Study
Design
No. of Patients
Radiation
Chemotherapy
Median follow up
PFS
OS
Karim et al. [18]
Prospective
379
45 Gy in 5 weeks
None
74 months
47 %
58 %
59.4 Gy in 6.6 weeks
50 %
59 %
Van den Bent et al. [17]
Prospective
314
54 Gy Early
Variable, given to some patients at relapse
5.3 years
7.4 years
54 Delayed until progression
3.4 years
7.2 years
Shaw et al. [20]
Prospective, Phase III
203
50.4 Gy
None
6.43 years
72 % at 5 years
64.8 Gy
64 % at 5 years
Shaw et al. [24]
Prospective
251a
54 Gy
Radiation alone
7.8 years
54 Gy
Radiation plus PCV
13.3 years
Pace et al. [28]
Prospective, Phase II
43
Variableb
TMZ at progression
39 % at 12 months
Quinn et al. [31]
Prospective, Phase II
46
Variableb
TMZ at progression
Median PFS 22 months; 76 % at 12 months
Hoang-Xuan et al. [27]
Prospective
60
Nonec
TMZ at progression
14 months
73.4 % at 12 months
Kaloshi et al. [10]
Retrospective
149
Nonec
TMZ at progression
30.4 months
Median PFS 28 months; 79.5 % at 12 months
European Organization for Research and Treatment of Cancer (EORTC) 22844 and 22845 are two of the largest phase III trials ever completed in adult patients with LGG and they were designed to investigate optimal timing and dose of radiation. As mentioned previously, the EORTC 22844 and 22845 studies were used as a construction set for evaluating and validating high risk features, ultimately involving over 600 patients. A high-risk designation was given to patients with three or more high risk factors (age over 40, astrocytoma histology, presence of neurologic deficits before surgery, tumor diameter of 6 cm or greater, and tumor crossing the midline) [5]. This provided a guideline on which patients needed early intervention, namely radiation.
The EORTC 22,845 study compared early RT with delayed RT. After surgery, 314 patients were randomly assigned to early RT or deferred RT until the time of progression (control group). Median PFS was 5.3 years in the early RT group and 3.4 years in the control group. However, OS was similar between groups: median survival in the early RT group was 7.4 years compared with 7.2 years in the control group. Interestingly, at 1 year, seizures were better controlled in the early RT group. This study demonstrated that early RT after surgery lengthens PFS but does not affect OS [17].
Multiple studies looked at the optimal dose of RT with the goal to optimize tumor kill while minimizing both acute and long-term radiation toxicity. The largest such trial was the EORTC 22844 trial mentioned above. In this trial, patients were treated with either 45 Gy over 5 weeks or 59.4 Gy over 6.6 weeks with no significant difference in outcome observed between the two treatment groups. At 74 months follow up, the low dose arm had an OS 58 % and the high dose with an OS of 59 %. Furthermore, no difference was found in PFS between the two arms and QOL was worse in the high dose arm [18, 19].
A North Central Cancer Treatment Group (NCCTG)/RTOG/Eastern Cooperative Oncology Group (ECOG) study looked at over 200 LGG patients (with either astrocytoma or mixed oligoastrocytoma) treated with either low dose (50.4 Gy in 28 fractions) versus high-dose (64.8 Gy in 36 fractions) localized radiation therapy for supratentorial low-grade glioma [20]. Survival at 5 years was not significantly different in the two RT doses (72 % with low-dose RT and 64 % with high-dose RT). This study confirmed the results of EORTC 22844 showing no OS benefit and a higher incidence of radiation necrosis in the high-dose RT arm.
5 Proton Therapy
Proton therapy is a novel technique that utilizes a heavier particle than standard photons to deliver radiation. Protons therefore enable dose reduction while sparing normal brain tissue during radiation given the lesser exit dose. There is limited data available for the efficacy of protons in adult LGG. Hauswald et al. [21] retrospectively analyzed patients with low-grade glioma (WHO grade I and II) treated with proton therapy. Proton beam therapy was administered to 19 patients total (median age 29). Median dose applied was 54 Gy in fractions of 1.8 Gy. Median follow up was only 5 months; therefore, PFS and OS data are incomplete. The treatment was tolerable with the most common complication of focal alopecia and fatigue. Further prospective trials with extended follow up are needed to determine the role of proton therapy in LGG treatment.
6 Chemotherapy
Much of the data for chemotherapy use in LGG has been extrapolated from data on the treatment of high-grade glioma (HGG). There is substantial overlap in the chemotherapeutic regimens utilized in HGG and LGG, which are most commonly TMZ and procarbazine/lomustine (CCNU)/vincristine (PCV). Chemotherapy can be given during radiation, adjuvantly following radiation treatment, or at progression. There is still controversy over the ideal chemotherapy agent, the ideal time of administration, and the optimal duration of chemotherapy treatment. Recent data suggests that chemotherapy can play a role in improving outcome of LGG patients (Table 1).
