Oligodendroglial tumors constitute between 5 and 20 percent of all glial tumors. The frequency of these lesions is higher in contemporary reports due primarily to modifications in subjective histopathologic criteria. Oligodendroglial tumors typically arise in the fourth to sixth decades, with low-grade ODs and OAs occurring at an earlier age than anaplastic lesions. Despite the prolonged clinical course seen with oligodendroglial tumors, the outcome is almost always fatal.

The clinical manifestations, pathology, and treatment of oligodendroglial tumors will be reviewed here. The approaches to patients with other low-and high-grade gliomas are discussed separately.

CLINICAL MANIFESTATIONS

Clinical features — The presenting signs and symptoms of oligodendroglial tumors are nonspecific and depend upon the location and extent of the tumor. Symptoms may be either generalized (eg, headaches, seizures, cognitive dysfunction) or focal (eg, weakness, sensory abnormalities, aphasia). There are no clinical features which distinguish these tumors from other gliomas. (See "Clinical presentation and diagnosis of brain tumors").

Most oligodendroglial tumors arise in the white matter of the cerebral hemispheres, predominantly in the frontal lobes. However, they can arise throughout the central nervous system, including infratentorial sites and the spinal cord. As with other glial tumors, ODs and OAs only rarely metastasize outside the central nervous system.

The majority of ODs and some OAs are characterized by a 1p/19q codeletion. The presence of this chromosomal abnormality defines a specific subset of patients with oligodendroglial tumor with important ramifications for patient management.

Radiographic features — The signs and symptoms that suggest a brain tumor can also be present in nonneoplastic conditions. Neuroradiologic imaging (magnetic resonance imaging [MRI] and/or computed tomography [CT]), generally confirms the diagnosis of a brain tumor and provides critical information needed for surgical planning. (See "Clinical presentation and diagnosis of brain tumors").

A number of features on neuroimaging can suggest a diagnosis of an oligodendroglial tumor or provide information about the possible status of chromosomes 1p and 19q. These include:

Low-grade oligodendroglial tumors — On MRI, typical low-grade tumors show increased signal intensity on T2-weighted images without enhancement [3]. On CT, these tumors appear as low-density masses without enhancement. Calcifications are suggestive, but not specific, for ODs and OAs, and may be more frequent in tumors with 1p/19q codeletion [4].
Anaplastic oligodendroglial tumors — On MRI and CT, most anaplastic ODs and OAs are characterized by enhancement, which presumably reflects microvascular proliferation. However, the absence of contrast enhancement does not exclude an anaplastic tumor and some contrast enhancement can be observed in low-grade tumors [4].
1p/19q codeleted tumors — Findings on MRI may suggest the presence of 1p/19q codeletion in oligodendroglial tumors [4]. Contrast enhancement of 1p/19q co-deleted tumors is often patchy and homogeneous, as opposed to the ring-like enhancement with necrosis typically seen in high-grade tumors without 1p/19q deletion. In addition, tumors with 1p/19q loss often have indistinct borders and a mixed signal intensity on T1- and T2-weighted images, whereas tumors without 1p/19q codeletion typically have a distinct border and a uniform signal on T1- and T2-weighted images [4,5].
None of these imaging findings are specific, and histopathologic examination is mandatory. Sampling error, particularly in patients who undergo only a biopsy, can lead to misclassification and undergrading of a brain tumor. The neuroimaging characteristics of a lesion may guide the surgeon in choosing a site for biopsy, and the presence of enhancement should be considered suggestive of a high-grade tumor.

PATHOLOGY AND BIOLOGY

Classification — Oligodendroglioma (OD) is defined by the World Health Organization (WHO) as a well-differentiated, diffusely infiltrating tumor of adults, that typically is located in the cerebral hemispheres and is composed predominantly of cells resembling oligodendroglia [6]. An OD with focal or diffuse histologic features of malignancy and a less favorable prognosis is defined as an anaplastic OD. Classical OD morphology is highly associated with the 1p/19q codeletion, although these chromosomal abnormalities are not always present.

