Clinical manifestations

Presentation and course

Oligodendrogliomas are often slow-growing tumors and may not be clinically evident for several years. Most oligodendrogliomas arise in adults between the fourth and sixth decade of life, though on occasion are diagnosed in adolescents and patients older than 65 years (36). Pediatric cases are rare and frequently lack IDH mutation and 1p/19q codeletion, suggesting genetic and biological distinction from the adult form (42). These tumors would no longer fit into the contemporary World Health Organization classification for oligodendrogliomas.

Symptoms depend on the size and location of the tumor, though seizures at initial presentation are common. The incidence of seizure at initial presentation is much higher in these IDH-mutated tumors when compared to the IDH wild-type astrocytomas (11). Headaches and focal neurologic deficits, such as limb weakness, can also occur. Rarely, these tumors are detected incidentally after neuroimaging for unrelated reasons, such as head trauma (58). Anaplastic oligodendrogliomas can present de novo or secondary to transformation of the low-grade form and may present with focal neurologic impairment, cognitive change, or signs of increased intracranial pressure or seizures.

Although the prognosis of oligodendroglioma is more favorable than its astrocytic counterpart, no cure exists. The pattern of recurrence is typically local, though leptomeningeal dissemination can occur in later stages of the disease. Metastasis outside the central nervous system is very rare (60; 55). The cause of death is usually attributed to local tumor progression, and it is uncertain whether oligodendroglioma exhibits the same disseminated pattern of spread described in glioblastoma (14).

Prognosis and complications

Although the clinical course in oligodendrogliomas tends to be more protracted than with astrocytomas, nearly all patients eventually succumb to the disease. The median overall survival rate for low-grade oligodendroglioma ranges from 10 to 15 years, and for anaplastic oligodendroglioma the median overall survival ranges from 5 to 9 years. Features associated with worse outcome in low-grade gliomas are neurologic deficits at baseline, shorter duration of symptoms prior to presentation, astrocytic histology, and tumor size larger than 5 cm. For anaplastic oligodendroglioma, recursive partitioning analysis identified 5 prognostic classes in which the group with the most favorable outcome was defined by younger age, tumor located in the frontal lobe, and 1p/19q codeletion (19; 38). It should be noted that the survival numbers are based on diagnoses that did not integrate molecular features into histopathologic findings and likely underestimate the prognosis of IDH-mutant, 1p/19q codeleted oligodendroglial tumors.

Leptomeningeal spread is a rare complication that occurs late in the disease course, and successful treatment with radiation or chemotherapy has been reported (43; 35).

Overall, children and adolescents with oligodendroglial tumors tend to survive longer than adults, though no significant difference in outcome was detected between adults with high-grade tumors and children with high-grade tumors. Factors associated with worse prognosis in the pediatric population include headaches at presentation, subtotal resection, tumor location in the parietal lobe, and large tumor size (18; 51). However, it should be kept in mind that many tumors previously categorized as pediatric oligodendrogliomas would not qualify based on the contemporary WHO classification system.

Clinical vignette

A 33-year-old woman initially presented in 2006 with vague disequilibrium and a feeling of left ear fullness. Symptoms did not improve after left myringotomy, and the following year she underwent a brain MRI. This was notable for a nonenhancing, infiltrating mass in the left frontal lobe characteristic of a low-grade glioma. Close observation with serial imaging was advised. In July 2011 her scan was notable for interval growth over the prior year, prompting gross total resection. Pathology was consistent with a WHO grade 2 oligodendroglioma, and she resumed observation with repeat MRIs at 3-month intervals. Neuroimaging in February 2013 indicated regrowth of tumor, now associated with some central enhancement, and she underwent a second resection. Pathology was again interpreted as WHO grade 2 oligodendroglioma, though mitotic proliferation index was elevated at 13.5%. She then received adjuvant radiation to the surgical site (50.4 Gy in 28 fractions).

