- Email: [email protected]
Image-guided radiosurgery of head and neck cancers
Image-guided radiosurgery of head and neck cancers SAMUEL RYU, MD, MUMTAZ KHAN, MD, FANG-FANG YIN, PhD, ADRIANNE CONCUS, MICHAEL S. BENNINGER, MD, and JAE HO KIM, MD, PhD, Detroit, Michigan
OBJECTIVES: Radiosurgery precisely delivers a single high dose or a few fractionated doses of radiation to a localized tumor via the stereotactic approach. Some head and neck sites are suitable for radiosurgery since there is minimal or no organ motion. The clinical studies were carried out to determine the accuracy of stereotactic radiosurgery and to demonstrate the effectiveness of radiosurgery in head and neck cancers. MATERIALS AND METHODS: Thirteen patients were treated with either single-dose or fractionated radiosurgery to the tumor. All patients except one with cancer of the lip had received prior treatments including surgery, radiotherapy, and chemotherapy for the primary cancers. The dose ranged 12 to 18 Gy for single-dose radiosurgery and 30 Gy in 5 or 6 fractions twice a week for fractionated radiosurgery. Tumor localization was achieved via the stereotactic approach. RESULTS: Accuracy of radiosurgery was within 1.5 mm. Despite the recurrent disease from previous heavy treatments, 9 patients (70%) showed a significant response (complete or >50% tumor reduction) to radiosurgery, and 3 patients had stable disease. Complete tumor response was achieved in 6 patients. All patients had excellent pain relief with functional and cosmetic preservation. There was no acute and subacute radiation toxicity detected clinically during the minimal follow-up of 6 months. CONCLUSION: Image-guided radiosurgery is effective in achieving the local tumor control and pain relief. Radiosurgery provided excellent functional and cosmetic preservation with minimal complication. The results indicate the potential of radiosurgery in the treatment of recurrent and selected pri-
From the Departments of Radiation Oncology (Drs Ryu, Yin, Ajlouni, and Kim) and Otolaryngology–Head and Neck Surgery (Drs Concus, Khan, and Benninger), Henry Ford Hospital. Reprint requests: Samuel Ryu, MD, Department of Radiation Oncology, Henry Ford Hospital, 2799 W Grand Blvd, Detroit, MI 48202; e-mail, [email protected]. 0194-5998/$30.00 Copyright © 2004 by the American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. doi:10.1016/j.otohns.2003.10.009 690
MD,
MUNTHER AJLOUNI,
MD,
mary head and neck cancers. (Otolaryngol Head Neck Surg 2004;130:690-7.)
T he major treatment modalities used for the local tumor control of primary head and neck cancers are surgery and radiotherapy, although the selection of treatment modality may vary significantly with geography and institutional preference.1 Specific surgical and radiotherapeutic approaches vary depending on the treatment philosophy of the institution and individual physicians. It is even more complex and individualized when it comes to the management of recurrent tumors. Often, patients receive palliative treatment for the control of any presenting symptoms. However, there is a subset of patients with recurrent head and neck tumors who may have excellent local tumor control and good quality of function as well as improved survival. With the development of more effective cancer therapy, the overall survival of cancer patients has increased for many malignant diseases. As cancer patients are now experiencing longer survival, quality of life has become an important factor not only in the decision to treat definitively but also in the decision to offer palliative treatment of recurrent or metastatic disease. This is particularly true in the treatment of recurrent tumors of the head and neck because proper local tumor control is directly related to the quality of life with respect to phonation, swallowing, and breathing. Many patients, if not treated, lose control of these functions, which results in significant morbidity. Treatments for recurrent head and neck tumors with surgery, external beam radiotherapy, or chemotherapy have been limited. Salvage surgery for recurrent or persistent tumors is often offered only to patients who have resectable disease, sometimes with increased risk of healing, functionality, and anesthesia. However, some patients who undergo postradiotherapy salvage laryngopharyngectomy develop major complications such as pharyngocutaneous fistula or functional disability.2 Chemotherapy is also offered in an attempt to
Otolaryngology– Head and Neck Surgery Volume 130 Number 6
RYU et al 691
Table 1. Summary of patient characteristics and outcome of radiosurgery Site
Single-dose radiosurgery Floor of mouth Pharyngeal wall Base of tongue Base of skull Parotid Fractionated radiosurgery Neck node Orbit Neck node Neck node Floor of mouth Orbit Orbit Lip
Dose (tumor periphery)
Pathology
Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Mucoepidermoid carcinoma Squamous cell carcinoma
18 16 18 12 16
Gy Gy Gy Gy Gy
Squamous cell carcinoma (from tonsil) Adenoid cystic carcinoma Adenoma (from stomach) Squamous cell carcinoma (from lung) Squamous cell carcinoma Adenoid cystic carcinoma Basal cell carcinoma Squamous cell carcinoma
30 30 36 36 30 30 30 36
Gy/6 Gy/6 Gy/6 Gy/6 Gy/6 Gy/6 Gy/6 Gy/6
achieve tumor shrinkage and palliation with or without radiotherapy.3 In certain circumstances, radiotherapy alone using different treatment techniques could be used.4,5 Although most of the radiotherapy for recurrent cancers is administered to the patient with palliative intent, this is also limited by previous heavy treatment and the volume of the tumor. Conventional radiotherapy uses large margins around the involved areas to compensate for infiltrating portion of the tumor and internal organ motion as well as patient motion during the treatment. In view of this, a significant amount of normal tissue, which has been already heavily treated, is included within the treatment volume. Therefore, there is a pressing need for a new approach to increase the tumor control without inflicting treatment on the already heavily treated normal tissues. Radiosurgery delivers a highly conformal, large radiation dose to a localized target tumor via a stereotactic approach. This requires accurate targeting and immobilization of patient during irradiation. Radiosurgery has not been applied to extracranial tumors due to organ motion associated with breathing and/or lack of immobilization techniques. Further, the irregularly shaped tumor would require a sophisticated conformal targeting method. With the progress of imaging and computer science and development of reliable noninvasive positioning devices,6 stereotactic target localization could be achieved in the extracranial
Response
CR (pathologic) Not evaluable CR (pathologic) Stable Stable fractions fractions fractions fractions fractions fractions fractions fractions
CR CR PR PR PR Stable CR CR
sites. We have demonstrated the use of extracranial radiosurgery for spinal tumors and reported the accuracy and precision of tumor localization and excellent clinical outcome.7,8 Among the extracranial organs, the head and neck organs have low or no breathing-related organ movement. Therefore, head and neck cancers could be ideal organ sites for stereotactic radiosurgery. We report the experience of radiosurgery for head and neck cancers with an excellent local tumor control and functional outcome. MATERIALS AND METHODS A total of 13 patients were treated with radiosurgery at Henry Ford Hospital from April 2001 to June 2002. Eleven patients had recurrent tumors that were diagnosed by clinical, radiologic, and pathologic examination. All patients had a prior diagnosis of pathologically proved primary malignant neoplasm that was previously treated with combined modalities of surgery, radiotherapy, and chemotherapy. One patient with primary lip cancer was treated with radiosurgery and was included in this study. The patient characteristics and the primary and recurrent sites of malignancy are shown in Table 1. All patients received stereotactic radiosurgery with either single or fractionated doses of radiation. The decision to use single or fractionated radiosurgery was based on several factors, including prior radiotherapy and/or surgery, location and volume of the tumor, and pa-