The natural history of ODs is a gradual progression from low-grade (well-differentiated) tumors into high-grade lesions with anaplastic features (high cell density, mitosis, nuclear atypia, microvascular proliferation, and necrosis). Since these morphologic changes appear gradually within a tumor, a distinction between OD (low-grade) and anaplastic OD (high-grade) is not always possible. Some patients present with an anaplastic OD without evidence of a prior low-grade lesion.

Tumors containing elements with characteristics of both ODs and astrocytomas are classified as mixed OAs. There are no widely accepted histologic criteria defining the percentage of oligodendroglial elements that need to be present in a predominantly astrocytic lesion before a tumor is classified as an OA [7]. European Organization for Research and Treatment of Cancer (EORTC) and Radiation Therapy Oncology Group (RTOG) trials have used the presence of more than 25 percent oligodendroglial elements as an arbitrary threshhold, but even this criterion is subject to interobserver variability [8,9].

Whether mixed anaplastic OA with necrosis is a distinct entity from glioblastoma (GBM) with oligodendroglial morphology is controversial [10]. These anaplastic mixed OAs usually behave clinically like a GBM. Most GBMs with oligodendroglial features and anaplastic OAs with necrosis have genetic lesions associated with high grade astrocytic tumors (eg, EGFR amplification, 10q loss) and do not have the 1p/19q codeletion. (See "Pathogenesis and biology of malignant gliomas", section on Molecular Biology).

The recognition of the chemosensitivity of oligodendroglial tumors emphasized the importance of diagnosing oligodendroglioma. Thus, the histologic criteria for oligodendroglial tumors were broadened since mixed OAs also tended to be sensitive to chemotherapy. The revised criteria emphasized features thought to be associated with an oligodendroglial origin, such as the presence of microgemistocytes, gliofibrillary oligodendrocytes, and protoplasmic astrocytes [11]. These changes in criteria for OAs led to an increase in the incidence of oligodendroglial tumors from about 5 percent of all glial tumors to about 20 percent. Clearly, distinguishing oligodendroglial tumors from astrocytomas is subject to interobserver variability.

Codeletion of 1p/19q — The characteristic chromosomal abnormalities in oligodendroglial tumors are deletions of the short arm of chromosome 1 (1p) and the long arm of chromosome 19 (19q). This abnormality arises from an unbalanced translocation of the short arm of chromosome 19 (19p) to the long arm of chromosome 1 (1q), after which the derivative chromosome with the short arm of 1 and the long arm of 19 is lost [12,13]. In tumors in which this abnormality is identified, the deletions of 1p and 19q persist as the tumor progresses, suggesting that this translocation is an early event in tumorigenesis [14]. Additional chromosomal abnormalities are frequent in anaplastic oligodendroglial tumors [14,15].

Codeletions of chromosomes 1p and 19q have been reported in 60 to 70 percent of classical anaplastic ODs [10,16]. The presence of astrocytic elements decreases the likelihood of a 1p/19q codeletion; in one report, only 39 percent of those with mixed anaplastic OAs and a predominance of oligodendroglial elements had the characteristic chromosomal deletion [14]. In contrast, these deletions are present in only 14 to 20 percent of those with anaplastic OAs without a predominance of oligodendroglial elements [10,16].

Additional evidence supporting a different origin for tumors with 1p/19q deletions comes from the anatomic distribution of these tumors. Tumors originating in the frontal, occipital and parietal lobe are more likely to possess these characteristic abnormalities, compared to tumors in the insular region, temporal lobes and diencephalon [16-18].

Routine testing for 1p/19q deletions is feasible on paraffin-embedded specimens and is recommended in all patients with ODs and OAs. The characteristic lesion involves loss of the entire short arm of chromosome 1, and not just the partial chromosome 1p deletion that is often observed in high grade astrocytic tumors. These partial deletions of 1p are not associated with 19q loss and are associated with a poor prognosis [19]. In anaplastic OAs or atypical anaplastic ODs, assessment of EGFR amplification and loss of 10 or 10q may identify tumors with poor prognosis and suggest a nonoligodendroglial origin of the tumor [14,20].