Radiographic recurrence was again noted 2 years later, and pathology from her third surgery was now characteristic of anaplastic oligodendroglioma. She underwent fractionated stereotactic radiation (35 Gy in 10 fractions) with concurrent temozolomide, followed by a year of adjuvant temozolomide. Evident on her surveillance MRI from September 2017 was a small focus of enhancement in the left frontal lobe that was not associated with increased blood volume on perfusion imaging or neurologic symptoms. Short-interval imaging was concerning for ongoing enlargement and interval development of hyperperfusion characteristic of recurrence. She commenced treatment with a combination of procarbazine, lomustine, and vincristine (PCV), which was discontinued after 3 cycles due to chemotherapy-induced peripheral neuropathy and hematotoxicity. As of March 2019, her scans showed stable disease, though she was lost to follow-up shortly afterwards.

The slow radiographic progression followed by frequent recurrence after anaplastic transformation 6 years later is typical. The new focus of enhancement in the original tumor location may represent treatment-related necrosis, especially in the context of decreased perfusion imaging, though tumor recurrence could not be completely ruled out. Close follow-up is essential to ensuring timely treatment and preservation of function.

Our knowledge base regarding oligodendrogliomas has evolved as the results of large prospective trials have matured. It is possible/probable that the clinical management of this case may be different from what is described if the patient were diagnosed today.

Biological basis

Etiology and pathogenesis

As with other cancers, the complex chain of events prompting tumorigenesis of glial cells can be attributed to defects at the genetic and molecular levels that result in aberrant regulation of cell survival and growth (21). Some of these defects provide diagnostic, prognostic, and predictive information.

Genetics. Familial oligodendroglioma and oligodendroglial tumors associated with hereditary cancer predisposition syndromes are very rare. Most oligodendrogliomas arise sporadically, and a number of influential somatic mutations have been identified. In early studies, molecular genetic analysis revealed that oligodendroglial tumors could be characterized by the deletion of chromosome 1 and chromosome 19. Further investigation showed that the 1p/19q codeletion represents a clinically relevant unbalanced translocation, resulting in combined whole-arm losses of chromosomes 1p and 19q, which were associated with a significantly longer overall survival (25). In contrast, partial 1p deletions acquired through other mechanisms in astrocytic tumors are associated with worse prognosis, warranting caution in relying on fluorescent in-situ hybridization (FISH) to detect total 1p/19q loss due to the risk of false-positive results from partial 1p and 19q losses (23).

1p/19q loss invariably occurs with missense mutations of IDH1 codon 132 or IDH2 codon 172, and it is presumed that the development of IDH mutation precedes chromosomal loss. IDH mutation leads to an accumulation of 2-hydroxyglutarate (2-HG), which has the downstream effects of inhibiting histone demethylation, subsequently impeding cell differentiation and leading to glioma formation. 2-HG also leads to epigenetic alterations and hypermethylation of CpG islands, including the MGMT promoter region that is a key site in DNA repair (56; 32; 48). Further investigation with whole-exome sequencing of oligodendrogliomas has identified other genetic mutations, such as those involving CIC, FUBP1, TCF12, NOTCH1, PIK3CA, and TERT (05; 09; 29; 46). CIC, which is located on chromosome 19q13, is speculated to play a role in tumor suppression by encoding a protein (cic), which blocks transcription by binding to regulatory regions. FUBP1, found on chromosome 1p, is also thought to be involved in tumor suppression though the exact mechanism is unclear. Another avenue of tumorigenesis includes increased telomerase activity secondary to point mutations in the TERT gene promoter, allowing infinite cell proliferation and tumorigenesis. Increased telomerase expression is also seen in astrocytomas, though via pathways involving the ATRX gene rather than TERT (05; 57; 40). In oligodendrogliomas that arise in children or teenagers, the molecular features differ. During the teenage years, tumors with classical histological appearance of oligodendroglioma are still IDH mutant and 1p/19q codeleted, though TERT promoter is absent (30). Pediatric tumors with a histologic appearance of oligodendroglioma lack 1p/19q codeletion or IDH mutations and instead demonstrate alterations in the FGFR1 oncogene (59).

Pathology. Macroscopically, oligodendrogliomas are ill-defined soft gray masses within white matter. Macrocalcification is present in 90% of cases. Histologically, oligodendrogliomas have a well-defined nucleus with abundant cytoplasm. In fixed tissue, postprocessing artifact produces a "fried egg" appearance to individual oligodendroglial cells. A fine intercellular matrix is abundant with small capillaries and frequently with areas of microcystic degeneration. The presence of astrocytic foci is not uncommon, and disagreement among reviewing pathologists can occur when relying solely on histologic features. The incorporation of specific molecular and genetic features to the diagnostic criteria improves interobserver variability: thus, a fully-characterized oligodendroglioma is defined by a histopathologic appearance of oligodendroglioma along with IDH-mutation and combined whole-arm loss of chromosomes 1p and 19q.