692 RYU et al
Otolaryngology– Head and Neck Surgery June 2004
Fig 1. Picture of stereotaxy. The image is reconstructed from the simulation CT scan with stereotactic Z-shaped localizer device (blue lines). This localizer device produces fiducial points on each side of axial CT image. The isocenter of the radiosurgery is determined at the midpoint of the target tumor volume, and the isocenter coordinate in each crosssectional plane is obtained by longitudinally connecting the same points in each plane. The distance of the isocenter from the Z-shaped localizer device gives the actual setting of the target in the anteroposterior and lateral planes. One example of focused radiosurgery beam (yellow) is directed to the target tumor (red).
tient’s condition and tolerability. For single-dose radiosurgery, a median dose of 17 Gy (range, 12 to 18 Gy) was delivered to the tumor. For fractionated radiosurgery, the radiation dose was between 30 and 36 Gy in 5 or 6 fractions with treatment twice weekly. Radiation dose was prescribed to the 90% isodose line that covers the entire target tumor as identified by either computed tomography (CT) or magnetic resonance imaging (MRI). The procedure of the radiosurgery is briefly described. Immobilization of the head and neck is accomplished by using a custom-made noninvasive mask frame with individual tongue bite. Sim-
ulation CT (Philips, Cleveland, OH) with contrast agent is obtained with cross-sections of 3 mm thickness without spacing, with a stereotactic localization box (BrainLab, Inc, Heimstetten, Germany). This localization box is mounted with a Z-shaped localizer device that is filled with aluminum strips with low CT artifact. A reconstructed image of localization system is shown in Figure 1. This localizer device produces small fiducial points on each side of the axial CT scan. The images are sent to the dedicated treatment planning computer system (BrainScan; BrainLab, Inc). The target tumor and the critical normal
Otolaryngology– Head and Neck Surgery Volume 130 Number 6
tissues are determined with consideration of clinical presentation of the tumor as well as radiologic appearance. The volumes of the tumor and normal tissues are drawn on each cross-sectional image of the simulation CT scan. An effective radiosurgery dose to the tumor with a limiting dose to the adjacent normal tissue is prescribed. The isocenter of the radiosurgery is usually determined at the midpoint of the target tumor volume, and the isocenter coordinate in each cross-sectional plane is obtained by longitudinally connecting the same points in each plane. The distance of the isocenter from the Z-shaped localizer device gives the actual setting of the target in the anteroposterior and lateral planes. The isocenter of the radiosurgery is printed as digitally reconstructed radiographs of the CT scans. Radiosurgery usually uses multiple numbers (5 to 9) of intensity-modulated radiation beams or multiple dynamic conformal arc beams to minimize the dose to the critical organs. The prescribed radiosurgery dose is again checked in relation to the final radiation isodose distribution. Once all the requirements are met, the patient is brought to the radiosurgery suite. Repositioning of the patient is achieved by using the already-made mask frame with the stereotactic localization box that was used for simulation. The isocenter position defined for each patient during the process of treatment planning is printed and attached to the localization box. Before the delivery of radiation, orthogonal portal films are obtained for the final verification and to determine the precision and accuracy of the radiosurgery. The image-guided shaped beam radiosurgery at Henry Ford Hospital uses the Novalis system (BrainLAB, Inc). It is equipped with a 3-mm-thick micromultileaf collimator that allows 3-dimensional beam shaping. The dosimetric characteristics of this treatment unit and the technique of radiation intensity modulation have been reported previously.9-12 The end points for evaluation were 1) to determine the precision and accuracy of the radiosurgery for head and neck cancers and 2) to demonstrate the local tumor control after radiosurgery by clinical and radiologic examination or by biopsy. The precision of the radiosurgery was defined as the degree of variation between the isocenters of the radiosurgery treatment plan and of the portal films taken at the time of radiation delivery. It was
RYU et al 693
measured in millimeters by image fusion of the digitally reconstructed radiograph of CT simulation and portal films that were taken at the actual treatment. For assessment of tumor response, complete response (CR) was defined as disappearance of the tumor by clinical and radiologic examination or by biopsy. Partial response (PR) was defined as reduction of the tumor size in its greatest dimension by more than 50%. Although it was not the primary end point, patients were also evaluated for relief of pain and symptom improvement. The median follow-up time was 8 months (range, 6 to 16 months). RESULTS Precision and Accuracy of Radiosurgery The precision for a given isocenter between the simulation and actual treatment position was 1.4 ⫾ 0.7 mm. We also measured the radiation dose at the isocenter using the same positioning parameters of the individual patients in a phantom with a micro–ion chamber. The average deviation of the measured dose from the estimated dose was 2%. These 2 independent measurements were mutually corroborative of the accuracy of radiosurgery in the head and neck. Local Tumor Control The patient characteristics with primary treatment and radiosurgery for the recurrent tumor are shown in Table 1. Nine of 13 patients (70%) showed significant response (complete or ⬎50% tumor reduction) to radiosurgery. Complete response was achieved in 6 patients (2 with singledose radiosurgery, 4 with fractionated radiosurgery). Partial response defined as ⬎50% reduction of tumor volume was seen in an additional 3 patients with fractionated radiosurgery. There were 3 patients with recurrent salivary gland tumors (1 with mucoepidermoid carcinoma involving the base of skull and 2 with adenoid cystic carcinoma involving the orbit). One patient with recurrent adenoid cystic adenocarcinoma of the orbit showed nearcomplete disappearance of the tumor at 10-month follow-up MRI. The other 2 patients were stable with significant symptom relief. An example of a patient with a complete pathologic response is shown in Figure 2. This patient initially had squa-
694 RYU et al
Otolaryngology– Head and Neck Surgery June 2004
Fig 2. Radiosurgery of a biopsyproved recurrent tumor at the base of the tongue. (A) Preradiosurgery CT scan shows an enhancing mass (arrow) with microscopic feature of invasive squamous carcinoma. (B) Radiosurgery plan with isodose distribution. The tumor was treated with singledose 18-Gy radiosurgery. The high radiation dose was delivered to the enhancing target tumor, whereas the dose to the surrounding normal tissue was minimal. (C) CT scan taken 6 months postradiosurgery showed the enhancing lesion reduced in size (arrowhead). Biopsy showed no tumor cells.
