1

Radiosurgery Practice Guideline Initiative

Stereotactic Radiosurgery for Patients with Metastatic Brain TumorsRadiosurgery Practice Guideline Report # 5-08

ORIGINAL GUIDELINE: May 2008MOST RECENT LITERATURE SEARCH: May 2008

This practice guideline, together with a report on “Metastatic Brain Tumor Management” is anoriginal guideline approved by The International RadioSurgery Association and issued in May2008.

PrefaceSummaryThe IRSA® (International RadioSurgery Association) Radiosurgery Practice Guideline Initiative aims to improve outcomes forbrain metastases radiosurgery by assisting physicians and clinicians in applying research evidence to clinical decisions whilepromoting the responsible use of health care resources.

CopyrightThis guideline is copyrighted by IRSA (2008) and may not be reproduced without the written permission of IRSA. IRSAreserves the right to revoke copyright authorization at any time without reason.

DisclaimerThis guideline is not intended as a substitute for professional medical advice and does not address specific treatments orconditions for any patient. Those consulting this guideline are to seek qualified consultation utilizing information specific totheir medical situation. Further, IRSA does not warrant any instrument or equipment nor make any representations concerningits fitness for use in any particular instance nor any other warranties whatsoever.

KEY WORDS • brain metastases • WBRT • stereotactic radiosurgery • Gamma Knife® • linear acceleratorKEY WORDS • Bragg peak proton therapy • irradiation

Metastatic Brain Tumors

Consensus StatementObjectiveTo develop a consensus-based radiosurgery practiceguideline for brain metastases treatment recommendationsto be used by medical and public health professionals whodiagnose and manage patients with brain metastatic disease.

ParticipantsThe working group included physicians and physicists fromthe staff of major medical centers that provide radiosurgery.

EvidenceThe first author (AN) conducted a literature search inconjunction with the preparation of this document anddevelopment of other clinical guidelines. The literatureidentified was reviewed and opinions were sought fromexperts in the diagnosis and management of brain metastasesincluding members of the working group.

Consensus ProcessThe initial draft of the consensus statement was a synthesisof research information obtained in the evidence gatheringprocess. Members of the working group provided formalwritten comments that were incorporated into the preliminarydraft of the statement. No significant disagreements existed.

The final statement incorporates extensive relevant evidenceobtained by the literature search in conjunction with the finalconsensus recommendations supported by all working groupmembers.

Group CompositionThe radiosurgery guidelines group is comprised ofneurosurgeons, neuro-oncologists, radiation and medicaloncologists and physicists. Community representatives didnot participate in the development of this guideline.

Names of Group Members: Ajay Niranjan, M.B.B.S.,M.Ch., Neurosurgeon, Chair; L. Dade Lunsford, M.D.,Neurosurgeon; Richard L. Weiner, M.D., Neurosurgeon; GailL. Rosseau, M.D., Neurosurgeon; Gene H. Barnett, M.D.,F.A.C.S., Neurosurgeon; Massaki Yamamoto, M.D.Neurosurgeon; Lawrence S. Chin, M.D., F.A.C.S.,Neurosurgeon; Paul J. Miller, M.D., Radiation Oncologist;Andrew E. Sloan, M.D., Neurosurgeon; Burton L. Speiser,M.D., Radiation Oncologist; Sandra S. Vermeulen, M.D.,Radiation Oncologist; Harish Thakrar, M.D., RadiationOncologist; Frank Lieberman, M.D., Neuro-Oncologist;David Schiff, M.D., Neuro-Oncologist; Sammie R. Coy,Ph.D., Medical Physicist; Tonya K. Ledbetter, M.S., M.F.S.,Editor; Rebecca L. Emerick, M.S., M.B.A., C.P.A., exofficio.

2

ConclusionsSpecific recommendations are made regarding targetpopulation, treatment alternatives, interventions and practicesand additional research needs. Appropriate use ofradiosurgery for patients with brain metastases isrecommended.

This guideline is intended to provide the scientific foundationand initial framework for patients who have been diagnosedwith brain metastases. The assessment and recommendationsprovided herein represent the best professional judgment ofthe working group at this time, based on clinical researchdata and expertise currently available. The conclusions andrecommendations will be regularly reassessed as newinformation becomes available.

Stereotactic RadiosurgeryBrain stereotactic radiosurgery (SRS) involves the use ofprecisely directed, closed skull, single session radiation tocreate a desired radiobiologic response within the brain targetwith acceptable minimal effects on surrounding structuresor tissues. In the case of brain metastases, highly conformal,precisely focused radiation is delivered to the metastatictumor in a single session under the direct supervision of aradiosurgery team. At Centers of Excellence, theradiosurgery team includes a neurosurgeon, a radiationoncologist, a physicist and a registered nurse.

