- Email: [email protected]
Results of carbon ion radiotherapy in 152 patients
Int. J. Radiation Oncology Biol. Phys., Vol. 58, No. 2, pp. 631– 640, 2004 Copyright © 2004 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/04/$–see front matter
doi:10.1016/j.ijrobp.2003.09.041
ICTR 2003
Translational Research in Clinics
RESULTS OF CARBON ION RADIOTHERAPY IN 152 PATIENTS DANIELA SCHULZ-ERTNER, M.D.,* ANNA NIKOGHOSYAN,† CHRISTOPH THILMANN, M.D.,† THOMAS HABERER, PH.D.,‡ OLIVER JA¨ KEL, PH.D.,† CHRISTIAN KARGER, PH.D.,† GERHARD KRAFT, PH.D.,‡ MICHAEL WANNENMACHER, M.D., D.D.S.,* AND JU¨ RGEN DEBUS, M.D., PH.D.*† *Department of Clinical Radiology, University of Heidelberg, Heidelberg, Germany; †German Cancer Research Center (dkfz), Heidelberg, Germany; ‡Department of Biophysics, GSI (Gesellschaft fu¨r Schwerionenforschung mbH), Darmstadt, Germany Purpose: This study summarizes the experience with raster scanned carbon ion radiation therapy (RT) at the Gesellschaft fu¨r Schwerionenforschung (GSI), Darmstadt, Germany since 1997. Methods and Materials: Between December 1997 and December 2002, 152 patients were treated at GSI with carbon ion RT. Eighty-seven patients with chordomas and low-grade chondrosarcomas of the skull base received carbon ion RT alone (median dose 60 GyE); 21 patients with unfavorable adenoid cystic carcinomas and 17 patients with spinal (n ⴝ 9) and sacrococcygeal (n ⴝ 8) chordomas and chondrosarcomas were treated with combined photon and carbon ion RT. Twelve patients received reirradiation with carbon ions with or without photon RT for recurrent tumors. Furthermore, 15 patients with skull base tumors other than chordoma and low-grade chondrosarcoma were treated with carbon ions. Results: Actuarial 3-year local control was 81% for chordomas, 100% for chondrosarcomas, and 62% for adenoid cystic carcinomas. Local control was obtained in 15/17 patients with spinal (8/9) and sacral (7/8) chordomas or chondrosarcomas and in 11/15 patients with skull base tumors other than chordomas and low-grade chondrosarcomas, respectively. Six of 12 patients who received reirradiation are still alive without signs of tumor progression. Common Toxicity Criteria Grade 4 or Grade 5 toxicity was not observed. Conclusion: Carbon ion therapy is safe with respect to toxicity and offers high local control rates for skull base tumors such as chordomas, low-grade chondrosarcomas, and unfavorable adenoid cystic carcinomas. © 2004 Elsevier Inc. Carbon ion radiotherapy, Chordomas, Chondrosarcomas, Adenoid cystic carcinomas.
Compared with photons, heavy charged particles such as protons or carbon ions provide a higher physical selectivity because of their finite range in tissue and, in the case of carbon ions, biologic advantages due to an increased relative biologic effectiveness (RBE) within the Bragg peak. These advantages lead to improved dose distributions, permitting dose escalation within the target and an optimal sparing of adjacent normal tissues. Therefore, radiotherapy with heavy charged particles suggests a clinical gain in the treatment of tumors characterized by poor radiosensitivity and critical location. Taking into account the favorable physical and biologic properties of carbon ions, a clinical benefit by means of improved cure rates and reduction of toxicity can be antic-
ipated for specific tumor entities. Potential indications for carbon ion radiation therapy (RT) are tumors with a low a/ ratio such as chordomas, low-grade chondrosarcomas, and adenoid cystic carcinomas. Furthermore, carbon ion RT might be advantageous in the treatment of bone and soft tissue sarcomas, lung cancer, and prostate cancer. However, clinical experience with carbon ion RT in the management of these tumors is still limited. Between 1977 and 1992, 223 patients with skull base tumors were treated with heavy ions at the Lawrence Berkeley Laboratory in Berkeley, California. Most patients were treated with helium or neon ions for skull base tumors (1). Charged particles heavier than protons are currently available at three centers worldwide. There are two hospitalbased facilities in Japan ( the National Institute of Radio-
Reprint requests to: Daniela Schulz-Ertner, M.D., Dept. of Clinical Radiology, University of Heidelberg, INF 400, 69120 Heidelberg, Germany. Tel: (⫹49) 6221-568201; Fax: (⫹49) 6221565353; E-mail: [email protected] Presented at ICTR 2003, Lugano, Switzerland, March 16 –19, 2003. Supported in part by a grant from the BMBF #01ZP9601/8.
Acknowledgments—The authors thank S. Kuhn, A. Joedicke, E. Rittinghausen, and K. Kuhn for their excellent technical support. This work is dedicated to Harald zur Hausen on the occasion of his retirement as head of the German Cancer Research Center (Deutsches Krebsforschungszentrum) in Heidelberg with gratitude and appreciation for 20 years of leadership. Received Sep 8, 2003. Accepted for publication Sep 15, 2003.
INTRODUCTION
631
632
I. J. Radiation Oncology
● Biology ● Physics
Volume 58, Number 2, 2004
Table 1. Histology and treatment modalities of 152 patients Histology Chordoma Skull base Cervical spine Sacrum Low-grade chondrosarcoma Skull base Cervical spine Adenoid cystic carcinoma Other skull base tumors Meningioma/WHO grade 2 Meningioma/WHO grade 3 Myoblastoma Malignant schwannoma Chondroblastoma High grade Chondrosarcoma Squamous carcinoma Undifferentiated, not other classified tumor Reirradiations Skull base chordoma Adenoid cystic carcinoma Meningioma/WHO grade 3
logic Science in Chiba and the Hyogo Ion Beam Medical Center in Hyogo). At both facilities, carbon ion therapy is provided throughout the year using passive beam shaping. At the facility in Chiba, carbon ion therapy is mainly offered to patients with head-and-neck tumors, bone and soft tissue sarcomas, tumors of the liver, lung cancer, and prostate cancer. Since 1994 more than 1000 patients have been treated with carbon ions. Initial results in the treatment of some of these tumors were promising. Excellent results were reported especially for Stage I lung cancer, bone and soft tissue sarcomas, prostate cancer, and malignant salivary gland tumors (2). At the Gesellschaft fu¨ r Schwerionenforschung (GSI) in Darmstadt, Germany, carbon ion RT has been available for patient treatments since December 1997. GSI is a worldwide cooperating basic physics institute. The heavy ion synchrotron (SIS) at GSI is capable of accelerating heavy charged particles up to uranium. The carbon ion beam is available for RT within three treatment blocks, with each consisting of 20 consecutive days. The responsibility for patient treatments rests with the University of Heidelberg. Beam delivery is performed with the intensity-controlled raster scan technique (3). Compared with passive dose delivery techniques, the dose to the entrance channel as well as to the exit channel of the beam is reduced, which further contributes to target conformality. The raster scan technique together with biologic plan optimization (4 –7) for carbon ion RT promises to be an optimal tool for the treatment of skull base lesions and spinal tumors which are characterized by a close vicinity to radiosensitive organs at risk. This study summarizes and updates our clinical results with carbon ion RT in the treatment of patients with skull base
No.
RT modality
54 8 8
C-ions C-ions or C-ions ⫹ photon RT C-ions ⫹ photon RT
33 1 21
C-ions C-ions C-ions ⫹ photon RT ⫹ ⫹ ⫹ ⫹
2 6 1 2 1 1 1 1
C-ions C-ions C-ions C-ions C-ions C-ions C-ions C-ions
photon photon photon photon
RT RT RT RT
7 3 2
C-ions or C-ions ⫹ photon RT C-ions or C-ions ⫹ photon RT C-ions or C-ions ⫹ photon RT
⫹ photon RT
tumors and spinal/sacral chordomas and chondrosarcomas between 1997 and 2002.
METHODS AND MATERIALS Patient characteristics Between December 1997 and December 2002, 152 patients were treated with carbon ions at GSI. Most patients had tumors originating from the skull base or tumors infiltrating the skull base. Median age was 53 years (range, 18 – 80 years). Table 1 summarizes the patient numbers treated with respect to histology and treatment modalities. The feasibility study and a clinical Phase I/II trial for carbon ion RT in skull base chordomas and low-grade chondrosarcomas have already been completed. The Phase I/II trials for combined photon RT and a carbon ion boost for extracranial chordomas and chondrosarcomas and locally advanced adenoid cystic carcinomas are still ongoing. Chordomas and low-grade chondrosarcomas of the skull base Since December 1997, we have treated 54 patients with histologically proven chordoma and 33 patients with lowgrade chondrosarcoma of the skull base. The first 67 patients (44 chordomas and 23 low-grade chondrosarcomas) were included in a clinical Phase I/II trial. Patients with history of a former irradiation with photons or protons were excluded from the study. Since completion of the trial in December 2001, patients with chordomas and low-grade chondrosarcomas of the skull base are offered carbon ion RT as an alternative to proton RT at foreign centers. Since this time, we have treated 20 further patients with chordomas and chondrosarcomas of the skull base. The presented
Carbon ion RT
● D. SCHULZ-ERTNER et al.