Otolaryngology– Head and Neck Surgery Volume 130 Number 6
mous cell carcinoma of an unknown primary neck node that was treated with radical neck dissection and radiotherapy. The patient developed a second primary tumor at the base of the tongue after 12 years. This lesion was treated with a single dose of radiosurgery of 18 Gy. A follow-up CT scan at 6 months after the radiosurgery showed a smaller residual enhancement; biopsy revealed no tumor cells. At 8 months after radiosurgery, the patient developed mucosal irregularity of the area with induration on clinical examination. This area was locally excised and pathologic examination showed no evidence of residual tumor. Functional Outcome Four patients who presented with pain had a significant relief of symptom. Three patients had complete pain relief. One patient who was treated for palliation of pain with single-dose 12-Gy radiosurgery for the recurrent mucoepidermoid cancer at the skull base had partial pain relief to the level that pain medication could be significantly reduced. This patient had 3 operations and heavy radiotherapy for the primary site at the maxillary/ palatal area and interstitial brachytherapy to the palatal area including the skull base before radiosurgery. After 10 months, the tumor progressed. Two patients who had metastatic neck nodes from stomach and lung cancers had partial tumor response with satisfactory pain relief but developed systemic metastasis. None of the patients had any long-term adverse effects. Acute side effects were temporary and minimal, with mild mucositis and skin irritation depending on the tumor location. Preservation of function and cosmetic outcome were well maintained after radiosurgery. In all of these patients, subjective functional preservation was satisfactory in the ability of swallowing and phonation. A patient with lip cancer at the oral commissure had the original tumor measured 3.5 cm with ulceration and subcutaneous infiltration. This patient refused surgical treatment. The tumor was treated with fractionated radiosurgery to a total dose of 36 Gy in 6 fractions twice a week. The patient had complete resolution of the tumor with excellent functional and cosmetic outcome at 2 months after radiosurgery.
RYU et al 695
DISCUSSION Radiosurgery delivers a highly conformal large radiation dose to a localized target tumor. It requires accurate targeting and immobilization of the target organ during irradiation. Therefore, the accuracy and precision for radiosurgery should be tested first before administering the large dose of highly conformal radiation. Most experience with stereotactic radiosurgery involves intracranial lesions such as arteriovenous malformations and malignant and benign tumors. For radiosurgery of intracranial tumors, the precision of the isocenter target has usually been within 1 mm. Radiosurgery has not been applied to extracranial tumors due to organ motion associated with breathing and/or lack of immobilization techniques. Among the extracranial organs, the spines and some head and neck sites have the least breathing-related movement. This makes the head and neck sites suitable for stereotactic radiosurgery. We have demonstrated the accuracy of the spinal radiosurgery within 1.3 mm, where the spinal cord is intimately located with the spinal tumors.8 In this study, we report the accuracy of the radiosurgery in the head and neck sites within 1.5 mm. This magnitude of accuracy is certainly acceptable for treatment of the head and neck cancers. One potential problem would be unexpected swallowing or coughing that may occur during the delivery of radiation. This may need to be explored to provide more accurate radiosurgery treatment. The tumors involving the head and neck can be irregular in shape and are often associated with significant infiltration to the surrounding soft tissue. This target irregularity makes the use of radiosurgery difficult. By including the irregular target within the radiosurgery volume, normal tissue immediately adjacent to the tumor also receives high radiation doses, particularly in recurrent tumors. The criteria of the target tumor delineation have to be defined as experience with radiosurgery is gained. In this study with recurrent tumors, margins of 2 to 3 mm from the radiographically and clinically visible or palpable tumor were allowed to account for the infiltrative disease. With this approach, tumor recurrence in the marginal zone of radiosurgery was not seen during the follow-up period. A representative example of an intensitymodulated treatment plan is shown in Figure 2.
