Clinical manifestations

Presentation and course

The clinical presentation of patients with primary brain tumors can include overt focal neurologic deficits and new-onset seizures (41; 24). The presentation can also be more subtle and manifest as memory impairment, personality changes, and symptoms of elevated intracranial pressure—progressive headaches, nausea, vomiting, blurred vision, and diplopia. Focal deficits depend on the affected region and result from compression of surrounding structures. Tumors of the brainstem, spinal cord, infratentorium, and eloquent cortices are identified earlier due to overt neurologic deficits, such as sensorimotor or cranial nerve deficits. IDH-mutated astrocytomas most often manifest in the cerebral cortex, especially the frontal lobe (26). Studies have also identified an association between the IDH-mutational status of gliomas and clinical presentation. Patients with IDH-mutated gliomas are more likely to present with seizures (59% to 74%) than IDH-wildtype gliomas (18% to 34%) (12).

Prognosis and complications

Prognosis. IDH-mutated diffuse astrocytomas carry a better prognosis than glioblastoma IDH-wildtype tumors, but they are usually less favorable than oligodendrogliomas. The histological grade of gliomas is less useful for prognosis than their mutational status. As such, many studies pool data between grade 2 (diffuse) and grade 3 (anaplastic) gliomas.

Complications. Complications experienced by a patient depend on the treatment modality employed. Surgical resection aims to maximize tumor excision while avoiding eloquent brain regions. There has been widespread adoption of awake craniotomy, fMRI, and fluorescent dyes to guide surgical resection to maximize tumor excision while avoiding eloquent brain regions. However, there continues to be a risk of damage to regions important for speech, judgment, motor function, etc. Radiation therapy has been associated with adverse effects on cognition (08; 15; 17). Nitrosoureas in specific and temozolomide to a lesser degree have been associated with toxicity of gonadal function and sperm production and quality, which can be particularly harmful for patients interested in having children (55; 06). In turn, patients of childbearing age may be referred for oncofertility consultation prior to initiating treatment.

Clinical vignette

A 30-year-old female with a history of schizoaffective disorder presented initially with loss of consciousness while driving, resulting in a motor vehicle collision and development of new-onset focal seizures with secondary generalization. The initial head CT demonstrated a left frontal hypodensity. Subsequently, an MRI demonstrated a 26 mm x 23 mm x 18 mm T2 and FLAIR (fluid-attenuated inversion recovery) bright mass, without enhancement on T1 post-contrast.

The patient had an awake craniotomy that achieved gross total resection. Pathology demonstrated astrocytoma, IDH1-mutated, WHO grade 2. Of note, 1p/19q was intact, demarcating this tumor as NOT an oligodendroglioma. Further treatment with chemotherapy or radiation was deferred, but monitoring was ongoing every 3 months. Nine months after the operation, there was no evidence of disease recurrence.

Biological basis

Etiology and pathogenesis

The brain has long been considered a postmitotic organ wherein all cells had already differentiated and there were no resident stem cells. As a result, gliomas were believed to originate from the dedifferentiation of mature glial cells. Subsequent advances revealed that neural stem cells and glial progenitor cells can be found in multiple regions of the brain. As such, there is suspicion that gliomas originate from these stem cells. As with other cancers, spontaneous mutations in oncogenes and tumor suppressor genes drive the formation of the initial glioma cells. Subsequently, additional mutations are acquired to facilitate the propagation, invasiveness, and angiogenic properties of these tumors (43).

Mechanistic understanding of the survival benefits associated with IDH mutations

Mutation in IDH1 or IDH2 is a common driver of gliomagenesis that appears to precede secondary mutations. IDH genes encode isocitrate dehydrogenase, an enzyme that catalyzes the conversion of isocitrate to alpha-ketoglutarate in the citric acid cycle. Gliomas have mutations in only one copy of the IDH genes, suggesting that mutations may not simply result in a loss of function (03; 16). Instead, Dang and colleagues demonstrated that mutations in arginine 132 lead to a gain of function in the ability to catalyze the conversion of alpha-ketoglutarate to 2-hydroxyglutarate (16). Patients with inborn errors of 2-hydroxyglutarate metabolism that result in its accumulation are at increased risk of malignant brain tumors (01). Mutations in IDH also lead to extensive DNA methylation, leading to deregulated gene expression and cellular dedifferentiation (48).