analysis was performed for the 67 patients entered into the clinical Phase I/II study. The prescribed median total tumor dose was 60 GyE (weekly fractionation 7 ⫻ 3.0 GyE, 10 patients were treated to 70 GyE) with carbon ion RT alone. Forty-two of 67 evaluable patients were treated for primary tumors; 25 patients were treated for recurrent tumors after surgery as initial therapy modality in the primary situation. Locally advanced adenoid cystic carcinomas Twenty-one patients with unfavorable, locally advanced adenoid cystic carcinoma were treated within an ongoing clinical Phase I/II study. Only patients with infiltration of the skull base and macroscopic tumor residual after surgery without history of a former irradiation were considered for participation in the study. Therapy consisted of combined stereotactically guided photon RT to the clinical target volume (CTV) (median CTV dose 54 Gy, weekly fractionation 5 ⫻ 1.8 Gy) and a carbon ion boost to the macroscopic tumor residual (median boost dose 18 GyE, weekly fractionation 6 ⫻ 3.0 GyE). Thirteen patients were treated for primary tumors, and 8 patients received combined photon RT with a carbon ion boost for recurrent tumors after radical surgery in the first-line treatment. Two patients were treated after resection of lung metastases without evidence of active pulmonary disease at the time of presentation for RT. One patient presented with solitary lymph node metastases. Spinal and sacrococcygeal chordoma and low-grade chondrosarcoma We treated 8 patients with chordoma of the cervical spine, 1 patient with low-grade chondrosarcoma of the cervical spine, and 8 patients with sacrococcygeal chordoma within an ongoing clinical Phase I/II study. Treatment consisted of combined photon intensity-modulated radiotherapy (IMRT) and a carbon ion boost to the macroscopic tumor residual after surgery. Surgery aimed in maximum removal of the tumor. The prescribed target dose and the ratio between photons and carbon ions was chosen with respect to the tumor site and the distance between tumor and radiosensitive structures such as the spinal cord. Median photon dose was 50.4 Gy and carbon ion boost dose was 18 GyE. Patients with metal implants within the irradiation field were excluded from the study. Nevertheless, 1 patient with metal implants outside the carbon ion boost volume was considered to be a candidate for combined photon RT and a carbon ion boost. Other skull base tumors Fifteen patients with skull base tumors other than chordomas and low-grade chondrosarcomas were treated with carbon ion RT within a feasibility study. Patients with atypical or malignant meningioma (8), malignant peripheral schwannoma (2), high grade chondrosarcoma (1), chondroblastoma (1), myoblastoma (1), squamous carcinoma (1), and 1 patient with undifferentiated, not otherwise classified skull base tumor were treated with carbon ion RT alone or with combined photon RT and a carbon ion boost.
633
Reirradiation of skull base and spinal/sacral tumors Reirradiation for recurrent tumors after a former course of irradiation with photons (10) or protons (2) was delivered in 12 patients. We treated 6 patients with skull base chordoma, 1 patient with a recurrent sacral chordoma, 3 patients with recurrent adenoid cystic carcinoma, and 2 patients with recurrent atypical meningioma. Patients with recurrent skull base chordoma received a median total tumor dose of 51 GyE (range 45 to 60 GyE) with carbon ions alone. Of the 3 patients with adenoid cystic carcinoma, 1 patient received a total tumor dose of 42 GyE with carbon ions alone. The other 2 patients received combined photon RT and a carbon ion boost with target doses of 47.4 GyE and 63 GyE, respectively. The patient with the recurrent sacral chordoma was treated with combined photon RT and a carbon ion boost (total target dose 40 Gy photons ⫹ 18 GyE carbon ions). One of the meningioma patients received carbon ion RT alone (prescribed target dose 45 GyE); the second patient received combined photon and carbon ion RT (prescribed target dose 36 Gy photons plus 18 GyE carbon ions). Treatment planning and delivery Patients with skull base tumors were immobilized with a precision head mask that ensured a repositioning accuracy of 1–2 mm (8). Patients with extracranial lesions were immobilized within a rigid immobilization device consisting of a wrap-around body mask and a precision head mask (Fig. 1) developed by Lohr et al. (9). Target point localization and positioning was performed using stereotactic methods. All patients received three-dimensional (3D) treatment planning based on computed tomographic (CT) and magnetic resonance imaging (MRI) scans. Dose calculation was based on a 3D-CT data set of continuous 3-mm CT slices. For delineation of the target volumes and organs at risk, the CT images were stereotactically correlated with T2weighted and contrast-enhanced T1-weighted MRI images. Target volumes included a safety margin of 1–2 mm for possible patient misalignment and uncertainties in particle range for skull base tumors and a margin of 3 to 5 mm for extracranial lesions. For setup verification during carbon ion RT, daily orthogonal X-rays were performed and compared with digitally reconstructed radiographs. Setup errors larger than 1 mm in the skull base region and 3 mm in extracranial tumor sites were corrected before RT. Treatment planning for carbon ions was performed using the treatment planning programs VOXELPLAN and TRiP and included biologic plan optimization taking into account the local RBE values. The planning procedure, including biologic plan optimization, as well as the quality assurance program are described elsewhere (4 –7, 10). The aim of biologic plan optimization was to achieve a homogenous biologically effective dose (Deff ⫽ RBE ⫻ absorbed dose) within the target volume. The calculated mean RBE was about 3 within the target volume. Values for RBE show large variances within the target volume and the different tissues. This was considered in the optimization process.
634
I. J. Radiation Oncology
● Biology ● Physics
Volume 58, Number 2, 2004
Fig. 1. Individually manufactured rigid immobilization device in a patient with sacral chordoma.
The dose constraints for specific organs at risk were 54 GyE for optic chiasm and optic nerves. The brainstem received less than 50 GyE at the central portion, but very small volumes of less than 1 cc at the surface were allowed to receive up to 60 GyE. The dose to the spinal cord was constrained to 45 GyE. The rectal wall was not included in the gross tumor volume (GTV) for treatment planning in sacral chordomas to minimize the carbon ion dose to the rectum. Carbon ion treatment plans consisted of 2– 4 isocentric fields and were applied using the horizontal beam line. The intensity-controlled raster scan system with active energy variation was used for beam delivery and allowed for active beam shaping without any additional compensators (3). Online verification of the particle range was realized by measurement of the positron emitter distributions during treatment and by comparison with a simulated emitter distribution based on the absorbed dose distributions (11).
Follow-up and statistical analysis All patients included in studies had follow-up examinations including neurologic examinations and MRI scans performed 6 weeks after RT, in 3-month to 6-month intervals during the first 2 years, and annually thereafter. Testing of visual and auditory function as well as endocrinologic examinations were regularly performed in patients at risk. Actuarial rates for local control, disease-free survival, and overall survival were calculated from the onset of RT using Kaplan-Meier methods (12). Local control was defined as absence of clinical or radiographic signs of tumor enlargement. Acute and late side effects of carbon ion RT were scored according to the Common Toxicity Criteria (CTC) (13).
RESULTS Chordomas and low-grade chondrosarcomas of the skull base The results for 67 patients with chordomas and low-grade chondrosarcomas treated within the clinical Phase I/II study were updated in February 2003. Median follow-up was 20 months (range 3 to 54 months). The actuarial 3-year local control rate was 100% for chondrosarcomas and 81% for chordomas of the skull base, respectively (Fig. 2). We observed four local recurrences after a full course of carbon ion RT in patients with chordomas. Two of the local recurrences occurred at the field margin 13 and 20 months after RT, respectively. Correlation with the treatment plan revealed that a lower dose had been applied because of normal tissue constraints. One patient developed a subcutaneous recurrence outside the former irradiation field within the surgical scar 6 months after RT, and 1 patient was diagnosed with an in-field recurrence 27 months after carbon ion RT. One patient developed spinal seeding 18 months after RT. Actuarial 3-year overall survival was 91% for chordomas and chondrosarcomas of the skull base (Fig. 3). One patient with local recurrence of a chordoma infiltrating the pons died of tumor progression 22 months after RT. Three patients died of intercurrent disease without signs of tumor progression on follow-up MRI scans. Objective tumor response by means of partial tumor remission was observed in about one-third of the chordoma patients, whereas it was a rare finding in chondrosarcoma patients. Toxicity was mild. Seven of 67 patients developed contrast-enhancing lesions within the temporal lobes on follow-up T1-weighted MRI. The lesions were circumscribed and in close vicinity to the target volume. Correlation with the treatment plans revealed that 5 of the 7 patients
Carbon ion RT
● D. SCHULZ-ERTNER et al.
635
Fig. 2. Actuarial local control by histology in 44 patients with chordomas and 23 with low-grade chondrosarcomas of the skull base treated with carbon ion RT (Kaplan-Meier curve). Chondrosarcomas ⫽ solid line; chordomas ⫽ dotted line.
who developed temporal lobe injury received a total target dose of 70 GyE with a weekly fractionation of 7 ⫻ 3.5 GyE. Contrast-enhancing lesions within the temporal lobes were observed in 5 of 10 patients (50%) who received a target dose of 70 GyE, but were found in only 2 of 57 patients (3.5%) who received target doses between 57 and 65 GyE. Four of the patients with temporal lobe injury did not show
any clinical symptoms attributable to the lesions, and the lesions diminished or remained stable without the need for steroid medication or surgery. In 3 patients, the lesions were associated with acute clinical symptoms such as cognitive deficits, weakness, and seizures requiring long-time steroid medication. Two other patients developed contrast-enhancing lesions
Fig. 3. Actuarial overall survival in 67 patients with chordomas and low-grade chondrosarcomas of the skull base treated with carbon ions (Kaplan-Meier curve).
