A Phase I/II Study of the Photon Radiosurgery System

Central nervous system tumors account for approximately 20% of all childhood neoplasms. The
treatment modalities used in the primary management of brain tumors are surgery, radiation
therapy and chemotherapy. In recent years, considerable progress has been made in each of
these therapeutic approaches. In spite of these advancements local tumor recurrence
continues to be an important reason for treatment failure in these children. The local tumor
recurrence rate varies according to the primary tumor type, treatment technique and
tumor-stage at initial presentation. After conventional treatment, the local tumor
recurrence rate ranges from 10% - 40% in tumors like medulloblastoma, craniopharyngioma,
ependymoma and low-grade gliomas . However in aggressive tumors like glioblastoma multiforme
the tumor recurrence rate in spite of the best modern treatments remains at 80-100%.

Radiation therapy has always played a key role in the management of adult and pediatric
brain tumors. There has been considerable interest in treating brain tumors using
stereotactic radiosurgery (SRS) using the Gamma knife or Linear accelerator and stereotactic
radiotherapy (SRT). The goal of stereotactic treatment is to deliver a high dose of
radiation with high geometric precision to a discrete tumor located in the brain. This is
accomplished by the use of rigid immobilization skull frames and CT / MRI information for
treatment planning and tumor targeting. Presently there are several therapeutic options
available for children with recurrent brain tumors. Reirradiation has been employed in
recurrent gliomas, medulloblastomas and ependymomas with stereotactic radiosurgery
stereotactic radiation and brachytherapy . Following reirradiation, tumor control rates of
50-70% have been obtained. The radiosurgery doses used in children with radiation recurrent
tumors have ranged from 10-24 Gy. The reirradiation has been generally well tolerated with
retreatment complications like transient edema, cranial neuropathy or radiation necrosis
observed in 10-15% of children. The results with high dose chemotherapy and bone marrow /
stem cell transplantation in children with recurrent malignant gliomas, medulloblastoma and
ependymoma have been disappointing with significant morbidity and mortality. Intraoperative
radiation has also been utilized for the treatment of primary and recurrent brain tumors. In
a report from Japan, 17 patients including two children with radiation recurrent malignant
brain tumors were treated with intraoperative radiation to doses of 20 - 40 Gy.
Intraoperative radiation was delivered using special applicators and electron beams. The
radiation was delivered after tumor resection and doses of 23 - 40 Gy were delivered to
depths of 0.5-1.5 cm. The median survival for patients with malignant gliomas and other
tumors (ependymoma, oligodendroglioma) was 12 months and 51 months respectively. The two
children with ependymoma were cured and are currently alive at 134 and 88 months after
intraoperative radiation. Three patients developed symptomatic brain necrosis, two of them
had relief of symptoms with surgery and one patient died. Three patients also developed
postoperative meningitis. In another report from University of Nebraska Medical Center, 49
patients with glioblastoma multiforme were treated with interstitial Cobalt 60 high
dose-rate irradiation to a dose of 20 Gy to the tumor with a 1-cm margin. The patients with
no prior radiation therapy (Group I) received an additional 40 Gy of external irradiation.
Nineteen of these patients (Group II) had been previously irradiated, and they received only
interstitial irradiation. The Cobalt 60 probe was guided into the tumor using CT scans and a
stereotactic frame. This treatment was well tolerated, one patient had a dural leak and
another had a subdural hematoma. There were no cases of meningitis or radiation necrosis.
The median survival for Group I and Group II patients were 7 months and 6 months
respectively.

The photon radiosurgery system (PRS) is an intraoperative irradiation device that is capable
of delivering high radiation doses to brain tumors. This system has recently been approved
for clinical use by the Food and Drug Administration (FDA).

Photon Radiosurgery System (PRS)

The Photon Radiosurgery System (PRS) incorporates a miniature, 40 KeV x-ray source capable
of delivering a prescribed radiation dose directly to a target volume. The PRS consists in
part of an electron beam-activated x-ray source with a sealed vacuum tube that is 10 cm long
and 3.2 mm in outer diameter that is designed for insertion into the body. This vacuum tube
incorporates an electron beam target on the inside surface of its tip. When an accelerated
electron beam is generated and sent down the tube to strike the target, Bremsstrahlung and
line x-rays are emitted from the tip of the tube in a nearly isotropic pattern.