As the use of TMZ has become standard of care for HGG [22], its use in LGG has increased among the neurooncology community. Prior to the widespread use of TMZ, PCV was the chemotherapy of choice for gliomas and in long-term analysis of prior studies, PCV may still have superior survival outcomes specifically in codeleted gliomas [23]. However, part of this regimen requires intravenous administration and can produce significant hematopoietic toxicity for patients. TMZ is usually better tolerated and easier to administer compared to PCV. The efficacy of TMZ has never been compared head-to-head with PCV in a prospective, randomized trial.
The largest trial evaluating long-term benefit from chemotherapy in upfront LGG is the Radiation Therapy Oncology Group (RTOG) 9802 trial. In this study, a total of 251 patients with LGG were enrolled and divided into high-risk and low-risk LGG. Patients with favorable risk included those less than 40 years old and those with gross-total resection. Unfavorable risk patients were those over 40 years with STR or biopsy only. Patients in the favorable risk group were observed postoperatively. Patients in the unfavorable risk group were randomly assigned to RT alone or RT followed by PCV chemotherapy. A significant improvement in OS was noted for study participants who received PCV chemotherapy plus radiation (13.3 years median survival time) compared to those receiving radiation therapy alone (7.8 years median survival time) [24, 25]. This was similar to results seen in patients with grade III anaplastic oligodendroglioma and oligoastrocytoma who also showed improved response to treatment with PCV chemotherapy [26]. In addition, chromosome 1p/19q deletions have been associated with favorable radiographic response rates and prolonged survival in patients with anaplastic oligodendrogliomas receiving PCV therapy [27]. Loss of the heterozygosity of chromosomes 1p and 19q have been shown to be powerful predictors of survival and chemosensitivity in grade 3 oligodendroglial tumors [28] and presumably have the same prognostic and predictive impact in their LGG counterparts.
RTOG 9802 trial showed efficacy of chemotherapy in conjunction with RT. However, optimal timing of chemotherapy relative to radiation is unclear as is the best choice of chemotherapy. One study utilized neoadjuvant PCV in large unresectable low-grade gliomas or gliomatosis cerebri to avoid large field radiation. The median time to disease progression in newly diagnosed patients was >24 months [29]. Another RTOG trial (RTOG 0424) recruited high-risk LGG patients at upfront diagnosis for treatment consisting of concurrent radiation (54 Gy/30 fractions) and TMZ 75 mg/m2 followed by 12 adjuvant cycles of TMZ 150–200 mg/m2 for 5 days. Preliminary results are available. 136 patients were accrued. Median follow-up time is 4.1 years. Median survival time has not yet been reached. Three-year OS rate was 73.1 %, significantly improved from historical controls. Long-term data from this trial is still pending [30].
Multiple trials have evaluated TMZ in the recurrent LGG setting and have showed disease response. Pace et al. [28] prospectively examined 43 patients with LGG who were treated with TMZ (5 day on/23 off at a dose of 150–200 mg/m2/day for 5 days per month) at the time of documented clinical and radiological progression. Median duration of response was 10 months with a PFS rate of 39 % at 12 months. Quinn et al. [31] conducted a phase II trial of TMZ (5 day on/23 off at a dose of 200 mg/m2) for 46 patients with progressive low-grade glioma. The objective response rate was 61 %. Median PFS was 22 months. And lastly, Hoang-Xuan et al. [27] conducted a third prospective study of 60 patients with LGG and progressive disease on MRI treated with TMZ (5 day on/23 off at a dose of 200 mg/m2) where objective radiologic response was 31 % and the median time to maximal tumor response was 12 months. Overall, there was a statistically significant positive correlation seen with the loss of chromosome 1p (with or without 19q deletion) and radiographic response to treatment. Retrospectively, Kaloshi et al. [10] examined 149 patients treated with TMZ (5 day on/23 off at a dose of 200 mg/m2) in which 53 % experienced an objective response. The median time to maximal response was 12 months. The median PFS was 28 months. Combined 1p/19q codeletion was associated with a higher rate of response, longer PFS, and longer OS. All of the above studies utilizing TMZ in LGG treated patients with 10–14 cycles of monthly TMZ. In LGGs treated with TMZ, it has been suggested that prolonged duration of treatment (aka metronomic or dose-dense therapy) may achieve a prolonged response. Similar to their HGG counterparts, the methylation status of LGG may impact its response to TMZ as well [32, 33]. Metronomic TMZ has been tried in the LGG setting with the goal to overcome MGMT resistance in patients with newly diagnosed LGG. Kesari et al. [2, 34] looked at patients with LGG who received TMZ dosed at 75 mg/m2 daily for 7 weeks followed by 4 weeks off treatment (11 week cycle). Treatment continued for a total of six cycles or until tumor progression and overall median PFS was 38 months. Molecular analysis revealed that patients who had a methylated MGMT promoter or deletion of either 1p or 19q chromosomes had longer OS. In LGG, dose-dense TMZ has not been compared to the conventional 5 day on/23 off regimen. However, in GBM patients, metronomic TMZ has been compared to the conventional “Stupp” regimen in the RTOG 0525 study which showed a lack of increased efficacy with dose-dense TMZ and increased toxicity for patients [35]. With the available data, the authors recommend using traditionally dosed adjuvant TMZ when indicated for 6–12 cycles.