High-Grade Glioma

Updated by Sarah McAvoy

BACKGROUND

What % of primary CNS tumors are malignant?

~40% of primary brain tumors are considered malignant.

In adults, what is the most common malignant CNS neoplasm?

~80% of CNS neoplasms in adults are glioblastoma (GBM), which constitutes 20% of all primary tumors. ∼22,000 new malignant primary brain tumors are diagnosed annually in the U.S.

What are the WHO classifications for high-grade CNS tumors?

WHO III: anaplastic astrocytoma (AA)/anaplastic oligodendroglioma (AO)/anaplastic oligoastrocytoma (AOA)

WHO IV: GBM

What are some common genetic changes seen in malignant brain tumors?

↑ EGFR (50%) and phosphatase and tensin homolog (PTEN) mutation (30%–40%)

What are the initial genetic changes associated with primary vs. secondary GBM?

Primary: ↑ EGFR/MDM2 amplification/LOH 10/p16 loss

Secondary: p53 mutation → low-grade glioma (LGG) → LOH 19q/p16 loss → AA → LOH 10, DCC → 2^nd GBM

What % of GBMs are multicentric?

<5% of GBMs are multicentric.

What are the 4 pathologic characteristics used for astrocytoma grading?

Nuclear Atypia, Mitoses, Endothelial proliferation, and Necrosis

(Mnemonic: AMEN)

What is the defining pathologic characteristic of GBM?

Necrosis

WORKUP/STAGING

What is the Cushing triad, and what does it represent in brain tumors?

HTN, bradycardia, respiratory irregularity. It represents ↑ ICP.

With what Sx do high-grade gliomas (HGGs) most commonly present?

HA (especially in the morning, 50%), seizures (20%), focal neurologic dysfunction, and mental status change

What are the common imaging characteristics of HGGs on MRI?

Hypodense on T₁, gadolinium enhancing, T₂ enhancing, and + T₂ FLAIR (edema)

TREATMENT/PROGNOSIS

What is the MS for LGG vs. HGG?

Low grade: pure oligodendroglioma: 10 yrs; oligoastrocytoma: 7 yrs; anaplastic oligodendroglioma (AO): 5 yrs

High grade: AA: 3 yrs; GBM: 14 mos

What are the most important factors used for the RTOG recursive partitioning analysis (RPA) stratification?

Age 50 yrs, histology (AA or GBM), Karnofsky performance status (KPS) of 70, MS changes, and Sx ≥3 mos

(Curran WJ et al., J Natl Cancer Inst 1993)

What is the MS of a pt with RPA classes I–II, III–IV vs. V–VI?

MS by RPA class:

Classes I–II: 40–60 mos (3–5 yrs)

Classes III–IV: 11–18 mos (1–1.5 yrs)

Classes V–VI: 5–9 mos

Under what RPA classes can GBM fall?

GBMs fall under classes III–VI:

Class III: <50 yo, KPS 90–100

Class IV: <50 yo, KPS <90 or >50 yo, good KPS

Class V: >50 yo, KPS <70 but no change in MS

Class VI: KPS <70 and MS change

On what is the current modified RPA based?

Outcomes with temozolomide (TMZ) (Mirimanoff RO, JCO 2006)

What is the 4-yr OS and MS for RT + TMZ vs. RT alone for the adapted RPA groups for malignant gliomas (per Mirimanoff RO, ASTRO 2007 update)?

Overall survival: class III (<50 yo, performance status [PS] 0): 28.4% vs. 6.4%; class IV: 11.3% vs. 3.3%; class V (>50 yo, Mini-Mental State Examination <27, Bx only): 6% vs. 1%

Median survival: class III: 21 mos vs. 15 mos; class IV: 16 mos vs. 13 mos; class V: 10 mos vs. 9 mos

What additional factors did the European Nomogram (European GBM Calculator) investigate for stratification purposes?

MGMT methylation status and extent of resection; only MGMT, age, PS, and MS were prognostic (Gorlia T et al., Lancet Oncol 2008)

What is MGMT, and why is it important?