Otolaryngology– Head and Neck Surgery June 2004
696 RYU et al
The radiation dose was prescribed for the volume included within the 90% radiation isodose line. To achieve tumor control and to overcome the potential complications of the adjacent normal tissue by treatment with high-dose radiosurgery, accurate tumor localization and precise shaping of the tumor by the radiation beam are required. Radiosurgery will provide a rapid radiation dose fall-off immediately outside of the target volume by shaping the radiation beam with micromultileaf collimation and by modulating the radiation intensity. These techniques increase the ability to conform to the irregular tumor shape. Selection of radiosurgery dose and tolerance of the surrounding normal tissue to the single or fractionated dose(s) of radiosurgery are not well defined. Experience with interstitial implantation showed the safe radiation dose of retreatment in the range of 50 to 60 Gy to the entire tumor volume with low-dose rate of radiation. This lowdose rate brachytherapy has been used for treatment of recurrent tumors with 30% to 70% salvage rates and 30% to 40% complication rates.4,13 Radiosurgery has been mostly used for recurrent nasopharyngeal cancers with small fractional doses and showed good local tumor control.5,14-16 These studies involved the use of fractionated radiosurgery/radiotherapy to a rather larger volume, encompassing the entire nasopharynx. Patient tolerance was reported to be excellent. We used the radiation dose of 12 to 18 Gy for singledose radiosurgery and 30 to 36 Gy in 5 or 6 fractions for fractionated radiosurgery. Radiation dose-response of specific tumors could not be evaluated due to the different tumor histology and small number of patients. In the case of the squamous cell carcinomas, a higher dose appears to achieve superior tumor response, based on our clinical experience. Radiosurgery should be carefully planned in patients who have had heavy prior treatment because of the higher risk of complications from the additional therapy in recurrent diseases. For dose selection of retreatment, careful consideration must be given to the previous radiation to normal tissues (dose and fields); the extent and pathology of the recurrent tumor; the shape, size, and location of the tumor; the condition of the surrounding normal tissues; the patient’s general status includ-
ing comorbidities; and the ultimate goal of treatment.17 We encountered no unacceptable acute or subacute effects on normal tissue and no complications. Although the follow-up was relatively short, there have not been any signs that might be suggestive of long-term complications. Patients who have advanced primary tumors or recurrent tumors that are locally controlled by radiation alone have chances to develop a major complication and a nonfunctioning larynx.18 However, the radiosurgery is delivered only to the local tumor rather than to the large volume containing lymph node areas, thereby reducing the chance of developing complications. CONCLUSION This study demonstrated the proof of principle and the potential of radiosurgery for the treatment of head and neck tumors. However, special consideration should be given in selection of treatment and patients, particularly with the lymphatic drainage and tumor infiltration of the primary head and neck cancers that may warrant broader radiotherapy with nodal irradiation. This study offers a potential to extend the use of radiosurgery to benign tumors,19 as a viable treatment option for recurrent head and neck cancers, and, possibly, for certain selected primary cancers of the head and neck. REFERENCES
1. Mendenhall WM, Hinderman RW, Morris CG, et al. Management of supraglottic carcinoma. Int J Radiat Oncol Biol Phys 2002;53:793. 2. Chee N, Siow JK. Pharyngocutaneous fistula after laryngectomy: incidence, predisposing factors and outcome. Singapore Med J 1999;40:130. 3. Wong ZW, Tan EH, Yap SP, et al. Chemotherapy with or without radiotherapy in patients with locoregionally recurrent nasopharyngeal carcinoma. Head Neck 2002;24:549. 4. Puthawala AA, Syed MN. Interstitial re-irradiation for recurrent and/or persistent head and neck cancers. Int J Radiat Oncol Biol Phys 1987;13:1113. 5. Xiao J, Xu G, Miao Y. Fractionated stereotactic radiosurgery for 50 patients with recurrent or residual nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2001;51:164. 6. Alheit H, Dornfeld S, Dawel M, et al. Patient position reproducibility in fractionated stereotactically guided conformal radiotherapy using the BrainLab mask system. Strahlenther Onkol 2001;177:264. 7. Rock J, Kole M, Yin FF, et al. Radiosurgical treatment for Ewing’s sarcoma of the lumbar spine. Spine 2002; 27:471.
Otolaryngology– Head and Neck Surgery Volume 130 Number 6
8. Ryu S, Yin FF, Rock J, et al. Image-guided and intensitymodulated radiosurgery for spinal metastasis. Cancer 2003;97:2013. 9. Cosgrove VP, Jahn U, Pfaender M, et al. Commissioning of a micro multi-leaf collimator and planning system for stereotactic radiosurgery. Radiother Oncol 1999;50:325-36. 10. Yin F-F, Zhu J, Hui Y, et al. Dosimetric characteristics of Novalis shaped beam surgery unit. Med Phys 2002;29: 1729-38. 11. Yin F-F, Ryu S, Ajlouni M, et al. A technique of intensity-modulated radiosurgery (IMRS) for spinal tumors. Med Phys 2002;29:2815-22. 12. Llacer J, Solberg TD, Promberger C. Comparative behavior of the Dynamically Penalized Likelihood algorithm in inverse radiation therapy planning. Phys Med Biol 2001;46:2637-63. 13. Nag S, Cano ER, Demanes DJ, et al. The American brachytherapy society recommendations for high dose rate brachytherapy for head and neck carcinoma. Int J Radiat Oncol Biol Phys 2001;50:1190-8.
RYU et al 697
14. Ahn YC, Lee KC, Kim DY, et al. Fractionated stereotactic radiation therapy for extracranial head and neck tumors. Int J Radiat Oncol Biol Phys 2000;48:501-5. 15. Chen HJ, Leung SW, Su CV. Linear accelerator based radiosurgery as a salvage treatment for skull base and intracranial invasion of recurrent nasopharyngeal carcinomas. Am J Clin Oncol 2001;24:255-8. 16. Chua DT, Sham JS, Hung KN, et al. Salvage treatment for persistent and recurrent T1-2 nasopharyngeal carcinoma by stereotactic radiosurgery. Head Neck 2001;23:791-8. 17. Ryu S, Gorty S, Kazee AM, et al. “Full dose” re-irradiation of human cervical spinal cord. Am J Clin Oncol 2000;23:29-31. 18. Mancuso AA, Mukherji SK, Schmalfuss I, et al. Preradiotherapy computed tomography as a predictor of local control in supraglottic carcinoma. J Clin Oncol 1999;17: 631-7. 19. Foote RL, Pollock BE, Gorman DA, et al. Glomus jugulare tumor: tumor control and complications after stereotactic radiosurgery. Head Neck 2002;24:332-8.
