C

Mwymabstemttes

P

tob

OTm

Int. J. Radiation Oncology Biol. Phys., Vol. 67, No. 2, pp. 521–524, 2007Copyright © 2007 Elsevier Inc.

Printed in the USA. All rights reserved0360-3016/07/$–see front matter

doi:10.1016/j.ijrobp.2006.08.081

LINICAL INVESTIGATION Brain

STEREOTACTIC RADIOSURGERY FOR BRAINSTEM METASTASES:SURVIVAL, TUMOR CONTROL, AND PATIENT OUTCOMES

AAMIR HUSSAIN, M.D.,* PAUL D. BROWN, M.D.,* SCOTT L. STAFFORD, M.D.,* AND

BRUCE E. POLLOCK, M.D.*†

*Departments of Radiation Oncology and †Neurological Surgery, Mayo Clinic College of Medicine, Rochester, MN

Purpose: Patients with brainstem metastases have limited treatment options. In this study, we reviewed outcomesafter stereotactic radiosurgery (SRS) in the management of patients with brainstem metastases.Methods and Materials: Records were reviewed of 22 consecutive patients presenting with brainstem metastaseswho underwent SRS. The most frequent primary malignancy was the lung (n � 11), followed by breast (n � 3)and kidney (n � 2). Three patients (14%) also underwent whole-brain radiation therapy (WBRT). The mediantumor volume was 0.9 mL (range, 0.1–3.3 mL); the median tumor margin dose was 16 Gy (range, 14–23 Gy).Results: Median survival time after SRS was 8.5 months. Although local tumor control was achieved in allpatients with imaging follow-up (n � 19), 5 patients died from development and progression of new brainmetastases. Two patients (9%) had symptom improvement after SRS, whereas 1 patient (5%) developed a newhemiparesis after SRS.Conclusions: Radiosurgery is safe and provides a high local tumor control rate for patients with small brainstemmetastases. Patients with limited systemic disease and good performance status should be strongly considered forSRS. © 2007 Elsevier Inc.

Brain metastasis, Stereotactic radiosurgery, Brainstem.

fPptolTburds

R

((owwimT(

A

INTRODUCTION

etastases represent the most common brain tumor in adultsith an estimated 98,000–170,000 new cases diagnosed eachear in the United States (1). The most common mechanism ofetastases to the brain is by hematogenous spread resulting indistribution of metastases that correlates with the relative

lood flow to each area of the brain (1, 2): cerebral hemi-pheres (80%), cerebellum (15%), and brainstem (5%). Al-hough a large number of studies have demonstrated thefficacy of stereotactic radiosurgery (SRS) in the manage-ent of patients with brain metastases (3–5), little informa-

ion is available specifically on patients with metastases tohe brainstem (6, 7). In this article, we review our experi-nce using SRS in the management of patents with brain-tem metastases.

METHODS AND MATERIALS

atient populationTwenty-two patients (12 men, 10 women) with brainstem me-

astases were identified from a prospectively maintained databasef patients having radiosurgery at Mayo Clinic, Rochester, MN,etween June 1991 and September 2004. No patients were lost to

Reprint requests to: Paul D. Brown, M.D., Department of Radiationncology, 200 First Street SW, Mayo Clinic, Rochester, MN 55905.el: (507) 284-2949; Fax: (507) 284-0079; E-mail: brown.paul@

ayo.edu.

