Proton Therapy for Craniopharyngioma in Children: the Orsay Proton Centre Experience
Reporter: Abigail Berman Milby, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: May 21, 2012
Presenting Author/Institution: Claire Alapetite, Proton Center of Orsay, Institut Curie, France
Craniopharyngiomas are benign intracranial tumors arising from Rathke's pouch in the sellar region. While they do not have the potential for malignant spread, they can become very large, creating mass effect in an eloquent area of the brain.
Craniopharyngiomas represent 3-9% of intracranial pediatric tumors. Over 50% of patients are less than 18 years of age.
Patients with craniopharyngiomas are at risk for complications including visual disturbances, endocrinopathies and neurocognitive dysfunction.
These complications can be part of the presenting symptoms of the tumor, or they can be caused by operative and/or radiotherapy damage.
Gross total resection (GTR) is the primary curative modality; however, GTR is often difficult to achieve because of the proximity of the tumor and normal brain structures. Therefore, many patients undergo a subtotal resection (STR) and are at higher risk of recurrence.
After STR, immediate post-operative radiotherapy reduces the recurrence rate; however, repeat surgery or salvage radiation is also a reasonable alternative. Therefore, the role of immediate post-operative radiation is debated, particularly in younger children who are most sensitive to the long-term effects of radiation (Merchant et al. 2008).
Patients with craniopharyngioma are particularly good candidates for proton therapy for three main reasons:
Tumors are in a quite eloquent location near the visual structures, hypothalamus, and temporal lobes.
Long-term survival is expected in these patients and therefore reducing long-term toxicity is at the utmost importance.
Many patients have pre-existing morbidities from either the tumor itself or surgical treatment.
Improving the dose delivered to critical structures including the brain may permit safer delivery of immediate post-operative radiotherapy.
From 1994-2009, 49 children with craniopharyngioma were treated.
A total of 19 of these patients were treated at relapse. 31% had one prior surgery but 69% had 2-4 surgeries (2 surgeries=4 patients, 3 surgeries=5 patients, 4 surgeries=3 patients).
One patient had prior gamma knife radiosurgery to 12 Gy.
Starting in 2004, 30 patients with initial hypothalamic involvement were treated post-operatively after a conservative subtotal resection (STR).
Median age was 10 years (range 3.8-16.1 years).
A median total dose of 54 Gy was delivered using conventional fractionation.
A combined photon/proton approach was used in 10 early patients and then proton therapy alone in later patients, using passive scattering delivery in fixed beam line.
Clinical target volume (CTV) was constructed via CT planning with image fusion with 3 MRI sequences (T1 with contrast, T2, FLAIR). Discussion with the neurosurgeon was also encouraged.
GTV median size 4.6 cc (0.28-15 cc); CTV margins of 3-5 mm were applied to the GTV.
PTV median size was 27 cc (6-18 cc).
Proton beam therapy was delivered as double scatter proton beam therapy.
5 beams were used with 2 alternating beams/day.
Comparative dosimetry was performed in two cases and showed a benefit of proton beams for critical organs including the optic chiasm, brainstem, cochlea, temporal lobes, and whole brain. The main difference is the low-dose bath (the 10 Gy isodose line).
Maximum dose of either 52 or 54 Gy was applied to the optic pathways were adapted to visual status.
Volume of temporal lobes receiving 20 Gy (V20) was maintained at less than 10%.
Median V10 of first 10 patients was 38%, but was 6% for later patients who were treated with proton beam therapy alone.
Two patients required general anesthesia.
Cyst size was monitored during proton therapy (Beltran et al 2012) and one patient had enlargement warranting replanning of collimators and compensators at week 3. One case required transient treatment interruption for surgery.
Follow-up included serial imaging and neurocognitive evaluation in all recent cases.
Median follow up was 53 months (range 24-156 months).
There were 4 in-field cystic relapses that were treated with surgery. Median time to relapse was 43 months (range 3-68 months).
1 relapse was categorized as early at 3 months and the other 3 as late
Three cases were not felt to be a geographical miss; however, one case was at the inner border of the field.
One additional relapse occurred along the surgical access route after 56 months. This area was not treated with proton beam therapy in an effort to minimize the volume.
All but one child had hypopituitarism prior to proton therapy.
The authors found a reduced rate of obesity (27% vs 60%) with STR and immediate post-operative RT compared to radiation at relapse.
There were no proton-related optic neuropathies.
15 patients had had neuropsychiatric assessments, telephone semi-structured interview with parent, questionnaires and quality of life assessments. In children irradiated after several surgeries, neuropsychiatric evaluation emphasized altered short-term memory, social and emotional functioning, and significant school difficulties. There were lower rates of behavioral disorders in patients treated with STR and radiation.
Preliminary results of a combined approach with conservative surgery for craniopharyngioma and proton therapy radiation suggest reduced morbidity without risking tumor control.
Long term follow up is required including neurocognitive testing and quality of life.
Proton therapy is a promising tool to decrease the risk of late sequelae and secondary malignancies.
The Curie Institute currently has an phase II study of dose escalation from 52.7/54.4 Gy (RBE) to 59.4 Gy (RBE), while maintaining dose constraints to optic pathways.
Although craniopharyngiomas are benign tumors, the destructive effects of the tumor and surgical and radiation treatments can be devastating.
Proton therapy is ideally suited to the treatment of craniopharyngiomas given eloquent location of these tumors within the brain.
This study addresses the controversial question of optimal management: conservative surgery followed by radiation immediately or more radical surgery with radiation only at relapse.
The authors find that the risk of obesity and neurocognitive deficits is lower with conservative surgery/radiation.
However, it should be noted that the follow up is shorter in the latter population than for relapsed patients as they were treated more recently.
In addition, most patients treated immediately post-operatively received proton therapy alone, whereas patients treated at relapse were treated with a mixture of proton and photon therapy; this may confound the adverse effects results.
The authors note that MRIs need to be performed while on treatment at least every 2 weeks to assure that the target volume still receives adequate doses of radiation.
It should be noted that if pencil beam scanning or intensity modulated proton therapy is used, MRI scans should be performed even more frequently as this treatment is very sensitive to small target volume changes.
Future studies need to be performed to see if margins can be safely decreased. The phase II dose escalation study underway at the Curie Institute will help determine if there is improved local control with dose escalation.
Further long-term follow up is needed on the visual, hypothalamic, and neurocognitive effects of radiotherapy in both the immediate post-operative and relapse settings.
May 11, 2012 - In children and adolescents with brain tumors treated with proton radiation, health-related quality of life scores are affected by both disease type and treatment, with assessments made by the patients correlating well with those of their parents, according to a study published online May 7 in the Journal of Clinical Oncology.