Overview of Brain MetastasesEpidemiologic FeaturesMetastatic brain tumors are the most common intracranialneoplasms in adults and are a significant cause of morbidityand mortality. They outnumber primary brain tumors by aratio of 10:1.107 Approximately 1.37 million individuals werediagnosed with cancer in 2005–2006. Conservativeestimates suggest that 100,000–170,000 new cases of brainmetastases are diagnosed every year in the United States(U.S.).59,108 Between 20% and 40% of all patients withmetastatic cancer will have brain metastases at autopsy.106

The estimate of the incidence rate of metastatic brain tumorsvaries from 8.3–11 per 100,000.91,127 In two large populationcohorts of patients who were diagnosed with colorectal, lung,breast or kidney carcinoma or melanoma, brain metastaseswere diagnosed in 8.5–9.6% of patients.6,108 The incidencevaried by primary tumor site. The cumulative incidence wasestimated at 16.3–19.9% in patients with lung carcinoma,6.5–9.8% in patients with renal carcinoma, 6.9–7.4% inpatients with melanoma, 5.0–5.1% in patients with breastcarcinoma, and 1.2–1.8% in patients with colorectalcarcinoma.

The majority of patients who develop brain metastases havea known primary cancer (metachronous presentation). Noprimary systemic site of cancer is detected in 5–10% ofpatients with brain metastases.58,95 Patients with a history oflung cancer have the shortest latency period between thetime of initial diagnosis and the diagnosis of brain metastases(median, 6–9 months). For renal cell carcinoma, the intervalis approximately one year. Patients with breast, melanoma

and colon cancer experience spread of their disease to thebrain at a median latency of approximately two years.106 Therate of breast cancer metastases to the brain may be higheramong patients treated with trastuzumab (Herceptin®).7 Thismay be due to the preference for the brain by HER-2-positivetumor cells, poor penetration of the CNS by the drug orimproved extracranial control resulting in improved survivaland late tumor spread to the CNS.7

The detection rate of brain metastases appears to beincreasing. This increase has been variably attributed toimprovements in systemic therapy leading to longer survival,an aging patient population, and the ability of magneticresonance imaging (MRI) to detect small metastases.130 Themajority of brain metastases are multiple, although thereported percentage of patients with solitary or multiplelesions may vary with the imaging modality used to makethe diagnosis. In the CT (computed tomography) era 50%of lesions were thought to be solitary at the time of neurologicdiagnosis.88 In an analysis of the Radiation TherapyOncology Group (RTOG) database of brain metastasespatients, 19% of patients presented with a single brainmetastasis on MRI, and 50% of patients had 1–3 brainmetastases. Melanoma has the greatest tendency to producemultiple lesions (75% of patients). Multiple lesions are alsofrequent in metastases from colon, breast and lung cancer.Renal cell metastases are more likely to be single.27

SexPredilection for gender follows that of the primary tumor.Lung cancer is the most common source of metastases inmale patients, while breast cancer is the most common sourcein female patients. As the frequency of lung cancer in womenincreases, it may become the most common primary tumorto metastasize to the brain in women as well.

AgeIncidence of brain metastases based on age parallels that ofprimary systemic tumors. Brain metastases are most commonin the fifth to seventh decades of life. Sarcomas and germcell tumors are the most common solid tumors to metastasizeto the brain in children.45

PathophysiologyMetastatic spread to the brain through blood circulationoccurs primarily via arterial circulation and less often viathe Batson venous plexus (pelvic and GI tumors).129 Arterialblood must pass through the lungs before entering the brainand larger clumps of tumor cells are filtered out in lungcapillaries. Many emboli traveling to the brain via the arterialroute originate either from a primary lung tumor or ametastatic site in the lung; however, single tumor cells maypass through the capillaries of the lung. The metastatic cellsthen get trapped in gray-white junction or watershed areas,because of the change in the size of the blood vessels inthese areas.28 Less than 0.1% of deposited cells ultimatelyform metastatic tumors.14 Once the metastatic tumor embolusreaches 1 mm in size, tumor-induced angiogenesis increasesvascular permeability and disrupts the blood-brain barrier(BBB).14 New capillary endothelial cells in metastatic brain

3

tumors display morphological and functional characteristicsassociated with the blood vessels of the primary cancer.65

Macroscopic FindingsIntracranial metastases can be categorized by location asskull, dura, leptomeninges and parenchymal brainmetastases. Lesions of the brain and leptomeninges accountfor 80% of intracranial metastases. The majority of brainmetastases (approximately 80%) are located in the cerebralhemispheres. The cerebellum (10–15%) and brainstem (2–3%) are less frequently involved.87 Most metastases areround, well-demarcated lesions located at the junction ofgray and white matter. Metastatic lesions in the brain displacesurrounding brain parenchyma as they grow. Somemetastases have a miliary type of distribution throughoutthe parenchyma. Leaky tumor vessels result in an extensivezone of edema surrounding the tumor. Cystic degeneration,necrosis and areas of hemorrhage are often seen. Specifictumors may have a more characteristic gross appearance.The metastatic lesions of melanoma, choriocarcinoma andrenal cell carcinoma often develop intratumoralhemorrhages.84 Edema of the adjacent brain parenchyma isoften prominent and sometimes disproportionate to the smallsize of the lesions.29 Meningeal carcinomatosis may occurin patients with lung and breast carcinoma, malignantmelanoma, and less commonly, with lymphoma, leukemiaand other tumors.

