- Email: [email protected]
Stereotactic radiosurgery for intracranial arteriovenous malformations using a standard linear accelerator
Inr. J. Radialion Oncobgy Biol Phys., Vol. Il. p. 673-677 Printed in the U.S.A. All rights reserved.
Copyright
0360-3016/89 $3.00 + .OO 0 1989 Peqmnon F’ws plc
??Technical Innovations and Notes
STEREOTACTIC RADIOSURGERY FOR INTRACRANIAL ARTERIOVENOUS MALFORMATIONS USING A STANDARD LINEAR ACCELERATOR JAY S. LOEFFLER, M.D.,1,2 EBEN ALEXANDER III, M.D.,2 ROBERT L. SIDDON, PHD.,‘~~ WILLIAM M. SAUNDERS, PH.D., M.D.,1,2 C. NORMAN COLEMAN, M.D.’ AND KEN R. WINSTON,
M.D.2
‘Joint Center for Radiation Therapy, Department of Radiation Therapy, Harvard Medical School; ‘Stereotactic Radiosurgery Program, The Neurosurgical Service, The Children’s Hospital and Brigham & Women’s Hospital and Department of Surgery, Harvard Medical School, Boston, MA We have previously described the development of a technique which utilizes a standard linear accelerator to provide stereotactic, limited field radiation. The radiation is delivered using a modified and carefully calibrated 6 MV linear accelerator. Precise target localization and patient immobilization is achieved using a Brown-Roberts-Wells (BRW) stereotactlc head frame which is in place during anglography, CT scanning, and treatment. Seventeen arterlovenous malformations (AVMs) have been treated in 16 patients from February 1986 to July 1988. Single doses of 15082508 cGy were delivered using multiple non-coplanar arcs with small, sharp edged x-ray beams to lesions ~2.7 cm in greatest diameter. The dose distribution from this technique has a very rapid dropoff of dose beyond the target volume. Doses were prescribed at the periphery of the AVMs, typically to the 8048% isodose line. Eleven of 16 patients have been followed by repeat angiography at least 1 year following treatment. Five of 11 have had complete obliteration of their AVM in 1 year and an additional three patients have achieved complete obliteration by 24 months. There have been no incidences of rebleeding or serious complications in any patient. We conclude that stereotactic radiosurgery using a standard linear accelerator is an effective and safe technique in the treatment of intracranial AVMs and the results compare favorably to the more expensive and elaborate systems that are currently available for stereotactic treatments. Arteriovenous
malformations,
Stereotactic radiation, Radiosurgery, Linear accelerator.
INTRODUCTION
untreated (12). A recent review from the Mayo Clinic of 168 patients with unruptured lesions found a mean risk of hemorrhage to be 2.2% per year, and the observed annual rates of hemorrhage increased over time. The risk of death was 29%, and 23% of the survivors had significant long-term morbidity. The treatment of choice for small peripherally located lesions is surgery. The use of ionizing radiation has been investigated for small, inoperable AVMs, mainly in the brainstem and deep nuclei (5, 9, 10, 19). Conventional, fractionated external beam irradiation has been used to treat inoperable AVMs ( 14). However, in general, results have been poor with only a small percentage of patients having angiographic resolution, despite high doses (50007500 cGy) and relatively large volumes. The development of stereotactic technology has led to the successful treatment of AVMs with radiation. The
Intracranial arteriovenous malformations (AVMs) are thought to be congenital clusters of abnormal arteries and veins. These lesions are known to be prone to spontaneous hemorrhage. The most common clinical presentation of patients with AVMs is bleeding. Considerable controversy has surrounded the issue of whether an unruptured AVM of the brain should be managed non-operatively or with surgery. The advent of more refined surgical techniques and stereotactic radiosurgery has led to greater optimism about aggressive management in recent years. In patients who have suffered bleeding from an AVM, the subsequent risk for rebleeding has been estimated at 2% to 6% per year, and the risk may increase with each additional hemorrhage (2, 6, 7). Also, patients with unruptured AVMs have a substantial risk of bleeding if left Dr. Winston is currently the Director of Pediatric Neurosurgery, University of Colorado Health Science Center, Denver, co. Reprint requests to: Jay S. Loeffler, M.D., Joint Center for Radiation Therapy, Department of Radiation Therapy, Harvard Medical School, 50 Binney St., Boston, MA 02 115.