Measurements of dose-rate in a water phantom have determined that the x-ray beam emanates
essentially, from a point source, with a nominal dose rate of 150 cGy/min at 10 mm, for a
beam current of 40 uA and a voltage of 40 kV. The absolute dose is estimated to be + 10%.
The dose distribution in water falls off approximately as a function of the third power of
the distance from the power source. The generator is light weighed, only 3.45 lbs. The
radiation dose is adjusted by accelerating voltage (ranging from 30 to 50kV), beam current
(ranging from 5 to 40 uA) and treatment time (0-60 minutes) through the control console that
weighs only 40lbs. The lightweight of PRS system readily allows us to carry the device to
the laboratory and the operating room.

For use of the PRS as an adjuvant treatment, treatment applicators made from a rigid
biocompatible plastic (ULTEM 1000) with known x-ray transmission characteristics are used.
The inside is hollowed out to allow introduction of the PRS x-ray probe to the epicenter of
the applicator so that the dose at its outer surface is uniform. The end of the applicators
is spherical with its diameter ranging from 1.5 cm to 4 cm. Treatment applicators will be
sterilized prior to each use. The applicator is inserted into the tumor-resected cavity to
deliver the prescribed dose of radiation.

The operation and dose characteristics of the PRS combine advantages of external beam
radiosurgery with those of brachytherapy (implantation of radiation seeds). As with
brachytherapy, the PRS can be located very precisely within the target volume, and can
improve the delivery of conformal therapy by irradiating the target volume precisely, with
little or no scatter of radiation. Due to its very rapid dose fall-off, the PRS
significantly reduces the radiation dose delivered to healthy tissues as compared with
external beam radiation and radiosurgery. Like radiosurgery, however, the PRS has a very
high dose rate and can deliver high radiation doses to the target volume. Another distinct
advantage of the PRS system is the ability to significantly decrease the radiation dose to
the normal structures in the brain adjacent to the tumor. All of the radiation treatment
techniques presently available deliver 10-50% of prescribed dose to the normal brain.
Intraoperative irradiation using PRS because of its direct application into the tumor or
tumor bed limits the dose to the normal tissue. This approach could result in a significant
decrease in radiation induced complications in vital structures such as the optic pathway,
brain stem and cerebral blood vessels. Another advantage of PRS is that unlike other types
of therapy, the PRS does not require the use of a radiation-shielded room. To summarize,
the advantages of the interstitial/surface application of radiation using the PRS are:

1. Direct access to the surgical bed of the tumor

2. Accurate delivery of a high single dose of radiation to the tumor

3. Superior protection of adjacent brain, cranial nerves or other critical structures by
the use of intraoperative shielding or intraoperative displacement of these organs

4. Superior radiobiological effectiveness (RBE) of low energy X-rays

5. Tumor dose inhomogeneity similar to brachytherapy and Gamma knife radiosurgery, with
the center of the tumor receiving a higher dose than the peripheral region that is
adjacent to normal structures.

Results of studies carried out with the PRS in brain tumors have demonstrated it to be
capable of delivering a lethal dose of radiation, in a single application to intracranial
tumors with minimal side effects. It has been used to treat primary and metastatic brain
tumors. In a report from Massachusetts General Hospital, 14 adults with primary and
metastatic brain tumors < 3.5 cm in greatest diameter were treated with a single fraction of
irradiation using PRS. The treated tumor diameter ranged from 10mm - 35 mm (mean 21mm), and
the tumor edge prescribed dose ranged from 10-20 Gy (average 12.5 Gy). The average treatment
time was 23 minutes (range, 7-45 minutes). Local control was obtained in 10 of the 13
patients with a follow-up of 1.5 - 36 months (mean 12 months). All patients tolerated the
procedure well, and most patients were discharged home the day after treatment. No new
neurological deficits were noted after irradiation. This study aims at determining the
maximum tolerated dose of irradiation using PRS in recurrent pediatric brain tumors.