These molecular genetic changes are retained in WHO grade III (anaplastic) oligodendrogliomas, which exhibit increased cellular density, pleomorphism, microvascular proliferation, and an elevated mitotic rate. Anaplastic oligodendrogliomas have traditionally been associated with a shorter survival rate.

Pathophysiology. A single cell of origin has not been identified, and it is possible that oligodendrogliomas arise from different neuronal cell types, including neural stem cells, glial precursor cells, astrocytes, and oligodendrocyte precursor cells. Oligodendrocyte progenitor cells have been demonstrated to develop into either astrocytic or oligodendroglial tumors in mouse models, depending on the genes driving transformation (31). It is presumed that somatic mutations in IDH1 and IDH2 occur early in diffuse gliomas and affect codons 132 (IDH1) and 172 (IDH2). A subsequent event resulting in the unbalanced translocation of chromosomes 1p and 19q prompts the development of oligodendroglioma, presumably due to loss or alterations in tumor suppressor genes located on these loci. The driver mutations associated with anaplastic transformation have not been fully elucidated either, though deletion of chromosome 9p (resulting in loss of the CDKN2A locus) correlated with genomic instability, high vascular endothelial growth factor (VEGF) expression, and poor survival (41; 34).

Epidemiology

According to the Central Brain Tumor Registry of the United States (CBTRUS), oligodendrogliomas are quite rare, constituting approximately 5% of all central nervous system tumors of neuroepithelial origin. The average age-adjusted annual incidence rate of oligodendroglioma is 0.48 for adults 40 years or older. Oligodendroglioma is rarer in children: the annual average age-adjusted incidence rate for patients younger than 15 years of age is 0.03% compared to 0.27% for patients aged 15 to 39 years. No patients younger than 15 years with anaplastic oligodendroglioma were documented in 2013 to 2017 (36).

Prevention

Risk factors for glioma include inheritable genetic syndromes (such as Li-Fraumini, neurofibromatosis type 1, and Maffucci syndrome) and exposure to high-dose ionizing radiation, though no risk factors specific for oligodendroglioma are known; thus, no known preventative measures have been identified (33; 01).

Differential diagnosis

The clinical and radiographic findings can mimic other brain tumors, such as astrocytoma and dysembryoplastic neuroepithelial tumor. Nonneoplastic entities (eg, infectious or autoimmune encephalitis), demyelination, and vascular aberrations (eg, thrombosed arteriovenous malformations) should also be considered.

Definitive diagnosis can generally be made on histologic examination of sample tissue, though features of some tumors, such as pilocytic astrocytoma, central neurocytoma, dysembryoplastic neuroepithelial tumor, and clear cell ependymoma may resemble oligodendroglioma. Examination for allelic loss of 1p and 19q, and for the presence of an IDH mutation can help distinguish oligodendroglial tumors from dysembryonic neuroepithelial tumors or neurocytomas (17).

Diagnostic workup

After taking a full history and performing thorough physical and neurologic examinations, obtaining neuroimaging with CT or MRI is the next step to establishing a pre-operative diagnosis. A cortical-subcortical lesion located in the frontal or parietal lobe is typical. On CT, an oligodendroglioma may appear as a rounded hypo- or isodense white matter lesion with punctate areas of hyperdensity indicative of calcification.

On MRI, lesions are T1-hypointense and T2-hyperintense with indistinct borders (27). An imaging biomarker that may assist in distinguishing IDH-mutant, non-codeleted astrocytomas from IDH-mutant, codeleted oligodendrogliomas prior to tissue diagnosis is the presence or absence of the T2-FLAIR-mismatch sign. This is a highly specific imaging characteristic in which an IDH-mutant, non-codeleted WHO grade 2 or 3 glioma appears predominantly hypointense on FLAIR and nearly completely hyperintense on T2. Further investigation to elucidate the role of such findings is warranted before incorporating them in pretreatment management, as it is poorly sensitive (39; 20).