Histopathologic FeaturesThe histopathologic features of metastatic lesions are usuallysimilar to those of the primary tumor from which theyoriginate. Although the majority of metastatic lesions mayappear clearly demarcated from adjacent brain on both grossand microscopic examination, microinvasion of tumor cellsis invariably present. This aspect is particularly evident inmetastases of small cell lung carcinomas (SCLC) andmelanomas. Brain metastases elicit a number of reactionsfrom the brain parenchyma. Reactive astrocytosis is oftenpresent surrounding the metastatic nodules. Vascularproliferation with variable degrees of endothelialproliferation may be seen within and surrounding the tumormasses. This rich neovasculature appears to play a significantrole in development and maintenance of the metastaticlesions, and is a major contributory factor to the vasogenicedema that accompanies brain metastases.106 Necrosis isfrequent and macrophage infiltration may be prominentaround areas of necrosis.

For a newly diagnosed brain metastasis of unknown origin,adjunctive morphological techniques such asimmunohistochemistry are valuable and can guide the searchfor a primary site.29 Immunohistochemistry, in conjunctionwith the clinical history, can define specific cell lineages inthe great majority of cases. In cases in which lightmicroscopy and immunohistochemistry are inconclusive,electron microscopy is a useful adjunctive tool to studysubcellular structures that may be diagnostic of cellularlineage. Molecular genetic analysis is the latest resourcefor the complete evaluation of metastatic tumors of unknownorigin.

Clinical PresentationApproximately two-thirds of brain metastases eventuallybecome symptomatic. The clinical presentation of brainmetastases is similar to any intracranial mass lesion.Presenting signs and symptoms of an intracranial massinclude headache (70%), seizures (30–60%), cognitiveimpairment (30%), papilledema (8%), miscellaneous focalneurological deficits and intracranial hemorrhage, amongothers.95,106 Patients may develop obstructive hydrocephalus,particularly those with lesions in proximity to crucial areasof narrow cerebrospinal fluid (CSF) flow such as the 3rd or4th ventricle or the foramen of Monro. Symptoms typicallyhave a gradual onset but 5–10% of patients present withsudden onset of focal neurological symptoms71,97 associatedwith intratumoral hemorrhage, particularly from metastasesoriginating from melanoma, choriocarcinoma or renal cellcarcinoma.119 Even though a relatively small percentage oflung metastases manifest with hemorrhage, these lesionsrepresent the most common source of hemorrhage becauseof their much greater overall occurrence. Seizures arecommon on initial presentation and are the first sign of abrain metastasis in approximately 10–20% of patients.61

Tumors that cause seizures are frequently located in thecerebral hemispheres and often involve the cerebral cortex.Up to one-third of patients presenting with brain metastasesdo not have a prior diagnosis of cancer.77 When a patientwithout a history of systemic cancer presents with newneurological findings and a brain mass is discovered on CTor MRI, a plain chest radiograph and chest CT are alwaysrecommended because the majority of tumors thatmetastasize to the brain are from lung primaries.106

Computed tomography of the abdomen and pelvisoccasionally reveals an unknown primary cancer but mayalso offer evaluation of the systemic disease burden byimaging the liver, adrenals and lymph nodes. Further searchfor a primary cancer is rarely productive without positivefeatures in the patient’s history or localizing signs on physicalexamination that suggest a primary tumor.123 Despiteextensive evaluations, at least 15% of patients with brainmetastases may remain without a definite primary cancersite.77 For these patients stereotactic biopsy or excision ofbrain neoplasm is recommended to determine the finaltreatment strategy.

ImagingMost patients with a known primary tumor undergo imagingstudies when neurologic signs and symptoms develop.Contrast-enhanced CT is used widely because of its easyaccessibility and low cost. On noncontrast CT, metastaticlesions may be of a density less than, equal to, or greaterthan adjacent brain parenchyma. Hyperdensity in ametastasis on noncontrast CT is more likely to be hemorrhagethan calcification. Administration of IV contrast (30–40 giodine) increases the diagnostic accuracy of CT. Mostmetastases enhance after a standard dose of IV contrast. Useof a higher dose of contrast (80–85 g iodine) and scanningdelayed 1–3 hours after injection of the contrast agent offersa further increase in the detection of multiple metastases,and is appropriate if MRI is not available. Contrast-enhancedCT can detect major leptomeningeal spread. Studies

4

comparing contrast-enhanced CT with contrast-enhancedMRI indicate that approximately 20% of patients whodemonstrate a single lesion on CT may demonstrate multiplelesions on MRI.

