C

CosoSdfrr8horacpm

sI2

M

Int. J. Radiation Oncology Biol. Phys., Vol. 60, No. 4, pp. 1120–1126, 2004Copyright © 2004 Elsevier Inc.

Printed in the USA. All rights reserved0360-3016/04/$–see front matter

doi:10.1016/j.ijrobp.2004.05.064

LINICAL INVESTIGATION Head and Neck

CLINICAL RESULTS OF PROTON BEAM THERAPY FOR SKULL BASECHORDOMA

HIROSHI IGAKI, M.D.,*† KOICHI TOKUUYE, M.D.,*† TOSHIYUKI OKUMURA, M.D.,‡

SHINJI SUGAHARA, M.D.,† KENJI KAGEI, M.D.,*† MASAHARU HATA, M.D.,*† KIYOSHI OHARA, M.D.,†

TAKAYUKI HASHIMOTO, M.D.,† KOJI TSUBOI, M.D.,§ SHINGO TAKANO, M.D.,§

AKIRA MATSUMURA, M.D.,§ AND YASUYUKI AKINE, M.D.*†

�Proton Medical Research Center and Departments of †Radiation Oncology and §Neurosurgery, University Hospital, University ofTsukuba, Ibaraki, Japan; and ‡Department of Radiology, Ibaraki Prefectural Central Hospital, Tomobe, Japan

Purpose: To evaluate clinical results of proton beam therapy for patients with skull base chordoma.Methods and Materials: Thirteen patients with skull base chordoma who were treated with proton beams withor without X-rays at the University of Tsukuba between 1989 and 2000 were retrospectively reviewed. A mediantotal tumor dose of 72.0 Gy (range, 63.0–95.0 Gy) was delivered. The patients were followed for a median periodof 69.3 months (range, 14.6–123.4 months).Results: The 5-year local control rate was 46.0%. Cause-specific, overall, and disease-free survival rates at 5 yearswere 72.2%, 66.7%, and 42.2%, respectively. The local control rate was higher, without statistical significance,for those with preoperative tumors <30 mL. Partial or subtotal tumor removal did not yield better local controlrates than for patients who underwent biopsy only as the latest surgery.Conclusion: Proton beam therapy is effective for patients with skull base chordoma, especially for those withsmall tumors. For a patient with a tumor of <30 mL with no prior treatment, biopsy without tumor removalseems to be appropriate before proton beam therapy. © 2004 Elsevier Inc.

Proton beams, Radiotherapy, Chordoma, Skull base.

oprORst(bbs

P

stT

O

aCf

INTRODUCTION

hordomas are rare bone tumors of notochord origin, whichften develop in the skull base (e.g., at the clivus or para-ellar regions). Skull base chordomas constitute only 0.4%f primary brain tumors in Japan (1) and 0.2% in the Unitedtates (2). Preferred treatment strategies for skull base chor-oma are definitive radical surgery or maximal resectionollowed by postoperative radiotherapy (3–6). The localecurrence rate after incomplete resection followed by X-ay radiotherapy has been reported to be approximately0%–100% (7–10). Even such aggressive local therapiesave not resulted in satisfactory local control rates becausef the proximity to critical neurovascular structures and theadioresistant nature of the tumor cells; cure can rarely bechieved (7, 8). Metastatic or disseminated diseases are notommon in patients with skull base chordoma, which im-lies that local tumor control by more aggressive treatmentight provide an opportunity for long-term survival.Proton beams are characterized by a well-defined range

Reprint requests to: Hiroshi Igaki, M.D., Proton Medical Re-earch Center, University of Tsukuba, 1-1-1, Tennohdai, Tsukuba,baraki 305-8575, Japan. Tel: (�81) 29-853-7117, Fax: (�81)9-853-7102; E-mail: igaki-tky@umin.ac.jpA portion of this study was presented at the 15th Annual

eeting of the Japanese Society for Therapeutic Radiology and

1120

f penetration; delivery of higher doses to the tumors isossible, whereas doses to adjacent normal structures areestricted below the tolerable levels by proton beam therapy.ur results of proton beam therapy at the Proton Medicalesearch Center, University of Tsukuba, initiated in 1983,

uggest its superiority over conventional radiotherapy inerms of both efficacy and toxicity for several malignancies11–14). In the present study, we report results of protoneam therapy for patients with skull base chordoma treatedetween 1989 and 2000 and discuss the new treatmenttrategy.