636
I. J. Radiation Oncology
● Biology ● Physics
Volume 58, Number 2, 2004
Table 2. Severe side effects of carbon ion radiotherapy CH/CS of the skull base
Temporal lobe injury Optic nerve neuropathy Myelitis Rectum/small bowel damage
70 GyE
57–65 GyE
ACC
Spinal CH/CS
Sacral CH/CS
5/10 0/10 0/0 —
2/57 2/57 0/57
0/21 0/21 0/21 —
— — 0/9 0/9
— — 0/8 0/8
Abbreviations: ACC ⫽ adenoid cystic carcinomas; CH ⫽ chordomas; CS ⫽ chondrosarcomas; — ⫽ not at risk.
within the optic pathway. In both patients the optic tract was included in the target volume unilaterally because of the tumor compressing it. The contrast-enhancing lesions were strictly confined to the target volume. Both patients developed homonymous hemianopsia which resolved with steroid medication in 1 patient and remained stable in the second patient. We did not observe late toxicity greater than CTC Grade 3. Severe side effects of RT are displayed in Table 2. Adenoid cystic carcinomas with infiltration of the skull base Median follow-up was 14 months (range 3– 44 months). Eight of 21 patients received a reduced photon dose of 45 to 52.2 Gy instead of 54 Gy to respect the tolerance doses of neighboring organs at risk. Since August 2001, intensitymodulated photon RT has been used for the photon treatment. The desired target dose of 54 Gy to the CTV could be delivered in all 14 patients who received their photon treatment as inversely planned photon IMRT. During the follow-up period, we observed four locore-
gional recurrences. Two patients developed a recurrence within the GTV 5 and 9 months after RT, respectively. In 1 patient, locoregional recurrence within the GTV occurred 11 months after RT and was associated with synchronous development of spinal seeding. The patient was known to have lung metastases treated with surgical removal before irradiation of the primary tumor. One patient developed a recurrence within the CTV and outside the GTV 24 months after RT. The 2 patients with recurrences outside the GTV received reirradiation. The actuarial locoregional control rate was 62% at 3 years (Fig. 4). Distant metastases progressed in both patients who presented with lung metastases before RT 11 and 43 months after RT. A third patient developed bone metastases 14 months after RT. Actuarial disease-free survival was 40% at 3 years (Fig. 5). Two of the 21 patients died of tumor progression 15 and 25 months after RT, respectively. Actuarial overall survival was 75% at 3 years (Fig. 6). Acute severe side effects CTC Grade 3 were observed in 2/21 patients (9.5%) and included an abscess within the
Fig. 4. Actuarial locoregional control in 21 patients with locally advanced adenoid cystic carcinoma treated with combined photon and carbon ion RT (Kaplan-Meier curve).
Carbon ion RT
● D. SCHULZ-ERTNER et al.
637
Fig. 5. Actuarial disease-specific survival in 21 patients with locally advanced adenoid cystic carcinoma treated with combined photon and carbon ion RT (Kaplan-Meier curve).
surgical bed 3 weeks after completion of irradiation in 1 patient and a mucositis CTC Grade 3 in another patient. We have not observed treatment-related late effects greater than CTC Grade 2 thus far (Table 2). Spinal and sacrococcygeal chordomas and chondrosarcomas A locoregional recurrence occurred in 1/8 patients treated with combined photon RT to the sacrum and a
carbon ion boost to the macroscopic tumor. The recurrence was located outside the GTV but within the former CTV. Salvage therapy consisted of surgical removal. Two of 8 patients showed distant metastases 6 and 19 months after RT. One of these patients received reirradiation after partial resection of a solitary metastasis at the level of the third lumbal spine. The second patient who was treated for a locally advanced sacral chordoma (tumor volume almost 3 L) died of multiple metastases
Fig. 6. Actuarial overall survival in 21 patients with locally advanced adenoid cystic carcinoma treated with combined photon and carbon ion RT (Kaplan-Meier curve).
638
I. J. Radiation Oncology
● Biology ● Physics
10 months after RT of the primary. Seven of the 8 patients with sacral chordoma are still alive. Carbon ion RT alone or combined with photon RT yielded local control in 8 of 9 patients with chordoma (8 patients) or low-grade chondrosarcoma (1 patient) of the cervical spine. One patient treated with combined photon RT and a carbon ion boost to a total target dose of 58.5 GyE developed a local recurrence within the target volume 21 months after RT. All but 1 patient are still alive without signs of tumor progression. We observed mucositis CTC Grade 3 in 3 patients with chordomas of the cervical spine. None of our patients showed severe late effects to the spinal cord. Patients treated for sacral chordomas did not show any side effects. Other skull base tumors Local control was achieved in 7/8 patients with atypical (2 patients) and malignant (6 patients) meningioma treated postoperatively with combined photon RT and a carbon ion boost. One of 2 patients treated for malignant schwannoma developed distant metastases 4 months after RT of her primary while the primary tumor was still controlled. She died 20 months after RT of progressive metastatic disease. The second patient is still alive without signs of tumor progression 48 months after RT. In 3 of 5 patients treated for rare skull base tumors such as myoblastoma (0/1), squamous carcinoma (1/1), chondroblastoma (1/1), high-grade chondrosarcoma (0/1), and undifferentiated, not otherwise classified tumor (1/1), local control was achieved. Reirradiation of skull base and spinal/sacral tumors Reirradiation of 12 patients with skull base and spinal/ sacral tumors resulted in effective tumor control in 4/7 patients treated for chordoma, in 1/3 patients treated for recurrent adenoid cystic carcinoma, and in 1 of 2 patients with atypical meningioma. The median follow-up was 17 months (range 3– 45 months). Six of 12 patients (50%) are still alive without further tumor progression, and 2 patients are alive with progressive tumors. Of 4 patients who died during follow-up, 3 patients died of local tumor progression and 1 patient died of malignant teratoma while her primary chordoma had started growing again as well. DISCUSSION Chordomas and low-grade chondrosarcomas of the skull base Best results in the treatment of chordomas and chondrosarcomas of the skull base are reported for surgery with postoperative high-dose proton RT. The actuarial 5-year local control rate was 73% using tumor doses between 66 and 83 cobalt Gray equivalent (CGE) for 519 patients with skull base chordomas treated with protons at the Massachusetts General Hospital, Boston (14). The temporal lobe damage rate was 13.2% after 5
Volume 58, Number 2, 2004
years (15). Hug et al. reported an actuarial 3-year local control rate of 67% for 58 chordoma patients treated at the Loma Linda University Medical Center, Loma Linda, California, USA, with proton RT. Tumor doses ranged between 64.8 and 79.2 CGE. Grade 3 and 4 late toxicities occurred in 7% of the patients (16). Noel et al. reported the results of combined photon and proton RT at the Centre de Protonthe´ rapie d’Orsay in 45 patients with chordomas and chondrosarcomas of the skull base treated between December 1995 and December 1998. They delivered a median total tumor dose of 67 CGE (range 60 to 70 CGE). Photons represented two-thirds of the total dose and protons one-third. At 3 years, local control rates for chordomas and chondrosarcomas were 83.1% and 90%. In this study, 2 of 45 patients developed Grade 3 or Grade 4 toxicities (17). Experience using particle therapy with particles heavier than protons such as helium and carbon ions is limited. Between 1977 and 1992, Castro et al. treated 223 patients with helium ions (target dose 65 CGE) at the Lawrence Berkeley Laboratory in Berkeley, California. Actuarial 5-year local control rates were 63% for chordomas and 78% for chondrosarcomas, respectively. Radiation-induced complications occurred in 27% of the patients (1). Compared with helium ions, carbon ions offer potential biologic advantages. Our data indicate that carbon ion radiotherapy is highly effective in the treatment of chordomas and low-grade chondrosarcomas. Initial results in 37 patients were already encouraging with 2-year local control rates of 83% for chordomas and 100% for lowgrade chondrosarcomas, respectively (18). The updated data presented here support these preliminary results. Local control rates at 3 years are at least comparable to proton RT while severe radiation-induced side effects were minimized. A target dose of 60 GyE may be delivered with a low risk for severe side effects, whereas dose escalation has to be considered with caution, because late toxicity might be increased. Spinal and sacrococcygeal chordomas and low-grade chondrosarcomas Extracranial chordomas and low-grade chondrosarcomas are more difficult to treat compared with tumors of the skull base. Radiosensitive structures as well as larger setup errors limit the prescription dose. As a consequence of a suboptimal RT dose, local control rates have been poor. Best results were yielded with heavy charged particles. Schoenthaler et al. treated 14 patients with helium or neon ions at the Lawrence Berkeley Laboratory. Four of 14 patients were treated after gross tumor resection. They reported a 5-year local control rate of 55% with a trend for better local control for patients treated with neon ions compared with helium ions and for 4 patients who received RT after gross tumor resection (19). Berson et al. reported a 5-year local control rate of 54% for 10 patients with chordomas and chondrosarcomas of the
Carbon ion RT
● D. SCHULZ-ERTNER et al.