521

ollow-up; however, 3 patients did not have imaging follow-up.atient and tumor characteristics are outlined in Table 1. Twelveatients (55%) had lesions located in the pons, 7 had midbrainumors, and 3 had tumors in the medulla. Nineteen patients hadne metastatic lesion in the brainstem, and 3 patients had twoesions. Thus a total of 25 metastases were treated in 22 patients.he most common primary was lung (n � 11, 50%), followed byreast cancer (n � 3) and renal cancer (n � 2). One patient had annknown primary malignancy. The median age at the time ofadiosurgery was 60 years (range, 39–78 years). The Mayo Foun-ation Institutional Review Board approved all aspects of thetudy.

adiosurgical dosimetryRadiosurgery was performed with the Leksell Gamma Unit

Elekta Instruments, Norcross, GA). Magnetic resonance imagingMRI) was used in all cases for dose planning. The median numberf isocenters used was 2 (range, 1–10). The median tumor volumeas 0.9 mL (range, 0.1–3.3 mL). The median tumor margin doseas 16 Gy (range, 14–23 Gy), with 16 patients treated to the 50%

sodose line, 3 to 80%, and 1 each to 67%, 40%, and 58%. Theedian maximal tumor dose was 32 Gy (range, 18.8–40.1 Gy).hree patients received adjuvant whole-brain radiation therapy

WBRT) within 1 month after radiosurgery.

Conflict of interest: none.Received June 6, 2006, and in revised form Aug 31, 2006.

ccepted for publication Aug 31, 2006.

F

elfenwostlcrcdmfms

P

m(pdv

T

(

ss

P

rOapt(at

cwHnfmtcpSeip

ery; KP

522 I. J. Radiation Oncology ● Biology ● Physics Volume 67, Number 2, 2007

ollow-upPatients typically had follow-up clinical examinations and MRI

very 3 months after radiosurgery. Nineteen patients (86%) had ateast one MRI after radiosurgery, whereas 3 patients had clinicalollow-up but no imaging available for review. Patients werevaluated with regard to local and intracranial tumor control,eurologic symptoms, and overall systemic status. Local controlas assessed by comparing the tumor diameter on MRI at the timef radiosurgery to the tumor diameter on MRI at follow-up. Stableize or regression of the metastasis was considered a success, andherefore local control was defined as the lack of progression of theesion. Karnofsky performance status ratings were assigned tolassify patients according to functional impairment at the time ofadiosurgery and at follow-up appointments (8–10). Patient out-omes were censored at the time of the last follow-up (n � 4) oreath (n � 18). The median follow-up after radiosurgery was 8.5onths (range, 1.4–78.6 months) for all patients (n � 22), whereas

or living patients (n � 4), it was 26.6 months (range, 13.0–78.6onths). Overall survival was calculated from the date of radio-

urgery using the Kaplan-Meier method (11).

RESULTS

atient survivalThe median survival time after radiosurgery was 8.5onths (range, 1.4–78.6 months) (Fig 1). Five patients

23%) died from distant intracranial progression. Twelveatients (55%) died from progression of their systemicisease. One patient’s cause of death was unknown. Sur-ival at 1 year was 32%.

umor controlLocal tumor control was 100% (Fig. 2). Five patients

Table 1. Patient chara

Patient GenderPrimary

malignancyMetastaticlocation

1 F Lung Pons2 M Lung Pons3 M Lung Pons4 M Lung Pons5 M Lung Pons6 F Lung Pons7 M Lung Pons8 F Lung Midbrain9 F Lung Midbrain

10 M Lung Midbrain11 F Lung Medulla12 F Breast Pons13 F Breast Pons14 F Breast Medulla15 M Kidney Midbrain16 M Kidney Midbrain17 M Melanoma Medulla18 F Colon Pons19 F Esophagus Midbrain20 M Spindle cell Midbrain21 M Thyroid Pons22 M Unknown Pons

Abbreviations: cGy � centigray; SRS � stereotactic radiosurg

23%) developed new distant tumors. Two of these patients c

ubsequently underwent WBRT 4 and 5 months after radio-urgery, respectively.

atient outcomesTwo patients (9%) had resolution of symptoms after

adiosurgery (facial numbness, n � 1; diplopia, n � 1).ne patient (5%) had a radiation-related complication

fter radiosurgery of a metastasis located in the pons. Thisatient received 18 Gy at the 50% isodose line to cover aumor (volume, 0.88 mL) that had progressed after WBRT3,000 cGy in 10 fractions 4 months prior). Three monthsfter SRS, the patient developed a right-sided hemiparesishat persisted despite treatment with corticosteroids.