MRI with contrast enhancement currently is the procedureof choice, since MRI is more sensitive and specific than otherimaging techniques in determining the presence, locationand number of metastases.50 On MR multiple lesions withmarked vasogenic edema and mass effect are typically seenin patients with brain metastases. Lesions are isointense tomildly hypointense on T1-weighted images and hyperintenseon T2-weighted images or fluid attenuation inversionrecovery images. Surrounding edema is relativelyhyperintense on fluid attenuation inversion recovery T2-weighted images and hypointense on T1-weighted images.A metastasis itself may have various signal intensitiesdepending on the types of tissue present within the lesion(e.g., blood products, necrosis and melanin). The appearanceof blood products on T1-weighted images depends on the“age” of bleeding, and ranges from isointense in the first 24hours to hyperintense after 24–72 hours. Melanin isparamagnetic and hyperintense on T1 and hypointense onT2-weighted images. Metastatic melanomas often have bothcomponents (blood and melanin), which appear bright onnoncontrast T1-weighted scans. T2-weighted images areused to estimate the extent of peritumoral edema. Followingcontrast administration, solid, nodular or irregular ringpatterns of enhancement are seen. Contrast-enhanced MRIis the best method for detection of meningeal tumor seeding.Carcinomatous meningitis is usually seen as an irregularbrightly enhancing pial surface on T1-weighted imaging withcontrast. The arachnoidal surfaces, ventricular ependymaor dura may enhance pathologically as well. The usefulnessof diffusion-weighted and perfusion-weighted imaging andproton-MR spectroscopy in the initial diagnosis of brainmetastases has not been established. Gadolinium-enhancedMRI detects smaller lesions, and provides better soft-tissuecontrast, relatively stronger enhancement with paramagneticcontrast agents, images that lack bone artifacts and directmultiplanar imaging. High-dose gadoteridol (ProHance®)detects additional smaller lesions compared with routine-dose gadopentetate dimeglumine (Magnevist®).Magnetization transfer used with routine-dose gadoliniumcontrast is closely comparable to the high-dose technique.For patients with multiple brain metastases, gradient recalledacquisition using double-dose contrast and 2 mm thick slicesis a reliable method to detect brain metastases.32

The interpretation of subsequent radiological studies inpatients who have undergone a complete surgical resectionof their tumors is fairly straightforward, as any newenhancement in the lesion site can be considered localrecurrence. This is not always the case after radiosurgery,in which recurrent or residual tumor must be differentiatedfrom radiation necrosis. Radiation necrosis and recurrentbrain tumor can manifest with similar symptoms and maybe indistinguishable on MRI. Fluorodeoxyglucose-positronemission tomography (FDG-PET) has been proposed as adiagnostic alternative, particularly when coregistered withMRI. For brain metastases with MRI coregistration, FDG-

PET has a sensitivity of 86% and specificity of 80%.19 Thereare few preliminary reports on the utilization of MRspectroscopy to distinguish between radiation necrosis andrecurrent brain metastasis.40,102

Management OptionsMetastatic brain tumors require multimodal managementincluding drugs, surgery, radiosurgery, radiation therapy,chemotherapy, gene therapy and other innovativeapproaches.18,78,100,103,114,133

Symptomatic Medical ManagementThe initial drug of choice for treating cerebral edemaassociated with a brain metastasis is a corticosteroid(dexamethasone or methylprednisolone). The benefit ofcorticosteroids is often dramatic and may be evident withinhours, but ultimately they are insufficient unless definitivetumor management is pursued. Control of vasogenic cerebraledema and removal of the offending mass can result inimproved control of the seizure disorder. For prophylaxisand maintenance therapy, phenytoin is the most frequentlyprescribed agent. There has been a trend away fromphenytoin toward levetiracetam, because of decreased rash,no need for lab monitoring and no induction of CYP3A4.Carbamazepine, phenobarbital, valproic acid andlevetiracetam are often added for breakthrough seizures orto replace phenytoin if toxicity or allergic reactions occur.Benzodiazepines are used for patients in status epilepticus(continuous seizure activity lasting longer than 30 minutes).Although there is agreement that patients who present withseizures require antiepileptic drug therapy, the situation isnot as clear for prophylactic anticonvulsants for brainmetastases. In a meta-analysis of 12 studies in patients withbrain tumors of all etiologies (including brain metastases),prophylactic anticonvulsants did not protect againstsubsequent seizures.44 The American Academy of Neurologyhas published a position paper against the use of prophylacticanticonvulsants.44

Tumor ResectionSome prospective studies have demonstrated that, inappropriately selected cases, surgical treatment caneffectively prolong survival in patients with one or rarelymore than one brain metastasis. Not all patients will benefitfrom surgical resection. The location and accessibility ofthe tumor is a crucial factor in surgical decision making.

Surgery for Single Brain MetastasisPatients experiencing significant, medically refractorysymptoms related to volume and mass effect may requirecraniotomy for tumor removal, regardless of the status oftheir extracranial disease. However, patients with small celllung carcinomas, germ cell tumors, and primary or secondaryCNS lymphomas may not need surgical resection despitebeing symptomatic because these lesions generally responddramatically to radiosurgery, fractionated radiotherapy orsystemic chemotherapy, leading to very rapid resolution ofsymptoms. Patients whose symptoms cannot be palliatedsuccessfully with medical management (including high dosesteroids) and whose lesions can be excised without significant

5

risk of producing or worsening a neurologic deficit arepotential candidates for craniotomy.

In the case of a single, asymptomatic brain metastasis froman undiagnosed primary site, the decision for surgerybecomes more complex. For most patients obtaining a tissuediagnosis is mandatory. If a potential extracranial source isidentified, then biopsy of such a lesion is often performedbefore the intracranial disease is addressed. For patients inwhom a “nonsurgical” tumor type such as small cell lungcarcinoma is diagnosed by biopsy, a potentially unnecessarycraniotomy can be avoided. In other cases, knowledge ofthe putative histologic features of the single brain metastasismay lead to a recommendation for whole-brain radiationtherapy (WBRT) and/or SRS rather than craniotomy fortumor removal.