MGMT is a DNA repair enzyme that removes alkyl groups from the O⁶ position of guanine. When methylated the MGMT gene is inactive and therefore, there is no ability to repair the damage caused by TMZ = chemosensitive. Methylated MGMT leads to increased OS regardless of the type of treatment.

What is the mechanism of action of TMZ?

Oral agent that cross links DNA (alkylating)

When should anticonvulsants be started?

Anticonvulsants should be started only if the pt is symptomatic or has a Hx of seizures.

What is the impact of resection extent in HGGs?

Data suggest that the extent of resection correlates with improved outcomes. (Sanai N et al., Neurosurgery 2008; Stummer W, Lancet Oncol 2006)

What is the Tx paradigm for AA and GBM?

AA Tx paradigm: surgery → RT to 57–59.4 Gy +/– TMZ or PCV

GBM Tx paradigm: surgery → RT to 60 Gy + TMZ → TMZ × 6 months

What is the dose of TMZ, and how is it administered/scheduled?

Oral pill; 7 days/wk at 75 mg/m² concurrent with RT → 1-mo break → 6 cycles of adj TMZ at 150–200 mg/m² given 5 days of every 28 days

With the current Tx paradigm, what additional pharmacologic therapies are often necessary?

Steroids, proton pump inhibitors, and PCP prophylaxis

Which early GBM studies demonstrated significant (doubled) survival with RT vs. supportive care and helped RT become a standard component of Tx?

GBM studies showing significant survival:

BTSG 69–01 (Walker MD et al.): randomized to observation, BCNU (carmustine), WBRT, and BCNU + WBRT. There was no difference between WBRT vs. WBRT + BCNU, but RT was better than no RT. (J Neurosurg 1978)

BTSG 72–01 (Walker MD et al.): Randomized to MeCCNU (semustine), RT alone, BCNU + RT, and semustine +RT. MS 3–6 mos without RT, 9–12 mos with RT. (NEJM 1980)

Scandinavian Glioblastoma Study Group (SGSG) (Kristiansen K et al.): 45 Gy + bleomycin vs. 45 Gy vs. observation: MS 10.8 mos vs. 10.8 mos vs. 5.2 mos (SS). (Cancer 1981)

Which randomized study supports the use of RT vs. best supportive care for the elderly with GBM?

French data (Keime-Guibert F et al.): pts >70 yo, KPS ≥70, to 50.4 Gy vs. observation. MS improved with RT: 6.7 mos vs. 3.9 mos. There was no difference in QOL or cognition. (NEJM 2007)

What studies support the use of hypofractionation in elderly GBM pts with a poor KPS?

Roa W et al. (JCO 2004), Bauman GS et al. (IJROBP 1994), and Malmström A et al. (Lancet Oncol 2012) suggest feasibility of hypofractionation. All 3 trials looked at hypofractionated courses vs. standard fractionation in patients >60–65 yrs or with poor KPS and showed hypofractionation was as good or better than standard fractionation in terms of OS. Dose fractionation schemes included 40 Gy in 15 fx (Roa), 30 Gy in 10 fx (Bauman), and 34 Gy in 10 fx (Malmström).

Which study suggested that WBRT is not required (i.e., that limited-field RT is sufficient) in the Tx of HGGs?

BTCG 80–01 (Shapiro WR et al., J Neurosurg 1989): prospective RCT, 510 pts, WBRT to 60 Gy vs. WBRT to 43 Gy → CD to 60 Gy. There was no difference in survival in the RT arms. This study also demonstrates that BCNU single agent is equivalent to a multiagent regimen.

What evidence supports current RT volumes being used in the Tx of HGGs?

Hochberg FH: CT correlation with postmortem tissue, CT abnl + 2-cm margin encompassed tumor extent by 83%. Recurrence by imaging also occurred within 2 cm of the margin in primary Dz in 90% of cases. (Neurology 1980)

Kelly PJ: correlated imaging (MRI + CT) with stereotactic Bx in untreated gliomas. The study found that isolated tumor cells extended at least as far as T2, suggesting that T1 enhancement is equivalent to GTV and a +T2 is equivalent to subclinical Dz. (J Neurosurg 1987)

What evidence supports the current RT dose of 60 Gy used for HGGs?