METHODS AND MATERIALS

atientsThirteen patients with histopathologically confirmed

kull base chordoma, who received proton beam therapy athe Proton Medical Research Center of the University ofsukuba between January 1989 and June 2000, were retro-

ncology, November 21–23, 2003, Tokyo, Japan.Supported in part by Grants-in-Aid for Cancer Research (15-9)

nd the Second Term Comprehensive 10-year Strategy for Cancerontrol (H-15-006) from the Ministry of Health, Labor and Wel-

are of the Japanese Government.

Received Feb 18, 2003. Accepted for publication May 10, 2004.

sfyptr

T

asrbwc

Stcmbrfibwsj

ai5paTGc

scmflbeG2roo

E

mmtfmATtpm(

1lmpg

�

1121Proton beam therapy for skull base chordoma ● H. IGAKI et al.

pectively reviewed. Of the 13 patients, 5 were male and 8emale; the patients’ ages ranged 14–74 years (median 61ears) (Table 1). All patients were considered to be inap-ropriate for definitive radical surgery because of either theumor proximity to critical structures or their refusal ofadical surgery.

reatmentProton beams with the energy of 250 MeV, generated bybooster synchrotron at the High Energy Accelerator Re-

earch Organization, were used for irradiation. Treatmentooms were equipped with either horizontal or verticaleams. The beam time allocated for proton beam therapyas 4 hours per day, 120 days per year, in three separated

lustered periods of 9–10 weeks.Patients were immobilized with custom-made masks.

ubsequently, 5-mm-thick CT images were obtained in thereatment position. The clinical target volume (CTV) wasontoured manually on serial CT images around gross tu-or volume (GTV) with a 5–10-mm margin. Dose distri-

utions were calculated with a simple broad-beam algo-ithm. Beam delivery devices that included a ridge filter, ane degrader, and a collimator were selected automaticallyy the treatment-planning system. The compensation bolusas fabricated with tissue-equivalent material (MixDP; Tai-

ei Medical, Osaka, Japan). The patient position was ad-usted with fluoroscopy in each treatment session.

Fractionation regimens for proton beam therapy varied asresult of irregular availability of the proton beams. X-ray

rradiation with total doses ranging 6–24 Gy were given topatients when the proton beams were unavailable. Doses

er fraction ranged 2.0–3.5 Gy in proton beam therapy, andll X-ray treatments were delivered at 2.0 Gy per fraction.he total doses given ranged 63.0–95.0 Gy (median, 72.0y). The doses were constrained under 55 Gy for the optic

Table 1. Patients an

Patientno.

Ageand sex Tumor status Latest surgery

Preopetumor v

(m

1 58 M Recurrent Partial 432 64 F Recurrent Partial 663 37 M Primary Biopsy 124 67 M Recurrent Partial 245 14 F Primary Biopsy 36 61 F Primary Subtotal 517 71 M Primary Biopsy 198 68 M Primary Biopsy 59 35 F Primary Subtotal 52

10 74 F Primary Biopsy 2711 44 F Primary Partial 3012 16 F Primary Biopsy 8813 63 F Primary Partial 12

Abbreviations: M � male; F � female; Fr � fractions.* Equivalent dose in conventional fractionation of 2.0 Gy per f2.0 and �/� � 7.0, respectively (15).