Peritumoral edema is uncommon and consistent with the slow-growing nature of the lower-grade tumor (27). Surrounding parenchymal edema is more likely to be seen in tumors that have undergone anaplastic transformation. Furthermore, central heterogeneity from intratumoral hemorrhage or internal cystic formation may be noted. Contrast enhancement, which is associated with malignant transformation in other gliomas, is not reliably applicable in the case of oligodendroglioma; approximately 50% of low-grade oligodendrogliomas exhibit patchy or lacy patterns following contrast administration.

Though neuroimaging may appear typical for oligodendroglioma, histopathological analysis of adequate tissue sampling obtained by stereotactic biopsy or maximal safe resection is still mandatory for establishing a definitive diagnosis. All diffuse gliomas should undergo immunohistochemical staining for IDH1 R132H, the most common mutant form of IDH1. If negative or equivocal, next-generation sequencing is recommended for detection of less common IDH1 and IDH2 mutations. If an IDH mutation is detected, further assessment of ATRX expression and 1p/19q codeletion status is warranted. A fully characterized oligodendroglioma or anaplastic oligodendroglioma is contingent on the presence of IDH mutation and 1p/19q codeletion. This is particularly relevant in the case of infiltrating gliomas that exhibit oligodendroglial and astrocytic morphology, for which a diagnosis of oligoastrocytoma was historically assigned. In general, molecular genetic analysis will favor either an astrocytic or oligodendroglial genetic profile, though a few cases of dual-genotype oligoastrocytoma have been reported (22; 54). If testing is unavailable or inconclusive, the diagnosis includes a designation of “not otherwise specified.”

Routine staging with craniospinal imaging and cerebrospinal analysis is unnecessary unless clinical symptoms are suggestive of spinal cord or nerve root involvement or when the clinical presentation and brain imaging are indistinct from lymphoma or a demyelinating etiology.

Management

The primary goals of treatment are to alleviate symptoms and prolong the time to tumor progression, which eventually occurs in all oligodendrogliomas. Providing optimal care often involves a multidisciplinary approach, requiring input from the neuroradiologist, neuropathologist, neuro-oncologist, neurosurgeon, and radiation oncologist.

Surgery. In addition to obtaining tissue for definitive diagnosis, surgery can also alleviate symptoms from mass effect. It is standard of care to achieve maximal resection provided that it can be done safely, though this approach is controversial in young patients with small, asymptomatic lesions. Furthermore, that extent of resection in oligodendroglioma or anaplastic oligodendroglioma leads to improved survival is debatable. According to an analysis of the Surveillance, Epidemiology, and End Results (SEER) database, gross total resection was not found to be associated with improved survival in oligodendroglioma and was attributed to the exquisite chemosensitive nature of the tumor compared to astrocytomas (02). However, there are strong arguments in favor of extensive resection as well (15; 53; 28). Close follow-up with neuroimaging done at regular intervals is recommended if a watch-and-wait approach is pursued because untreated low-grade gliomas will demonstrate slow growth over a period of time.

Radiation. Because surgery alone is not curative, additional treatment modalities, such as radiation and chemotherapy, need to be considered. The optimal timing of such treatments is undefined, particularly in low-grade gliomas, as the benefits of early radiation must be weighed against the potential for cognitive impairment in the future. A follow-up of long-term survivors with low-grade glioma who received radiation showed a progressive increase in attentional deficits at a mean follow-up of 12 years, in contrast to patients who did not receive radiation and demonstrated stable neuroimaging and cognition and despite there being no detectable difference previously at a mean of 6 years after diagnosis (13). Therefore, it may be appropriate to delay radiation and favor close follow-up with serial neuroimaging in a young patient (40 years of age or younger) who is asymptomatic following complete resection of a fully characterized oligodendroglioma (what has been termed a low-risk, low-grade glioma). In contrast, immediate postoperative therapy is most likely to be recommended to patients who have high-risk features, such as those aged 40 years or older or those who have preoperative tumor size 5 cm or larger, subtotal resection, or high-grade features on tumor histology.

For low-grade gliomas, the delivery of 50.4 Gy in 28 fractions is standard; high-grade gliomas are typically treated with 59.4 Gy in 33 fractions. The ongoing NRG BN005 randomized cooperative group trial is investigating whether photon versus proton radiotherapy is associated with a difference in neurocognitive decline.