OBJECTIVES: Radiosurgery precisely delivers a single high dose or a few fractionated doses of radiation to a localized tumor via the stereotactic approach. Some head and neck sites are suitable for radiosurgery since there is minimal or no organ motion. The clinical studies were carried out to determine the accuracy of stereotactic radiosurgery and to demonstrate the effectiveness of radiosurgery in head and neck cancers. MATERIALS AND METHODS: Thirteen patients were treated with either single-dose or fractionated radiosurgery to the tumor. All patients except one with cancer of the lip had received prior treatments including surgery, radiotherapy, and chemotherapy for the primary cancers. The dose ranged 12 to 18 Gy for single-dose radiosurgery and 30 Gy in 5 or 6 fractions twice a week for fractionated radiosurgery. Tumor localization was achieved via the stereotactic approach. RESULTS: Accuracy of radiosurgery was within 1.5 mm. Despite the recurrent disease from previous heavy treatments, 9 patients (70%) showed a significant response (complete or >50% tumor reduction) to radiosurgery, and 3 patients had stable disease. Complete tumor response was achieved in 6 patients. All patients had excellent pain relief with functional and cosmetic preservation. There was no acute and subacute radiation toxicity detected clinically during the minimal follow-up of 6 months. CONCLUSION: Image-guided radiosurgery is effective in achieving the local tumor control and pain relief. Radiosurgery provided excellent functional and cosmetic preservation with minimal complication. The results indicate the potential of radiosurgery in the treatment of recurrent and selected pri-
From the Departments of Radiation Oncology (Drs Ryu, Yin, Ajlouni, and Kim) and Otolaryngology–Head and Neck Surgery (Drs Concus, Khan, and Benninger), Henry Ford Hospital. Reprint requests: Samuel Ryu, MD, Department of Radiation Oncology, Henry Ford Hospital, 2799 W Grand Blvd, Detroit, MI 48202; e-mail, [email protected]. 0194-5998/$30.00 Copyright © 2004 by the American Academy of Otolaryngology–Head and Neck Surgery Foundation, Inc. doi:10.1016/j.otohns.2003.10.009 690
MD,
MUNTHER AJLOUNI,
MD,
mary head and neck cancers. (Otolaryngol Head Neck Surg 2004;130:690-7.)
T he major treatment modalities used for the local tumor control of primary head and neck cancers are surgery and radiotherapy, although the selection of treatment modality may vary significantly with geography and institutional preference.1 Specific surgical and radiotherapeutic approaches vary depending on the treatment philosophy of the institution and individual physicians. It is even more complex and individualized when it comes to the management of recurrent tumors. Often, patients receive palliative treatment for the control of any presenting symptoms. However, there is a subset of patients with recurrent head and neck tumors who may have excellent local tumor control and good quality of function as well as improved survival. With the development of more effective cancer therapy, the overall survival of cancer patients has increased for many malignant diseases. As cancer patients are now experiencing longer survival, quality of life has become an important factor not only in the decision to treat definitively but also in the decision to offer palliative treatment of recurrent or metastatic disease. This is particularly true in the treatment of recurrent tumors of the head and neck because proper local tumor control is directly related to the quality of life with respect to phonation, swallowing, and breathing. Many patients, if not treated, lose control of these functions, which results in significant morbidity. Treatments for recurrent head and neck tumors with surgery, external beam radiotherapy, or chemotherapy have been limited. Salvage surgery for recurrent or persistent tumors is often offered only to patients who have resectable disease, sometimes with increased risk of healing, functionality, and anesthesia. However, some patients who undergo postradiotherapy salvage laryngopharyngectomy develop major complications such as pharyngocutaneous fistula or functional disability.2 Chemotherapy is also offered in an attempt to
Otolaryngology– Head and Neck Surgery Volume 130 Number 6
RYU et al 691
Table 1. Summary of patient characteristics and outcome of radiosurgery Site
Single-dose radiosurgery Floor of mouth Pharyngeal wall Base of tongue Base of skull Parotid Fractionated radiosurgery Neck node Orbit Neck node Neck node Floor of mouth Orbit Orbit Lip
Dose (tumor periphery)
Pathology
Squamous cell carcinoma Squamous cell carcinoma Squamous cell carcinoma Mucoepidermoid carcinoma Squamous cell carcinoma
18 16 18 12 16
Gy Gy Gy Gy Gy
Squamous cell carcinoma (from tonsil) Adenoid cystic carcinoma Adenoma (from stomach) Squamous cell carcinoma (from lung) Squamous cell carcinoma Adenoid cystic carcinoma Basal cell carcinoma Squamous cell carcinoma
30 30 36 36 30 30 30 36
Gy/6 Gy/6 Gy/6 Gy/6 Gy/6 Gy/6 Gy/6 Gy/6
achieve tumor shrinkage and palliation with or without radiotherapy.3 In certain circumstances, radiotherapy alone using different treatment techniques could be used.4,5 Although most of the radiotherapy for recurrent cancers is administered to the patient with palliative intent, this is also limited by previous heavy treatment and the volume of the tumor. Conventional radiotherapy uses large margins around the involved areas to compensate for infiltrating portion of the tumor and internal organ motion as well as patient motion during the treatment. In view of this, a significant amount of normal tissue, which has been already heavily treated, is included within the treatment volume. Therefore, there is a pressing need for a new approach to increase the tumor control without inflicting treatment on the already heavily treated normal tissues. Radiosurgery delivers a highly conformal, large radiation dose to a localized target tumor via a stereotactic approach. This requires accurate targeting and immobilization of patient during irradiation. Radiosurgery has not been applied to extracranial tumors due to organ motion associated with breathing and/or lack of immobilization techniques. Further, the irregularly shaped tumor would require a sophisticated conformal targeting method. With the progress of imaging and computer science and development of reliable noninvasive positioning devices,6 stereotactic target localization could be achieved in the extracranial
Response
CR (pathologic) Not evaluable CR (pathologic) Stable Stable fractions fractions fractions fractions fractions fractions fractions fractions
CR CR PR PR PR Stable CR CR
sites. We have demonstrated the use of extracranial radiosurgery for spinal tumors and reported the accuracy and precision of tumor localization and excellent clinical outcome.7,8 Among the extracranial organs, the head and neck organs have low or no breathing-related organ movement. Therefore, head and neck cancers could be ideal organ sites for stereotactic radiosurgery. We report the experience of radiosurgery for head and neck cancers with an excellent local tumor control and functional outcome. MATERIALS AND METHODS A total of 13 patients were treated with radiosurgery at Henry Ford Hospital from April 2001 to June 2002. Eleven patients had recurrent tumors that were diagnosed by clinical, radiologic, and pathologic examination. All patients had a prior diagnosis of pathologically proved primary malignant neoplasm that was previously treated with combined modalities of surgery, radiotherapy, and chemotherapy. One patient with primary lip cancer was treated with radiosurgery and was included in this study. The patient characteristics and the primary and recurrent sites of malignancy are shown in Table 1. All patients received stereotactic radiosurgery with either single or fractionated doses of radiation. The decision to use single or fractionated radiosurgery was based on several factors, including prior radiotherapy and/or surgery, location and volume of the tumor, and pa-
