PII S0360-3016(00)00612-X

3D-CRT

FRACTIONATED STEREOTACTIC RADIATION THERAPY FOREXTRACRANIAL HEAD AND NECK TUMORS

YONG CHAN AHN, M.D., KYU CHAN LEE, M.D., DAE YONG KIM, M.D., SEUNG JAE HUH, M.D.,IN HWAN YEO, PH.D., DO HOON LIM, M.D., MOON KYUNG KIM, M.D., KYUNG HWAN SHIN, M.D.,

SUKWON PARK, M.D., AND SEUNG HEE CHANG, M.D.

Department of Radiation Oncology, Samsung Medical Center, Sungkyunkwan University, School of Medicine, Seoul, Korea

Background: This study is to report the clinical experiences of fractionated stereotactic radiation therapy (FSRT)for extracranial head and neck tumors.Methods and Materials: Between the period of July 1995 and November 1998, 48 patients with extracranial headand neck tumors were given FSRT as a boost and sole modality. Individualized treatment planning wasperformed using XKnife-3 system with relocatable Gill-Thomas-Cosman frame. In 24 patients, FSRT wasapplied as a boost technique following the 2-dimensional conventional external radiation therapy (ERT); in 24patients FSRT was the sole radiotherapy modality. The primary diseases in the boost group consisted ofnasopharynx cancer (19), lacrimal gland adenoid cystic carcinoma (3), orbital lymphoma (1), and skull-baserecurrence of maxillary sinus adenoid cystic carcinoma (1). The primary diseases in the sole modality groupconsisted of recurrent nasopharynx cancer (12), orbital pseudotumor (4), skull-base recurrence of maxillarysinus, submandibular gland, and hypopharynx cancers (3), orbital rhabdomyosarcoma (2), orbital lymphoma (1),orbital metastasis of neuroblastoma (1), and nasal cavity melanoma (1). The fractionation schedule was to give5 treatments per one week and the fractional doses were 2.0–3 Gy depending on the treatment aim and the FSRTvolume. The FSRT doses varied depending on the nature of the primary diseases.Results: The local tumor response in nasopharynx cancer patients was excellent compared to retrospective datawithout occurrence of unexpectedly severe complication. FSRT to other regions was well tolerated by the patients andresulted in good to excellent local tumor responses with no unacceptable side effects as expected by the authors.Conclusion: Based on the current observations, FSRT is a very effective and safe modality in the treatment ofextracranial head and neck tumors. © 2000 Elsevier Science Inc.

Stereotactic radiotherapy; Head/neck tumors.

INTRODUCTION

Fractionated stereotactic radiation therapy (FSRT) is a modi-fication of stereotactic radiosurgery (SRS). SRS contains themerits of the mechanical accuracy of stereotaxy, the high andhomogeneous dose distribution within a small target volume,and the rapid dose fall-off in the surrounding normal tissues(1). SRS was originally developed for the treatment of theintracranial benign lesions including pituitary adenoma, arte-riovenous malformations, and acoustic neuromas. It did nottake long for the SRS’s indications to expand to the intracranialmalignancies, either primary or metastatic, and to the tumors ofthe skull-base and extracranial locations (2–5). The substantialrisk of radiation-induced morbidity of the single high-doseradiation, however, has motivated the development of FSRTtechnique (6–8). FSRT requires the technology of SRS, whichincludes the sophisticated treatment planning system, the high-energy therapy machine, and the stereotactic immobilization

devices. The latter, however, must be uniquely designed for therelocatable daily uses, and the relocatable stereotactic framesystem is the essential component for FSRT technique. Severalrelocatable frame systems are currently available. FSRT hasthe radiobiologic advantage of conventional fractionation inaddition to the mechanical precision by stereotactic devices.Samsung Medical Center adopted the FSRT technique for thefirst time in Korea and 48 patients with extracranial head andneck tumors were treated with FSRT between July of 1995 andNovember of 1998. This is to report the authors’ clinicalexperience with FSRT for the treatment of extracranial tumorsat Samsung Medical Center.