Prognostic factors

Histopathologic correlations — The presence of a major oligodendroglial component is associated with favorable effect on survival compared to pure astrocytic tumors. As an example, the median survival in a retrospective series of 167 patients was significantly longer in those with low-grade ODs or OAs (13 versus 7.5 years with low-grade astrocytoma). The prognosis of mixed OAs is intermediate between pure ODs and low-grade astrocytomas, possibly reflecting the frequency of the 1p/19q codeletion in these groups.

Patients with anaplastic ODs have a substantially worse prognosis than those with low-grade ODs. As an example, the median survival in one report was shorter with anaplastic ODs (4.5 versus 9.8 years with low-grade ODs). The presence of necrosis and endothelial proliferation are poor prognostic factors, although this observation may be true only for anaplastic OAs and not for anaplastic ODs

Codeletion of 1p/19q — The presence of a combined loss of chromosomes 1p and 19q is the most important factor associated with improved survival in patients with oligodendroglial tumors . This observation may reflect both the natural history of the disease and the better response to therapy  Some data suggest that the favorable prognostic implications of chromosomal 1p/19q deletions are restricted to tumors with classical OD morphology, but those observations require confirmation.

The impact of the 1p/19q codeletion was illustrated in a study in which the median survival was prolonged in patients with low-grade oligodendroglial tumors who had the characteristic abnormality (11.9 versus 10.3 years for the patients without the deletion). In a prospective trial of patients with anaplastic ODs and OAs, the presence of the 1p/19q codeletion was an even more important factor (median survival >7 versus 2.8 years in those without this abnormality).

In both low-grade and high-grade OAs, other molecular abnormalities are frequently identified. Abnormalities typically associated with astrocytoma (eg, TP53 mutations, EGFR amplification, 10q loss, PTEN mutations) are usually not present in tumors with the 1p/19q codeletion
These observations suggest that anaplastic OAs represent two somewhat different tumor types that arise from either an oligodendroglial or astrocytic lineage. This difference is reflected in the prognosis. Patients whose tumors are characterized by deletion of 1p/19q have a substantially better prognosis, while those whose tumors have loss of 10q and/or the amplification of EFGR do substantially worse

MGMT overexpression — The responsiveness of anaplastic ODs to chemotherapy was initially observed in studies using the procarbazine, lomustine, plus vincristine (PCV) chemotherapy regimen in patients with recurrent tumors. Subsequent studies have shown a high response rate with single agent temozolomide as well. The mechanism underlying the chemosensitivity of these tumors is not well understood.

The nuclear enzyme methylguanine methyl transferase (MGMT) is responsible for DNA repair following alkylating agent chemotherapy and may mediate part of the cellular resistance to alkylating agents. MGMT expression can be silenced by methylation of its promoter [33]. Expression of MGMT in oligodendroglial tumors is lower than in other glial tumors, and its underexpression may be more pronounced in tumors with codeletion of 1p/19q [34].

Methylation of the MGMT promoter has been reported in 47 percent of ODs [33]. Although this study did not identify a correlation between promoter methylation and codeletion of 1p/19q, other studies have found such a correlation [33,35-37]. Additional studies are required to understand the role of MGMT in the chemosensitivity of oligodendroglial tumors.

TREATMENT — Historically, the management of patients with oligodendroglial tumors was based upon results from studies in patients with gliomas, carried out prior to the recognition of the differences between oligodendroglial and other glial tumors. As discussed below, surgery remains the initial treatment modality for most patients. Additional treatment (RT and/or chemotherapy) is an important component of the initial management for patients whose tumors are not completely resected and for those with anaplastic lesions.

At least two contemporary trials focused specifically on patients with oligodendroglial tumors . In the future, trials in patients with ODs and OAs will need to incorporate information on the status of chromosomes 1p and 19q to facilitate optimal patient treatment.