692 RYU et al
Otolaryngology– Head and Neck Surgery June 2004
Fig 1. Picture of stereotaxy. The image is reconstructed from the simulation CT scan with stereotactic Z-shaped localizer device (blue lines). This localizer device produces fiducial points on each side of axial CT image. The isocenter of the radiosurgery is determined at the midpoint of the target tumor volume, and the isocenter coordinate in each crosssectional plane is obtained by longitudinally connecting the same points in each plane. The distance of the isocenter from the Z-shaped localizer device gives the actual setting of the target in the anteroposterior and lateral planes. One example of focused radiosurgery beam (yellow) is directed to the target tumor (red).
tient’s condition and tolerability. For single-dose radiosurgery, a median dose of 17 Gy (range, 12 to 18 Gy) was delivered to the tumor. For fractionated radiosurgery, the radiation dose was between 30 and 36 Gy in 5 or 6 fractions with treatment twice weekly. Radiation dose was prescribed to the 90% isodose line that covers the entire target tumor as identified by either computed tomography (CT) or magnetic resonance imaging (MRI). The procedure of the radiosurgery is briefly described. Immobilization of the head and neck is accomplished by using a custom-made noninvasive mask frame with individual tongue bite. Sim-
ulation CT (Philips, Cleveland, OH) with contrast agent is obtained with cross-sections of 3 mm thickness without spacing, with a stereotactic localization box (BrainLab, Inc, Heimstetten, Germany). This localization box is mounted with a Z-shaped localizer device that is filled with aluminum strips with low CT artifact. A reconstructed image of localization system is shown in Figure 1. This localizer device produces small fiducial points on each side of the axial CT scan. The images are sent to the dedicated treatment planning computer system (BrainScan; BrainLab, Inc). The target tumor and the critical normal
Otolaryngology– Head and Neck Surgery Volume 130 Number 6
tissues are determined with consideration of clinical presentation of the tumor as well as radiologic appearance. The volumes of the tumor and normal tissues are drawn on each cross-sectional image of the simulation CT scan. An effective radiosurgery dose to the tumor with a limiting dose to the adjacent normal tissue is prescribed. The isocenter of the radiosurgery is usually determined at the midpoint of the target tumor volume, and the isocenter coordinate in each cross-sectional plane is obtained by longitudinally connecting the same points in each plane. The distance of the isocenter from the Z-shaped localizer device gives the actual setting of the target in the anteroposterior and lateral planes. The isocenter of the radiosurgery is printed as digitally reconstructed radiographs of the CT scans. Radiosurgery usually uses multiple numbers (5 to 9) of intensity-modulated radiation beams or multiple dynamic conformal arc beams to minimize the dose to the critical organs. The prescribed radiosurgery dose is again checked in relation to the final radiation isodose distribution. Once all the requirements are met, the patient is brought to the radiosurgery suite. Repositioning of the patient is achieved by using the already-made mask frame with the stereotactic localization box that was used for simulation. The isocenter position defined for each patient during the process of treatment planning is printed and attached to the localization box. Before the delivery of radiation, orthogonal portal films are obtained for the final verification and to determine the precision and accuracy of the radiosurgery. The image-guided shaped beam radiosurgery at Henry Ford Hospital uses the Novalis system (BrainLAB, Inc). It is equipped with a 3-mm-thick micromultileaf collimator that allows 3-dimensional beam shaping. The dosimetric characteristics of this treatment unit and the technique of radiation intensity modulation have been reported previously.9-12 The end points for evaluation were 1) to determine the precision and accuracy of the radiosurgery for head and neck cancers and 2) to demonstrate the local tumor control after radiosurgery by clinical and radiologic examination or by biopsy. The precision of the radiosurgery was defined as the degree of variation between the isocenters of the radiosurgery treatment plan and of the portal films taken at the time of radiation delivery. It was
RYU et al 693
measured in millimeters by image fusion of the digitally reconstructed radiograph of CT simulation and portal films that were taken at the actual treatment. For assessment of tumor response, complete response (CR) was defined as disappearance of the tumor by clinical and radiologic examination or by biopsy. Partial response (PR) was defined as reduction of the tumor size in its greatest dimension by more than 50%. Although it was not the primary end point, patients were also evaluated for relief of pain and symptom improvement. The median follow-up time was 8 months (range, 6 to 16 months). RESULTS Precision and Accuracy of Radiosurgery The precision for a given isocenter between the simulation and actual treatment position was 1.4 ⫾ 0.7 mm. We also measured the radiation dose at the isocenter using the same positioning parameters of the individual patients in a phantom with a micro–ion chamber. The average deviation of the measured dose from the estimated dose was 2%. These 2 independent measurements were mutually corroborative of the accuracy of radiosurgery in the head and neck. Local Tumor Control The patient characteristics with primary treatment and radiosurgery for the recurrent tumor are shown in Table 1. Nine of 13 patients (70%) showed significant response (complete or ⬎50% tumor reduction) to radiosurgery. Complete response was achieved in 6 patients (2 with singledose radiosurgery, 4 with fractionated radiosurgery). Partial response defined as ⬎50% reduction of tumor volume was seen in an additional 3 patients with fractionated radiosurgery. There were 3 patients with recurrent salivary gland tumors (1 with mucoepidermoid carcinoma involving the base of skull and 2 with adenoid cystic carcinoma involving the orbit). One patient with recurrent adenoid cystic adenocarcinoma of the orbit showed nearcomplete disappearance of the tumor at 10-month follow-up MRI. The other 2 patients were stable with significant symptom relief. An example of a patient with a complete pathologic response is shown in Figure 2. This patient initially had squa-
694 RYU et al
Otolaryngology– Head and Neck Surgery June 2004
Fig 2. Radiosurgery of a biopsyproved recurrent tumor at the base of the tongue. (A) Preradiosurgery CT scan shows an enhancing mass (arrow) with microscopic feature of invasive squamous carcinoma. (B) Radiosurgery plan with isodose distribution. The tumor was treated with singledose 18-Gy radiosurgery. The high radiation dose was delivered to the enhancing target tumor, whereas the dose to the surrounding normal tissue was minimal. (C) CT scan taken 6 months postradiosurgery showed the enhancing lesion reduced in size (arrowhead). Biopsy showed no tumor cells.