Chemotherapy

Low-grade gliomas (WHO grade 2). Long-term results of the Radiation Therapy Oncology Group (RTOG) 9802 phase III randomized trial indicate a survival benefit in patients who receive a combination of radiation and chemotherapy with PCV in patients with high-risk (older than 40 years of age or presence of residual tumor after resection) supratentorial grade II astrocytoma, oligoastrocytoma, or oligodendroglioma compared to radiation alone. The median overall survival was 13.3 years in the combination arm compared to 7.8 years in the radiation-only arm (06). It should be kept in mind that there was significant cross-over in this trial (ie, patients randomized to radiation alone often received chemotherapy at disease progression), supporting the idea that the timing (earlier vs. later) of maximal therapy matters. Post hoc molecular analysis of this trial demonstrated improvements in overall survival in IDH-mutant groups, with significantly better outcome in the IDH-mutant/1p/19q intact codeleted group, in contrast to the IDH-wild type groups (04). Thus, PCV chemotherapy clearly improves outcomes of oligodendroglioma when added to radiation, though treatment-related toxicity, such as myelosuppression, may limit one’s ability to complete the recommended 6 cycles. The efficacy of temozolomide, an oral DNA alkylating agent, has been reported along with a better tolerability profile, though whether it is better, worse, or similar in efficacy to PCV when used upfront remains to be determined. A multicenter prospective trial comparing radiation and PCV with radiation and temozolomide in 1p/19q codeleted low-grade and anaplastic gliomas (CODEL, NCT00887146) is currently underway to answer this question. Analysis of the original CODEL design, in which patients with WHO grade 3 oligodendroglioma were randomized to temozolomide alone, radiotherapy alone, or radiotherapy with concurrent and adjuvant temozolomide, showed that treatment with temozolomide alone was associated with significantly shorter progression-free survival (24). It will take more time for the study to adequately accrue, and several more years after that for the data to mature. Meanwhile, there is growing concern that temozolomide may encourage development of an aggressive phenotype. In one study, genome sequence analysis was performed on primary and recurrent human gliomas treated with temozolomide. Tumors initially treated with temozolomide were more likely to demonstrate hypermutated phenotypes at recurrence, and genetic alterations included driver mutations, which activate the Akt-mammalian target of rapamycin (mTOR) pathway and disrupt the retinoblastoma-associated protein tumor suppressor pathway (RB)—phenotypes also found in glioblastoma (26). Whether this phenomenon is unique to temozolomide is unknown, due to the lack of PCV arm in this study.

Efforts to maximize treatment efficacy while minimizing toxicity have prompted studies exploring the option of deferring postoperative radiation in favor of chemotherapy-only regimens. Results from the EORTC 22033-26033 trial, which assessed the response of patients with high-risk low-grade glioma to either 12 cycles of postoperative dose-intense temozolomide (75 mg/m2 daily x 21 days per each 28-day cycle) or radiation (50.4 Gy), showed no difference in median progression-free survival between the chemotherapy-only and radiation-only arms. This was also observed in the IDH-mutant and 1p/19q codeleted subgroups; however, a subgroup of patients with IDH-mutant, 1p/19q non-codeleted tumors demonstrated longer progression-free survival in patients randomized to the radiation arm compared to the temozolomide arm (03).

High-grade gliomas (WHO grade 3). The survival advantage with postoperative radiation and PCV chemotherapy is also observed in anaplastic oligodendroglioma. Two large randomized trials comparing radiation and PCV chemotherapy to radiation only helped clarify this. EORTC 26951 utilized RT followed by 6 cycles of PCV. RTOG 9402 utilized 4 cycles of PCV (with a higher CCNU dose) followed by radiation therapy. Both EORTC 26951 and RTOG 9402 initially reported a significant delay in progression with combined-modality therapy with no overall survival benefit. However, long-term follow-up results of both indicate significant prolongation of both progression-free and overall survival after radiation plus PCV. Furthermore, the benefit was more pronounced in patients with 1p/19q codeleted tumors (07; 50). Thus, postoperative treatment with radiation and chemotherapy (with either PCV or temozolomide) has become the standard of care for anaplastic gliomas.