Acknowledgements-The authors wish to thank the past and present members of the Physics Division of the Joint Center for Radiation Therapy who helped bring this clinical project to fruition. Accepted for publication 13 April 1989. 673
614
I. J. Radiation Oncology 0 Biology0 Physics
I
September 1989, Volume 17, Number 3
1
JCRT - SEBI Proposal Collimator
Dose
12.5 mm
2750 cGy
15.0
2500
17.5
2250
20.0
2000
22.5
1875
25.0
1750
27.5
1650
1500 Collimator
(mm)
Fig. 1. A modification of Kjellberg’s isoeffect analysis (9, 10) for patients treated with proton beam therapy. We have substituted our collimator diameter for “beam diameter” in that figure, and have chosen to prescribe the corresponding dose at the margin of the target volume. The black boxes represent our doses as they relate to collimator diameter. Our dose-volume selection would predict for a less than 1% incidence of symptomatic radiation necrosis based on Kjellberg’s analysis. (JCRT = Joint Center for Radiation Therapy. SEBI = Stereotactic External Beam Irradiation).
three systems that have been used extensively are: (a) proton radiosurgery (9, IO), (b) helium ion radiosurgery (5), and, (c) multiple 6oCo beams (the “Gamma knife”) (1 I, 19). All three techniques have produced excellent results with protection against hemorrhage after a latent period of 12-24 months in the majority of patients. Technical improvements as well as dose reductions have led to very few serious complications for all three programs. A major disadvantage of the three radiosurgery techniques is that they each require very complex and expensive equipment available in only a few areas of the world. We, therefore, decided to develop a system for stereotactic radiation therapy based on a standard linear accelerator, as have other groups outside the United States (1, 3,4, 8, 15). The purpose of this report is to summarize our initial experience using this technique. METHODS
AND MATERIALS
Since January 1988, patients were treated only after the Stereotactic Radiation Therapy Committee concluded that this was the most appropriate therapy. The committee consists of neurosurgeons, neurologists, neuroradiologists, and radiation therapists. Between February 1986 and July 1988, 16 patients were treated for 17 AVMs with stereotactic radiation therapy using a linear accelerator.* Twelve patients presented with bleeding, three patients presented with seizures and unruptured AVMs, and one patient had an AVM found on MRI that was performed to follow a previously resected craniopharyngioma. The median age of our patients was 22 years with a range between 5 and 53 years. Eleven lesions were located within the cerebral cortex, three lesions in the thalamus, and one each in the pons, corpus callosum, and basal ganglia. * Clinac 6/100, Varian Corporation,
Palo Alto, California.
We have utilized a conservative approach to the selection of the dose prescribed to treat AVMs. Our goal was to strive for the minimum dose that has a high probability of success and low risk of late complications. We have used Kjellberg’s estimate of the dose for a 1% risk of symptomatic radiation necrosis as a function of volume treated (9, 10). We substituted our collimator diameter for “beam diameter” in that figure, and to prescribe the corresponding dose to the margin of the target volume. The dose-volume relationship is graphically displayed in Figure 1. The preparation of the linear accelerator, patient alignment and immobilization, target localization, and treatment planning have been previously published (13, 1618, 20). The entire sequence of localization and treatment is accomplished in a single day. The BRW frame? is applied just prior to angiography. Once localization angiograms are complete, a CT scan is performed using 3-5 mm slices. The center and dimensions of the AVM relative to the frame are obtained from the angiogram. A treatment plan is calculated using the CT scan, but using the position and size of the AVM as determined from the angiogram, taking care to include arterial feeding vessels as well as the nidus. Vital structures such as the eyes are outlined on the treatment plan CT so that beams from the various arcs used for treatment will not enter or exit through these structures. While the treatment plan is being finalized, the verification procedure is carried out to certify that the setting on the BRW floor stand will place the center of the AVM at the isocenter, and that the system is correctly aligned (13,20). The patient is then brought to the treatment area, and the BRW ring is attached to the specially adapted lloor stand. To date, treatments have required three to six non-coplanar arc rotations. Children less than t Radionics Inc., Burlington, Massachusetts.