Otolaryngology– Head and Neck Surgery Volume 130 Number 6
mous cell carcinoma of an unknown primary neck node that was treated with radical neck dissection and radiotherapy. The patient developed a second primary tumor at the base of the tongue after 12 years. This lesion was treated with a single dose of radiosurgery of 18 Gy. A follow-up CT scan at 6 months after the radiosurgery showed a smaller residual enhancement; biopsy revealed no tumor cells. At 8 months after radiosurgery, the patient developed mucosal irregularity of the area with induration on clinical examination. This area was locally excised and pathologic examination showed no evidence of residual tumor. Functional Outcome Four patients who presented with pain had a significant relief of symptom. Three patients had complete pain relief. One patient who was treated for palliation of pain with single-dose 12-Gy radiosurgery for the recurrent mucoepidermoid cancer at the skull base had partial pain relief to the level that pain medication could be significantly reduced. This patient had 3 operations and heavy radiotherapy for the primary site at the maxillary/ palatal area and interstitial brachytherapy to the palatal area including the skull base before radiosurgery. After 10 months, the tumor progressed. Two patients who had metastatic neck nodes from stomach and lung cancers had partial tumor response with satisfactory pain relief but developed systemic metastasis. None of the patients had any long-term adverse effects. Acute side effects were temporary and minimal, with mild mucositis and skin irritation depending on the tumor location. Preservation of function and cosmetic outcome were well maintained after radiosurgery. In all of these patients, subjective functional preservation was satisfactory in the ability of swallowing and phonation. A patient with lip cancer at the oral commissure had the original tumor measured 3.5 cm with ulceration and subcutaneous infiltration. This patient refused surgical treatment. The tumor was treated with fractionated radiosurgery to a total dose of 36 Gy in 6 fractions twice a week. The patient had complete resolution of the tumor with excellent functional and cosmetic outcome at 2 months after radiosurgery.
RYU et al 695
DISCUSSION Radiosurgery delivers a highly conformal large radiation dose to a localized target tumor. It requires accurate targeting and immobilization of the target organ during irradiation. Therefore, the accuracy and precision for radiosurgery should be tested first before administering the large dose of highly conformal radiation. Most experience with stereotactic radiosurgery involves intracranial lesions such as arteriovenous malformations and malignant and benign tumors. For radiosurgery of intracranial tumors, the precision of the isocenter target has usually been within 1 mm. Radiosurgery has not been applied to extracranial tumors due to organ motion associated with breathing and/or lack of immobilization techniques. Among the extracranial organs, the spines and some head and neck sites have the least breathing-related movement. This makes the head and neck sites suitable for stereotactic radiosurgery. We have demonstrated the accuracy of the spinal radiosurgery within 1.3 mm, where the spinal cord is intimately located with the spinal tumors.8 In this study, we report the accuracy of the radiosurgery in the head and neck sites within 1.5 mm. This magnitude of accuracy is certainly acceptable for treatment of the head and neck cancers. One potential problem would be unexpected swallowing or coughing that may occur during the delivery of radiation. This may need to be explored to provide more accurate radiosurgery treatment. The tumors involving the head and neck can be irregular in shape and are often associated with significant infiltration to the surrounding soft tissue. This target irregularity makes the use of radiosurgery difficult. By including the irregular target within the radiosurgery volume, normal tissue immediately adjacent to the tumor also receives high radiation doses, particularly in recurrent tumors. The criteria of the target tumor delineation have to be defined as experience with radiosurgery is gained. In this study with recurrent tumors, margins of 2 to 3 mm from the radiographically and clinically visible or palpable tumor were allowed to account for the infiltrative disease. With this approach, tumor recurrence in the marginal zone of radiosurgery was not seen during the follow-up period. A representative example of an intensitymodulated treatment plan is shown in Figure 2.