cervical spines treated with charged particles at the Lawrence Berkeley Laboratory (20). Similar results were obtained with protons in 20 patients with spinal and sacral chordomas (14 patients) and chondrosarcomas (6 patients) by Hug et al. at the Massachusetts General Hospital in Boston. They treated 5 patients with tumors of the thoracic spine, 5 patients with tumors of the lumbar spine, and 10 patients with sacral tumors. Five-year local control rates were 100% for chondrosarcomas and 53% for chordomas (16). Our results in 17 patients with sacral and spinal chordomas and chondrosarcomas demonstrate that combined photon and carbon ion RT is well tolerated and offers effective local control. Locally advanced unfavorable adenoid cystic carcinomas Surgery is the mainstay to therapy in adenoid cystic carcinoma, but adjuvant high-dose RT is generally recommended even after complete resection. Local control rates at 5 years exceed 90% after complete resection and postoperative RT (21), but remain poor after incomplete resection with macroscopic tumor residuals (22, 23). In the past, best results in this unfavorable group of adenoid cystic carcinoma have been achieved with neutrons. The advantage of neutrons over photon RT for malignant salivary gland tumors was proven within a randomized Phase III trial of the Radiation Therapy Oncology Group (RTOG) and the Medical Research Council (RTOG-MRC trial). Actuarial local control rates and overall survival rates at 2 years were 67% and 62% with neutrons compared with 17% and 25% with photons, respectively (24). At 10 years, local control was 56% for the neutron arm compared with 17% for patients treated with conventional RT, respectively. The improvement in local control was attributed to the high–linear energy transfer effect of neutrons, while survival rates showed no difference (25). The main disadvantage of neutron RT was the relatively high rate of severe late effects containing bone and soft tissue necrosis and fibrosis in up to 20% of the patients (26 –28). Our data indicate that postoperative combined photon IMRT and a carbon ion boost to the macroscopic tumor represents a therapy approach which offers not only high local control rates but also minimizes severe side effects. Nevertheless, disease-free survival remains poor due to a high rate of distant metastases in locally advanced adenoid cystic carcinomas which is rarely influenced with common chemotherapy regimes. Atypical and malignant meningioma Benign meningiomas of the skull base can be treated effectively with modern photon techniques. Actuarial 10-year local control rates in excess of 90% are yielded with fractionated stereotactic techniques even in large en plaque meningiomas of the skull base (29, 30). Delivery of fractionated stereotactic photon techniques is safe; severe late effects occurred in less than 5% of the patients
639
(29). On the other hand, treatment of atypical and malignant meningiomas with photons still remains unsatisfactory. Milosevic et al. reported a 5-year cause-specific survival rate of 34% in 59 patients with atypical or malignant meningioma treated with conventional photon RT (31). Hug et al. treated 31 patients with atypical/ malignant meningiomas with combined photon and proton RT or photons alone at the Massachusetts General Hospital in Boston. Significantly improved local control was observed for proton vs. photon RT (80% vs. 17% at 5 years). RT with target doses exceeding 60 Gy resulted in significantly better local control rates compared with doses ⬍60 Gy. The authors conclude that high-dose RT should be recommended in atypical/malignant meningiomas, which is rarely possible with photon RT (32). Our results show that the physical and biologic advantages of carbon ion RT can be utilized for the treatment of atypical/malignant meningiomas as well. With a combined approach using modern photon RT and a carbon ion boost, a high target dose of 68.4 Gy can safely be delivered. Reirradiation of skull base tumor Reirradiation of skull base tumors has rarely been considered in the past. Using raster scanned carbon ion RT, patients with recurrent tumors may be treated again with effective target doses because the dose gradient toward radiosensitive structures is steeper than with any other RT modality. The benefit of reirradiation of skull base tumors with carbon ions depends on the extension of the recurrent tumor, the distance to neighboring radiosensitive structures, and the dose applied to organs at risk during the first course of RT. Carbon ion RT might be considered in individual patients with circumscribed recurrences where effective carbon ion doses can be delivered to small volumes with an acceptable risk for radiation-induced damage to normal tissue. CONCLUSION With raster scanned carbon ion beams, high local control rates can be achieved with relatively low toxicity compared with photon and proton RT. Carbon ion RT may therefore be considered alternatively to proton RT which is currently the standard of care in patients with chordomas and chondrosarcomas of the skull base. A total target dose of 60 GyE delivered within 20 days proved to be the optimal prescription dose with respect to effectiveness and avoidance of radiation-induced side effects. A hospital-based facility will be built in Heidelberg, Germany until 2006 and will allow for prospective clinical Phase III studies comparing carbon ion RT with protons and modern photon RT such as intensity-modulated RT for specific tumors. Besides chordomas, low-grade chondrosarcomas, and malignant salivary gland tumors, more common tumors such as prostate cancer, lung cancer, and soft tissue sarcomas might be treated effectively with carbon ions.
640
I. J. Radiation Oncology
● Biology ● Physics
Volume 58, Number 2, 2004
REFERENCES 1. Castro JR, Linstadt DE, Bahary J-P, et al. Experience in charged particle irradiation of tumors of the skull base: 1977– 1992. Int J Radiat Oncol Biol Phys 1994;29:647–655. 2. Tsujii H, Morita S, Miyamoto T, et al. Experiences of carbon ion radiotherapy at NIRS. In: Kogelnik HD, Lukas P, Sedlmayer F, editors. Progress in radio-oncology VII. Bologna, Italy: Monduzzi Editore, 2002; p. 393–405. 3. Haberer T, Becher W, Schardt D, et al. Magnetic scanning system for heavy ion therapy. Nucl Instr Meth Phys Res 1993;330:296–305. 4. Kra¨ mer M, Ja¨ kel O, Haberer T, et al. Treatment planning for heavy ion radiotherapy: Physical beam model and dose optimization. Phys Med Biol 2000;45:3299–3317. 5. Kra¨ mer M, Scholz M. Treatment planning for heavy-ion radiotherapy: Calculation and optimization of biologically effective dose. Phys Med Biol 2000;45:3319–3330. 6. Ja¨ kel O, Hartmann GH, Karger CP, et al. Quality assurance for a treatment planning system in scanned ion beam therapy. Med Phys 2000;27:1588–1600. 7. Scholz M, Kellerer AM, Kraft-Weyrather W, et al. Computation of cell survival in heavy ion beams for therapy. The model and its approximation. Radiat Environ Biophys 1997; 36:59–66. 8. Karger CP, Ja¨ kel O, Debus J, et al. Three dimensional accuracy and interfractional reproducibility of patient fixation using a stereotactic head mask system. Int J Radiat Oncol Biol Phys 2001;49:1493–1504. 9. Lohr F, Debus J, Frank C, et al. Noninvasive patient fixation for extracranial stereotactic radiotherapy. Int J Radiat Oncol Biol Phys 1999;45:521–527. 10. Ja¨ kel O, Kra¨ mer M, Karger CP, et al. Treatment planning for heavy ion radiotherapy: Clinical implementation and application. Phys Med Biol 2001;46:1101–1116. 11. Enghardt W, Debus J, Haberer T, et al. The application of PET to quality assurance of heavy-ion tumor therapy. Strahlenther Onkol 1999;175(Suppl. 2):33–36. 12. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457–481. 13. National Cancer Institute (NCI). Common toxicity criteria. Bethesda, MD: Division of Cancer Treatment, National Cancer Institute; 1988. 14. Munzenrider JE, Liebsch NJ. Proton therapy for tumors of the skull base. Strahlenther Onkol 1999;175(Suppl. II):57–63. 15. Santoni R, Liebsch N, Finkelstein DM, et al. Temporal lobe (TL) damage following surgery and high-dose photon and proton irradiation in 96 patients affected by chordomas and chondrosarcomas of the base of the skull. Int J Radiat Oncol Biol Phys 1998;41:59–68. 16. Hug EB, Loredo LN, Slater JD, et al. Proton radiation therapy for chordomas and chondrosarcomas of the skull base. J Neurosurg 1999;91:432–439. 17. Noel G, Habrand JL, Mammar H, et al. Combination of photon and proton radiation therapy for chordomas and chondrosarcomas of the skull base: The Centre de Protonthe´ rapie d’Orsay experience. Int J Radiat Oncol Biol Phys 2001;51: 392–398.
18. Schulz-Ertner D, Haberer T, Ja¨ kel O, et al. Radiotherapy for chordomas and low-grade chondrosarcomas of the skull base with carbon ions. Int J Radiat Oncol Biol Phys 2002;53:36– 42. 19. Schoenthaler R, Castro JR, Petti PL, et al. Charged particle irradiation of sacral chordomas. Int J Radiat Oncol Biol Phys 1993;26:291–298. 20. Berson AM, Castro JR, Petti P, et al. Charged particle irradiation of chordoma and chondrosarcoma of the base of skull and cervical spine: The Lawrence Berkeley Laboratory experience. Int J Radiat Oncol Biol Phys 1988;15:559–565. 21. Garden AS, Weber RS, Morrison WH, et al. The influence of positive margins and nerve invasion in adenoid cystic carcinoma of the head and neck treated with surgery and radiation. Int J Radiat Oncol Biol Phys 1995;32:619–626. 22. Regine WF, Mendenhall WM, Parsons JT, et al. Radiotherapy for adenoid cystic carcinoma of the palate. Head Neck 1993; 15:241–244. 23. Vikram B, Strong EW, Shah JP, et al. Radiation therapy in adenoid cystic carcinoma. Int J Radiat Oncol Biol Phys 1984; 10:221–223. 24. Griffin TW, Pajak TF, Laramore GE, et al. Neutron vs photon irradiation of inoperable salivary gland tumors: Results of an RTOG-MRC cooperative randomized study. Int J Radiat Oncol Biol Phys 1988;15:1085–1090. 25. Laramore GE, Krall JM, Griffin TW, et al. Neutron versus photon irradiation for unresectable salivary gland tumors: Final report of an RTOG-MRC randomized clinical trial. Radiation Therapy Oncology Group. Medical Research Council. Int J Radiat Oncol Biol Phys 1993;27:235–240. 26. Saroja KR, Mansell J, Hendrickson FR, et al. An update on malignant salivary gland tumors treated with neutrons at Fermilab. Int J Radiat Oncol Biol Phys 1987;13:1319–1325. 27. Schultheiss TE, Cohen L, Mansell J. Normal tissue reactions and complications following high-energy neutron beam therapy II: Complication rates adjusted for censoring. Int J Radiat Oncol Biol Phys 1990;18:165–171. 28. Buchholz TA, Laramore GE, Griffin BR, et al. The role of fast neutron radiation therapy in the management of advanced salivary gland malignant neoplasms. Cancer 1992;69:2779– 2788. 29. Debus J, Wuendrich M, Pirzkall A, et al. High efficacy of fractionated stereotactic radiotherapy of large base-of-skull meningiomas: Long-term results. J Clin Oncol 2001;19:3547– 3553. 30. Pirzkall A, Debus J, Haering P, et al. Intensity modulated radiotherapy (IMRT) for recurrent, residual, or untreated skull-base meningiomas: Preliminary clinical experience. Int J Radiat Oncol Biol Phys 2003;55:362–372. 31. Milosevic MF, Frost PJ, Laperriere NJ, et al. Radiotherapy for atypical or malignant intracranial meningioma. Int J Radiat Oncol Biol Phys 1996;34:817–822. 32. Hug EB, DeVries A, Thornton AF, et al. Management of atypical and malignant meningiomas: Role of high-dose, 3Dconformal radiation therapy. J Neurooncol 2000;48:151–160.