676
I. J. Radiation Oncology 0 Biology 0 Physics
at least 12 months of follow-up and have undergone at least one repeat angiogram. Our plan is to not obtain angiograms until 12 months after treatment because of the pattern of response of AVMs to radiosurgery. There has been no rebleeding in the 16 patients treated. Of the eleven patients studied with angiography at 1 year, five patients have had a complete obliteration of their lesions, five patients have had a greater than 50% reduction in the volume of the AVM, and one patient has had a 25% reduction. Figure 2 demonstrates the before and after angiograms in one patient who achieved a complete obliteration 1 year following radiosurgery. Three patients with residual AVM at 1 year had a repeat angiogram 2 years after treatment and were found to have a complete obliteration. Thus, eight of eleven AVMs (72%) have had complete obliteration as documented by angiography by 2 years and the remaining three lesions have decreased significantly (>75%) at 2 year and await their 2-year angiogram. We have not observed any serious side-effects of our treatment. One patient with a small AVM in the corpus callosum developed headaches and lethargy 7 months after receiving 2000 cGy with a 2 cm collimator normalized to the 80% isodose line. An IV contrast enhanced CT scan showed a 1.5 cm enhancing ring corresponding to the exact treatment area with a great deal of surrounding cerebral edema (Fig. 3). The patient was treated with a short course of oral dexamethasone therapy and is currently free of symptoms. A repeat CT scan showed complete resolution of this enhancing area and repeat angiogram showed obliteration of the AVM. Another patient received 2000 cGy to an AVM within the pons with a 2 cm collimator normalized to the 80%
September 1989, Volume 17, Number 3
isodose line. Six months after therapy, the patient developed problems with coordination of the right lower extremity. This has slowly improved without therapy and the patient is performing well 17 months after therapy. Alopecia developed in one young child 1 month following treatment for a very superficial AVM of the left parietal lobe. The scalp in this region received approximately 400 cGy. Six months after radiation, the patient has full return of hair in the region. DISCUSSION The results from our initial experience demonstrate that stereotactic radiation therapy for AVMs with a standard linear accelerator is an effective and safe technique. As mentioned previously, we have been conservative in our selection of radiation doses. We have elected to investigate the minimum dose needed rather than the maximal dose tolerated in order to maximize the therapeutic index. Even with our conservative doses, our results are similar to those reported by other groups using various forms of stereotactic radiosurgery. Investigators from the Lawrence Berkeley Laboratory, Berkeley, have treated 130 patients with helium ion radiosurgery since 1980 (5). Doses biologically equivalent to 2500 cGy to 4500 cGy were given in one or two sessions. With a follow-up of over 2 years for the majority of patients, only four patients have bled. This hemorrhage rate was felt by the investigators to be well below the rate expected in an untreated population of AVM patients. Although the follow-up for these patients is short, no cases of symptomatic radiation injury have been reported. Kjellberg et al., from the Massachusetts General Hos-
Fig. 3. Lateral projection of angiogram showing a relatively small AVM located near the corpus callosum on the right side. The patient was treated with 2000 cGy prescribed at the 80% isodose line with a 20 mm collimator. The CT scan on the right was taken 7 months after the completion of treatment because of the development of headaches and lethargy showing an enhancing ring corresponding to the treatment volume with surrounding cerebral edema. After a short course of oral dexamethasone therapy, the symptoms resolved. Angiogram repeated at 13 months showed a completed resolution of the AVM and repeat CT showed a resolution of the enhancing pattern and edema.
Stereotactic radiation 0 J. S. LOEFF’LERet al.
the age of 10 have required general anesthesia for the duration of the procedure. Treatment usually takes 3045 minutes to complete once the machine has been set up.
615
RESULTS At the time of this analysis (July 1988), eleven of the 16 patients treated with stereotactic radiosurgery have had
Fig. 2. (A) Anterior and lateral projections of an angiogram of a patient in this series. The right parietaJ lesion was fed from both the anterior and middle cerebral circulations. The four arrows represent the paths of the four noncoplanar arcs used in treatment. (B) Anterior and lateral projections of the angiogram taken 1 year after treatment showing complete obliteration of the AVM. The patient was treated with a 30 mm collimator and the dose was 1500 cGy prescribed at the 80% isodose line.
Stereotactic radiation 0 J. S. LOEFFLER et al.
pital, reported the results of treating 439 patients with AVMs with proton radiosurgery (9, 10). Doses in the proton beam project have been lowered to 1000 cGy for large lesions (treated with a 50 mm beam) to 5000 cGy for small lesions (treated with a 7 mm beam). Eight patients died of hemorrhage within 18 months of proton radiosurgery, a rate considered to be consistent with the rate for untreated patients. Only one patient died of hemorrhage after the 18 month period. Patients presenting initially with hemorrhage bled at a rate of 7.3% per year prior to treatment and at a rate of 2.4% per year beyond 2 years of treatment (p < .005). Neurological deficits were seen in the initial patients treated with large fields and high doses, but the incidence was small. After the modification of radiation dose for the large lesions, complications have been uncommon and Kjellberg believes there is protection against hemorrhage after a latent period of 18-24 months following proton radiosurgery. Since 1970, 135 patients have been treated with the Leksell technique (“Gamma knife”) at the Karolinska Institute in Stockholm (19). The radiation dose for the majority of patients was between 5000 cGy and 10,000 cGy, but ranged from 3500 cGy to 12,500 cGy. Complete obliteration was achieved in 84% of this group as docu-
677
mented by angiography 2 years following treatment. Only two of the patients hemorrhaged after treatment, both within 6 months of therapy. Four of the 135 patients developed new neurological deficits, a rate considered acceptable considering the very high success rate of this program. In summary, for patients with AVMs, stereotactic radiation therapy or radiosurgery appears to protect against hemorrhage after a latent period of 12-24 months following treatment using any of the four techniques described (protons, helium ions, “Gamma knife,” linear accelerator). Symptomatic radiation injury or necrosis of the brain has been uncommon despite single doses that are considered extremely high by conventional radiation therapy standards. We have developed a relatively simple and inexpensive technique for stereotactic radiosurgery using a standard linear accelerator. Our preliminary clinical data suggest that this is an efficacious and safe technique with results that equal the more expensive and elaborate systems that are available to provide stereotactic radiation therapy (radiosurgery). Furthermore, it may be possible to use a relatively low radiation dose, although more experience is needed.