Stereotactic biopsy of a solitary metastasis in unresectablelocations is indicated to obtain a tissue diagnosis in patientswhen no other source has been identified that can be biopsiedmore safely. Biopsy is also indicated when the chance of acoincidental nonmalignant intracranial lesion is greater thanthe probability of a significant biopsy-related complication.In most series the rate of such complications approximates1%. Many specialists also strongly consider a stereotacticbiopsy in patients whose primary and systemic malignantdisease has been in remission for a length of time sufficientto raise the question of a new malignancy rather than ametastasis. Stereotactic or open biopsy of a single brainmetastasis may also be indicated after treatment to distinguishbetween recurrence and treatment effects such as radiationnecrosis.

Several retrospective series report benefits of surgery inselected cases of single brain metastases from lung,67 breast,colon36,47,131 and renal cell origins.48,116,132 The potentialbenefit of resection to long-term survival is significantlyinfluenced by the presence of extracranial metastases.25,122

Surgery for brain metastases from esophageal, gastric,pancreatic and hepatocellular primary sites is rarelyrecommended due to poor overall survival.135 Craniotomymay be indicated in patients whose lesions have progressedafter radiosurgery and/or fractionated radiation and provenunresponsive to systemic chemotherapy.

Single choriocarcinoma metastases to brain are usuallydetected after childbirth and are generally very sensitive toradiation and/or systemic chemotherapy. Occurrence duringpregnancy or lesions unresponsive to systemic therapy orradiotherapy may mandate resection of a solitarychoriocarcinoma metastasis to the brain.68 Germ cell tumors,particularly those of testicular primary origin, are often verysensitive to radiosurgery or WBRT and systemicchemotherapy. Thus, even relatively large lesions may besuccessfully managed without craniotomy. Progressivesymptoms in the face of medical management or presentationin extremis may, however, demand surgical resection.

Three randomized trials of craniotomy followed by WBRTversus WBRT alone for single brain metastases have beenconducted. Two demonstrate that surgery prolongs high-quality survival,90,125 and one does not.79 Resection of singlemetastases may be indicated for symptomatic lesions or inthe hope of providing quick relief from pressure effect.

Table 1: Randomized Trials of Resection Followed by WBRT Versus WBRT Alone

a Karnofsky Performance Statusb World Health Organization Performance Status

First Author Management Modalities PatientNumber

PatientEligibility

MedianSurvival(Months)

StatisticalSignificance

Patchell 199090 WBRT 23 Age > 18KPSa < 70

3.5 p < 0.01

WBRT + Resection 25 9.2

Vecht 1993125 WBRT 31 Age > 18

WHO PSb > 26 p = 0.04

WBRT + Resection 32 10

Mintz 199679 WBRT 43 Age > 50KPS < 80

6.3 p = 0.24

WBRT + Resection 41 5.6

6

Surgery for Multiple MetastasesTraditionally, multiple brain metastases were considered acontraindication to surgical intervention and were treatedexclusively with WBRT.87,99 Resection of multiple metastaseswas undertaken only in the case of life-threatening lesionsor in cases in which the diagnosis was in question.86 Thereare few recent reports recommending surgical resection ofmultiple brain metastases in properly selected patients.11,52

Whole-Brain Radiation TherapyWhole-brain radiation therapy has been a widely appliedmanagement for patients with brain metastases. Various doseand fractionation schedules have been tested in earlier clinicaltrials but no single schedule was proven superior in terms ofoverall survival. Concern about potential late neurotoxicitywith WBRT and the development of radiosurgery have ledto the introduction of newer approaches which do not favorupfront WBRT for all patients. In recent years clinical researchhas attempted to identify various management strategiesappropriate for patients with metastatic brain cancers.

Studies Investigating Dose and Fractionation Schedulesof WBRTTo investigate the effects of WBRT on brain metastases theRTOG conducted several studies using various dose andfractionation schedules.12,57,81 Borgelt et al. reported theoutcome of various dose and fractionation schedules usedin RTOG 69-01 and 73-61.12 Doses ranging from 30–40 Gyin 6–20 fractions (30 Gy in 6 fractions, 40 Gy in 20 fractions)were tested. Although there was no survival difference, themedian time to neurologic progression was longer when moreprotracted schedules were used. No survival difference wasseen when patients were randomized to 30 Gy in 10 fractionsversus 30 Gy in 6 fractions in RTOG 79-16.55 One-third ofpatients died of brain disease suggesting the need for furtherimprovement in CNS control (but also emphasizing thatstudies that use survival as an endpoint in brain metastasestrials will have difficulty remaining relevant since most deathis non-neurologic). No single WBRT dose schedule has beenproven superior for overall survival because most patientsdied of systemic disease. Gaspar et al. performed a recursivepartitioning analysis (RPA) on patients enrolled for WBRTin RTOG trials from 1979 to 1993.41 The median survivalwas 7.1 months for RPA class I (controlled primary tumor,age < 65 years, Karnofsky Performance Status (KPS) > 70,and absence of non-CNS metastases). The median survivalwas 2.4 months for RPA class III (KPS < 70, age > 65, andpresence of other systemic disease). Other patients (RPAclass II) had an intermediate median survival of 4.2 months.The RTOG 85-28 trial evaluated higher doses of radiation(twice per day with accelerated hyperfractionation) inpatients who had a single brain metastasis.33 A total radiationdose of 32 Gy was delivered at 1.6 Gy per fraction whichwas given twice a day. In addition patients received focalradiation boosts to total doses of 48 Gy, 54.4 Gy, 64 Gy and70.4 Gy. Patients treated to 54.4 Gy or higher showedimproved survival and better neurologic function. Whenthis regimen (54.4 Gy) was compared with 30 Gy in 10fractions, no benefit to overall survival with the acceleratedhyperfractionated regimen was noted.81 The toxicity wassimilar in both arms.