Epidemiology

According to the Central Brain Tumor Registry of the United States (CBTRUS), primary central nervous system tumors have an average annual age-adjusted incidence of 24.71 per 100,000; 28.4% of these are malignant, with an annual incidence of 7.02 per 100,000 (32). Glioblastoma accounted for 50.1% of all malignant tumors, and the remaining 49.9% have multiple histopathologic types. IDH-mutated diffuse astrocytoma account for approximately 6% of malignant tumors (31). Brain tumors as a whole have an overall rate in females of 27.62 versus males at 21.60 per 100,000. This is driven in large part by the higher incidence of meningiomas in women. The overall rate in non-Hispanics versus Hispanics was 25.09 versus 22.95 per 100,000. Glioblastoma is 1.39-fold more common in males than females, whereas diffuse astrocytoma is 1.27-fold more common in males.

Prevention

There are currently no clear preventative strategies. Exposure to ionizing radiation has been associated with an increased risk of developing gliomas (32). On the other hand, individuals with allergies or atopic diseases are at decreased risk.

Differential diagnosis

Confusing conditions

Associated or underlying disorders

Diagnostic workup

Contrast-enhanced MRI is the standard approach to identify and characterize mass lesions in the brain. Diffuse gliomas are characterized by a poorly marginated mass that infiltrates the surrounding tissue. For IDH-mutated low-grade astrocytomas, an imaging modality known as the T2-FLAIR (fluid-attenuated inversion recovery) mismatch sign has been developed (36; 07). A meta-analysis examining 1053 patients demonstrated that T2-FLAIR has a specificity of 100% but a poor sensitivity of 42% (34). This mismatch sign is defined as a complete or near-complete homogeneous high signal intensity on T2-weighted images and a relative suppression of the signal intensities on the FLAIR sequence (36). Nevertheless, the current standard for tumor classification includes surgical resection, histological grading, and molecular genotyping (27), regardless of which infiltrating glioma type is suspected. Prior to resection, it is helpful to consider whether the brain lesion could be a metastasis from an extra-CNS cancer or whether the lesion is due to an infectious or inflammatory process.

Gross total resection and possibly supratotal resection, if it can be safely done, is the next step in glioma management. Sometimes, it is necessary to perform a subtotal resection, which can be used for diagnostic purposes, particularly if the tumor invades eloquent brain regions. As discussed below, gross total resection has been associated with improved survival compared to subtotal resection.

Beginning with the WHO 2016 classification, it has become standard management to genotype the tumor for IDH mutations and 1p19q co-deletion (26). Testing is often also performed for additional driver mutations in H3K27M, MYBL1, FGFR1, BRAF, MAPK, and CDK2A/B. CKDN2A is of particular relevance as it is a strong adverse prognostic factor in patients with IDH-mutated diffuse astrocytoma (02; 47). CKDN2A mutation carries a hazard ratio of 2.12 for progression-free survival and 1.72 for overall survival (02). The prognostic value of the other aforementioned mutations is not as strong (25). Ideally, this testing is performed via large, efficient next-generation sequencing panels as opposed to ad hoc piecemeal fashion. Methylation studies may also be performed as the MGMT promoter methylation status has been associated with improved survival (48; 05); however, this has only been definitively validated in glioblastoma, IDH-wildtype. More extensive DNA methylation profiling can also be performed. This may have the greatest value when the diagnosis remains uncertain.

Management

Surgical resection. Per the National Comprehensive Cancer Network (NCCN) guidelines, gross total resection has long been established as a mainstay of glioma treatment. Shaw and colleagues found that patients with diffuse gliomas in whom resection resulted in greater than 1 cm of residual tumor had a worse survival compared to those who had less than 1 cm residual (HR: 3.54, 95% CI 1.83-6.84, p=0.0002) (45). In patients with diffuse grade 2 gliomas, gross total resection has been found to result in improved outcome compared to observation (22). Jakola and colleagues compared two hospitals--one with a conservative observational approach to treatment and the other with a preference for early resection--and found that there was improved survival in patients at the hospital that favored early resection (22). In fact, this effect was observed across all three glioma subgroups analyzed (1p19q/IDH-mutated oligodendroglioma, IDH-mutated astrocytoma, and IDH-wildtype astrocytoma). Bauman and colleagues compared gross total resection with subtotal resection in patients with diffuse gliomas and found worse survival in patients with subtotal resection (HR: 1.59, 95% CI 1.06-2.38, p=0.026) (04). There may be promise for supratotal resection, ie, resection of both the contrast-enhancing tumor and the surrounding T2-FLAIR signal that represents edema, although this is not established as the standard of care (Leeuw and Vogelbaum 2018; 42).