REFERENCES I. Ektti, 0. 0.; Derechinsky,
2.
3.
4.
5.
6.
7.
8.
9.
10.
V. E. Hyperselective encephalic irradiation with linear accelerator. Acta. Neurochirurgica. 33:385-390; 1984. Brown, R. D.; Wieber, D. 0.; Forbes, G.; O’Fallon, W. M.; Piepgras, D. G.; Marsh, W. R.; Maciunas, R. J. The natural history of unruptured intracranial arteriovenous malformations. J. Neurosurg. 68:352-357; 1988. Chierego, G.; Marchetti, C.; Avanzo, R. C.; Pozza, F.; Colombo, F. Dosimetric considerations of multiple arc stereotaxic radiotherapy. Radiother. Oncol. 12: 141-152; 1988. Colombo, F.; Benedetti, A.; Pozza, F., Avanzo, R. C.; Marchetti, C.; Chierego, G.; Zandardo, A. External stereotactic irradiation by linear accelerator. Neurosurg. 16: 154- 159; 1985. Fabrikant, J. I.; Lyman, J. T.; Frankel, K. A. Heavy chargedparticle Bragg peak radiosurgery for intracranial vascular disorder. Radiat. Res. 104:244-258; 1985. Forster, D. M. C.; Steiner, L.; Hakanson, S. Arteriovenous malformations of the brain. A long-term clinical study. J. Neurosurg. 37:562-570; 1972. Graf, C. J.; Perret, G. E.; Turner, J. C. Bleeding from cerebral arteriovenous malformations as part of their natural history. J. Neurosurg. 58:331-337; 1983. Hartmann, G. H.; Schlegel, W.; Stunn, V.; Kober, B.; Postyr, 0.; Lorenzo, W. J. Cerebral radiation surgery using a moving field irradiation of a linear accelerator facility. Int. J. Radiat. Oncol. Biol. Phys. 11:1185-l 192; 1985. KjelIberg, R. N. Bragg peak proton beam therapy for arteriovenous malformations of the brain. Clin. Neurosurg. 3 1: 248-268; 1983. Kjellberg, R. N.; Hanamura, T.; Davis, K. R.; Lyons, S. L.; Adams, R. D. Bragg-peak proton-beam therapy for arteriovenous malformations of the brain. N. Engl. J. Med. 309: 269-274; 1983.
11. Leksell, L. The stereotactic method and radiosurgery of the
brain. Acta. Chir. Stand. 102:3 16-3 19; 195 1. 12. Luessenhop, A. J. Natural history of cerebral arteriovenous malformations. In: Wislon, C. B., Stein, B. M., eds. Intracranial arteriovenous malformations. Baltimore: Williams and Wilkins; 1984: 12-23. 13. Lutz, W.; Winston, K. R.; Maleki, N. A system for stereotactic radiosurgery with a linear accelerator. Int. J. Radiat. Oncol. Biol. Phys. 14:373-381; 1988. summary 14. Paulsen, M. G. Arteriovenous malformations-a of 6 cases treated with radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 13:1553-1557; 1988. 15. Podgursak, E. B.; Oliver, A.; Pla, M.; Lefebvre, P. Y.; Hazel, J. Dynamic stereotactic radiosurgery. Int. J. Radiat. Oncol. Biol. Phys. 14:115-126; 1988. 16. Rice, R. K.; Hansen, J. L.; Svensson, G. K.; Siddon, R. L. Measurement of dose distributions in small beams of 6 MV x-rays. Phys. Med. Biol. 32: 1087-1099; 1987. 17. Saunders, W. M.; Winston, K. R.; Siddon, R. L.; Svensson, G. H.; Kijewski, P. K.; Rice, R. K.; Hansen, J. L.; Barth, N. H. Radiosurgery for arteriovenous malformations of the brain using a standard linear accelerator: rationale and technique. Int. J. Radiat. Oncol. Biol. Phys. 15:441-447; 1988. 18. Siddon, R. L.; Barth, N. H. Stereotactic localization of intracranial targets. Int. J. Radiat. Oncol. Biol. Phys. 13: 12411246; 1987. 19. Steiner, L. Treatment of arteriovenous malformations by radiosurgery. In: Wilson, C. B., Stein, B. M., eds. Intracranial arteriovenous malformations. Baltimore: Williams and Wilkins; 1984:295-3 13. 20. Winston, K. R.; Lutz, W. Linear accelerator as a tool for stereotactic radiosurgery. Neurosurgery 22:454-464; 1988.