FSRT PROCEDURE

For the stereotactic three-dimensional coordinate def-inition, the authors used the individually made relocat-

Address reprint requests to: Dr. Yong Chan Ahn, Department ofRadiation Oncology, Samsung Medical Center, SungkyunkwanUniversity, School of Medicine, 50 Ilwon-dong, Kangnam-Ku,Seoul, 135-710, Korea. E-mail: ycahn@smc.samsung.co.kr

This work was presented at the 2nd S. Takahashi Memorial 3-DConformal Radiation Therapy Symposium in Nagoya, Japan, inDecember 1998.

Accepted for publication 24 June 1999.

Int. J. Radiation Oncology Biol. Phys., Vol. 48, No. 2, pp. 501–505, 2000Copyright © 2000 Elsevier Science Inc.Printed in the USA. All rights reserved

0360-3016/00/$–see front matter

501

able Gill-Thomas-Cosman (GTC) stereotactic frame (9 –10). The FSRT dose planning was done using theXKnife-3 system. The contrast enhancing CT scan wasobtained with the GTC frame on, and was transferred tothe XKnife-3 system for the patient’s contour and ana-tomical information. The radiation therapy dose planningwas undertaken under the concept of multiple noncopla-nar arc rotations on an individual basis considering thesize, shape, and location of the lesions after the three-dimensional reconstruction of the CT anatomical data.The size of FSRT cone, three-dimensional stereotacticcoordinates of the isocenter, and treatment beam param-eters, including couch angles and arc ranges, were deter-mined through the dose planning. The authors used 4 MVX-rays from Clinac 600C (Varian Association, Palo Alto,CA). The fractionation schedule was to give 5 dailytreatments during 1 week, and the fractional dose rangedfrom 2.0 Gy to 3.0 Gy depending on the treatment aimand the FSRT volume. Before each treatment, the veri-fication of mechanical accuracy of the linear acceleratorand the quality-assurance procedure were performed ac-cording to the recommendation (11).

FSRT AS A BOOST MODALITY

In 24 patients, FSRT was applied as a boost techniquefollowing conventional external radiation therapy (ERT).The primary diseases consisted of nasopharynx cancer(19), lacrimal gland adenoid cystic carcinoma (3), orbitallymphoma (1), and skull-base recurrence of maxillarysinus adenoid cystic carcinoma (1).

The median boost FSRT dose in nasopharynx cancerpatients was 16 Gy (8 – 40 Gy) after the median conven-tional ERT dose of 55.8 Gy (36 – 61.2 Gy). The fractionaldose by FSRT was 2.0 Gy and the median total dose tothe primary disease site was 74.4 Gy (67.4 – 80 Gy). Allpatients except one also received systemic chemotherapyin addition to the current radiation therapy, where 16received concurrent chemoradiotherapy. With the medianfollow-up of 28 months (3– 45 months), all except onepatient who died of chemotherapy related sepsis are aliveand well without evidence of disease. The local tumorresponse was excellent and complete response wasachieved in 18 patients (94.7%). There was one incidenceof local mucosal necrosis, which eventually recovered,and there were three recurrences outside the FSRT vol-ume (anterior nasal cavity, anterior ethmoid sinus, andmiddle ear cavity), all of which had been successfullysalvaged by FSRT re-irradiation with follow-up ofgreater than 1 year. The local control rates and survivalrates calculated by Kaplan–Meyer method at 4 years were89% and 75%, respectively.

Three lacrimal gland adenoid cystic carcinoma patientsreceived FSRT boost after conventional two-dimensionalERT of 40 Gy. The fractional dose by FSRT was 2.0 Gy andthe boost doses were 20 Gy and 26 Gy for 2 patients whosegross tumors were removal before radiation therapy, while

it was 30 Gy for one patient who received definitive RTwithout surgery. All three patients are alive and well for7–11 months.

One patient with orbital lymphoma received FSRTboost of 24 Gy after conventional two-dimensional ERTof 16 Gy. The fractional dose was 2.0 Gy and the mainaim of FSRT in this patient was to prevent cataractformation that might not be prevented by two-dimen-sional ERT.