Dose Selection

The radiation dose delivered by the PRS and radiosurgery are similar with regard to
dose-rate and total dose. The RTOG (Radiation Therapy Oncology Group) has performed a dose
escalation study to assess the maximum tolerated dose of radiosurgery in adults with
previously irradiated brain tumors and brain metastases. Based on acute and late toxicity,
the maximum tolerated radiosurgery doses were 24 Gy, 18 Gy and 15 Gy for tumors < 20 mm,
21-30 mm and 31-40 mm respectively.

In this study we had intended to perform a similar dose escalation study with doses ranging
from 10-19 Gy, 10-16 Gy and 10 - 14 Gy for tumors < 20 mm, 21-25 mm and 26-40 mm
respectively. These doses are lower than the established maximum tolerated doses for brain
reirradiation in adults with radiosurgery. These doses are also lower than the 20-40 Gy
doses utilized for intraoperative irradiation of adult brain tumors with electrons and
Cobalt 60 sources.

An interim analysis of patients entered on the study was performed. Based on the occurrence
of treatment -related complications in ONE patient who required 2 applicators and in another
patient in whom the dose was prescribed to 5 mm depth, the protocol has been modified as
follows:

1. No patient will have PRS treatment using more than one applicator

2. The depth of prescribed dose should not exceed 2 mm

3. Patients who have received prior RT and those who have not received prior RT would be
classified into Group A and Group B respectively.

4. The dose levels will for these two groups would be as shown in the table in Section
8.0. The dose escalation for non-brainstem (10-19 Gy) remains the same, but dose
escalation will no longer be stratified by tumor size. The dose levels for sites
adjacent to the brainstem and/or cranial nerves (10-14 Gy) also remain the same. A
minimum of 3 patients will have to be accrued in each dose level and only if there are
no complications observed, accrual at the next dose level would begin.

Dose escalation will be based on the incidence of acute CNS toxicity defined by RTOG
criteria. Unacceptable toxicity will be considered to be irreversible grade 3 (severe), any
grade 4 (life threatening) or grade 5 (fatal) RTOG CNS toxicity occurring within 3 months of
reirradiation. If no patient developed an unacceptable CNS toxicity as defined below, the
dose for that tumor size was then escalated.

The brain stem is very important part of the brain that controls most bodily functions like
blood pressure, respiration etc. In this study, we have adopted a gentler dose escalation
scheme for tumors in and around the brain stem. The three doses to be studied for tumors in
this location are10 Gy, 12 Gy and 14 Gy. These doses will be delivered independent of tumor
size.

Inclusion Criteria:

- All patients must be between 2 and 32 yrs. of age. Patients who are pregnant or
lactating will not be included in this study.

- Patients must have an estimated survival time of > 3 months, and a Karnofsky
performance status of > 50% or and ECOG performance status of 0-2.

- Patient must have radiographic imaging evidence of tumor recurrence except for
glioblastoma multiforme.

- The recurrent tumor prior to irradiation with PRS must be < 4 cm.

- All patients must be > 3 weeks from cytotoxic chemotherapy, except if patient
received < 1 week of non-myelotoxic chemotherapy. In that case, patient may be
enrolled with permission of the principal investigator.

- All patients must be > 6 weeks from high dose chemotherapy with stem cell rescue.

- All patients must be > 3 weeks from prior radiation therapy.

- Patients with a history of prior irradiation will be included.

- Informed consent must be obtained prior to registration on the study.

Tumor Sites:

- Brain

- Spinal Cord

Tumor Types:

- Recurrent Medulloblastoma

- Recurrent Ependymoma (recurrence regardless of primary treatment)

- Recurrent Glioma

- Glioblastoma Multiforme

- Recurrent Craniopharyngioma

- Recurrent PNET (Primitive Neuroectodermal tumor)

- Recurrent Meningioma

- Recurrent Pineoblastoma

Tumor Size:

- < 4 cm in maximum diameter