Foregoing resection in favor of biopsy can further complicate care as it can result in the misdiagnosis in as many as 49% of cases (21). Another study supported the importance of gross total resection by demonstrating that it was associated with improved survival over subtotal resection, partial resection, and biopsy (13). In contrast to the aforementioned studies, Wijnenga and colleagues performed a retrospective cohort study examining radiologically presumed low-grade gliomas to compare observation, biopsy, and early resection (52). They found that observation had no difference in survival compared to early resection. However, consistent with above, they did find that only a biopsy was associated with worse survival (HR: 2.69, 95% CI 1.19-6.06, p=0.02) (52). This may be influenced by the patient selection for biopsy versus resection.

Laser interstitial thermal therapy (LITT) has emerged as a viable alternative to surgical resection. This minimally invasive approach utilizes lasers that are absorbed by tumor tissue, heating to around 42°C and coagulating the tissue (19). It is particularly useful in the treatment of patients who have deep-seated tumors or if the patients have comorbidities that make them poor surgical candidates. One small study specifically examined the use of LITT in the treatment of IDH-mutated low-grade gliomas (astrocytoma and oligodendroglioma) and found that these patients had a 72.5% progression-free survival at 3 years and 54.4% at 5 years, which is on par with results in patients with surgical resection (23). There is a need for a prospective clinical trial comparing LITT to surgical resection in order to assess whether it truly is a viable alternative.

Radiation therapy. Radiotherapy is also often utilized in the treatment of diffuse gliomas--as well as higher-grade gliomas. In 2005, van den Bent and colleagues demonstrated that radiotherapy in addition to gross total resection increased patients’ progression-free survival from 3.4 years without radiotherapy to 5.3 years with it (54). However, the study did not demonstrate an overall survival benefit with radiotherapy (54). Another trial compared low-dose versus high-dose radiotherapy for patients with grade 2 gliomas (44). It found that patients with high-dose radiotherapy had lower survival and increased radiation necrosis (44).

Chemotherapy. RTOG 9802 examined the utility of procarbazine, CCNU (lomustine), and vincristine (PVC) chemotherapy in patients with diffuse gliomas. It demonstrated that patients who received adjuvant PVC after radiotherapy had an improved median overall survival versus radiotherapy alone (not reached vs. 7.5 years) (45). As a substantial percentage of patients in the radiotherapy-only arm subsequently received chemotherapy, the trial results could be interpreted to support earlier combinatorial over sequential therapy. Examination of molecular subtypes of gliomas in this study demonstrated that adjuvant PCV chemotherapy led to improved survival in patients with 1p19q codeletion/IDH-mutated oligodendrogliomas and IDH-mutated astrocytomas (05). There was no survival benefit in patients with IDH-wildtype astrocytomas (05), although the study was not specifically powered to answer questions for molecular subtypes.

Temozolomide is another chemotherapy that has widely been used for high-grade glioblastoma; however, there are limited data on its efficacy in patients with low-grade gliomas, and molecular subtypes were not assessed. Initial studies demonstrated that low-grade gliomas respond to temozolomide, but these studies had a limited sample size (33; 39). A more robust study, RTOG 0424, was a single-arm trial examining the efficacy of temozolomide alongside radiotherapy in high-risk low-grade gliomas (both astrocytoma and oligodendroglioma); the 3-year overall survival of 73.5% was better than the historical 3-year overall survival of patients who did not receive temozolomide (18). Another study directly tested the efficacy of temozolomide alone in IDH-mutant astrocytomas (grade 2 and 3) compared to radiation alone or even a “wait-and-scan” approach (50). It found that radiation alone was superior to temozolomide alone (OS 14.4 vs. 10.7 years; p=0.02). Interestingly, this study even demonstrated that the wait-and-scan approach was superior to early temozolomide treatment (OS not reached vs. 10.7 years; p< 0.01). These effects applied to both grade 2 and 3 tumors. There is an ongoing randomized controlled trial that will be assessing the efficacy of temozolomide in patients with low-grade gliomas, and it will be interesting to see if these findings are replicated (49).