doi:10.1016/j.ijrobp.2003.09.041
ICTR 2003
Translational Research in Clinics
RESULTS OF CARBON ION RADIOTHERAPY IN 152 PATIENTS DANIELA SCHULZ-ERTNER, M.D.,* ANNA NIKOGHOSYAN,† CHRISTOPH THILMANN, M.D.,† THOMAS HABERER, PH.D.,‡ OLIVER JA¨ KEL, PH.D.,† CHRISTIAN KARGER, PH.D.,† GERHARD KRAFT, PH.D.,‡ MICHAEL WANNENMACHER, M.D., D.D.S.,* AND JU¨ RGEN DEBUS, M.D., PH.D.*† *Department of Clinical Radiology, University of Heidelberg, Heidelberg, Germany; †German Cancer Research Center (dkfz), Heidelberg, Germany; ‡Department of Biophysics, GSI (Gesellschaft fu¨r Schwerionenforschung mbH), Darmstadt, Germany Purpose: This study summarizes the experience with raster scanned carbon ion radiation therapy (RT) at the Gesellschaft fu¨r Schwerionenforschung (GSI), Darmstadt, Germany since 1997. Methods and Materials: Between December 1997 and December 2002, 152 patients were treated at GSI with carbon ion RT. Eighty-seven patients with chordomas and low-grade chondrosarcomas of the skull base received carbon ion RT alone (median dose 60 GyE); 21 patients with unfavorable adenoid cystic carcinomas and 17 patients with spinal (n ⴝ 9) and sacrococcygeal (n ⴝ 8) chordomas and chondrosarcomas were treated with combined photon and carbon ion RT. Twelve patients received reirradiation with carbon ions with or without photon RT for recurrent tumors. Furthermore, 15 patients with skull base tumors other than chordoma and low-grade chondrosarcoma were treated with carbon ions. Results: Actuarial 3-year local control was 81% for chordomas, 100% for chondrosarcomas, and 62% for adenoid cystic carcinomas. Local control was obtained in 15/17 patients with spinal (8/9) and sacral (7/8) chordomas or chondrosarcomas and in 11/15 patients with skull base tumors other than chordomas and low-grade chondrosarcomas, respectively. Six of 12 patients who received reirradiation are still alive without signs of tumor progression. Common Toxicity Criteria Grade 4 or Grade 5 toxicity was not observed. Conclusion: Carbon ion therapy is safe with respect to toxicity and offers high local control rates for skull base tumors such as chordomas, low-grade chondrosarcomas, and unfavorable adenoid cystic carcinomas. © 2004 Elsevier Inc. Carbon ion radiotherapy, Chordomas, Chondrosarcomas, Adenoid cystic carcinomas.
Compared with photons, heavy charged particles such as protons or carbon ions provide a higher physical selectivity because of their finite range in tissue and, in the case of carbon ions, biologic advantages due to an increased relative biologic effectiveness (RBE) within the Bragg peak. These advantages lead to improved dose distributions, permitting dose escalation within the target and an optimal sparing of adjacent normal tissues. Therefore, radiotherapy with heavy charged particles suggests a clinical gain in the treatment of tumors characterized by poor radiosensitivity and critical location. Taking into account the favorable physical and biologic properties of carbon ions, a clinical benefit by means of improved cure rates and reduction of toxicity can be antic-
ipated for specific tumor entities. Potential indications for carbon ion radiation therapy (RT) are tumors with a low a/ ratio such as chordomas, low-grade chondrosarcomas, and adenoid cystic carcinomas. Furthermore, carbon ion RT might be advantageous in the treatment of bone and soft tissue sarcomas, lung cancer, and prostate cancer. However, clinical experience with carbon ion RT in the management of these tumors is still limited. Between 1977 and 1992, 223 patients with skull base tumors were treated with heavy ions at the Lawrence Berkeley Laboratory in Berkeley, California. Most patients were treated with helium or neon ions for skull base tumors (1). Charged particles heavier than protons are currently available at three centers worldwide. There are two hospitalbased facilities in Japan ( the National Institute of Radio-
Reprint requests to: Daniela Schulz-Ertner, M.D., Dept. of Clinical Radiology, University of Heidelberg, INF 400, 69120 Heidelberg, Germany. Tel: (⫹49) 6221-568201; Fax: (⫹49) 6221565353; E-mail: [email protected] Presented at ICTR 2003, Lugano, Switzerland, March 16 –19, 2003. Supported in part by a grant from the BMBF #01ZP9601/8.
Acknowledgments—The authors thank S. Kuhn, A. Joedicke, E. Rittinghausen, and K. Kuhn for their excellent technical support. This work is dedicated to Harald zur Hausen on the occasion of his retirement as head of the German Cancer Research Center (Deutsches Krebsforschungszentrum) in Heidelberg with gratitude and appreciation for 20 years of leadership. Received Sep 8, 2003. Accepted for publication Sep 15, 2003.
INTRODUCTION
631
632
I. J. Radiation Oncology
● Biology ● Physics
Volume 58, Number 2, 2004
Table 1. Histology and treatment modalities of 152 patients Histology Chordoma Skull base Cervical spine Sacrum Low-grade chondrosarcoma Skull base Cervical spine Adenoid cystic carcinoma Other skull base tumors Meningioma/WHO grade 2 Meningioma/WHO grade 3 Myoblastoma Malignant schwannoma Chondroblastoma High grade Chondrosarcoma Squamous carcinoma Undifferentiated, not other classified tumor Reirradiations Skull base chordoma Adenoid cystic carcinoma Meningioma/WHO grade 3
logic Science in Chiba and the Hyogo Ion Beam Medical Center in Hyogo). At both facilities, carbon ion therapy is provided throughout the year using passive beam shaping. At the facility in Chiba, carbon ion therapy is mainly offered to patients with head-and-neck tumors, bone and soft tissue sarcomas, tumors of the liver, lung cancer, and prostate cancer. Since 1994 more than 1000 patients have been treated with carbon ions. Initial results in the treatment of some of these tumors were promising. Excellent results were reported especially for Stage I lung cancer, bone and soft tissue sarcomas, prostate cancer, and malignant salivary gland tumors (2). At the Gesellschaft fu¨ r Schwerionenforschung (GSI) in Darmstadt, Germany, carbon ion RT has been available for patient treatments since December 1997. GSI is a worldwide cooperating basic physics institute. The heavy ion synchrotron (SIS) at GSI is capable of accelerating heavy charged particles up to uranium. The carbon ion beam is available for RT within three treatment blocks, with each consisting of 20 consecutive days. The responsibility for patient treatments rests with the University of Heidelberg. Beam delivery is performed with the intensity-controlled raster scan technique (3). Compared with passive dose delivery techniques, the dose to the entrance channel as well as to the exit channel of the beam is reduced, which further contributes to target conformality. The raster scan technique together with biologic plan optimization (4 –7) for carbon ion RT promises to be an optimal tool for the treatment of skull base lesions and spinal tumors which are characterized by a close vicinity to radiosensitive organs at risk. This study summarizes and updates our clinical results with carbon ion RT in the treatment of patients with skull base
No.
RT modality
54 8 8
C-ions C-ions or C-ions ⫹ photon RT C-ions ⫹ photon RT
33 1 21
C-ions C-ions C-ions ⫹ photon RT ⫹ ⫹ ⫹ ⫹
2 6 1 2 1 1 1 1
C-ions C-ions C-ions C-ions C-ions C-ions C-ions C-ions
photon photon photon photon
RT RT RT RT
7 3 2
C-ions or C-ions ⫹ photon RT C-ions or C-ions ⫹ photon RT C-ions or C-ions ⫹ photon RT
⫹ photon RT
tumors and spinal/sacral chordomas and chondrosarcomas between 1997 and 2002.
METHODS AND MATERIALS Patient characteristics Between December 1997 and December 2002, 152 patients were treated with carbon ions at GSI. Most patients had tumors originating from the skull base or tumors infiltrating the skull base. Median age was 53 years (range, 18 – 80 years). Table 1 summarizes the patient numbers treated with respect to histology and treatment modalities. The feasibility study and a clinical Phase I/II trial for carbon ion RT in skull base chordomas and low-grade chondrosarcomas have already been completed. The Phase I/II trials for combined photon RT and a carbon ion boost for extracranial chordomas and chondrosarcomas and locally advanced adenoid cystic carcinomas are still ongoing. Chordomas and low-grade chondrosarcomas of the skull base Since December 1997, we have treated 54 patients with histologically proven chordoma and 33 patients with lowgrade chondrosarcoma of the skull base. The first 67 patients (44 chordomas and 23 low-grade chondrosarcomas) were included in a clinical Phase I/II trial. Patients with history of a former irradiation with photons or protons were excluded from the study. Since completion of the trial in December 2001, patients with chordomas and low-grade chondrosarcomas of the skull base are offered carbon ion RT as an alternative to proton RT at foreign centers. Since this time, we have treated 20 further patients with chordomas and chondrosarcomas of the skull base. The presented
Carbon ion RT
● D. SCHULZ-ERTNER et al.