hiasm and optic nerves and 55 Gy and 50 Gy for the d

urface and the center of the brainstem, respectively. Be-ause various fractionation regimens were used for treat-ent, the equivalent dose when delivered at 2.0 Gy per

raction was calculated for comparison according to theinear quadratic model (15) with the �/� ratio of 2.0 Gy forrain tissues (16) and 7.0 Gy for the tumor (6). Totalquivalent doses for 2.0 Gy per fraction were 64.0–113.3y (median, 74.8 Gy) when the �/� ratio was assumed to be.0 Gy and 64.0–103.1 Gy (median, 73.5 Gy) when the �/�atio was 7.0 Gy. The relative biological effectiveness valuef 1.0 was adopted according to our experimental databtained with fibrosarcoma NFSa cells (17).

valuation and statistical analysesAfter completion of therapy, patients were followed byeans of hospital visits, visits to their referring physicians,ail, or phone to monitor for recurrence and late radiation

oxicity. Survival and local control rates were calculatedrom the start of proton beam therapy by the Kaplan-Meierethod. Statistical analyses were made by a log–rank test.cute reactions were scored according to the Commonerminology Criteria for Adverse Events, version 3.0, by

he National Cancer Institute (18). Late normal-tissue com-lications were evaluated according to the late effects nor-al tissues–subjective, objective, management, and analytic

LENT-SOMA) scoring system (19).

RESULTS

No patients underwent gross total tumor removal (Table). All patients tolerated the treatment well; no patient wasost to follow-up. The median follow-up period was 69.3onths (range, 14.6–123.4 months; Table 2). Of the 13

atients, 6 suffered from recurrence; all of them had locore-ional recurrence, and none had disseminated or metastatic

ent characteristics

Irradiation dose Equivalent dose*

Proton X-ray �/� � 2.0 �/� � 7.0

75.5 Gy/26 Fr (�) 92.8 83.295 Gy/35 Fr (�) 113.3 103.172 Gy/24 Fr (�) 90.0 80.063 Gy/20 Fr (�) 81.4 71.265 Gy/26 Fr (�) 73.1 68.6

57.2 Gy/26 Fr 16 Gy/8 Fr 76.1 74.557.5 Gy/33 Fr 6 Gy/3 Fr 70.7 66.748.4 Gy/22 Fr 24 Gy/12 Fr 74.8 73.5

64 Gy/32 Fr 12 Gy/6 Fr 76.0 76.054 Gy/27 Fr 12 Gy/6 Fr 66.0 66.072 Gy/36 Fr (�) 72.0 72.066 Gy/33 Fr (�) 66.0 66.064 Gy/32 Fr (�) 64.0 64.0

was calculated based on a linear quadratic model assuming �/�

d treatm

rativeolume

L)

.7

.0

.7

.5

.3

.0

.7

.8

.0

.4

.8

.4

.9

raction

iseases. Patterns of failure were marginal local recurrences

ofsGrci

4p6rptad

s

r7rrTgt2

qcppat3rp

Fc

F

1122 I. J. Radiation Oncology ● Biology ● Physics Volume 60, Number 4, 2004

utside the CTVs for 2 patients, marginal local recurrencesrom the dose-limiting regions near the radiation-sensitivetructures for 2 patients, in-field local recurrence within theTV for 1 patient, and out-of-field regional nodal recur-

ence for 1 patient (Table 2). Seven patients died: 3 byhordoma, 1 by radiation-induced brain necrosis, and 3 byntercurrent diseases.

Local control rates at 3 and 5 years were 67.1% and6.0%, respectively (Fig. 1). Cause-specific, overall, androgression-free survival rates were 91.7%, 84.6%, and1.5% at 3 years, and 72.2%, 66.7%, and 42.2% at 5 years,espectively (Fig. 2). For patients with tumors �30 mLreoperatively, the local control rate seemed to be superioro that for patients with tumors �30 mL (75.0% vs. 50.0%t 3 years and 60.0% vs. 0% at 5 years; Fig. 3), but theifference was not statistically significant (p � 0.31).For 6 patients who underwent biopsy only as the latest

urgery before proton irradiation (the biopsy group), no

Table 2. T

Patientno.