One patient with skull-base recurrence of maxillarysinus cancer received FSRT of 25 Gy after two-dimen-sional ERT of 25 Gy. The fractional dose was 2.5 Gy andthe reason of applying FSRT was to give less radiation tothe brain parenchyma and to prevent radiation necrosis asthis patient was previously managed by high dose con-ventional ERT to the maxillary sinus and the orbit.

FSRT AS A SOLE MODALITY

In 24 patients, FSRT was applied as the sole radiotherapymodality without the aid of conventional ERT. The primarydiseases consisted of recurrent nasopharynx cancer (12),orbital pseudotumor (4), skull-base recurrence of maxillarysinus, submandibular gland, and hypopharynx cancers (3),orbital rhabdomyosarcoma (2), orbital lymphoma (1), or-bital metastasis of neuroblastoma (1), and nasal cavity mel-anoma (1)

FSRT was applied as re-irradiation modality for 12locally recurrent nasopharynx cancer patients. The me-dian total dose of FSRT was 54 Gy (45– 65 Gy) and thefractional FSRT dose was 2.5 Gy or 3.0 Gy depending onthe FSRT volume and the patients’ general condition. In3 patients, FSRT was their third radiotherapy applicationto the same region throughout their disease course. Threepatients received systemic chemotherapy in addition toFSRT, and 1 patient with middle ear recurrence receiveddebulking operation before FSRT. With the median fol-low-up of 13 months (8 –24 months), there have been 8survivors for 8 –24 months and 4 deaths, of which 3 diedof distant metastases and 1 of radiation myelopathy.Among the survivors, 3 patients have distant metastases,and the remaining 5 patients are well without evidence ofdisease. The local tumor response was excellent andcomplete response was achieved in 11 patients (91.7%).Figure 1 demonstrates a patient with recurrent nasophar-ynx cancer who was successfully treated with FSRT. Thelocal control rates and survival rates calculated byKaplan–Meyer method at 2 years were 92% and 60%,respectively.

The authors applied FSRT to 4 steroid-refractory orbitalpseudotumor patients in whom one or two extra-ocularmuscles were involved by the inflammatory process. Themain purpose of employing FSRT in this setting was to giveless radiation dose to lens and the visual pathways. 20 Gyover 10 fractions was successfully delivered with symptom-atic improvement in all.

Three patients with symptomatic skull-base recur-

502 I. J. Radiation Oncology● Biology ● Physics Volume 48, Number 2, 2000

rences were given FSRT for palliation. The fractionaldose was 2.5 Gy, and the total doses were 50 Gy in 2patients and 61.5 Gy in 1 patient. The goal of symptom-atic relief was achieved in all 3 patients without severeside effects.

Two childhood orbital rhabdomyosarcoma patientswere given FSRT to the primary mass lesion during theircombined chemoradiotherapy protocol. The total radia-tion doses were 46 Gy and 50 Gy at 2Gy/fraction, and theauthors expected less RT volume by FSRT which wouldlead to less bone growth retardation. Other indications ofFSRT included orbital lymphoma, orbital metastasis fromneuroblastoma, and nasal cavity mailgnant melanoma.

DISCUSSION

The fact that the risk of late neurological damageproportionally increases with the increment of the frac-tional dose is evident on the radiobiological as well as onthe clinical grounds. The idea of combining the stereo-tactic methods and the concept of fractionation was mo-tivated after the observation of neurological complica-tions in the single high dose SRS practices. FSRT is amodification of SRS enabling conventional fractionationwithout losing the advantage of the mechanical accuracyby stereotaxy. FSRT may have advantages over SRS in a

few respects. First, more radiation damage to the tumortissue and less radiation damage to the normal neuraltissue can be expected from FSRT, as this has been theradiobiological principle in usual fractionated radiationtherapy. Second, the tumors located close to the criticalstructures may be more safely treated by FSRT, as thefractional dose is the most important factor in late neu-rological damage. Third, relatively larger targets may betreated more safely by FSRT than SRS, as the volumeeffect is much more problematic in single high dose SRSthan FSRT. Indications for FSRT may also be extendedto the benign and malignant lesions located not onlywithin the cranium but also in the skull base and theextracranial structures including the nasopharynx, thenasal cavity, the orbit, and the paranasal sinuses. How-ever, FSRT may not be an ideal recommendation forthose lesions that benefit more from the single high-doseradiation (i.e., arteriovenous malformations, functionalSRS, etc.).