Several randomized trials have studied the effect of radiationsensitizers on local tumor control in patients receivingWBRT. Bromodeoxyuridine (BrdUrd) was evaluated inRTOG 89-05 in patients with brain metastases.93 Patientswere randomized to 37.5 Gy in 15 fractions WBRT versusWBRT plus BrdUrd (0.8 g/m2 four days per week duringWBRT). No survival difference was noted. In another studypatients were randomized to WBRT (30 Gy in 10 fractions)or WBRT plus motexafin gadolinium (MGd).75 Motexafingadolinium selectively targets tumor cells where it generatesreactive oxygen species. Motexafin gadolinium improvedthe time to neurologic progression and neurocognitivefunction in the subset of lung cancer patients, but had noeffect on overall survival. A follow-up trial with this agentwas restricted to non-small cell lung cancer (NSCLC)patients. This randomized, controlled Phase 3 trial, knownas the SMART (Study of Neurologic Progression withMotexafin Gadolinium And Radiation Therapy) trial wasdesigned to compare the safety and efficacy of WBRT aloneto WBRT plus MGd. The primary endpoint of the studywas time to neurologic progression (TNP). The trial enrolled554 patients from 94 centers in North America, Europe andAustralia. Although patients receiving MGd had a longertime to neurologic progression, the study’s primary endpoint,the difference compared with patients in the control arm,did not reach statistical significance.74 Another phase IIIstudy tested RSR13 (efaproxiral) in combination withWBRT. RSR13 binds to hemoglobin and reduces its affinityfor oxygen thus increasing tissue pO

2. In this study of WBRT

plus supplemental oxygen with or without RSR13, overallsurvival increased from 4.6 months to 8.7 months in thesubset of breast cancer patients. The role of RSR13 iscurrently being evaluated for patients with brain metastasesfrom breast cancer.110

The combined effect of temozolomide and WBRT has alsobeen studied. In a phase III trial Antonadou et al. randomizedpatients to WBRT alone versus WBRT plus temozolomide(75 mg/m2 daily during WBRT).3 These investigatorsreported significantly improved radiographic response ratesin patients treated with WBRT and temozolomide comparedwith WBRT alone (96% vs. 67%). In addition, patientstreated with combined therapies had greater neurologicalimprovements and lower requirements for corticosteroids.In another randomized study Verger et al. reported similarresults.126 The addition of temozolomide to WBRT did notaffect overall survival; however, patients treated withcombined therapies had better 90-day progression-freesurvival and a lower rate of death secondary to neurologicalcauses. Although there was a trend toward a modest benefitin median survival, no significant improvement in survivalwas documented.

WBRT as an Adjuvant to Surgical ResectionPatchell et al. evaluated the role of WBRT in addition tosurgical resection in a prospective randomized study.Postoperative MRI was used to document that resection wascomplete and no other brain metastases were present.89

Patients who had complete resection of single brainmetastases were randomized to adjuvant WBRT (50.4 Gy at1.8 Gy per fraction) or observation. The crude risk of local

7

recurrence with surgery alone was 46%, and the actuarialrisk of local recurrence at one year was 70%. The resultsshowed that WBRT reduced the incidence of recurrence ofbrain tumors anywhere in the brain from 70% in the surgeryalone group to 18% in the adjuvant WBRT group. Whole-brain radiation therapy reduced the incidence of local failureat the primary resection site and reduced the incidence ofnew brain tumors elsewhere in the brain. The WBRT grouphad a reduced risk of neurologic death. The overall survivalin both arms was not different, although the study was notstatistically powered to detect that endpoint. No differencewas seen in functional independence as measured by KPS.Although this study demonstrated a benefit in terms ofreduction in brain recurrence, it can be argued that becauseno overall survival benefit was seen, WBRT can be delayed.Although WBRT would reduce intracranial relapse aftersurgery or radiosurgery, the toxicity of this therapy has beena concern. DeAngelis et al. reported 12 patients who werecured from brain metastases but developed dementia andsevere disability.26 Eight of the twelve patients were alsotreated with surgery. All patients who experienced neurologicmorbidity received a high dose per fraction (3–6 Gy),whereas no patients treated with lower daily doses had thiscomplication. The studies with a lower dose per fraction orlower total dose seemed to have a much lower risk of toxicity,although many of these studies did not use sensitiveinstruments to assess neurocognitive function. AlthoughWBRT reduces the incidence of progression of disease inthe brain following surgery or radiosurgery, and has beenshown in prospective randomized trials to improve qualityof life, the role of adjuvant WBRT remains controversial inpatients with oligometastatic disease since it does not clearlytranslate into a survival advantage. On the other hand thereare tumors such as SCLC that tend to seed in a miliary fashionand are controllable with low doses of WBRT, suggesting arole for “prophylactic” WBRT.