Copyright
0360-3016/89 $3.00 + .OO 0 1989 Peqmnon F’ws plc
??Technical Innovations and Notes
STEREOTACTIC RADIOSURGERY FOR INTRACRANIAL ARTERIOVENOUS MALFORMATIONS USING A STANDARD LINEAR ACCELERATOR JAY S. LOEFFLER, M.D.,1,2 EBEN ALEXANDER III, M.D.,2 ROBERT L. SIDDON, PHD.,‘~~ WILLIAM M. SAUNDERS, PH.D., M.D.,1,2 C. NORMAN COLEMAN, M.D.’ AND KEN R. WINSTON,
M.D.2
‘Joint Center for Radiation Therapy, Department of Radiation Therapy, Harvard Medical School; ‘Stereotactic Radiosurgery Program, The Neurosurgical Service, The Children’s Hospital and Brigham & Women’s Hospital and Department of Surgery, Harvard Medical School, Boston, MA We have previously described the development of a technique which utilizes a standard linear accelerator to provide stereotactic, limited field radiation. The radiation is delivered using a modified and carefully calibrated 6 MV linear accelerator. Precise target localization and patient immobilization is achieved using a Brown-Roberts-Wells (BRW) stereotactlc head frame which is in place during anglography, CT scanning, and treatment. Seventeen arterlovenous malformations (AVMs) have been treated in 16 patients from February 1986 to July 1988. Single doses of 15082508 cGy were delivered using multiple non-coplanar arcs with small, sharp edged x-ray beams to lesions ~2.7 cm in greatest diameter. The dose distribution from this technique has a very rapid dropoff of dose beyond the target volume. Doses were prescribed at the periphery of the AVMs, typically to the 8048% isodose line. Eleven of 16 patients have been followed by repeat angiography at least 1 year following treatment. Five of 11 have had complete obliteration of their AVM in 1 year and an additional three patients have achieved complete obliteration by 24 months. There have been no incidences of rebleeding or serious complications in any patient. We conclude that stereotactic radiosurgery using a standard linear accelerator is an effective and safe technique in the treatment of intracranial AVMs and the results compare favorably to the more expensive and elaborate systems that are currently available for stereotactic treatments. Arteriovenous
malformations,
Stereotactic radiation, Radiosurgery, Linear accelerator.
INTRODUCTION
untreated (12). A recent review from the Mayo Clinic of 168 patients with unruptured lesions found a mean risk of hemorrhage to be 2.2% per year, and the observed annual rates of hemorrhage increased over time. The risk of death was 29%, and 23% of the survivors had significant long-term morbidity. The treatment of choice for small peripherally located lesions is surgery. The use of ionizing radiation has been investigated for small, inoperable AVMs, mainly in the brainstem and deep nuclei (5, 9, 10, 19). Conventional, fractionated external beam irradiation has been used to treat inoperable AVMs ( 14). However, in general, results have been poor with only a small percentage of patients having angiographic resolution, despite high doses (50007500 cGy) and relatively large volumes. The development of stereotactic technology has led to the successful treatment of AVMs with radiation. The
Intracranial arteriovenous malformations (AVMs) are thought to be congenital clusters of abnormal arteries and veins. These lesions are known to be prone to spontaneous hemorrhage. The most common clinical presentation of patients with AVMs is bleeding. Considerable controversy has surrounded the issue of whether an unruptured AVM of the brain should be managed non-operatively or with surgery. The advent of more refined surgical techniques and stereotactic radiosurgery has led to greater optimism about aggressive management in recent years. In patients who have suffered bleeding from an AVM, the subsequent risk for rebleeding has been estimated at 2% to 6% per year, and the risk may increase with each additional hemorrhage (2, 6, 7). Also, patients with unruptured AVMs have a substantial risk of bleeding if left Dr. Winston is currently the Director of Pediatric Neurosurgery, University of Colorado Health Science Center, Denver, co. Reprint requests to: Jay S. Loeffler, M.D., Joint Center for Radiation Therapy, Department of Radiation Therapy, Harvard Medical School, 50 Binney St., Boston, MA 02 115.