As IDH mutations have been associated with increased survival in gliomas, studies have begun to assess the therapeutic utility of IDH inhibitors. Current work is beginning to show promise (51). The phase 3 INDIGO trial has completed accrual, and results are awaited. Another promising approach makes use of a peptide vaccine targeting the IDH mutations. Phase 1 results have demonstrated that this vaccine induced immune responses and gave hints that it may lead to survival benefits (38). Lastly, other trials are beginning to explore PARP inhibition and immunotherapy for low-grade gliomas as they have shown promise in other cancers (28). One study may have also discovered a unique vulnerability of IDH-mutant gliomas (46). IDH-mutant glioma cells and tumors were found to rely on de novo pyrimidine synthesis, and inhibition of dihydroorotate dehydrogenase—a key enzyme—was effective in limiting tumor growth and improving survival in mouse models. This approach may represent yet another promising approach to the treatment of these tumors.

Observation. Because surgical resection, radiotherapy, and chemotherapy carry risks and adverse effects, observation of less malignant gliomas has been examined. As discussed above, gross total resection is the standard of care, and observation is not a standard approach prior to resection. According to the NCCN, observation may, however, play a role in patients with low-risk features—those younger than 40 years who have had a gross total resection (30). This is further supported by the results of the Observation arm of the RTOG 9802 trial. However, continued clinical and radiographic follow-up is important.

Observation has also been examined as an approach for patients with incidentally identified low-grade gliomas. No randomized controlled trials or prospective studies have examined incidentally identified low-grade gliomas, but a retrospective study examining grade 2 gliomas demonstrated that these tumors can become symptomatic approximately 48 months after discovery (20). This study showed that patients with incidentally identified low-grade gliomas have survival benefits associated with surgical resection. As a result, this study provides indirect evidence that resection may be a better approach than observation for the management of grade 2 gliomas.

In summary, it is clear that radiotherapy and chemotherapy prolong survival, but the exact timing is not well understood. Early radiotherapy is associated with improved progression-free survival, but its exact impact on overall survival is unknown (EORTC 22845). PCV chemotherapy improved survival (RTOG 9802), though it is likely that temozolomide does as well. The results of (RTOG 9802) showed that there is a significant improvement in median overall survival in patients with high-risk low-grade glioma treated with radiotherapy followed by PCV chemotherapy compared with radiotherapy alone. As most patients randomized to the radiation arm subsequently received chemotherapy at the time of progression, one could interpret these study results to suggest that earlier combinatorial treatment with radiation plus chemotherapy is superior to sequential treatment with second-line therapy at the time of progression. However, observation may still be a reasonable option for a subset of patients with high-risk low-grade gliomas who are asymptomatic or stable, or patients with low-risk low-grade gliomas, keeping in mind that regular imaging will be needed. Oftentimes, the clinical management decisions are based on a number of patient-specific factors.

Outcomes

Adverse effects. As described above, there are risks associated with treatment. Surgical resection may damage important brain regions, radiotherapy may adversely impact cognition and quality of life (08; 15; 17), and chemotherapy can impair fertility or have other side effects depending on which drug is used (55; 06).

Recurrence. This high recurrence rate is suspected to be due to the infiltrative nature of these tumors, whereby single tumor cells can be found far from the original mass (11), as well as the relative treatment resistance of a subset of tumor cells. It is likely that the recurrence rate is approximately 100% and that any lower reported numbers represent inadequately short reporting follow-up.

Special considerations

Fertility and pregnancy

Chemotherapy has been associated with toxicity of gonadal function and sperm production and quality, which can be particularly impactful for patients interested in having children (55; 06). Current chemotherapeutic options of temozolomide and PCV have been shown to be teratogenic, and the United States Food and Drug Administration recommends against their use during pregnancy. As patients with IDH-mutated diffuse astrocytoma have relatively good survival compared to other gliomas, shared decision-making is important, and patients should be educated to weigh the risks and benefits of each treatment modality. Oftentimes, these patients will be referred to oncofertility for consultation prior to initiation of treatment.

Anesthesia

Risks associated with anesthesia during tumor resection are the same as those for other surgical procedures.

Age range of presentation

The median age of presentation of diffuse astrocytomas is 45 years. It is unclear whether IDH-mutated diffuse astrocytomas differ in age of onset from IDH-wildtype diffuse astrocytomas (31).

Sex preponderance

Diffuse astrocytoma is 1.27-fold more prevalent in males than females (31).

Heredity

The vast majority of gliomas are not believed to be heritable, although some rare inherited genetic syndromes may predispose patients to gliomas. These syndromes include Li Fraumeni syndrome, neurofibromatosis, Lynch syndrome, melanoma-neural system tumor syndrome, Maffucci syndrome, Ollier disease, and tuberous sclerosis (29).