Module 7: Clinical Outcomes by Disease Site - The Use of Proton Therapy in the Treatment of Cancers of Central Nervous System
Introduction
Tumors of the central nervous system (CNS) exhibit a wide range of clinical behavior, from asymptomatic benign meningiomas to high-grade glioblastomas. For decades, radiation has been incorporated into the management of many CNS tumors given 1) the inability to completely resect lesions that involve critical structures, and 2) a need to improve upon the high recurrence rates seen in some tumors treated with surgery alone. Given the fundamental importance of the CNS, efforts to spare radiation dose to healthy brain tissue are of paramount importance. The lower integral dose characteristic of proton therapy is one valuable means of achieving this goal. While there is a rich body of literature on proton radiation for CNS malignancies in pediatrics, evidence for the benefit to proton therapy in adult CNS malignancies is more limited. The following review will focus on the use of proton therapy in adult glial tumors as well as more benign lesions as such meningioma and pituitary adenoma.
Proton therapy in low grade glioma
Although the clinical behavior and prognosis of low grade glioma can be highly variable, patients tend to be diagnosed at a young age and with modern treatment, many will remain long term survivors. Therefore, carefully assessing the risks and benefits of radiotherapy, and attempting to minimize toxicity by limiting dose to the surrounding brain, are critical considerations. Proton therapy is being explored as a means of widening the therapeutic index of radiotherapy in low grade glioma.
A recent study out of Heidelberg by Harrabi and colleagues attempted to quantify the dosimetric benefit of proton radiation over photon radiotherapy in low grade glioma. 3D conformal plans for 74 patients who had previously undergone proton radiation for low grade glioma were developed to compare dosimetric outcomes between the two modalities. Investigators found that target volume coverage was equivalent between protons and photons, but that maximum doses to optic nerves, optic chiasm, and brainstem, as well as mean doses to hypothalamus, inner ear, hippocampus, and pituitary were all improved with proton therapy. In particular, contralateral structures derived the largest dosimetric benefit. One caveat to this study is that intensity modulated radiotherapy (IMRT), rather than 3D conformal radiation, is more often the photon modality used in treating CNS malignancies. Therefore, the relevance of the comparison between 3D conformal photon therapy and proton therapy in this study is brought into question. Nevertheless, there clearly appears to be a dosimetric benefit to proton therapy compared to photon radiation in many cases.
Regarding clinical outcomes, a report from Massachusetts General Hospital by Shih and colleagues recently detailed results from a prospective trial of proton therapy for low grade glioma. 20 patients were treated on study. The median follow up was 5.1 years for patients who were alive and progression-free. The median patient age was 37.5 years old. The population was generally unfavorable as 1p19q deletion was present in only 10% of patients, a gross total resection had only been achieved in 20% of patients, and seizure was present at diagnosis in 85% of patients. Progression-free survival was 100%, 85%, and 40%, at 1, 3, and 5 years, respectively. This corresponded to an overall survival of 100%, 95%, and 84% over the same time intervals. New endocrine abnormalities developed in 6 patients, all but one of whom were treated to the hypothalamic pituitary axis. Adding to this experience, a separate study of 58 patients treated across 6 institutions for low grade glioma was recently presented in abstract form at the 2016 ASTRO Annual Meeting by Wilkonson and colleagues. 78% of patients were treated to 50.4-54 cCGE and most tumors were located in the frontal lobe. There were no grade 3+ toxicities, with common acute side effects included dermatitis, fatigue, and alopecia, and all acute side effects improved within 1-3 months of completion. This study is limited, however, by a lack of long term follow up.
Long term neurocognitive outcomes of proton therapy for low grade glioma are available from a companion study by Sherman and colleagues to the MGH prospective trial described above. Their 20 patient cohort was assessed with a battery of 25 neurocognitive tests at baseline and then at regular intervals until 5 years after treatment. Cognitive outcomes were found to be stable over time and in fact improved in visuospatial and executive functioning following treatment. Improvements were greater for patients with left-sided compared to right-sided tumors. Average emotional well-being remained stable over time as well. Investigators concluded that these outcomes were a consequence of normal tissue sparing from proton therapy. While no comparison to patients treated with photons is provided, their findings do represent the most rigorous evaluation of neurocognitive outcomes with proton therapy for low grade glioma to date.
Finally, dose-escalation with proton therapy for low grade glioma was attempted in a Phase I/II trial in the 1990’s from MGH by Fitzek and colleagues as well. This trial included grades 2 and 3 gliomas according to the Daumas-Duport classification, an older classification scheme which would incorporate low grade gliomas. Patients in this study were treated to doses approaching 80 CGE with combined proton-photon plans. Consistent with previous classical studies of dose escalation in low grade glioma using photon radiation, higher radiation doses with proton therapy failed to improve outcomes, providing yet more evidence that dose-escalation in low grade glioma is unlikely to improve survival. The potential benefit from proton therapy may therefore more likely lie in the possibility for improved neurocognitive outcomes rather than better tumor control.
Proton Therapy in Glioblastoma
Historically, given the consistently poor survival in glioblastoma despite maximal resection and adjuvant radiotherapy, there have been many attempts to improve survival by increasing local radiation doses to tumor. Previous strategies have included dose-escalated radiation alone with fractionated external beam radiation, brachytherapy, stereotactic radiosurgery and even proton therapy. Unfortunately, none of these approaches have improved outcomes. However, in the setting of chemoradiation with concurrent and adjuvant temozolamide, survival in glioblastoma was improved, particularly in MGMT methylated patients. With this achievement, there was renewed interest in attempting again dose-escalation in glioblastoma in the context of concurrent temozolamide chemotherapy.