analysis was performed for the 67 patients entered into the clinical Phase I/II study. The prescribed median total tumor dose was 60 GyE (weekly fractionation 7 ⫻ 3.0 GyE, 10 patients were treated to 70 GyE) with carbon ion RT alone. Forty-two of 67 evaluable patients were treated for primary tumors; 25 patients were treated for recurrent tumors after surgery as initial therapy modality in the primary situation. Locally advanced adenoid cystic carcinomas Twenty-one patients with unfavorable, locally advanced adenoid cystic carcinoma were treated within an ongoing clinical Phase I/II study. Only patients with infiltration of the skull base and macroscopic tumor residual after surgery without history of a former irradiation were considered for participation in the study. Therapy consisted of combined stereotactically guided photon RT to the clinical target volume (CTV) (median CTV dose 54 Gy, weekly fractionation 5 ⫻ 1.8 Gy) and a carbon ion boost to the macroscopic tumor residual (median boost dose 18 GyE, weekly fractionation 6 ⫻ 3.0 GyE). Thirteen patients were treated for primary tumors, and 8 patients received combined photon RT with a carbon ion boost for recurrent tumors after radical surgery in the first-line treatment. Two patients were treated after resection of lung metastases without evidence of active pulmonary disease at the time of presentation for RT. One patient presented with solitary lymph node metastases. Spinal and sacrococcygeal chordoma and low-grade chondrosarcoma We treated 8 patients with chordoma of the cervical spine, 1 patient with low-grade chondrosarcoma of the cervical spine, and 8 patients with sacrococcygeal chordoma within an ongoing clinical Phase I/II study. Treatment consisted of combined photon intensity-modulated radiotherapy (IMRT) and a carbon ion boost to the macroscopic tumor residual after surgery. Surgery aimed in maximum removal of the tumor. The prescribed target dose and the ratio between photons and carbon ions was chosen with respect to the tumor site and the distance between tumor and radiosensitive structures such as the spinal cord. Median photon dose was 50.4 Gy and carbon ion boost dose was 18 GyE. Patients with metal implants within the irradiation field were excluded from the study. Nevertheless, 1 patient with metal implants outside the carbon ion boost volume was considered to be a candidate for combined photon RT and a carbon ion boost. Other skull base tumors Fifteen patients with skull base tumors other than chordomas and low-grade chondrosarcomas were treated with carbon ion RT within a feasibility study. Patients with atypical or malignant meningioma (8), malignant peripheral schwannoma (2), high grade chondrosarcoma (1), chondroblastoma (1), myoblastoma (1), squamous carcinoma (1), and 1 patient with undifferentiated, not otherwise classified skull base tumor were treated with carbon ion RT alone or with combined photon RT and a carbon ion boost.
633
Reirradiation of skull base and spinal/sacral tumors Reirradiation for recurrent tumors after a former course of irradiation with photons (10) or protons (2) was delivered in 12 patients. We treated 6 patients with skull base chordoma, 1 patient with a recurrent sacral chordoma, 3 patients with recurrent adenoid cystic carcinoma, and 2 patients with recurrent atypical meningioma. Patients with recurrent skull base chordoma received a median total tumor dose of 51 GyE (range 45 to 60 GyE) with carbon ions alone. Of the 3 patients with adenoid cystic carcinoma, 1 patient received a total tumor dose of 42 GyE with carbon ions alone. The other 2 patients received combined photon RT and a carbon ion boost with target doses of 47.4 GyE and 63 GyE, respectively. The patient with the recurrent sacral chordoma was treated with combined photon RT and a carbon ion boost (total target dose 40 Gy photons ⫹ 18 GyE carbon ions). One of the meningioma patients received carbon ion RT alone (prescribed target dose 45 GyE); the second patient received combined photon and carbon ion RT (prescribed target dose 36 Gy photons plus 18 GyE carbon ions). Treatment planning and delivery Patients with skull base tumors were immobilized with a precision head mask that ensured a repositioning accuracy of 1–2 mm (8). Patients with extracranial lesions were immobilized within a rigid immobilization device consisting of a wrap-around body mask and a precision head mask (Fig. 1) developed by Lohr et al. (9). Target point localization and positioning was performed using stereotactic methods. All patients received three-dimensional (3D) treatment planning based on computed tomographic (CT) and magnetic resonance imaging (MRI) scans. Dose calculation was based on a 3D-CT data set of continuous 3-mm CT slices. For delineation of the target volumes and organs at risk, the CT images were stereotactically correlated with T2weighted and contrast-enhanced T1-weighted MRI images. Target volumes included a safety margin of 1–2 mm for possible patient misalignment and uncertainties in particle range for skull base tumors and a margin of 3 to 5 mm for extracranial lesions. For setup verification during carbon ion RT, daily orthogonal X-rays were performed and compared with digitally reconstructed radiographs. Setup errors larger than 1 mm in the skull base region and 3 mm in extracranial tumor sites were corrected before RT. Treatment planning for carbon ions was performed using the treatment planning programs VOXELPLAN and TRiP and included biologic plan optimization taking into account the local RBE values. The planning procedure, including biologic plan optimization, as well as the quality assurance program are described elsewhere (4 –7, 10). The aim of biologic plan optimization was to achieve a homogenous biologically effective dose (Deff ⫽ RBE ⫻ absorbed dose) within the target volume. The calculated mean RBE was about 3 within the target volume. Values for RBE show large variances within the target volume and the different tissues. This was considered in the optimization process.
634
I. J. Radiation Oncology
● Biology ● Physics
Volume 58, Number 2, 2004
Fig. 1. Individually manufactured rigid immobilization device in a patient with sacral chordoma.
The dose constraints for specific organs at risk were 54 GyE for optic chiasm and optic nerves. The brainstem received less than 50 GyE at the central portion, but very small volumes of less than 1 cc at the surface were allowed to receive up to 60 GyE. The dose to the spinal cord was constrained to 45 GyE. The rectal wall was not included in the gross tumor volume (GTV) for treatment planning in sacral chordomas to minimize the carbon ion dose to the rectum. Carbon ion treatment plans consisted of 2– 4 isocentric fields and were applied using the horizontal beam line. The intensity-controlled raster scan system with active energy variation was used for beam delivery and allowed for active beam shaping without any additional compensators (3). Online verification of the particle range was realized by measurement of the positron emitter distributions during treatment and by comparison with a simulated emitter distribution based on the absorbed dose distributions (11).
Follow-up and statistical analysis All patients included in studies had follow-up examinations including neurologic examinations and MRI scans performed 6 weeks after RT, in 3-month to 6-month intervals during the first 2 years, and annually thereafter. Testing of visual and auditory function as well as endocrinologic examinations were regularly performed in patients at risk. Actuarial rates for local control, disease-free survival, and overall survival were calculated from the onset of RT using Kaplan-Meier methods (12). Local control was defined as absence of clinical or radiographic signs of tumor enlargement. Acute and late side effects of carbon ion RT were scored according to the Common Toxicity Criteria (CTC) (13).
RESULTS Chordomas and low-grade chondrosarcomas of the skull base The results for 67 patients with chordomas and low-grade chondrosarcomas treated within the clinical Phase I/II study were updated in February 2003. Median follow-up was 20 months (range 3 to 54 months). The actuarial 3-year local control rate was 100% for chondrosarcomas and 81% for chordomas of the skull base, respectively (Fig. 2). We observed four local recurrences after a full course of carbon ion RT in patients with chordomas. Two of the local recurrences occurred at the field margin 13 and 20 months after RT, respectively. Correlation with the treatment plan revealed that a lower dose had been applied because of normal tissue constraints. One patient developed a subcutaneous recurrence outside the former irradiation field within the surgical scar 6 months after RT, and 1 patient was diagnosed with an in-field recurrence 27 months after carbon ion RT. One patient developed spinal seeding 18 months after RT. Actuarial 3-year overall survival was 91% for chordomas and chondrosarcomas of the skull base (Fig. 3). One patient with local recurrence of a chordoma infiltrating the pons died of tumor progression 22 months after RT. Three patients died of intercurrent disease without signs of tumor progression on follow-up MRI scans. Objective tumor response by means of partial tumor remission was observed in about one-third of the chordoma patients, whereas it was a rare finding in chondrosarcoma patients. Toxicity was mild. Seven of 67 patients developed contrast-enhancing lesions within the temporal lobes on follow-up T1-weighted MRI. The lesions were circumscribed and in close vicinity to the target volume. Correlation with the treatment plans revealed that 5 of the 7 patients
Carbon ion RT
● D. SCHULZ-ERTNER et al.
635
Fig. 2. Actuarial local control by histology in 44 patients with chordomas and 23 with low-grade chondrosarcomas of the skull base treated with carbon ion RT (Kaplan-Meier curve). Chondrosarcomas ⫽ solid line; chordomas ⫽ dotted line.
who developed temporal lobe injury received a total target dose of 70 GyE with a weekly fractionation of 7 ⫻ 3.5 GyE. Contrast-enhancing lesions within the temporal lobes were observed in 5 of 10 patients (50%) who received a target dose of 70 GyE, but were found in only 2 of 57 patients (3.5%) who received target doses between 57 and 65 GyE. Four of the patients with temporal lobe injury did not show
any clinical symptoms attributable to the lesions, and the lesions diminished or remained stable without the need for steroid medication or surgery. In 3 patients, the lesions were associated with acute clinical symptoms such as cognitive deficits, weakness, and seizures requiring long-time steroid medication. Two other patients developed contrast-enhancing lesions
Fig. 3. Actuarial overall survival in 67 patients with chordomas and low-grade chondrosarcomas of the skull base treated with carbon ions (Kaplan-Meier curve).