RadiosurgeryStereotactic radiosurgery is a surgical technique that employshighly focused radiation to treat intracranial targets withsubmillimeter precision in a single session. Stereotacticradiosurgery offers many of the benefits of craniotomywithout its risks. Stereotactic radiosurgery does not causethe physiological stress of an open operation and can beperformed in an outpatient setting. In the same session, SRSmay be used to treat multiple lesions in widely disparatelocations and in eloquent locations not conducive to opensurgical approaches. Three types of devices are commonlyused for radiosurgery: the multisource cobalt60 unit knownas the Gamma Knife,® specially modified or dedicated linearaccelerators like Novalis Tx™ and Axesse,™ or charged-particle irradiators. The Leksell Gamma Knife® consists of192 (Leksell Gamma Knife® Perfexion™) or 201 (ModelsU, B, C and 4-C) cobalt60 sources that emit gammairradiation. The epicenter of delivery of each dose (or “shot”)of radiation is always in the center of the sphere defined bythe helmet. The Leksell Gamma Knife® is the onlytechnology designed for brain tumors to provide the higherlevel of accuracy necessary in the brain. Stereotacticradiosurgery allows delivery of radiation to only a smallvolume with a rapid dose fall-off.

The second method for delivering SRS is by irradiationproduced by modified or dedicated linear accelerator(Novalis Tx™ and Axesse™), a machine that generates high-energy photons. Linear accelerators can be used for SRS byfocusing the beams through a variety of fixed, shaped fieldsat the target, or by a variety of arcs at the axis of rotation.Linear accelerator technology is made to target within thewhole body and may not provide the same accuracy withinthe brain as the cobalt60 based technology. Proton beamsystems use fixed high-energy beams that are either crossfired(non-Bragg peak) or use Bragg peak effect to depositradiation in the tumor.

Radiosurgery as the sole initial management or as a boostbefore or after WBRT has emerged as a widely practicedtreatment modality for brain metastases. The goal ofradiosurgery without WBRT is to achieve brain controlwithout the possible long term neurotoxic or cognitive sideeffects of WBRT.17 The rationale for radiosurgery, whenused as a boost after WBRT, is to achieve improved localbrain tumor control. Radiosurgery boost improves survivalin selected patients in whom the predominant problem isbrain disease rather than extracranial disease. Radiosurgeryis also used as salvage treatment for progressive intracranialdisease after surgery or WBRT. Traditionally radio-insensitive histologies tend to be more responsive to SRSthan to conventional fractionated radiation treatment. Inaddition, SRS causes indirect vascular injury and subsequentsclerosis of blood vessels, and eventual compromise of theblood supply and circulation within the tumor.121 The overallside effects of SRS are limited but can occasionally beserious. There are very few acute side effects of SRS relatedto the radiation. Stereotactic radiosurgery may cause mildfatigue and sometimes a temporary patch of hair loss if thetumor is close to the skull and scalp. There is a risk of lateside effects that can develop, the most common and seriousof which is tumor radionecrosis.134 Radiation necrosis isdamage to the tumor and or adjacent brain in the high-dosearea. This can result in edema and additional side effectsproduced by the mass including seizures and neurologicaldeficits. Radionecrosis can often be managed withcorticosteroids. Occasionally surgical intervention isrequired to reduce the mass effect. The risk of symptomaticradionecrosis is usually less than 5%.2,5,56 A multicenter phaseI RTOG trial involving SRS documented safe SRS in patientspreviously treated with standard external beam radiationtherapy.111 Early publications showed good control rates andled to further investigation.24,64,76,120 Retrospective series haveconsistently revealed local control of the target lesions inthe range of 80–85% or even higher with a very acceptableside effect profile.5,10,20,30,37,51,70 Prospective randomized trialshave demonstrated that the one-year local control rate oftarget lesions with radiosurgery is 73%, which increases to82–89% with the addition of WBRT.2,4

Retrospective Studies for SRSPatients treated with conventional open surgical resectionwithout WBRT had a 46% risk of failure at the site of theresection in a randomized trial evaluating the role of WBRTafter surgical resection.89 In subsequent studies patients weretreated with SRS alone (without WBRT). These studies

8

found excellent local control (70–80% at one year).21,83 Otherpublished series of patients treated with SRS havedemonstrated a risk of distant brain failure at one year,ranging from 43% to 57%.22,49,66,117 In general, the risk ofnew metastasis in patients with solitary tumors isapproximately 37% (crude), but the actuarial risk is 50% atone year.62,89 The histologic features or tumor type may playa role, with melanoma being more likely to be associatedwith multiple metastases than some other tumor types.95