A dosimetric analysis from the University of Florida by Ramakrishna et al comparing proton therapy to IMRT for unilateral glioblastoma was recently presented at ASTRO 2016 in an abstract. Comparison plans for 19 patients were analyzed. Relative to IMRT, proton plans delivered less low dose to uninvolved brain, particularly in the range of V5-V20. Contralateral hippocampal dose was also lower with proton therapy, as was the V54 for the optic chiasm planning risk volume. The clinical impact of these differences is uncertain.
However, clinical trials for proton therapy with concurrent chemotherapy in glioblastoma are in progress. An initial report from Japan by Mizumoto and colleagues of hyperfractionated concomitant boost proton radiotherapy for glioblastoma was published in 2010, with an update of long term outcomes released in 2015. In this study, 23 patients with supratentorial lesions without evidence of enhancement in the brainstem, hypothalamus, or thalamus were enrolled on a Phase I/II trial of dose-escalated twice-daily radiation with concurrent nimustine. The regimen consisted of 50.4 Gy delivered with photon therapy to the T2-hyperintense region, followed by two sequential proton therapy cone downs (the enhancing volume with a margin, then the enhancing volume alone) to 96 CGE. Encouraging survival outcomes were seen with a median survival of 21 months. Six of 23 patients experienced radiation necrosis, all of whom had survived 4 years without recurrence. Two of 23 experienced fatal leukoencephalopathy, though also without evidence of recurrence. With these results in mind, the NRG BN001 randomized Phase II trial (“RANDOMIZED PHASE II TRIAL OF HYPOFRACTIONATED DOSE-ESCALATED PHOTON IMRT OR PROTON BEAM THERAPY VERSUS CONVENTIONAL PHOTON IRRADIATION WITH CONCOMITANT AND ADJUVANT TEMOZOLOMIDE IN PATIENTS WITH NEWLY DIAGNOSED GLIOBLASTOMA”) is under accrual. Patients are randomized to a reference arm treating to 60 CGE with concurrent temozolamide, or to a dose-escalated arm treating to 75 CGE with either proton or IMRT, depending on the modalities available at the enrolling facility. The primary endpoint of this trial will be to compare standard versus escalated dose radiotherapy, but the trial will also compare proton versus IMRT dose-escalated therapy as a secondary endpoint.
Meningioma
A number of retrospective series have been published describing proton therapy for meningioma. As many of these lesions are slow-growing, optimizing long term control while minimizing toxicity is of paramount importance. Retrospective data from 2000-2015 have detailed outcomes treating both benign and atypical meningiomas either with proton therapy or combined proton-photon plans (reviewed by McDonald and colleagues). Most prior experiences included only passive scattering, though in some more recent studies patients were treated with active scanning. Doses in these studies were generally similar to those used in conventional fractionated photon radiation, with the primary advantage of enhanced normal tissue sparing and opportunity for dose escalation in more aggressive disease with protons. 5 year local control in these experiences ranged from 85-100% for benign meningiomas and 47-71% for atypical or malignant disease. These results compare favorably with historical controls. To highlight one recent experience, McDonald and colleagues reported on their institutional outcomes of fractionated proton therapy in atypical meningioma. 22 patients were treated to a median of 63 CGE, 6 of whom had also been previously treated with prior radiation. Actuarial local control was 87.5% at 5 years for patients who were treated to doses >60 CGE, compared to 50% for patients treated to lower doses. Prospective outcomes with proton therapy will be forthcoming, as both University of Pennsylvania and Massachusetts general Hospital are recruiting patients in prospective studies of dose-escalated proton therapy in meningioma.
Stereotactic radiosurgery with proton therapy has also been attempted for meningiomas. Halasz and colleagues from Harvard recently reported on their experience of 50 patients treated to a median dose of 13 CGE using proton stereotactic radiosurgery for benign meningiomas. Median follow up was 32 months. Three year actuarial local control rate was 94% with only 3 patients experiencing long term permanent adverse outcomes (2 seizure and 1 panhypopituitarism). Authors note that the slight decrease in local control in their population relative to similar studies relates to the presence of multiple patients with tumors having ‘atypical features’ that did not meet full criteria for atypical meningioma. Their report builds on prior older experience from the Uppsala facility in Sweden, who published results on over 140 patients with residual grade I skull base meningiomas treated with proton therapy with fractionated stereotactic radiotherapy. Eight year progression free survival was 85-90%. Therefore, proton-based stereotactic radiosurgery may have a role in the treatment of certain meningiomas, though no modern direct comparison to photon-based SRS is available. This strategy may be of particular value in the setting of re-irradiation, especially given that meningiomas classically can arise in previously irradiated patients.
Pituitary Adenoma
There are very limited modern data looking at proton therapy for pituitary adenoma. Ironically, pituitary irradiation was one of the first applications of proton therapy, at the time used to modulate hormone levels in women with metastatic breast cancer. There is only one large recent experience describing proton radiation for pituitary adenoma was published by Wattson and colleagues out of MGH in 2015. Investigators detailed outcomes of 165 patients with functional pituitary adenomas refractory to initial surgery, treated subsequently with proton therapy. 92% of patients were treated with proton SRS to a median to 20 CGE, the remainder treated with conventional fractionation to a median of 50.4 CGE. In the 140 patients with available follow up, tumor control was 98% at a median follow up time of 43 months. 62% of patients developed new hypopituitarism at 5 years, with larger target volume being associated with this complication. Four patients also experienced seizures following radiosurgery.
Conclusions
The potential benefits of proton therapy for CNS disease includes improvement of long term neurocognitive outcomes in the setting of more indolent disease, as well as the possibility for improved dose escalation in certain settings such as glioblastoma and atypical/malignant meningioma. Longer follow up from completed studies and results from ongoing prospective trials will hopefully shed light on these questions.