636
I. J. Radiation Oncology
● Biology ● Physics
Volume 58, Number 2, 2004
Table 2. Severe side effects of carbon ion radiotherapy CH/CS of the skull base
Temporal lobe injury Optic nerve neuropathy Myelitis Rectum/small bowel damage
70 GyE
57–65 GyE
ACC
Spinal CH/CS
Sacral CH/CS
5/10 0/10 0/0 —
2/57 2/57 0/57
0/21 0/21 0/21 —
— — 0/9 0/9
— — 0/8 0/8
Abbreviations: ACC ⫽ adenoid cystic carcinomas; CH ⫽ chordomas; CS ⫽ chondrosarcomas; — ⫽ not at risk.
within the optic pathway. In both patients the optic tract was included in the target volume unilaterally because of the tumor compressing it. The contrast-enhancing lesions were strictly confined to the target volume. Both patients developed homonymous hemianopsia which resolved with steroid medication in 1 patient and remained stable in the second patient. We did not observe late toxicity greater than CTC Grade 3. Severe side effects of RT are displayed in Table 2. Adenoid cystic carcinomas with infiltration of the skull base Median follow-up was 14 months (range 3– 44 months). Eight of 21 patients received a reduced photon dose of 45 to 52.2 Gy instead of 54 Gy to respect the tolerance doses of neighboring organs at risk. Since August 2001, intensitymodulated photon RT has been used for the photon treatment. The desired target dose of 54 Gy to the CTV could be delivered in all 14 patients who received their photon treatment as inversely planned photon IMRT. During the follow-up period, we observed four locore-
gional recurrences. Two patients developed a recurrence within the GTV 5 and 9 months after RT, respectively. In 1 patient, locoregional recurrence within the GTV occurred 11 months after RT and was associated with synchronous development of spinal seeding. The patient was known to have lung metastases treated with surgical removal before irradiation of the primary tumor. One patient developed a recurrence within the CTV and outside the GTV 24 months after RT. The 2 patients with recurrences outside the GTV received reirradiation. The actuarial locoregional control rate was 62% at 3 years (Fig. 4). Distant metastases progressed in both patients who presented with lung metastases before RT 11 and 43 months after RT. A third patient developed bone metastases 14 months after RT. Actuarial disease-free survival was 40% at 3 years (Fig. 5). Two of the 21 patients died of tumor progression 15 and 25 months after RT, respectively. Actuarial overall survival was 75% at 3 years (Fig. 6). Acute severe side effects CTC Grade 3 were observed in 2/21 patients (9.5%) and included an abscess within the
Fig. 4. Actuarial locoregional control in 21 patients with locally advanced adenoid cystic carcinoma treated with combined photon and carbon ion RT (Kaplan-Meier curve).
Carbon ion RT
● D. SCHULZ-ERTNER et al.
637
Fig. 5. Actuarial disease-specific survival in 21 patients with locally advanced adenoid cystic carcinoma treated with combined photon and carbon ion RT (Kaplan-Meier curve).
surgical bed 3 weeks after completion of irradiation in 1 patient and a mucositis CTC Grade 3 in another patient. We have not observed treatment-related late effects greater than CTC Grade 2 thus far (Table 2). Spinal and sacrococcygeal chordomas and chondrosarcomas A locoregional recurrence occurred in 1/8 patients treated with combined photon RT to the sacrum and a
carbon ion boost to the macroscopic tumor. The recurrence was located outside the GTV but within the former CTV. Salvage therapy consisted of surgical removal. Two of 8 patients showed distant metastases 6 and 19 months after RT. One of these patients received reirradiation after partial resection of a solitary metastasis at the level of the third lumbal spine. The second patient who was treated for a locally advanced sacral chordoma (tumor volume almost 3 L) died of multiple metastases
Fig. 6. Actuarial overall survival in 21 patients with locally advanced adenoid cystic carcinoma treated with combined photon and carbon ion RT (Kaplan-Meier curve).
638
I. J. Radiation Oncology
● Biology ● Physics
10 months after RT of the primary. Seven of the 8 patients with sacral chordoma are still alive. Carbon ion RT alone or combined with photon RT yielded local control in 8 of 9 patients with chordoma (8 patients) or low-grade chondrosarcoma (1 patient) of the cervical spine. One patient treated with combined photon RT and a carbon ion boost to a total target dose of 58.5 GyE developed a local recurrence within the target volume 21 months after RT. All but 1 patient are still alive without signs of tumor progression. We observed mucositis CTC Grade 3 in 3 patients with chordomas of the cervical spine. None of our patients showed severe late effects to the spinal cord. Patients treated for sacral chordomas did not show any side effects. Other skull base tumors Local control was achieved in 7/8 patients with atypical (2 patients) and malignant (6 patients) meningioma treated postoperatively with combined photon RT and a carbon ion boost. One of 2 patients treated for malignant schwannoma developed distant metastases 4 months after RT of her primary while the primary tumor was still controlled. She died 20 months after RT of progressive metastatic disease. The second patient is still alive without signs of tumor progression 48 months after RT. In 3 of 5 patients treated for rare skull base tumors such as myoblastoma (0/1), squamous carcinoma (1/1), chondroblastoma (1/1), high-grade chondrosarcoma (0/1), and undifferentiated, not otherwise classified tumor (1/1), local control was achieved. Reirradiation of skull base and spinal/sacral tumors Reirradiation of 12 patients with skull base and spinal/ sacral tumors resulted in effective tumor control in 4/7 patients treated for chordoma, in 1/3 patients treated for recurrent adenoid cystic carcinoma, and in 1 of 2 patients with atypical meningioma. The median follow-up was 17 months (range 3– 45 months). Six of 12 patients (50%) are still alive without further tumor progression, and 2 patients are alive with progressive tumors. Of 4 patients who died during follow-up, 3 patients died of local tumor progression and 1 patient died of malignant teratoma while her primary chordoma had started growing again as well. DISCUSSION Chordomas and low-grade chondrosarcomas of the skull base Best results in the treatment of chordomas and chondrosarcomas of the skull base are reported for surgery with postoperative high-dose proton RT. The actuarial 5-year local control rate was 73% using tumor doses between 66 and 83 cobalt Gray equivalent (CGE) for 519 patients with skull base chordomas treated with protons at the Massachusetts General Hospital, Boston (14). The temporal lobe damage rate was 13.2% after 5
Volume 58, Number 2, 2004
years (15). Hug et al. reported an actuarial 3-year local control rate of 67% for 58 chordoma patients treated at the Loma Linda University Medical Center, Loma Linda, California, USA, with proton RT. Tumor doses ranged between 64.8 and 79.2 CGE. Grade 3 and 4 late toxicities occurred in 7% of the patients (16). Noel et al. reported the results of combined photon and proton RT at the Centre de Protonthe´ rapie d’Orsay in 45 patients with chordomas and chondrosarcomas of the skull base treated between December 1995 and December 1998. They delivered a median total tumor dose of 67 CGE (range 60 to 70 CGE). Photons represented two-thirds of the total dose and protons one-third. At 3 years, local control rates for chordomas and chondrosarcomas were 83.1% and 90%. In this study, 2 of 45 patients developed Grade 3 or Grade 4 toxicities (17). Experience using particle therapy with particles heavier than protons such as helium and carbon ions is limited. Between 1977 and 1992, Castro et al. treated 223 patients with helium ions (target dose 65 CGE) at the Lawrence Berkeley Laboratory in Berkeley, California. Actuarial 5-year local control rates were 63% for chordomas and 78% for chondrosarcomas, respectively. Radiation-induced complications occurred in 27% of the patients (1). Compared with helium ions, carbon ions offer potential biologic advantages. Our data indicate that carbon ion radiotherapy is highly effective in the treatment of chordomas and low-grade chondrosarcomas. Initial results in 37 patients were already encouraging with 2-year local control rates of 83% for chordomas and 100% for lowgrade chondrosarcomas, respectively (18). The updated data presented here support these preliminary results. Local control rates at 3 years are at least comparable to proton RT while severe radiation-induced side effects were minimized. A target dose of 60 GyE may be delivered with a low risk for severe side effects, whereas dose escalation has to be considered with caution, because late toxicity might be increased. Spinal and sacrococcygeal chordomas and low-grade chondrosarcomas Extracranial chordomas and low-grade chondrosarcomas are more difficult to treat compared with tumors of the skull base. Radiosensitive structures as well as larger setup errors limit the prescription dose. As a consequence of a suboptimal RT dose, local control rates have been poor. Best results were yielded with heavy charged particles. Schoenthaler et al. treated 14 patients with helium or neon ions at the Lawrence Berkeley Laboratory. Four of 14 patients were treated after gross tumor resection. They reported a 5-year local control rate of 55% with a trend for better local control for patients treated with neon ions compared with helium ions and for 4 patients who received RT after gross tumor resection (19). Berson et al. reported a 5-year local control rate of 54% for 10 patients with chordomas and chondrosarcomas of the
Carbon ion RT
● D. SCHULZ-ERTNER et al.