Combined analysis of 3 BTSG trials (Walker MD, IJROBP 1979): 4 doses (<45 Gy, 50 Gy, 55 Gy, and 60 Gy). MS was 4 mos, 7 mos, 9 mos, and 10 mos, respectively. MRC data (Bleehen NM et al., Br J Cancer 1991): RCT, 474 pts, 45 Gy vs. 60 Gy (no chemo). MS was 9 mos vs. 12 mos.

Is there evidence for a dose escalation benefit beyond 60 Gy in HGGs?

No. There is no evidence for a dose escalation benefit.

In RTOG 7401, there was no benefit for 70 Gy vs. 60 Gy in >600 pts. (Chang CH, Cancer 1983)

Chan JL et al. escalated the dose to 90 Gy without survival benefit. Of those who failed at 90 Gy, 91% failed in-field. (JCO 2002)

Is there evidence supporting RT hyperfractionation for GBM?

No. RTOG 8302: >700 pts, randomized phase I, 64.8 vs. 81 Gy bid. There was no benefit. RTOG 9006 also showed no benefit.

Is there a benefit to radiosurgery boost for HGGs?

No. RTOG 9305 showed no benefit and higher toxicity. (Souhami L et al., IJROBP 2004)

Before the TMZ data, was there any benefit to CRT for HGGs?

Yes. This was shown by evidence from 2 large meta-analyses: The MRC Glioma Meta-analysis Trialist Group showed a small improved median PFS (7.5 mos vs. 6 mos) with chemo, reduced risk of death by 15%, and ↑ 1-yr OS by 6%. There was no RT dose response with less or more than 60 Gy. (Stewart LA et al., Lancet 2002)

What is the evidence that supports the current gold standard in GBM Tx with TMZ?

EORTC/NCIC data (Stupp R et al., NEJM 2005 and 5-yr update Stupp R et al., Lancet Oncol 2009): 5-yr OS 10% (+ TMZ) vs. 2% (– TMZ), MS 14.6 mos (+ TMZ) vs. 12 mos (– TMZ). Benefit in all groups including ages 60–70. In pts with methylated MGMT there is a PFS benefit to TMZ + RT vs. RT alone.

Which modified RPA class did TMZ + RT not benefit significantly (per Mirimanoff RO et al., JCO 2006)?

Class V. MS per RPA: class III, 17 mos; class IV, 15 mos; and class V, 10 mos. The only significant benefit of TMZ + RT vs. RT alone was in classes III–IV.

What is the role of MGMT methylation in terms of response to Tx with HGGs?

Greater response to TMZ + RT in those with methylated MGMT (Hegi ME et al., NEJM 2005; Mirimanoff RO, ASTRO 2007): 4-yr OS unmethylated (RT alone vs. RT + TMZ): 0% vs. 11% and methylated 5% vs. 22%, all SS

What are the options for recurrent GBM?

TMZ, bevacizumab, reirradiation to ∼36 Gy (Combs SE et al., BMC Cancer 2007), radiosurgery, brachytherapy (GliaSite), Gliadel, or clinical trial

What is the dose used for GliaSite in GBM? What is the radioisotope used?

60 Gy to 5–10 mm at a dose rate of 0.5 Gy/hr (Chan TA, IJROBP 2005); I-125

What are the approved uses of Gliadel?

FDA approval: recurrent dz with re-resection improved survival advantage 31 vs. 23 wks. (Brem H et al., Lancet 1995)

In newly diagnosed adj setting: MS was 13.9 mos vs. 11.6 mos. (Westphal M, Neurooncol 2003)

What are the general guidelines for RT target volume delineation in GBMs?