Surgery — Surgery provides tissue to establish the diagnosis and is used to relieve symptoms due to mass effect in patients with oligodendroglial tumors. In addition, gross total resection of the tumor is generally recommended if possible as part of the initial treatment in an effort to improve the long-term prognosis.

There are no randomized trials that have established the benefit of maximal surgical resection, and such studies are unlikely to be conducted. Evidence supporting this approach in oligodendroglial tumors comes from secondary analyses of two large trials, which demonstrated a positive association between a more extensive resection and prolonged survival. However, in such retrospective studies, small, superficial tumors with a relatively favorable prognosis are more likely to have been extensively resected, while large, deep-seated, or midline tumors with a poorer prognosis will not have been completely resected

Following surgery, further adjuvant treatment (radiotherapy and/or chemotherapy) is recommended for patients with focal deficits, residual lesions with mass effect, or anaplastic tumors.

A "wait and see" approach following initial surgery may be followed in young patients with a favorable prognosis whose symptoms are limited to seizures and who have undergone an extensive resection for a low-grade tumor, especially if molecular studies show the presence of a 1p/19q codeletion.

One possible exception to the initial use of surgical resection is for patients with small, presumably low-grade, tumors, whose only symptoms are seizures that can be managed medically. In this setting some advocate delaying treatment until there is evidence of progression.

Radiation therapy — Randomized trials that included high-grade gliomas of all types demonstrated that adjuvant RT provides a significant survival benefit.

The effectiveness of RT specifically for patients with oligodendroglial tumors has not been assessed in randomized trials. Although retrospective studies have given conflicting results about whether postoperative RT prolongs survival  RT is considered an integral component of the treatment for patients with anaplastic lesions or a large residual tumor.

The optimal timing, dose, and schedule are not fully established, but the approach is influenced by the histology and extent of disease:

Anaplastic ODs and OAs — For patients with anaplastic ODs and OAs, doses of 60 to 65 Gy in 30 to 35 fractions are recommended based upon extrapolation from studies in patients with high-grade gliomas. However, the optimal integration of RT with chemotherapy remains uncertain and is an area of active study.

Low-grade ODs and OAs
- Unresectable, or incompletely resected tumors and those that are associated with focal deficits or enhancing lesions require additional treatment. If RT is used, a cumulative dose of 50 to 54 Gy in fractions of 1.8 Gy is recommended. Higher cumulative doses do not improve outcome and may increase toxicity

- For younger patients who have undergone a complete or almost complete resection for low-grade ODs and OAs, the optimal approach is unclear. Rather than treat these immediately, it may be reasonable to withhold additional treatment until there is evidence of progression.

- As with patients with anaplastic ODs and OAs, the optimal integration of RT with chemotherapy for patients with low-grade ODs and OAs remains an area of uncertainty.

Chemotherapy — Oligodendroglial tumors are markedly more sensitive to chemotherapy than nonoligodendroglial tumors. This chemosensitivity was initially demonstrated for the PCV regimen (procarbazine, lomustine, and vincristine) in patients who had recurred or progressed after RT. Subsequent studies showed that temozolomide was also highly active in this setting. Both PCV and temozolomide are active when administered prior to RT.

Temozolomide is generally preferred over the PCV regimen, based upon its ease of administration and better patient tolerance. However, there are no randomized comparisons between PCV and temozolomide in patients with oligodendrogliomal tumors.

The timing of chemotherapy and its integration with RT are areas of active study.

PCV regimen — Multiple retrospective studies demonstrated the activity of PCV in patients with oligodendroglial tumors [1,46-50]. This regimen is active in patients with both low-grade and anaplastic lesions.