DISCUSSION

Stereotactic radiosurgery is proven to provide local tumorontrol (12) and provide survival benefit in combinationith WBRT for selected patients with brain metastases (3).owever, few data are available to evaluate the effective-ess of SRS for patients with metastases to the brainstem. Inact, despite the fact that SRS is commonly employed toanage patients with brainstem metastases, some question

he value of SRS for this patient group (13). Suki andolleagues reviewed the MD Anderson Cancer Center ex-erience and found complications in 37% of patients afterRS for brainstem metastases; this is in sharp contrast to ourxperience with only 1 patient (5%) of 22 having a signif-cant complication. This discrepancy may be partially ex-lained by how Suki and colleagues accounted for compli-

cs and treatment data

e atRS KPS

Totallesions

Volume(mL)

Margindose

54 100 1 0.16 1,80078 100 1 0.96 1,50060 90 1 1.90 1,50052 90 1 0.35 1,80055 90 1 1.20 1,50060 100 1 1.60 1,60072 100 1 0.90 1,50072 90 1 3.30 1,50066 80 2 0.59 1,80060 100 1 0.20 2,33056 100 1 0.55 1,50046 90 1 0.88 1,80066 90 1 0.81 1,60053 90 2 0.70 1,60068 70 1 1.40 1,80055 90 2 2.80 1,40039 90 1 2.78 1,80063 100 1 0.26 1,80048 90 1 0.26 1,50043 90 1 3.30 1,50072 90 1 2.20 1,60068 90 1 0.06 1,600

S � Karnofsky performance scale.

cteristi

AgS

ations including events such as tumor failures, death,

sc

ma

(cdpd

om tim

523Brainstem metastases radiosurgery ● A. HUSSAIN et al.

eizures, and deep venous thromboses under the category ofomplications.

Two previous series described the use of SRS in theanagement of patients with brainstem metastasis and hadmore favorable experience. Both Huang (Pittsburgh) et al.

Fig. 2. Preradiosurgery axial gadolinium-enhanced T1-wewith a pontine lung metastasis (a). Follow-up MRI 2 mothe 50% isodose) demonstrates response to treatment (band maintained local control of the pontine lesion, but de

0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1.0

0 2 4 6 8 10

Surv

ival

%

Fig. 1. Overall survival for all patients fr

months later.

6) and Fuentes (Marseilles) et al. (7) noted high localontrol rates (92–95%) similar to our experience. The me-ian tumor margin dose in Huang et al. was 16 Gy com-ared with 19.6 Gy in the report by Fuentes et al. Thisifference in radiosurgical dosing likely relates to the num-

magnetic resonance imaging (MRI) scan of patient (#6)fter gamma knife radiosurgery (16 Gy to the margin atpatient received adjuvant whole-brain radiation therapyd distant brain failure requiring further radiosurgery 11

14 16 18 20 22 24

s

e of stereotactic radiosurgery procedure.

ightednths a

). Thevelope

12

Month

bW(wMwT8mte

mrttdphsCp

wnaa

m(cudGic

cSts

524 I. J. Radiation Oncology ● Biology ● Physics Volume 67, Number 2, 2007

er of patients that either underwent prior or else plannedBRT in conjunction with SRS. Twenty-four of 26 patients

92%) underwent WBRT in the Pittsburgh study (6),hereas only 6 of 28 (21%) patients received WBRT in thearseilles study (7). Survival times and complication ratesere also quite similar in these articles compared to our results.he median survival time observed after SRS ranged from.5 to 12 months in these two articles. Similar to brainetastasis patients with tumors not located in the brainstem,

he most important predictor of survival after SRS is thextent and management of extracranial disease (14).