Survival(months)

Site ofrecurrence

Time to progression(months)

1 123.4 Paraspinal 44.62 69.3 Nodal 30.8

3 113.4 (�)4 53.6 Paraspinal 20.95 111.1 (�)6 85.9 Marginal 49.77 82.7 (�)8 76.5 (�)9 38.6 Central 15.2

10 46.8 (�)11 38.1 (�)12 14.6 (�)13 28.1 Marginal 11.8

Abbreviation: G � grade.* Suicide is considered an intercurrent disease.

ig. 1. Actuarial local control rate for 13 patients with skull base

hordoma treated with proton beam therapy. p

ecurrence was observed with a median follow-up period of9.6 months; among 7 patients who underwent a tumoremoval operation (the tumor-removal group), all but 1 hadecurrence within 4 years after proton beam therapy (Fig. 4).he mean tumor volume before surgery for the biopsyroup was less than that for the tumor-removal group, buthe difference was not significant by unpaired t test (26.2 �9.0 mL vs. 32.8 � 13.8 mL, p � 0.34; Table 1).Acute reactions were acceptable (Table 2). No patient re-

uired treatment breaks attributable to acute reactions. Lateomplications greater than Grade 3 were seen in 2 patients: 1atient suffered from Grade 4 brain necrosis in bilateral tem-oral lobes, and the other showed Grade 4 oral mucosal ulcer-tion plus Grade 5 brain necrosis (Table 2). Both patients werereated by large daily fraction of irregular sizes, ranging 2.5–.5 Gy; the total equivalent doses were 92.8 Gy and 113.3 Gy,espectively, by conventional fractionation for �/� � 2.0. Bothatients underwent necrotomy.

nt results

Cause of death

Morbidities

Acute Late

Pneumonia Headache (G1) Brain necrosis (G4)Brain necrosis Oral ulceration (G4)

Brain necrosis (G5)(Alive)Tumor bleeding(Alive)(Alive) Headache (G1)(Alive)PneumoniaTumor growth Nausea (G1)(Alive)(Alive)Suicide* Headache (G2)Tumor growth

ig. 2. Cause-specific, overall, and disease-free survival rates after

reatme

roton beam therapy.

rtvpsdit

b1tmtXas

d

sa(r6dlaLrmttcACfiprcacFwThfiptgl

wtig

gg(dossiaootwmrIroF

t

1123Proton beam therapy for skull base chordoma ● H. IGAKI et al.

DISCUSSION

The treatment of choice for skull base chordoma is maximaladical resection with postoperative radiotherapy (3–6). Grossotal removal of the tumor is impossible in most cases; con-entional X-ray radiotherapy offers little possibility of cure foratients with overt residual tumors. Results of the present studyuggest that a curative dose to overt residual skull base chor-omas can be given safely and that the better local control rates obtained for those with smaller tumor volumes with frac-ionated proton beam therapy.

X-ray radiotherapy for overt residual chordoma haveeen reported to produce progression-free survival rates of7%–39% at 5 years (7, 9, 10, 20–22; Table 3). In addition,he survival rates continue to decline thereafter, and theajority of patients in these studies finally died of local

umor progression. Most previous reports have shown that-ray radiotherapy is not a sufficient treatment modality

fter incomplete resections (7–10), although a dose–re-ponse relationship is noted in some series (23–25).

Charged-particle radiotherapy offers the advantage inose distribution. In fact, better results for patients with

Fig. 3. Local control rates by preoperative tumor volume.

ig. 4. Local control rates by type of surgery before proton beam

therapy.

