross “total resection.” In this setting, so-called disabling resections in patients with astrocytomas or oligodendrogliomas are neither wise nor helpful. This is especially true when dealing with tumors in eloquent cerebral cortical areas, including language and memory, function, as well as those portions essential to use of extremities, particularly motor structures within the dominant hemisphere where preservation of functional mobility is particularly important.

Subtotal resection is indicated in most gliomas remediable to decompression without leaving a significant disability (such as aphasia) and especially when the tumor’s mass effect is causing disability. In patients with pilocytic astrocytoma, surgical indications differ slightly; a complete resection may provide a cure, and a more aggressive surgical approach is often indicated.

The next therapeutic decision is whether to recommend external beam RT. Although RT does not prolong overall survival, there is a significant increase in progression-free survival in the treated group. Unfortunately, this benefit may be offset by a higher incidence of long-term cognitive impairment in the RT-treated group. Survival is not the only factor when considering RT. There are some clinical predictors suggesting which patients will benefit from RT. If more than two answers to the five questions listed below are yes, the patient is likely to benefit from RT: (1) Is the patient older than age 40 years? (2) Is the tumor symptomatic (other than seizures)? (3) Does the tumor cross the midline? (4) Is the tumor an astrocytoma (as opposed to an oligodendroglioma)? (5) Is the tumor larger than 5 cm?

The dose of RT for LG is usually 54 Gy given in 30 fractions. Higher doses resulted have not shown a clear benefit and should not be used.

Until recently chemotherapy has not been employed for treatment of LGG. However the recent widespread use of temozolomide, an oral alkylating agent for GBM, raises the question of whether there are a selected group of patients with LGG who potentially may also benefit from this therapy. Temozolomide is currently used in patients who do not meet criteria for RT, as listed above, but whose tumor has a mitotic index of greater than 3%. Although patients with low-grade tumors have a much better prognosis than those with anaplastic glioma and GBM, low-grade gliomas are still usually fatal. The median survival is 5–7 years for astrocytoma and 7–10 years for oligodendroglioma.

Anaplastic Glioma

Clinical Vignette

A 49-year-old right-handed man presented with 5 weeks of numbness and weakness in the left leg. Examination revealed decreased strength and slowed rapid movements of his left foot. There was extinction to touch and loss of joint position sense. Brain MRI revealed a 3 × 4-cm cystic mass centered in the medial aspect of the right parietal lobe and heterogeneous enhancement with gadolinium (Fig. 52-5).

Figure 52-5 Glioma, Anaplastic Astrocytoma.

Lesion resection revealed an anaplastic astrocytoma (Fig. 52-5). Postoperatively, he was treated with a combination of RT and concomitant temozolomide. He was treated monthly for 1 year and is now receiving the same dose at 8-week intervals. He is neurologically intact and radiographically stable 3 years after diagnosis.