The response rates and duration of effect in patients with oligodendroglial tumors are illustrated by three retrospective studies:

In a series of 52 patients with recurrent oligodendroglial tumors, there were 33 objective responses (63 percent), including nine complete responses (CRs) (17 percent overall). The median time to progression (TTP) was 10 months.
In a series of 37 patients, objective responses were observed in 22 (59 percent), and one-half of these were CRs. The median TTP was 12 months, and was 30 months in patients who achieved a CR. Activity was higher in patients with anaplastic ODs (response rate 77 versus 23 percent with anaplastic OAs and median TTP 19 versus 6 months).

The PCV regimen has also shown activity in patients with low-grade ODs and OAs. As an example, objective responses were observed in 16 of 26 patients (62 percent) in one series, including three with CRs [48]. The median TTP was longer significantly longer for patients with ODs (32 versus 12 months for those with OAs).
The PCV regimen is limited by cumulative hematologic and gastrointestinal toxicity and most patients do not tolerate a full course of six cycles. More intensive regimens of these agents are more toxic and have not demonstrated an improved outcome.

The responsiveness to PCV appears to be greater in patients with combined loss of chromosomes 1p/19q. The lower frequency of 1p/19q codeletion may contribute to the lower response rates observed in patients with OAs.

Temozolomide — The spectrum of activity of single agent temozolomide in chemotherapy-naive patients with recurrent disease following surgery and RT is illustrated by two prospective series:

In one series of 67 patients with anaplastic ODs or OAs, 14 partial and 17 complete responses were observed for an overall response rate of 46 percent. The outcomes were significantly better in patients with codeletion of chromosome 1p/19q (response rate 59 versus 34 percent in those without the deletion and progression-free survival 22 versus 8 months). A significantly higher response rate was observed in patients with anaplastic ODs (62 versus 25 percent in those with anaplastic OAs), although the difference in progression-free survival (PFS) was not significant (13 versus 10 months).
In a phase II study conducted by the EORTC, 38 patients with recurrent or progressive disease after RT were treated with temozolomide, including 24 with ODs and 14 with OAs [51]. All patients had measurable, contrast-enhancing lesions at least 1 cm in diameter. Objective responses were seen in 20 patients (53 percent), including 10 with CRs. Nine of the 10 CRs were seen in patients with ODs. The median TTP was 10 months.
In both studies, treatment with temozolomide was well-tolerated. The primary toxicity was hematologic.

Adjuvant chemotherapy — The role of chemotherapy in addition to adjuvant RT in the initial management of patients with anaplastic oligodendroglial tumors was addressed in two multicenter randomized phase III trials that compared early versus delayed chemotherapy. Despite the sensitivity of oligodendroglial tumors to chemotherapy, these trials did not provide evidence of a survival advantage when adjuvant PCV was added to RT, rather than delayed until there was evidence of disease progression.

RTOG-9402 — In RTOG 9402, 289 patients with anaplastic ODs or mixed anaplastic OAs were randomly assigned to four cycles of intensified PCV followed by RT or to immediate RT without chemotherapy. The cumulative dose of RT on both treatment arms was 60 Gy. Although there was a statistically significant prolongation of progression-free survival in patients treated with PCV group (2.6 versus 1.7 years in those treated with RT only as the initial therapy), the overall median survival was not significantly improved with the initial use of PCV (4.9 versus 4.7 years). As in other studies, patients whose tumors had the 1p/19q codeletion had significantly longer median survival times independent of the assigned treatment arm (>7 versus 2.8 years compared to those without these chromosomal deletions).
EORTC 26951 — A different approach to adjuvant treatment was evaluated in EORTC 26951, in which 368 patients were randomly assigned to immediate RT only or RT followed by six cycles of PCV . The total dose of RT on both treatment arms was 60 Gy. At a median follow-up of 60 months, the increase in median survival was not statistically significant in those receiving PCV (40 versus 31 months without adjuvant chemotherapy), although PFS was significantly prolonged (23 versus 13 months). The 1p/19q deletion was present in 25 percent of patients, and approximately 75 percent of these patients were alive at five years, regardless of treatment assignment.
In both trials, approximately 80 percent of patients who had been randomly assigned to the RT only arm subsequently received PCV when progressive disease was diagnosed. The response to delayed chemotherapy may account for the absence of an increase in overall survival.