kull base chordoma have been reported from hadron ther-py institutions (26–30; Table 3). Munzenrider and Liebsch26) reported that the 5-year local recurrence-free survivalate was 73% for 169 patients with chordoma treated with6–83 cobalt gray equivalent (CGE) of proton beam irra-iation with or without X-ray irradiation at a median fol-ow-up of 41 months. In a series of 33 patients treated withmean total dose of 70.7 CGE of proton beam irradiation atoma Linda University, 3-year and 5-year local control

ates were reported to be 67% and 59%, respectively, at aean follow-up of 33 months (27). Noël et al. (28) reported

hat the 3-year local control rate was 71% with a medianotal dose of 67 CGE of proton beam irradiation. The localontrol rate in our series seemed slightly lower (Table 3).mong 6 recurrent cases, 3 failed in the regions outside theTVs, which included 2 cases of local recurrences at theeld border and a case of regional nodal recurrence in theresent series (Tables 2 and 4). Austin et al. (31), in aeview of 15 patients with chordoma that had locally re-urred, reported that 10 of 15 patients had recurrent diseasest the region given lower doses because of normal tissueonstraints: only 2 patients had marginal recurrences. Arench group found that 4 and 4 of 10 recurrent cases wereithin the GTVs and within the CTVs, respectively (28).he median total dose of 72.0 Gy in our series is somewhatigher than that of the previous reports of proton beam therapyor chordoma, considering the adoption of the relative biolog-cal effectiveness value of 1.0 for proton beam in our clinicalractice instead of 1.1 (26–28). This might lead to the rela-ively high rate of local recurrence outside the CTV. Moreenerous margins or treatment planning with detailed radio-ogic examinations might have improved our results.

There was a trend toward better local control for patientsith tumors �30 mL preoperatively than for those with larger

umors (Fig. 3), but the difference was not statistically signif-cant. Several studies with charged particle therapy have sug-ested better results for smaller tumors (27, 28, 32, 33).

In the present study, the local control rate for the biopsyroup was apparently superior to that for the tumor-removalroup (Fig. 4). Possible explanations include the following:1) patients with large volume tumors who have to undergoebulking surgery tend to be locally uncontrollable becausef the close location of the tumor to critical structures; (2)urgical procedures provide opportunities for tumor celleeding and increase the rate of marginal failure or dissem-nation; and (3) there were other risk factors for recurrencemong our patients who underwent tumor removal. First, inur 13 patients, we found no relationship between the typef operation and the proximity to radiation-sensitive struc-ures. Second, some investigators mentioned surgical path-ay recurrences of skull base chordoma after tumor re-oval operations. However, the rates of surgical pathway

ecurrence were reported to be as low as 4%–7% (34–36).n addition, we have not experienced any patients whoseecurrent mass had developed evidently along the pathwayf the prior surgical access. To the best of our knowledge,

here has been no report indicating that tumor removal

opdesrX1rg

ttmglimhriamipp

psfwtd�r

P

P

H

C

rrdoma

WNANUT

c

1124 I. J. Radiation Oncology ● Biology ● Physics Volume 60, Number 4, 2004

peration is a poor prognostic factor for local control com-ared with biopsy only in patients with skull base chor-oma. Our results might be attributable to the third hypoth-sis above, owing to the small number of patients in oureries. All 3 patients who received proton beam therapy foregrowing tumors after the primary surgery with or without-ray radiotherapy were in the tumor-removal group (Table). Some investigators have showed poorer prognosis afteradiotherapy for patients with recurrent diseases after sur-ery (32, 37). In addition, the proportion of patients with

Table 3. Treatment results

Authors (reference) N Total dos

hotonMagrini et al. (20) 12 48–60

(median,Forsyth et al. (21) 39 22.93–67

(median,Romero et al. (22) 18 29.9–64

(median, 5Catton et al. (7) 24 25–60

(median,Zorlu et al. (9) 18 50–64

(median,roton with or without photonMunzenriden and Liebsch (26) 169 66–83†

Hug et al. (27) 33 66.6–79(mean, 71

Noël et al. (28) 47 60–73†

(median,Present series 13 63–95

(median,elium ionCastro et al. (29) 53 60–80†

(mean, 6arbon ionShults-Ertner et al. (30) 44 median, 6

* Total doses are shown in cobalt gray equivalent for all series eelative biological effectiveness of unity.