Acknowledgements-The authors wish to thank the past and present members of the Physics Division of the Joint Center for Radiation Therapy who helped bring this clinical project to fruition. Accepted for publication 13 April 1989. 673
614
I. J. Radiation Oncology 0 Biology0 Physics
I
September 1989, Volume 17, Number 3
1
JCRT - SEBI Proposal Collimator
Dose
12.5 mm
2750 cGy
15.0
2500
17.5
2250
20.0
2000
22.5
1875
25.0
1750
27.5
1650
1500 Collimator
(mm)
Fig. 1. A modification of Kjellberg’s isoeffect analysis (9, 10) for patients treated with proton beam therapy. We have substituted our collimator diameter for “beam diameter” in that figure, and have chosen to prescribe the corresponding dose at the margin of the target volume. The black boxes represent our doses as they relate to collimator diameter. Our dose-volume selection would predict for a less than 1% incidence of symptomatic radiation necrosis based on Kjellberg’s analysis. (JCRT = Joint Center for Radiation Therapy. SEBI = Stereotactic External Beam Irradiation).
three systems that have been used extensively are: (a) proton radiosurgery (9, IO), (b) helium ion radiosurgery (5), and, (c) multiple 6oCo beams (the “Gamma knife”) (1 I, 19). All three techniques have produced excellent results with protection against hemorrhage after a latent period of 12-24 months in the majority of patients. Technical improvements as well as dose reductions have led to very few serious complications for all three programs. A major disadvantage of the three radiosurgery techniques is that they each require very complex and expensive equipment available in only a few areas of the world. We, therefore, decided to develop a system for stereotactic radiation therapy based on a standard linear accelerator, as have other groups outside the United States (1, 3,4, 8, 15). The purpose of this report is to summarize our initial experience using this technique. METHODS
AND MATERIALS
Since January 1988, patients were treated only after the Stereotactic Radiation Therapy Committee concluded that this was the most appropriate therapy. The committee consists of neurosurgeons, neurologists, neuroradiologists, and radiation therapists. Between February 1986 and July 1988, 16 patients were treated for 17 AVMs with stereotactic radiation therapy using a linear accelerator.* Twelve patients presented with bleeding, three patients presented with seizures and unruptured AVMs, and one patient had an AVM found on MRI that was performed to follow a previously resected craniopharyngioma. The median age of our patients was 22 years with a range between 5 and 53 years. Eleven lesions were located within the cerebral cortex, three lesions in the thalamus, and one each in the pons, corpus callosum, and basal ganglia. * Clinac 6/100, Varian Corporation,
Palo Alto, California.
We have utilized a conservative approach to the selection of the dose prescribed to treat AVMs. Our goal was to strive for the minimum dose that has a high probability of success and low risk of late complications. We have used Kjellberg’s estimate of the dose for a 1% risk of symptomatic radiation necrosis as a function of volume treated (9, 10). We substituted our collimator diameter for “beam diameter” in that figure, and to prescribe the corresponding dose to the margin of the target volume. The dose-volume relationship is graphically displayed in Figure 1. The preparation of the linear accelerator, patient alignment and immobilization, target localization, and treatment planning have been previously published (13, 1618, 20). The entire sequence of localization and treatment is accomplished in a single day. The BRW frame? is applied just prior to angiography. Once localization angiograms are complete, a CT scan is performed using 3-5 mm slices. The center and dimensions of the AVM relative to the frame are obtained from the angiogram. A treatment plan is calculated using the CT scan, but using the position and size of the AVM as determined from the angiogram, taking care to include arterial feeding vessels as well as the nidus. Vital structures such as the eyes are outlined on the treatment plan CT so that beams from the various arcs used for treatment will not enter or exit through these structures. While the treatment plan is being finalized, the verification procedure is carried out to certify that the setting on the BRW floor stand will place the center of the AVM at the isocenter, and that the system is correctly aligned (13,20). The patient is then brought to the treatment area, and the BRW ring is attached to the specially adapted lloor stand. To date, treatments have required three to six non-coplanar arc rotations. Children less than t Radionics Inc., Burlington, Massachusetts.