It has been generally accepted that there is a positivedose–response relationship in the local control of nasophar-ynx cancer with respect to the tumor size. The usuallyrecommended radiation doses to the primary tumor are60–65 Gy for T1–2 lesions, and 65–70 Gy for T3–4 le-sions, though there may be some variations depending onthe histologic subtypes. The local recurrences, on the aver-

Fig. 1. Example case of recurrent nasopharynx cancer: A 57-year-old woman presented with recurrent tumor at the rightnasopharynx and upper neck node 52 months after definitive conventional external radiation therapy and chemotherapy(upper figures). She was managed with FSRT (50 Gy/20 fractions) alone and the follow-up MR imaging taken at 1month post-FSRT showed near complete disappearance of the tumor (lower figures). This woman has been alive over13 months with asymptomatic multiple lung metastases.

503FSRT for head/neck tumors● Y. C. AHN et al.

age, have been reported in one-third of the total patientswith the range of 18–54% (12, 13). Re-irradiation for localrecurrences has been accepted as the most optimal treatmentmodality, which is without the immediate life-threateningmorbidity or mortality as that of surgery, and has frequentlypromising treatment results. Summarizing the reports onre-irradiation, the local control was achieved in 30–60%,and the 5-year survival rates were approximately 20% (14,15). The nasopharynx is surrounded by several importantanatomic structures, such as the brainstem, the spinal cord,the pituitary gland, the optic pathway, the temporomandib-ular joint, the parotid gland, etc. These structures are oftenvulnerable to high-dose radiation and have been the mainobstacles in the conventional two-dimensional radiationtherapy techniques. The authors have employed FSRT togive adequate radiation dose to the small volume of localtumor without causing severe radiation morbidity in thetreatment of nasopharynx cancer both in the boost andre-irradiation settings. The local control rates and survivalrates calculated by Kaplan–Meyer method at 4 years were89% and 75% in the boost setting, and the correspondingfigures at 2 years were 92% and 60% in the re-irradiationsetting, respectively. The authors’ experience of FSRT asre-irradiation to the first 3 cases with locally recurrentnasopharynx cancer was reported previously (16). Eventhough the follow-up has not been matured yet, these in-terim results of local control and survival are fairly goodcompared to the historic data (17–20).

Other main indications for FSRT in this report were theorbital tumors and the recurrences after previous radiationtherapy to the same region. As the radiation volumes byFSRT are, in general, much smaller than those by theconventional two-dimensional ERT techniques, the authorsanticipated less volume of normal tissue adjacent to thetumor volume to receive a high radiation dose and thus beprone to radiation morbidity. This, in turn, throws light onthe possibility of dose escalation and improved local tumorcontrol. The authors could achieve fairly good local tumorresponses without occurrences of severe morbidity by FSRTto the head and neck tumors. The dose–volume relationshipmay be even more important and critical in the re-irradiationsetting, where quite a large amount of normal tissues havealready received radiation dose up to their tolerance limits.The traditional conservative approach for treatment of re-currences at a previously irradiated area has been just toobserve and give supportive management but not to re-irradiate in fear of radiation complications. Even though thenumber of patients is rather small and the follow-up periodis short, the authors believe that a well-designed re-irradi-ation treatment plan using FSRT technique is worthwhilebased on our current experiences.

In brief, the authors strongly believe that FSRT is a veryeffective and safe radiation therapy modality in the treat-ment of extracranial head and neck tumors, and that furtherlarge-scale prospective clinical trials are desirable.

REFERENCES

1. Shrieve DC, Tarbell NJ, Kooy HM,et al. Fractionated (relo-catable) stereotactic radiotherapy. In DeVita VT, Hellman S,Rosengerg SA, editors. Cancer—Principles and practices ofoncology. 5th ed. Philadelphia: Lippincott-Raven Publishers;1997. p. 3107–3114.