† Total doses and follow-up periods were for patients with cho

Table 4. Sites of recurrence after proton beam therapy inthree studies

SiteAustin etal. (31)

Noël et al.(28)*

Presentseries

ithin GTV 3 4 1ear critical structure 10 4 2round CTV† 2 0 2ode 0 0 1nknown 0 2 0otal 15 10 6

* Chondrosarcomas were included.† Recurrence at the field border outside the CTV, excluding

pervical lymph node recurrence.

umors �30 mL was larger in the biopsy group than in theumor-removal group (5 of 6 vs. 2 of 7; Table 1), whichight have affected the favorable results for the biopsy

roup. However, it is notable in the present series that noocal recurrences have been observed among the 6 patientsn the biopsy group with a median follow-up period of 79.6onths (Fig. 4). Other studies from hadron therapy centers

ave showed no significant difference in local tumor controlate by the type of surgical interventions (26–29), and somenvestigators drew similar conclusions regarding postoper-tive X-ray radiotherapy (8–10). We hypothesized that tu-or biopsy for pathologic diagnosis is a sufficient surgical

ntervention for a patient with a tumor of �30 mL with norior treatment, provided the patient is scheduled to undergoroton beam therapy.Incidence of postirradiation brain necrosis was high in the

resent series. One reason might be the large daily fractionizes, up to 3.5 Gy. We were obliged to use such a largeraction size to complete a series of proton irradiationsithin a certain period of time under limited availability of

he proton beams. The tolerance dose of the brain is highlyependent on the daily fraction size; daily fractions of2.2–2.5 Gy are associated with increased risk of late

adiation injury (38). In fact, all of our patients suffering

adiotherapy in 11 studies

Follow-up(y)

Local control rate (%)

3-y 5-y 10-y

1–25(median, 6)

Not stated 25 25

5.6–29.7(median, 8.3)

Not stated 39 31

0.1–8.6(median, 3.1)

Not stated 17 —

0.3–20(median, 5.2)

Not stated 23 15

1.0–8.0(median, 3.6)

Not stated 23 —

0.1–21.2†

(median, 3.4)Not stated 73 54

0.6–6.3†

(mean, 2.8)67 59 —

0.3–5.9†

(median, 2.4)71 — —

1.2–10.3(median, 5.8)

67.1 46.0 —

0.3–15.9†

(median, 4.3)Not stated 63 —

0.3–38.3†

(median, 1.3)87.1 — —

for the present series, in which the dose is shown in Gy, assuming

and other diseases (mostly chondrosarcoma).

after r

e*

58).4250).80.1)

50)

60)

.2.9)

67)

72)

5)

0†

xcept

ostirradiation brain injury, including a patient with asymp-

to

rtaaorfssopto

ptqtctrbfahar

1

1

1

1

1

1

1125Proton beam therapy for skull base chordoma ● H. IGAKI et al.

omatic brain injury, received daily fraction doses of 2.2 Gyr more.The main limitations of this study are (1) the nature of a

etrospective review on a relatively small number of pa-ients irradiated over a long period with various fraction-tion schedules; and (2) that biologically effective dosesccording to a linear quadratic model could not explain allf the events in different fractionation regimens. For thoseeasons, it is impossible to determine the optimal dose or theractionation regimen for skull base chordoma or to predictequelae from this study. However, these limitations did noteem to influence the outcome of our study strongly becauseur results demonstrated excellent local control rate foratients who received biopsy only before proton irradiation,hus indicating the superiority of proton beam therapy for

vert residual skull base chordoma. o

REFEREN

tionated radiotherapy. Br J Radiol 1989;62:679–694.

1

1

1

1

2

2

2

2

2

2

2

2

2

2

3

CONCLUSION

Proton beam therapy is an effective treatment option foratients with skull base chordoma, especially for those withumors �30 mL, and tumor removal seems not to be re-uired for patients with tumors �30 mL with no priorreatment before proton beam therapy. Further studies in-orporating a larger number of patients should be conductedo test this hypothesis. An optimal dose and a fractionationegimen also have yet to be defined. In September 2001, weegan to treat patients at the new proton beam therapyacility dedicated to the treatment, where the proton beam isvailable throughout the year. We are currently testing aigh-dose irradiation regimen in the conventional fraction-tion schedule for chordoma in the skull base to reduce lateadiation injuries while maintaining the higher probability

f local control.