RTOG:

Initial volume (46 Gy): GTV1 = T₁/tumor bed + T₂/FLAIR, CTV1 = GTV1 + 2 cm

Boost volume (14 Gy): GTV2 = T₁/tumor bed, CTV2 = GTV2 + 2 cm. PTV adds 0.5 cm to all CTVs. Postop imaging (with MRI fusion) should be used for target delineation.

Alternative MDACC technique with simultaneous integrated boost (Chang et al., IRJOBP 2007): GTV = T₁/tumor bed, CTV = GTV + 2 cm. PTV 50 = CTV + 5-mm dose to 50 Gy in 30 fractions, PTV 60 = GTV + 5-mm dose 60 Gy in 30 fx

Which recent study showed similar survival outcomes with adj RT vs. adj chemo with procarbazine/lomustine/vincristine (PCV) or TMZ in WHO III gliomas (AA)?

German NOA-04 study (Wick W et al., JCO 2009): same PFS/OS for all arms (RT alone or 2 chemo agents alone). Good predictors: extent of resection, oligo component (oligodendroglioma or oligoastrocytoma), IDH1 mutation, MGMT promoter hypermethylation, 1p19q codeletion. Toxicity: grades 3–4 hematologic toxicity was significantly higher for PCV than for TMZ.

Which study investigated sequential PCV → RT vs. RT alone in oligodendroglial tumors?

RTOG 9402/INT-0149 (Cairncross G et al., JCO 2006 and Cairncross G et al., JCO 2013): AO or AOA s/p resection randomized to PCV × 4 → RT 59.4 Gy vs. RT alone. No median OS benefit. There was improved median PFS (8.4 yrs vs. 2.9 yrs) with chemo. Pt with 1p19q co-deletion (unplanned subgroup analysis) had improved median OS with PCV vs. RT alone (14.7 vs. 7.3 yrs). PCV increased toxicity.

Which recent study investigated sequential RT → BCNU vs. RT alone in AA? What did it find?

EORTC 26882 (Hildebrand J et al., Eur J Cancer 2008): no OS or PFS difference

What study tested the role of adj PCV after RT in oligodendroglial tumors?

EORTC 26951 (Van den Bent MJ et al., JCO 2006 and JCO 2013): With long-term follow-up OS better with RT + PCV (24.3 mos vs. 13.2 mos). 1p19q deleted pts did better. There was no long-term difference in QOL after PCV.

What phase III study investigated the efficacy of combining RT with either TMZ or nitrosourea in anaplastic gliomas?

RTOG 9813 (Chang S, Neurooncol 2008): Study closed early with 201 pts enrolled. Showed increased toxicity with BCNU vs. TMZ.

What studies are investigating the role of TMZ in A0/A0A?

NCCTG N0577: pts with 1p19q codeletion randomized to RT + PCV vs. RT + TMZ → TMZ vs. TMZ × 12 cycles.

EORTC 26053: anaplastic gliomas without 1p19q del 2 × 2 randomization RT vs. RT + TMZ then observation vs. adj TMZ.

What is the Tx paradigm for gliosarcoma?

Gliosarcoma Tx paradigm: treat like GBM (surgery → RT + TMZ → TMZ)

Which study tested dose-intensified TMZ after TMZ + RT?

RTOG 0525 (Gilbert M et al., ASCO 2011). This study randomized pts after TMZ + RT (after a 1-mo break) to TMZ on days 1–21 vs. standard days 1–5 for up to 12 cycles (max) depending on the response. Prelim results showed more toxicity in dose-dense arm and no difference in OS or PFS. Final data are pending as are neurocognitive and QOL results.

What ongoing phase III studies looked at bevacizumab in the upfront setting for GBM?

RTOG 0825 (Gilbert M et al., ASCO 2013): all treated with RT + TMZ then randomized to concurrent bevacizumab vs. placebo. Preliminary results show no OS or PFS benefit, but bevacizumab arm had worse QOL, cognitive function, and Sx burden.

AVAglio (ASCO 2013): all treated with RT + TMZ then randomized to bevacizumab vs. placebo. Early results suggest PFS benefit.