cervical spines treated with charged particles at the Lawrence Berkeley Laboratory (20). Similar results were obtained with protons in 20 patients with spinal and sacral chordomas (14 patients) and chondrosarcomas (6 patients) by Hug et al. at the Massachusetts General Hospital in Boston. They treated 5 patients with tumors of the thoracic spine, 5 patients with tumors of the lumbar spine, and 10 patients with sacral tumors. Five-year local control rates were 100% for chondrosarcomas and 53% for chordomas (16). Our results in 17 patients with sacral and spinal chordomas and chondrosarcomas demonstrate that combined photon and carbon ion RT is well tolerated and offers effective local control. Locally advanced unfavorable adenoid cystic carcinomas Surgery is the mainstay to therapy in adenoid cystic carcinoma, but adjuvant high-dose RT is generally recommended even after complete resection. Local control rates at 5 years exceed 90% after complete resection and postoperative RT (21), but remain poor after incomplete resection with macroscopic tumor residuals (22, 23). In the past, best results in this unfavorable group of adenoid cystic carcinoma have been achieved with neutrons. The advantage of neutrons over photon RT for malignant salivary gland tumors was proven within a randomized Phase III trial of the Radiation Therapy Oncology Group (RTOG) and the Medical Research Council (RTOG-MRC trial). Actuarial local control rates and overall survival rates at 2 years were 67% and 62% with neutrons compared with 17% and 25% with photons, respectively (24). At 10 years, local control was 56% for the neutron arm compared with 17% for patients treated with conventional RT, respectively. The improvement in local control was attributed to the high–linear energy transfer effect of neutrons, while survival rates showed no difference (25). The main disadvantage of neutron RT was the relatively high rate of severe late effects containing bone and soft tissue necrosis and fibrosis in up to 20% of the patients (26 –28). Our data indicate that postoperative combined photon IMRT and a carbon ion boost to the macroscopic tumor represents a therapy approach which offers not only high local control rates but also minimizes severe side effects. Nevertheless, disease-free survival remains poor due to a high rate of distant metastases in locally advanced adenoid cystic carcinomas which is rarely influenced with common chemotherapy regimes. Atypical and malignant meningioma Benign meningiomas of the skull base can be treated effectively with modern photon techniques. Actuarial 10-year local control rates in excess of 90% are yielded with fractionated stereotactic techniques even in large en plaque meningiomas of the skull base (29, 30). Delivery of fractionated stereotactic photon techniques is safe; severe late effects occurred in less than 5% of the patients
639
(29). On the other hand, treatment of atypical and malignant meningiomas with photons still remains unsatisfactory. Milosevic et al. reported a 5-year cause-specific survival rate of 34% in 59 patients with atypical or malignant meningioma treated with conventional photon RT (31). Hug et al. treated 31 patients with atypical/ malignant meningiomas with combined photon and proton RT or photons alone at the Massachusetts General Hospital in Boston. Significantly improved local control was observed for proton vs. photon RT (80% vs. 17% at 5 years). RT with target doses exceeding 60 Gy resulted in significantly better local control rates compared with doses ⬍60 Gy. The authors conclude that high-dose RT should be recommended in atypical/malignant meningiomas, which is rarely possible with photon RT (32). Our results show that the physical and biologic advantages of carbon ion RT can be utilized for the treatment of atypical/malignant meningiomas as well. With a combined approach using modern photon RT and a carbon ion boost, a high target dose of 68.4 Gy can safely be delivered. Reirradiation of skull base tumor Reirradiation of skull base tumors has rarely been considered in the past. Using raster scanned carbon ion RT, patients with recurrent tumors may be treated again with effective target doses because the dose gradient toward radiosensitive structures is steeper than with any other RT modality. The benefit of reirradiation of skull base tumors with carbon ions depends on the extension of the recurrent tumor, the distance to neighboring radiosensitive structures, and the dose applied to organs at risk during the first course of RT. Carbon ion RT might be considered in individual patients with circumscribed recurrences where effective carbon ion doses can be delivered to small volumes with an acceptable risk for radiation-induced damage to normal tissue. CONCLUSION With raster scanned carbon ion beams, high local control rates can be achieved with relatively low toxicity compared with photon and proton RT. Carbon ion RT may therefore be considered alternatively to proton RT which is currently the standard of care in patients with chordomas and chondrosarcomas of the skull base. A total target dose of 60 GyE delivered within 20 days proved to be the optimal prescription dose with respect to effectiveness and avoidance of radiation-induced side effects. A hospital-based facility will be built in Heidelberg, Germany until 2006 and will allow for prospective clinical Phase III studies comparing carbon ion RT with protons and modern photon RT such as intensity-modulated RT for specific tumors. Besides chordomas, low-grade chondrosarcomas, and malignant salivary gland tumors, more common tumors such as prostate cancer, lung cancer, and soft tissue sarcomas might be treated effectively with carbon ions.
640
I. J. Radiation Oncology
● Biology ● Physics
Volume 58, Number 2, 2004
REFERENCES 1. Castro JR, Linstadt DE, Bahary J-P, et al. Experience in charged particle irradiation of tumors of the skull base: 1977– 1992. Int J Radiat Oncol Biol Phys 1994;29:647–655. 2. Tsujii H, Morita S, Miyamoto T, et al. Experiences of carbon ion radiotherapy at NIRS. In: Kogelnik HD, Lukas P, Sedlmayer F, editors. Progress in radio-oncology VII. Bologna, Italy: Monduzzi Editore, 2002; p. 393–405. 3. Haberer T, Becher W, Schardt D, et al. Magnetic scanning system for heavy ion therapy. Nucl Instr Meth Phys Res 1993;330:296–305. 4. Kra¨ mer M, Ja¨ kel O, Haberer T, et al. Treatment planning for heavy ion radiotherapy: Physical beam model and dose optimization. Phys Med Biol 2000;45:3299–3317. 5. Kra¨ mer M, Scholz M. Treatment planning for heavy-ion radiotherapy: Calculation and optimization of biologically effective dose. Phys Med Biol 2000;45:3319–3330. 6. Ja¨ kel O, Hartmann GH, Karger CP, et al. Quality assurance for a treatment planning system in scanned ion beam therapy. Med Phys 2000;27:1588–1600. 7. Scholz M, Kellerer AM, Kraft-Weyrather W, et al. Computation of cell survival in heavy ion beams for therapy. The model and its approximation. Radiat Environ Biophys 1997; 36:59–66. 8. Karger CP, Ja¨ kel O, Debus J, et al. Three dimensional accuracy and interfractional reproducibility of patient fixation using a stereotactic head mask system. Int J Radiat Oncol Biol Phys 2001;49:1493–1504. 9. Lohr F, Debus J, Frank C, et al. Noninvasive patient fixation for extracranial stereotactic radiotherapy. Int J Radiat Oncol Biol Phys 1999;45:521–527. 10. Ja¨ kel O, Kra¨ mer M, Karger CP, et al. Treatment planning for heavy ion radiotherapy: Clinical implementation and application. Phys Med Biol 2001;46:1101–1116. 11. Enghardt W, Debus J, Haberer T, et al. The application of PET to quality assurance of heavy-ion tumor therapy. Strahlenther Onkol 1999;175(Suppl. 2):33–36. 12. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958;53:457–481. 13. National Cancer Institute (NCI). Common toxicity criteria. Bethesda, MD: Division of Cancer Treatment, National Cancer Institute; 1988. 14. Munzenrider JE, Liebsch NJ. Proton therapy for tumors of the skull base. Strahlenther Onkol 1999;175(Suppl. II):57–63. 15. Santoni R, Liebsch N, Finkelstein DM, et al. Temporal lobe (TL) damage following surgery and high-dose photon and proton irradiation in 96 patients affected by chordomas and chondrosarcomas of the base of the skull. Int J Radiat Oncol Biol Phys 1998;41:59–68. 16. Hug EB, Loredo LN, Slater JD, et al. Proton radiation therapy for chordomas and chondrosarcomas of the skull base. J Neurosurg 1999;91:432–439. 17. Noel G, Habrand JL, Mammar H, et al. Combination of photon and proton radiation therapy for chordomas and chondrosarcomas of the skull base: The Centre de Protonthe´ rapie d’Orsay experience. Int J Radiat Oncol Biol Phys 2001;51: 392–398.
18. Schulz-Ertner D, Haberer T, Ja¨ kel O, et al. Radiotherapy for chordomas and low-grade chondrosarcomas of the skull base with carbon ions. Int J Radiat Oncol Biol Phys 2002;53:36– 42. 19. Schoenthaler R, Castro JR, Petti PL, et al. Charged particle irradiation of sacral chordomas. Int J Radiat Oncol Biol Phys 1993;26:291–298. 20. Berson AM, Castro JR, Petti P, et al. Charged particle irradiation of chordoma and chondrosarcoma of the base of skull and cervical spine: The Lawrence Berkeley Laboratory experience. Int J Radiat Oncol Biol Phys 1988;15:559–565. 21. Garden AS, Weber RS, Morrison WH, et al. The influence of positive margins and nerve invasion in adenoid cystic carcinoma of the head and neck treated with surgery and radiation. Int J Radiat Oncol Biol Phys 1995;32:619–626. 22. Regine WF, Mendenhall WM, Parsons JT, et al. Radiotherapy for adenoid cystic carcinoma of the palate. Head Neck 1993; 15:241–244. 23. Vikram B, Strong EW, Shah JP, et al. Radiation therapy in adenoid cystic carcinoma. Int J Radiat Oncol Biol Phys 1984; 10:221–223. 24. Griffin TW, Pajak TF, Laramore GE, et al. Neutron vs photon irradiation of inoperable salivary gland tumors: Results of an RTOG-MRC cooperative randomized study. Int J Radiat Oncol Biol Phys 1988;15:1085–1090. 25. Laramore GE, Krall JM, Griffin TW, et al. Neutron versus photon irradiation for unresectable salivary gland tumors: Final report of an RTOG-MRC randomized clinical trial. Radiation Therapy Oncology Group. Medical Research Council. Int J Radiat Oncol Biol Phys 1993;27:235–240. 26. Saroja KR, Mansell J, Hendrickson FR, et al. An update on malignant salivary gland tumors treated with neutrons at Fermilab. Int J Radiat Oncol Biol Phys 1987;13:1319–1325. 27. Schultheiss TE, Cohen L, Mansell J. Normal tissue reactions and complications following high-energy neutron beam therapy II: Complication rates adjusted for censoring. Int J Radiat Oncol Biol Phys 1990;18:165–171. 28. Buchholz TA, Laramore GE, Griffin BR, et al. The role of fast neutron radiation therapy in the management of advanced salivary gland malignant neoplasms. Cancer 1992;69:2779– 2788. 29. Debus J, Wuendrich M, Pirzkall A, et al. High efficacy of fractionated stereotactic radiotherapy of large base-of-skull meningiomas: Long-term results. J Clin Oncol 2001;19:3547– 3553. 30. Pirzkall A, Debus J, Haering P, et al. Intensity modulated radiotherapy (IMRT) for recurrent, residual, or untreated skull-base meningiomas: Preliminary clinical experience. Int J Radiat Oncol Biol Phys 2003;55:362–372. 31. Milosevic MF, Frost PJ, Laperriere NJ, et al. Radiotherapy for atypical or malignant intracranial meningioma. Int J Radiat Oncol Biol Phys 1996;34:817–822. 32. Hug EB, DeVries A, Thornton AF, et al. Management of atypical and malignant meningiomas: Role of high-dose, 3Dconformal radiation therapy. J Neurooncol 2000;48:151–160.