The results in our small series, of a median survival of 8.5onths, local control of 100%, and a 5% incidence of a

adiation-related complication is not remarkably differenthan most radiosurgical series of brain metastasis patients inhe literature. However, one should not assume there is noifference in the outcomes after SRS for brainstem com-ared to cerebral metastases, because this is obviously aighly selected population with only 22 patients with brain-tem metastases treated over a 14-year period at the Mayolinic. Nonetheless, SRS should be strongly considered for

atients with brainstem metastases, especially those patients t

REFEREN

2006;58:37–40.

1

1

1

1

1

ith limited systemic disease and good performance status,oting SRS provides better local control than WBRT (3)nd concerns regarding significant morbidity and mortalityssociated with microsurgery for brainstem metastases.

It should be emphasized that the treatment of brainstemetastases larger than the median size described in this study

0.9.mL), and especially those larger than 1.5 mL, should beonsidered for a dose reduction relative to the SRS dosessed in this sample of patients. Similarly, radiosurgeryelivered with less conformal methods than the Leksellamma Knife should also consider dose reduction or send-

ng the patient to an appropriate facility capable of highlyonformal radiosurgery.

CONCLUSIONS

Radiosurgery is safe and provides a high local tumorontrol rate for patients with small brainstem metastases.tereotactic radiosurgery should be considered an important

reatment option in selected patients with good performancetatus and limited systemic disease with brainstem metas-

ases.

CES

1. Patchell R. Brain metastases. In: Vinken PJ, Bruyn GW,Klarvams HL, editors. Handbook of neurology. New York:Elsevier Science; 1997. p. 135–149

2. Delattre JY, Krol G, Thaler HT, et al. Distribution of brainmetastases. Arch Neurol 1988;45:741.

3. Andrews DW, Scott CB, Sperduto PW, et al. Whole brainradiation therapy with or without stereotactic radiosurgery boostfor patients with one to three brain metastases: Phase III results ofthe RTOG 9508 randomised trial. Lancet 2004;363:1665–1672.

4. Sneed PK, Suh JH, Goetsch, SJ, et al. A multi institutionalreview of radiosurgery alone vs. radiosurgery with wholebrain radiotherapy as the initial management of brain metas-tases. Int J Radiat Oncol Biol Phys 2002;53:519–526.

5. Auchter RM, Lamond JP, Alexander E, et al. A multi-insti-tutional outcome and prognostic factor analysis of radiosur-gery for resectable single brain metastasis. Int J Radiat OncolBiol Phys 1996;35:27–35.

6. Huang CF, Kondziolka D, Flickinger JC, et al. Stereotactic radio-surgery for brainstem metastases. J Neurosurg 1999;91:563–568.

7. Fuentes S, Delsanti C, Metellus P, et al. Brainstem metastases:Management using Gamma-Knife radiosurgery. Neurosurgery

8. Evans C, McCarthy M. Prognostic uncertainty in terminalcare: Can the Karnofsky index help? Lancet 1985;1:12104.

9. Yates JW, Chalmer B, McKegney FP. Evaluation of patientswith advanced cancer using the Karnofsky performance status.Cancer 1980;45:2220.

0. Mor V, Laliberte L, Morris JN, et al. The Karnofsky Perfor-mance Status Scale: An examination of its reliability andvalidity in a research setting. Cancer 1984;53:2002.

1. Kaplan EL, Meier P. Nonparametric estimation from incom-plete observations. J Am Stat Assoc 1958;53:457–481.

2. Kondziolka D, Patel A, Lunsford LD, et al. Stereotactic ra-diosurgery plus whole brain radiotherapy versus radiotherapyalone for patients with multiple brain metastases. Int J RadiatOncology Biol Phys 1999;45:427–434.

3. Suki D, Lang FF, et al. Complications of stereotactic radio-surgery in patients with brain metastases; considerations fordeep and functional brain tumor locations [Abstract]. Neuro-Oncology 2005;7:393.

4. Kondziolka D, Martin JJ, Flickinger JC, et al. Long-termsurvivors after gamma knife radiosurgery for brain metastases.

Cancer 2005;104:2784–2791.