2. Kondziolka D, Lunsford LD. Stereotactic radiosurgery forsquamous cell carcinoma of the nasopharynx.Laryngoscope1991;101:519-522.

3. Buatti JM, Friedman WA, Bova FJ,et al. Linac radiosurgeryfor locally recurrent nasopharyngeal carcinoma: rationale andtechnique.Head Neck1995;17:14–19.

4. Firlik KS, Kondziolka D, Lunsford LD,et al.Radiosurgery forrecurrent cranial base cancer arising from the head and neck.Head Neck1996;18:160–166.

5. Cmelak AJ, Cox RS, Adler JR,et al. Radiosurgery for skullbase malignancies and nasopharyngeal carcinoma.Int J Ra-diat Oncol Biol Phys1997;37(5):997–1003.

6. Nedzi LA, Kooy H, Alexander E III,et al. Variables associ-ated with the development of complications from radiosurgeryof intracranial tumors.Int J Radiat Oncol Biol Phys1991;21:591–599.

7. Flickinger JC, Lunsford LD, Kondziolka D,et al. Radiosur-gery and brain tolerance: An analysis of neurodiagnostic im-aging changes after gamma knife radiosurgery for arterio-venous malformations.Int J Radiat Oncol Biol Phys1992;23:19–26.

8. Souhami L, Oliver A, Podgorsak EB. Fractionated stereotacticradiation therapy for intracranial tumors.Cancer 1991:68:2101–2108.

9. Gill SS, Thomas DGT, Warrington AP,et al. Relocatableframe for stereotactic external beam radiotherapy.Int J RadiatOncol Biol Phys1991;20:599–603.

10. Kooy HM, Dunbar SF, Tarbell NJ,et al. Adaptation andverification of the relocatable Gill-Thomas-Cosman frame instereotactic radiotherapy.Int J Radiat Oncol Biol Phys1994;30:685–691.

11. XKnife-3 user’s guide. Version 3.0, Radionics software ap-plications, Inc.

12. Lee AWM, Poon YF, Foo W,et al. Retrospective analysis of5037 patients with nasopharyngeal carcinoma treated during1976-1985: Overall survival and patterns of failure.Int JRadiat Oncol Biol Phys1992;23(2):261–270.

13. Yang TS, Ng KT, Wang HM,et al. Prognostic factors oflocoregionally recurrent nasopharyngeal carcinoma—A retro-spective review of 182 cases.Am J Clin Oncol1996;19(4):337–343.

14. Lee AWM, Law SCK, Foo W,et al.Retrospective analysis ofpatients with nasopharyngeal carcinoma treated during 1976–1985: Survival after local recurrence.Int J Radiat Oncol BiolPhys1993;26(5):773–782.

15. Pryzant RM, Wendt CD, Delclos L,et al. Re-treatment ofnasopharyngeal carcinoma in 53 patients.Int J Radiat OncolBiol Phys1992;22(5):941–947.

16. Ahn YC, Kim DY, Huh SJ,et al. Fractionated stereotacticradiation therapy for locally recurrent nasopharynx cancer:Report of three cases.Head Neck1999;21:338 –345.

17. Morton RP, Liavaag PG, McLean M,et al. Transcervico-mandibulo-palatal approach for surgical salvage of recurrentnaspopharyngeal cancer.Head Neck1996;18:352–358.

504 I. J. Radiation Oncology● Biology ● Physics Volume 48, Number 2, 2000

18. Choy D, Sham JST, Wei WI,et al. Transpalatal insertion ofradioactive gold grain for the treatment of persistent andrecurrent nasopharyngeal carcinoma.Int J Radiat Oncol BiolPhys1993;25(3):505–512.

19. Leung TW, Tung SY, Wong VYW,et al. High doserate intracavitary brachytherapy in the management of na-

sopharyngeal carcinoma.Acta Oncologica1996;35(1):43–47.

20. Feehan PE, Castro JR, Phillips TL,et al. Recurrent locallyadvanced nasopharyngeal carcinoma treated with heavycharged particle irradiation.Int J Radiat Oncol Biol Phys1992;23(4):881–884.

505FSRT for head/neck tumors● Y. C. AHN et al.