The lack of benefit from the immediate use of chemotherapy with RT is in contrast to the observations in patients with glioblastoma, where the administration of temozolomide during and after the initial RT is associated with a survival benefit. Current trials are investigating the role of combined chemoirradiation in anaplastic ODs and OAs, with separate trials for tumors with and without combined 1p/19q loss.

In view of the outcome of these trials, adjuvant or neoadjuvant chemotherapy cannot be recommended.

Upfront chemotherapy — The sensitivity of oligodendroglial tumors to chemotherapy led to the evaluation of chemotherapy rather than RT as the initial treatment after surgery The goal of this approach is to minimize any late neurotoxicity caused by the RT. (See "Complications of cranial irradiation").

The feasibility of this approach is illustrated by three uncontrolled studies:

A nonrandomized study that included 20 patients with anaplastic ODs found that upfront temozolomide chemotherapy was active in patients with deletion of 1p, although not in those without this chromosomal abnormality. All seven patients with chromosome 1p loss were progression-free at 24 months, while the median time to progression was eight months in those without loss of 1p.  A more aggressive chemotherapy approach with RT was used in a phase II study in 69 patients with newly diagnosed anaplastic or aggressive ODs. Patients were initially treated with an intensive PCV regimen. High-dose thiotepa with autologous stem cell rescue was then administered to the 39 patients (57 percent) who responded to the initial chemotherapy [55]. Among the subset receiving high-dose thiotepa, the median PFS was 78 months and 21 patients (54 percent) had not relapsed at a median follow-up of 80 months [53]. Although this study showed that prolonged PFS is possible without initial RT, definitive conclusions are not possible without a randomized trial.

Upfront chemotherapy without RT has also been studied in patients with low-grade oligodendroglial tumors. As an example, in one series of 60 patients treated with temozolomide following surgical resection, 17 percent of patients had a partial response, 14 percent had a minor response, and an additional 61 percent had stable disease. The median time to maximal response was 12 months. Loss of chromosome 1p correlated with radiographic response. A similar level of responsiveness was observed in 16 previously untreated patients given PCV.
Although these results are promising, the clinical benefits of upfront chemotherapy have not been defined in randomized trials. Although this approach avoids the early use of RT, the concerns about neurotoxicity from RT are based upon older studies using whole brain RT or relatively large RT portals, and the adverse effects of focal RT have not been demonstrated in large prospective trials.

The choice of chemotherapy or RT as the initial adjuvant treatment must balance the side-effects of a short period of six weeks of RT treatments against the systemic side-effects of chemotherapy for one year. If upfront chemotherapy with deferred RT is considered, it should be restricted to patients whose tumors contain the chromosomal codeletion 1p/19q. The role of immediate versus deferred RT in patients who have had a favorable response to upfront chemotherapy remains uncertain.

Initial management of low-grade tumors — The optimal management of patients with small, minimally symptomatic low-grade glial tumors remains controversial. Many of these patients will remain stable for a protracted period without treatment, and some physicians recommend deferring treatment until there is clinical or imaging evidence of progression. This may result in a deferred tissue diagnosis, and many of these patients will ultimately be found to have oligodendroglial tumors. The advantages and disadvantages of deferring treatment for patients with presumed low-grade glioma are discussed elsewhere. (See "Management of low-grade glioma", section on Surgery).

For patients with do undergo immediate surgical resection, the decision whether or not to include additional treatment (RT or chemotherapy) depends upon the presence of focal deficits, a residual lesion with mass effect, or the identification of foci containing anaplastic tumor. (See "Surgery" above).

Recurrent disease — Both PCV and temozolomide have activity in patients who have failed an initial chemotherapy regimen, although response rates are lower and the duration of disease control is generally shorter.