CES

1. The Committee of Brain Tumor Registry of Japan, The Jap-anese Pathological Society. The classification and incidence ofbrain tumors [in Japanese]. In: General rules for clinical andpathological studies on brain tumors. 2nd ed. Tokyo: Kane-hara Shuppan; 2002. p. 9–12.

2. The Central Brain Tumor Registry of the United States. 2002–2003 statistical report: Primary brain tumors in the UnitedStates, 1995–1999. Available at: http://www.cbtrus.org/2002/2002report.pdf. Accessed January 2004.

3. Soo MY. Chordoma: Review of clinicoradiological featuresand factors affecting survival. Australas Radiol 2001;45:427–434.

4. Tai PT, Craighead P, Liem SK, et al. Management issues inchordoma: A case series. Clin Oncol (R Coll Radiol) 2000;12:80–86.

5. Gay E, Sekhar LN, Rubinstein E, et al. Chordomas andchondrosarcomas of the cranial base: Results and follow-up of60 patients. Neurosurgery 1995;36:887–896.

6. Tai PT, Craighead P, Bagdon F. Optimization of radiotherapyfor patients with cranial chordoma. A review of dose-responseratios for photon techniques. Cancer 1995;75:749–756.

7. Catton C, O’Sullivan B, Bell R, et al. Chordoma: Long-termfollow-up after radical photon irradiation. Radiother Oncol1996;41:67–72.

8. Cummings BJ, Hodson DI, Bush RS. Chordoma: The resultsof megavoltage radiation therapy. Int J Radiat Oncol BiolPhys 1983;9:633–642.

9. Zorlu F, Gürkaynak M, Yildiz F, et al. Conventional externalradiotherapy in the management of clivus chordomas withovert residual disease. Neurol Sci 2000;21:203–207.

0. Fuller DB, Bloom JG. Radiotherapy for chordoma. Int JRadiat Oncol Biol Phys 1988;15:331–339.

1. Shioyama Y, Tokuuye K, Okumura T, et al. Clinical evalua-tion of proton radiotherapy for non-small-cell lung cancer. IntJ Radiat Oncol Biol Phys 2003;56:7–13.

2. Kagei K, Tokuuye K, Okumura T, et al. Long-term results ofproton beam therapy for carcinoma of the uterine cervix. Int JRadiat Oncol Biol Phys 2003;55:1265–1271.

3. Matsuzaki Y, Osuga T, Saito Y, et al. A new, effective, andsafe therapeutic option using proton irradiation for hepatocellularcarcinoma. Gastroenterology 1994;106:1032–1041.

4. Tsujii H, Tsuji H, Inada T, et al. Clinical results of fractionatedproton therapy. Int J Radiat Oncol Biol Phys 1993;25:49–60.

5. Fowler JF. The linear-quadratic formula and progress in frac-

6. Smit BJ. Radiobiology. In: Breuer H, Smit BJ, editors. Protontherapy and radiosurgery. Berlin: Springer-Verlag; 2000. p.121–131.

7. Ando K, Koike S, Kawachi K, et al. Relative biologicaleffectiveness of the therapeutic proton beams at NIRS andTsukuba University. Nippon Acta Radiol 1985;45:531–535.

8. National Cancer Institute. Common terminology criteria foradverse events version 3.0. Cancer Therapy Evaluation Pro-gram, National Cancer Institute. Available at: http://ctep.can-cer.gov/forms/CTCAEv3.pdf. Accessed January 2004.