676
I. J. Radiation Oncology 0 Biology 0 Physics
at least 12 months of follow-up and have undergone at least one repeat angiogram. Our plan is to not obtain angiograms until 12 months after treatment because of the pattern of response of AVMs to radiosurgery. There has been no rebleeding in the 16 patients treated. Of the eleven patients studied with angiography at 1 year, five patients have had a complete obliteration of their lesions, five patients have had a greater than 50% reduction in the volume of the AVM, and one patient has had a 25% reduction. Figure 2 demonstrates the before and after angiograms in one patient who achieved a complete obliteration 1 year following radiosurgery. Three patients with residual AVM at 1 year had a repeat angiogram 2 years after treatment and were found to have a complete obliteration. Thus, eight of eleven AVMs (72%) have had complete obliteration as documented by angiography by 2 years and the remaining three lesions have decreased significantly (>75%) at 2 year and await their 2-year angiogram. We have not observed any serious side-effects of our treatment. One patient with a small AVM in the corpus callosum developed headaches and lethargy 7 months after receiving 2000 cGy with a 2 cm collimator normalized to the 80% isodose line. An IV contrast enhanced CT scan showed a 1.5 cm enhancing ring corresponding to the exact treatment area with a great deal of surrounding cerebral edema (Fig. 3). The patient was treated with a short course of oral dexamethasone therapy and is currently free of symptoms. A repeat CT scan showed complete resolution of this enhancing area and repeat angiogram showed obliteration of the AVM. Another patient received 2000 cGy to an AVM within the pons with a 2 cm collimator normalized to the 80%
September 1989, Volume 17, Number 3
isodose line. Six months after therapy, the patient developed problems with coordination of the right lower extremity. This has slowly improved without therapy and the patient is performing well 17 months after therapy. Alopecia developed in one young child 1 month following treatment for a very superficial AVM of the left parietal lobe. The scalp in this region received approximately 400 cGy. Six months after radiation, the patient has full return of hair in the region. DISCUSSION The results from our initial experience demonstrate that stereotactic radiation therapy for AVMs with a standard linear accelerator is an effective and safe technique. As mentioned previously, we have been conservative in our selection of radiation doses. We have elected to investigate the minimum dose needed rather than the maximal dose tolerated in order to maximize the therapeutic index. Even with our conservative doses, our results are similar to those reported by other groups using various forms of stereotactic radiosurgery. Investigators from the Lawrence Berkeley Laboratory, Berkeley, have treated 130 patients with helium ion radiosurgery since 1980 (5). Doses biologically equivalent to 2500 cGy to 4500 cGy were given in one or two sessions. With a follow-up of over 2 years for the majority of patients, only four patients have bled. This hemorrhage rate was felt by the investigators to be well below the rate expected in an untreated population of AVM patients. Although the follow-up for these patients is short, no cases of symptomatic radiation injury have been reported. Kjellberg et al., from the Massachusetts General Hos-
Fig. 3. Lateral projection of angiogram showing a relatively small AVM located near the corpus callosum on the right side. The patient was treated with 2000 cGy prescribed at the 80% isodose line with a 20 mm collimator. The CT scan on the right was taken 7 months after the completion of treatment because of the development of headaches and lethargy showing an enhancing ring corresponding to the treatment volume with surrounding cerebral edema. After a short course of oral dexamethasone therapy, the symptoms resolved. Angiogram repeated at 13 months showed a completed resolution of the AVM and repeat CT showed a resolution of the enhancing pattern and edema.
Stereotactic radiation 0 J. S. LOEFF’LERet al.
the age of 10 have required general anesthesia for the duration of the procedure. Treatment usually takes 3045 minutes to complete once the machine has been set up.
615
RESULTS At the time of this analysis (July 1988), eleven of the 16 patients treated with stereotactic radiosurgery have had
Fig. 2. (A) Anterior and lateral projections of an angiogram of a patient in this series. The right parietaJ lesion was fed from both the anterior and middle cerebral circulations. The four arrows represent the paths of the four noncoplanar arcs used in treatment. (B) Anterior and lateral projections of the angiogram taken 1 year after treatment showing complete obliteration of the AVM. The patient was treated with a 30 mm collimator and the dose was 1500 cGy prescribed at the 80% isodose line.
Stereotactic radiation 0 J. S. LOEFFLER et al.
pital, reported the results of treating 439 patients with AVMs with proton radiosurgery (9, 10). Doses in the proton beam project have been lowered to 1000 cGy for large lesions (treated with a 50 mm beam) to 5000 cGy for small lesions (treated with a 7 mm beam). Eight patients died of hemorrhage within 18 months of proton radiosurgery, a rate considered to be consistent with the rate for untreated patients. Only one patient died of hemorrhage after the 18 month period. Patients presenting initially with hemorrhage bled at a rate of 7.3% per year prior to treatment and at a rate of 2.4% per year beyond 2 years of treatment (p < .005). Neurological deficits were seen in the initial patients treated with large fields and high doses, but the incidence was small. After the modification of radiation dose for the large lesions, complications have been uncommon and Kjellberg believes there is protection against hemorrhage after a latent period of 18-24 months following proton radiosurgery. Since 1970, 135 patients have been treated with the Leksell technique (“Gamma knife”) at the Karolinska Institute in Stockholm (19). The radiation dose for the majority of patients was between 5000 cGy and 10,000 cGy, but ranged from 3500 cGy to 12,500 cGy. Complete obliteration was achieved in 84% of this group as docu-
677
mented by angiography 2 years following treatment. Only two of the patients hemorrhaged after treatment, both within 6 months of therapy. Four of the 135 patients developed new neurological deficits, a rate considered acceptable considering the very high success rate of this program. In summary, for patients with AVMs, stereotactic radiation therapy or radiosurgery appears to protect against hemorrhage after a latent period of 12-24 months following treatment using any of the four techniques described (protons, helium ions, “Gamma knife,” linear accelerator). Symptomatic radiation injury or necrosis of the brain has been uncommon despite single doses that are considered extremely high by conventional radiation therapy standards. We have developed a relatively simple and inexpensive technique for stereotactic radiosurgery using a standard linear accelerator. Our preliminary clinical data suggest that this is an efficacious and safe technique with results that equal the more expensive and elaborate systems that are available to provide stereotactic radiation therapy (radiosurgery). Furthermore, it may be possible to use a relatively low radiation dose, although more experience is needed.