Temozolomide — Most trials of second-line chemotherapy have evaluated the activity of temozolomide in patients who received prior PCV, either given as an adjuvant or at first recurrence [58,59]. The activity of temozolomide after failure with PCV was illustrated by a retrospective series of 48 patients with anaplastic ODs and OAs [59]. In this series, 21 patients (44 percent) had an objective response, include eight with a CR. The median PFS was 7 months and the median overall survival was 10 months. Treatment was well-tolerated, and the primary toxicity was thrombocytopenia. In other series on similar patients, an objective response rate of 25 percent was noted.
PCV regimen — Experience with PCV after progression on temozolomide is more limited. Nonetheless, there is evidence that PCV is active in some patients who have progressed after previous treatment with temozolomide. In a retrospective study of 24 patients, second-line PCV induced an objective response in 17 percent of cases, 50 percent were progression-free at six months, and 21 percent were progression-free at 12 months [60].
Other agents — The management of patients who have progressed on both temozolomide and PCV is experimental. Other agents that have shown some activity as second-line chemotherapy in patients with anaplastic oligodendroglial tumors include paclitaxel, irinotecan, carboplatin, and the combination of etoposide plus cisplatin. Response rates with these agents have generally been low (less than 15 percent), and almost all patients' progress in less than 12 months. (See "Experimental treatment approaches for malignant gliomas").

SUMMARY AND RECOMMENDATIONS — Oligodendrogliomas (ODs) and oligoastrocytomas (OAs) have important differences from other glial tumors, with significant ramifications for patient management. Many of these tumors contain a characteristic codeletion of the short arm of chromosome 1 (1p) and the long arm of chromosome 19 (19q), which has been correlated with striking sensitivity to chemotherapy and a more prolonged natural history. Tissue should be tested whenever possible to determine whether codeletion of chromosomes 1p and 19q is present.

Anaplastic ODs and OAs

For patients with a newly-diagnosed anaplastic OD or OA, we recommend maximal surgical resection consistent with preservation of neurologic function (Grade 1C). Although gross total resection is preferred whenever possible, subtotal resection or stereotactic biopsy may be required depending upon the location and extent of the tumor. Following surgery, we recommend adjuvant RT (Grade 1B). Upfront chemotherapy with either temozolomide or the PCV regimen is an alternative in patients with the 1p/19q chromosome deletion, but its activity compared to adjuvant radiation therapy (RT) remains uncertain. Following surgery and adjuvant RT, we suggest deferring chemotherapy until there is evidence of progressive disease (Grade 2B). Although adjuvant chemotherapy either immediately before or after RT prolongs the disease-free interval, it does not improve overall survival.
For patients with recurrent or progressive disease following RT, we recommend chemotherapy (Grade 1B). Although both temozolomide and PCV have similar activity, temozolomide generally is better tolerated and can be administered orally.

Low-grade ODs and OAs

For young patients with transient symptoms and a small tumor not creating a mass effect on adjacent structures, we suggest maximal surgical resection if safely possible (Grade 2C). (See "Initial management of low-grade tumors" above).
- For patients who have undergone surgical resection, we suggest that further treatment with either RT or chemotherapy be deferred until there is evidence of recurrence or progression of symptoms (Grade 2C).
- Careful observation is an alternative to immediate surgery in these patients, with surgery reserved until there is clinical or imaging evidence of progressive disease. (See "Initial management of low-grade tumors" above).

For patients presenting with a large mass, extensive neurologic symptoms, or an age above 40 to 50 years, we recommend immediate surgery to debulk the tumor and establish the diagnosis (Grade 1C).
- In those patients whose tumors contain the chromosome 1p/19q deletions, we suggest RT (Grade 2C). Alternatively, chemotherapy (temozolomide or PCV) for the initial postoperative treatment of residual or progressive disease can be considered in lieu of RT, because of the prolonged natural history of the disease in these patients. RT is then reserved for progression after chemotherapy. (See "Upfront chemotherapy" above).

- For patients whose tumors do not contain the 1p/19q codeletion, we suggest that RT be administered postoperatively (Grade 2C). Chemotherapy should be reserved for patients with progressive symptoms following RT.