9. LENT SOMA scales for all anatomic sites. Int J Radiat OncolBiol Phys 1995;31:1049–1091.

0. Magrini SM, Papi MG, Marletta F, et al. Chordoma—naturalhistory, treatment and prognosis. The Florence RadiotherapyDepartment experience (1956–1990) and a critical review ofthe literature. Acta Oncol 1992;31:847–851.

1. Forsyth PA, Cascino TL, Shaw EG, et al. Intracranial chor-domas: A clinicopathological and prognostic study of 51cases. J Neurosurg 1993;78:741–747.

2. Romero J, Cardenes H, la Torre A, et al. Chordoma: Resultsof radiation therapy in eighteen patients. Radiother Oncol1993;29:27–32.

3. Pearlman AW, Friedman M. Radical radiation therapy ofchordoma. Am J Roentgenol Radium Ther Nucl Med 1970;108:332–341.

4. Higinbotham NL, Phillips RF, Farr HW, et al. Chordoma.Thirty-five-year study at Memorial Hospital. Cancer 1967;20:1841–1850.

5. Tewfik HH, McGinnis WL, Nordstrom DG, et al. Chordoma:Evaluation of clinical behavior and treatment modalities. Int JRadiat Oncol Biol Phys 1977;2:959–962.

6. Munzenrider JE, Liebsch NJ. Proton therapy for tumors of theskull base. Strahlenther Onkol 1999;175(Suppl. 2):57–63.

7. Hug EB, Loredo LN, Slater JD, et al. Proton radiation therapyfor chordomas and chondrosarcomas of the skull base. J Neu-rosurg 1999;91:432–439.

8. Noël G, Habrand JL, Jauffret E, et al. Radiation therapy forchordoma and chondrosarcoma of the skull base and thecervical spine. Prognostic factors and patterns of failure.Strahlenther Onkol 2003;179:241–248.

9. Castro JR, Linstadt DE, Bahary JP, et al. Experience incharged particle irradiation of tumors of the skull base: 1977–1992. Int J Radiat Oncol Biol Phys 1994;29:647–655.

0. Schulz-Ertner D, Nikoghosyan A, Thilmann C, et al. Carbon

ion radiotherapy for chordomas and low-grade chondrosarco-

3

3

3

3

3

3

3

3

1126 I. J. Radiation Oncology ● Biology ● Physics Volume 60, Number 4, 2004

mas of the skull base. Results in 67 patients. StrahlentherOnkol 2003;179:598–605.

1. Austin JP, Urie MM, Cardenosa G, et al. Probable causes ofrecurrence in patients with chordoma and chondrosarcoma ofthe base of skull and cervical spine. Int J Radiat Oncol BiolPhys 1993;25:439–444.

2. Berson AM, Castro JR, Petti P, et al. Charged particle irradi-ation of chordoma and chondrosarcoma of the base of skulland cervical spine: The Lawrence Berkeley Laboratory expe-rience. Int J Radiat Oncol Biol Phys 1988;15:559–565.

3. O’Connell JX, Renard LG, Liebsch NJ, et al. Base of skullchordoma. A correlative study of histologic and clinical fea-tures of 62 cases. Cancer 1994;74:2261–2267.

4. Fagundes MA, Hug EB, Liebsch NJ, et al. Radiation therapy

for chordomas of the base of skull and cervical spine: Patternsof failure and outcome after relapse. Int J Radiat Oncol BiolPhys 1995;33:579–584.

5. Fischbein NJ, Kaplan MJ, Holliday RA, et al. Recurrence ofclival chordoma along the surgical pathway. Am J Neurora-diol 2000;21:578–583.

6. Arnautovic KI, Al-Mefty O. Surgical seeding of chordomas.J Neurosurg 2001;95:798–803.

7. Carpentier A, Polivka M, Blanquet A, et al. Suboccipital andcervical chordomas: The value of aggressive treatment at firstpresentation of the disease. J Neurosurg 2002;97:1070–1077.

8. Schultheiss TE, Kun LE, Ang KK, et al. Radiation response ofthe central nervous system. Int J Radiat Oncol Biol Phys

1995;31:1093–1112.