REFERENCES I. Ektti, 0. 0.; Derechinsky,
2.
3.
4.
5.
6.
7.
8.
9.
10.
V. E. Hyperselective encephalic irradiation with linear accelerator. Acta. Neurochirurgica. 33:385-390; 1984. Brown, R. D.; Wieber, D. 0.; Forbes, G.; O’Fallon, W. M.; Piepgras, D. G.; Marsh, W. R.; Maciunas, R. J. The natural history of unruptured intracranial arteriovenous malformations. J. Neurosurg. 68:352-357; 1988. Chierego, G.; Marchetti, C.; Avanzo, R. C.; Pozza, F.; Colombo, F. Dosimetric considerations of multiple arc stereotaxic radiotherapy. Radiother. Oncol. 12: 141-152; 1988. Colombo, F.; Benedetti, A.; Pozza, F., Avanzo, R. C.; Marchetti, C.; Chierego, G.; Zandardo, A. External stereotactic irradiation by linear accelerator. Neurosurg. 16: 154- 159; 1985. Fabrikant, J. I.; Lyman, J. T.; Frankel, K. A. Heavy chargedparticle Bragg peak radiosurgery for intracranial vascular disorder. Radiat. Res. 104:244-258; 1985. Forster, D. M. C.; Steiner, L.; Hakanson, S. Arteriovenous malformations of the brain. A long-term clinical study. J. Neurosurg. 37:562-570; 1972. Graf, C. J.; Perret, G. E.; Turner, J. C. Bleeding from cerebral arteriovenous malformations as part of their natural history. J. Neurosurg. 58:331-337; 1983. Hartmann, G. H.; Schlegel, W.; Stunn, V.; Kober, B.; Postyr, 0.; Lorenzo, W. J. Cerebral radiation surgery using a moving field irradiation of a linear accelerator facility. Int. J. Radiat. Oncol. Biol. Phys. 11:1185-l 192; 1985. KjelIberg, R. N. Bragg peak proton beam therapy for arteriovenous malformations of the brain. Clin. Neurosurg. 3 1: 248-268; 1983. Kjellberg, R. N.; Hanamura, T.; Davis, K. R.; Lyons, S. L.; Adams, R. D. Bragg-peak proton-beam therapy for arteriovenous malformations of the brain. N. Engl. J. Med. 309: 269-274; 1983.
11. Leksell, L. The stereotactic method and radiosurgery of the
brain. Acta. Chir. Stand. 102:3 16-3 19; 195 1. 12. Luessenhop, A. J. Natural history of cerebral arteriovenous malformations. In: Wislon, C. B., Stein, B. M., eds. Intracranial arteriovenous malformations. Baltimore: Williams and Wilkins; 1984: 12-23. 13. Lutz, W.; Winston, K. R.; Maleki, N. A system for stereotactic radiosurgery with a linear accelerator. Int. J. Radiat. Oncol. Biol. Phys. 14:373-381; 1988. summary 14. Paulsen, M. G. Arteriovenous malformations-a of 6 cases treated with radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 13:1553-1557; 1988. 15. Podgursak, E. B.; Oliver, A.; Pla, M.; Lefebvre, P. Y.; Hazel, J. Dynamic stereotactic radiosurgery. Int. J. Radiat. Oncol. Biol. Phys. 14:115-126; 1988. 16. Rice, R. K.; Hansen, J. L.; Svensson, G. K.; Siddon, R. L. Measurement of dose distributions in small beams of 6 MV x-rays. Phys. Med. Biol. 32: 1087-1099; 1987. 17. Saunders, W. M.; Winston, K. R.; Siddon, R. L.; Svensson, G. H.; Kijewski, P. K.; Rice, R. K.; Hansen, J. L.; Barth, N. H. Radiosurgery for arteriovenous malformations of the brain using a standard linear accelerator: rationale and technique. Int. J. Radiat. Oncol. Biol. Phys. 15:441-447; 1988. 18. Siddon, R. L.; Barth, N. H. Stereotactic localization of intracranial targets. Int. J. Radiat. Oncol. Biol. Phys. 13: 12411246; 1987. 19. Steiner, L. Treatment of arteriovenous malformations by radiosurgery. In: Wilson, C. B., Stein, B. M., eds. Intracranial arteriovenous malformations. Baltimore: Williams and Wilkins; 1984:295-3 13. 20. Winston, K. R.; Lutz, W. Linear accelerator as a tool for stereotactic radiosurgery. Neurosurgery 22:454-464; 1988.