The NOA-04 trial evaluated the efficacy of chemotherapy alone (either PCV or temozolomide) against radiation alone in patients with anaplastic astrocytomas, anaplastic oligoastrocytomas, and oligodendrogliomas. Unlike EORTC 22033-26033, no difference between either single-modality intervention was demonstrated in any subgroup in this study (52).

Another chemotherapy-only approach using induction chemotherapy with PCV or temozolomide followed by myeloablative chemotherapy (HDC) with autologous stem cell transplant (ASCT) in newly diagnosed anaplastic oligodendroglioma can be safely administered; progression-free survival and overall survival rates are promising although ongoing follow-up is requisite to fully characterize treatment efficacy (47).

At this time, there is sufficient evidence to support the combination of radiation and chemotherapy in glioma patients.

Recurrent oligodendroglioma. Salvage treatment with surgery, radiation, chemotherapy, or some combination thereof is often considered at the time of tumor relapse or malignant transformation. Surgical intervention to address a symptomatic focus or confirm disease progression is justifiable. Re-irradiation may be given in selected patients, but in general the risk of neurotoxicity and radiation necrosis increases with repeat conventional external beam radiotherapy unless special care is given to limit the treatment volume and minimize beam exposure to the normal surrounding structures. Stereotactic radiotherapy can be used to improve conformal delivery.

The use of temozolomide is safe and effective for oligodendrogliomas that have recurred after radiation and PCV chemotherapy and in those that have undergone malignant transformation (12). Paclitaxel, irinotecan, carboplatin, and etoposide plus cisplatin are other agents that have demonstrated some activity in recurrent anaplastic oligodendrogliomas, though response rates are low and durability is less than ideal. The benefit of adding bevacizumab, a monoclonal antibody against vascular endothelial growth factor (VEGF), to salvage chemotherapy is not well established.

Outcomes

Randomized clinical trials assessing the extent of resection are lacking though observational studies suggest an association between improved survival and more extensive resection. A prospective series of 111 patients who underwent gross total resection for low-grade glioma demonstrated a 5-year overall survival rate of 93%, though 5-year progression-free survival was just under 50% (45). Early postoperative radiation improves progression-free but not overall survival, based on the EORTC 22845 trial, a large randomized trial of early versus delayed radiotherapy for patients with low-grade glioma (49). Results indicated a progression-free survival advantage with early radiation, although no improvement in overall survival was detected. The long-term results of RTOG 9802 strongly support the administration of postoperative radiation followed by adjuvant PCV chemotherapy. Although most patients receiving sequential therapy for anaplastic oligodendroglioma or oligoastrocytoma undergo radiation followed by chemotherapy, a retrospective analysis of the National Cancer Database suggests no difference in survival if chemotherapy precedes radiation (44).

Late effects of treatment, including delayed compromise of cognition and quality of life, are particularly important in tumors with prolonged survival, such as oligodendroglioma and anaplastic oligodendroglioma. An analysis of neurocognitive and structural brain changes on MRI was performed on 48 long-term survivors with an oligodendroglial tumor and previously received radiation with or without chemotherapy. Cognitive deterioration, particularly in the realms of visual memory and executive functioning, was reported as early as 2 years following treatment. Structural changes such as leukoencephalopathy and cortical atrophy lagged behind the cognitive change and were more significant in patients more than 5 years out from treatment (10). These findings emphasize the need to anticipate increased social support for long-term survivors and improve future treatment approaches to minimize cognitive side effects while optimizing tumor control.

Special considerations

Pregnancy

A cohort of 36 women with gliomas who gave birth following tumor surgery were followed, and their clinical courses were compared to that of 18 nulliparous women. Patients were matched for tumor diagnosis (which included WHO grade 2 or 3 astrocytoma, oligodendroglioma, and ependymoma), including molecular markers, extent of resection, and tumor location (16). Neither pregnancy nor delivery seemed to impact disease progression or overall survival.

Results from another small study of 12 pregnancies in 11 women with grade II gliomas suggests that radiological “velocity of diametric expansion” increased during pregnancy, compared with pre-pregnancy and post-pregnancy growth rates (37). However, there were no matched controls. If an operation is needed during pregnancy, the risks to the fetus are those of any intracranial operation requiring general anesthesia.

Anesthesia

Precautions in the use of anesthesia are the same as for any intracranial surgery.