Otolaryngology– Head and Neck Surgery June 2004
696 RYU et al
The radiation dose was prescribed for the volume included within the 90% radiation isodose line. To achieve tumor control and to overcome the potential complications of the adjacent normal tissue by treatment with high-dose radiosurgery, accurate tumor localization and precise shaping of the tumor by the radiation beam are required. Radiosurgery will provide a rapid radiation dose fall-off immediately outside of the target volume by shaping the radiation beam with micromultileaf collimation and by modulating the radiation intensity. These techniques increase the ability to conform to the irregular tumor shape. Selection of radiosurgery dose and tolerance of the surrounding normal tissue to the single or fractionated dose(s) of radiosurgery are not well defined. Experience with interstitial implantation showed the safe radiation dose of retreatment in the range of 50 to 60 Gy to the entire tumor volume with low-dose rate of radiation. This lowdose rate brachytherapy has been used for treatment of recurrent tumors with 30% to 70% salvage rates and 30% to 40% complication rates.4,13 Radiosurgery has been mostly used for recurrent nasopharyngeal cancers with small fractional doses and showed good local tumor control.5,14-16 These studies involved the use of fractionated radiosurgery/radiotherapy to a rather larger volume, encompassing the entire nasopharynx. Patient tolerance was reported to be excellent. We used the radiation dose of 12 to 18 Gy for singledose radiosurgery and 30 to 36 Gy in 5 or 6 fractions for fractionated radiosurgery. Radiation dose-response of specific tumors could not be evaluated due to the different tumor histology and small number of patients. In the case of the squamous cell carcinomas, a higher dose appears to achieve superior tumor response, based on our clinical experience. Radiosurgery should be carefully planned in patients who have had heavy prior treatment because of the higher risk of complications from the additional therapy in recurrent diseases. For dose selection of retreatment, careful consideration must be given to the previous radiation to normal tissues (dose and fields); the extent and pathology of the recurrent tumor; the shape, size, and location of the tumor; the condition of the surrounding normal tissues; the patient’s general status includ-
ing comorbidities; and the ultimate goal of treatment.17 We encountered no unacceptable acute or subacute effects on normal tissue and no complications. Although the follow-up was relatively short, there have not been any signs that might be suggestive of long-term complications. Patients who have advanced primary tumors or recurrent tumors that are locally controlled by radiation alone have chances to develop a major complication and a nonfunctioning larynx.18 However, the radiosurgery is delivered only to the local tumor rather than to the large volume containing lymph node areas, thereby reducing the chance of developing complications. CONCLUSION This study demonstrated the proof of principle and the potential of radiosurgery for the treatment of head and neck tumors. However, special consideration should be given in selection of treatment and patients, particularly with the lymphatic drainage and tumor infiltration of the primary head and neck cancers that may warrant broader radiotherapy with nodal irradiation. This study offers a potential to extend the use of radiosurgery to benign tumors,19 as a viable treatment option for recurrent head and neck cancers, and, possibly, for certain selected primary cancers of the head and neck. REFERENCES
1. Mendenhall WM, Hinderman RW, Morris CG, et al. Management of supraglottic carcinoma. Int J Radiat Oncol Biol Phys 2002;53:793. 2. Chee N, Siow JK. Pharyngocutaneous fistula after laryngectomy: incidence, predisposing factors and outcome. Singapore Med J 1999;40:130. 3. Wong ZW, Tan EH, Yap SP, et al. Chemotherapy with or without radiotherapy in patients with locoregionally recurrent nasopharyngeal carcinoma. Head Neck 2002;24:549. 4. Puthawala AA, Syed MN. Interstitial re-irradiation for recurrent and/or persistent head and neck cancers. Int J Radiat Oncol Biol Phys 1987;13:1113. 5. Xiao J, Xu G, Miao Y. Fractionated stereotactic radiosurgery for 50 patients with recurrent or residual nasopharyngeal carcinoma. Int J Radiat Oncol Biol Phys 2001;51:164. 6. Alheit H, Dornfeld S, Dawel M, et al. Patient position reproducibility in fractionated stereotactically guided conformal radiotherapy using the BrainLab mask system. Strahlenther Onkol 2001;177:264. 7. Rock J, Kole M, Yin FF, et al. Radiosurgical treatment for Ewing’s sarcoma of the lumbar spine. Spine 2002; 27:471.
Otolaryngology– Head and Neck Surgery Volume 130 Number 6
8. Ryu S, Yin FF, Rock J, et al. Image-guided and intensitymodulated radiosurgery for spinal metastasis. Cancer 2003;97:2013. 9. Cosgrove VP, Jahn U, Pfaender M, et al. Commissioning of a micro multi-leaf collimator and planning system for stereotactic radiosurgery. Radiother Oncol 1999;50:325-36. 10. Yin F-F, Zhu J, Hui Y, et al. Dosimetric characteristics of Novalis shaped beam surgery unit. Med Phys 2002;29: 1729-38. 11. Yin F-F, Ryu S, Ajlouni M, et al. A technique of intensity-modulated radiosurgery (IMRS) for spinal tumors. Med Phys 2002;29:2815-22. 12. Llacer J, Solberg TD, Promberger C. Comparative behavior of the Dynamically Penalized Likelihood algorithm in inverse radiation therapy planning. Phys Med Biol 2001;46:2637-63. 13. Nag S, Cano ER, Demanes DJ, et al. The American brachytherapy society recommendations for high dose rate brachytherapy for head and neck carcinoma. Int J Radiat Oncol Biol Phys 2001;50:1190-8.
RYU et al 697
14. Ahn YC, Lee KC, Kim DY, et al. Fractionated stereotactic radiation therapy for extracranial head and neck tumors. Int J Radiat Oncol Biol Phys 2000;48:501-5. 15. Chen HJ, Leung SW, Su CV. Linear accelerator based radiosurgery as a salvage treatment for skull base and intracranial invasion of recurrent nasopharyngeal carcinomas. Am J Clin Oncol 2001;24:255-8. 16. Chua DT, Sham JS, Hung KN, et al. Salvage treatment for persistent and recurrent T1-2 nasopharyngeal carcinoma by stereotactic radiosurgery. Head Neck 2001;23:791-8. 17. Ryu S, Gorty S, Kazee AM, et al. “Full dose” re-irradiation of human cervical spinal cord. Am J Clin Oncol 2000;23:29-31. 18. Mancuso AA, Mukherji SK, Schmalfuss I, et al. Preradiotherapy computed tomography as a predictor of local control in supraglottic carcinoma. J Clin Oncol 1999;17: 631-7. 19. Foote RL, Pollock BE, Gorman DA, et al. Glomus jugulare tumor: tumor control and complications after stereotactic radiosurgery. Head Neck 2002;24:332-8.