Despite a relatively high risk of new metastases outside theradiosurgery volume in patients who have SRS alone,retrospective studies have not confirmed a survival benefitto adjuvant WBRT.94,117,118 Freedom from local progressionin the brain at one year was significantly superior in patientswho received both SRS and WBRT compared with SRS alone(28% vs. 69%), although the overall survival rate was notsignificantly different.49 A retrospective, multi-institutionalstudy in which patients were treated with SRS alone (n =268) or SRS + WBRT (n = 301) also reported no significantdifference in the overall survival rate.161 Despite the higherrate of new lesions developing in patients treated with SRSalone, the overall survival appears to be equivalent to SRS+ WBRT since salvage therapies are fairly effective andpatients’ extracranial disease is frequently the cause ofdeath.117 Only 24% of patients managed initially withradiosurgery alone required salvage WBRT. Pirzkall et al.reported that there was no survival benefit for an overallgroup of 236 patients with adjuvant WBRT but these authorsnoted a trend toward improved survival in a subset of patientswith no extracranial tumor (15.4 vs. 8.3 months, p = 0.08).94

Chidel et al. reported on 78 patients managed initially withSRS alone and 57 patients treated with SRS and adjuvantWBRT.157 Whole-brain radiation therapy did not improvethe overall survival rate but was useful in preventing boththe local progression and the development of new brainmetastases (74% vs. 48%, p = 0.06). These retrospectivestudies suggest that WBRT will improve local and distantcontrol in the brain, but do not clearly demonstrate a survivaladvantage.117

A multicenter retrospective analysis was performed with 502patients treated at 10 institutions in which all of the patientswere treated with WBRT and SRS. The patients werestratified by the recursive partitioning analysis and comparedwith similar patients from the RTOG database who had beentreated with WBRT alone.104 The study revealed that patientswith higher KPS, controlled primary tumor, absence ofextracranial metastases and lower RPA class had statisticallysuperior survival. The addition of an SRS boost resulted ina median survival of 16.1, 10.3 and 8.7 months, respectively,for RPA classes I, II and III. This is in comparison to 7.1,4.2 and 2.3 months for similar RPA class patients from theRTOG database. This improvement in overall survival,stratified by RPA class with an SRS boost, was statisticallysignificant.104 In a recent study SRS alone was found to beas effective as resection plus WBRT in the treatment of oneor two brain metastases for patients in RPA classes I andII.98

Table 2: Selected Recent Retrospective Studies of Radiosurgery for Brain Metastases

First Author Year No. of Patients Primary Organ

MedianSurvival(Months)

MedianMargin Dose

(GY)Local Control

Gaudy-Marqueste42 2006 106 Melanoma 5.09 25 84%

Bhatnagar9 2006 205 Mixed 8 16 71% at 1 yr

Chang16 2005 109 NSCLC 7.5 18 64% renal 47% melanoma

Serisawa109 2005 521 Mixed 9 20 95.70%

Nam82 2005 130 Mixed 8.75 17.9 63.90%

Pan85 2005 191 all lung 14 18 91%

Gerosa43 2005 504 all NSCLC 14.5 21.4 95%

Lippitz63 2004 215 Mixed 7.8–13.7 22 93.90%

Hasegawa49

2003 121 Mixed 8 18.5 79%

Petrovich92

2002 458 Mixed 9 18 87%

Sheehan112

2002 273 all NSCLC 7 16 86%

Amendola1

2000 68 all breast 7.8 15–24 94%

Simonova115

2000 237 Mixed 6–12 21.5 95%

9

Radiosurgery Versus Resection for Single BrainMetastasesThe available data indicate that SRS and open surgicalresection (where feasible) are both excellent treatmentoptions for patients with solitary brain metastases. Tumorresection offers immediate relief of symptoms in patientswith a large symptomatic tumor causing mass effect.Stereotactic radiosurgery has a number of advantages incomparison to open surgical resection. Stereotacticradiosurgery does not require open resection and can beperformed as an outpatient procedure. Compared with opensurgical resection SRS is a more cost-effective procedure,73

and can be performed for any tumor location. Severalinvestigators have discussed the need to evaluate the outcomeof SRS and open surgical resection in a prospectivefashion.5,80 However, it is unlikely that such a study willever be performed due to the large number of patientsrequired and very difficult randomization, because of currentoutcome data, widely variant risks and eligibility forresection. The prospective trial by the University ofKentucky demonstrated a 48% local control rate with surgeryalone90 whereas the prospective trial by Aoyamademonstrated a local control rate of 73% with SRS alone.4

However no difference in survival has ever beendemonstrated.5

Role of SRS for Multiple Brain MetastasesStereotactic radiosurgery is an effective treatment for patientswith multiple brain metastases. A substantial amount of

published literature now supports use of radiosurgery in thetreatment of multiple brain metastases. Stereotacticradiosurgery offers a very high control rate with a low riskof serious side effects. The RTOG 95-08 study authorsconcluded that addition of stereotactic radiosurgery to WBRTimproved functional autonomy for all patients; thereforeWBRT and stereotactic radiosurgery should be consideredfor patients with two or three brain metastases. For patientswith good performance status up to three brain metastases,SRS in addition to WBRT is reasonable.

Role of Radiosurgery and Resection for Multiple BrainMetastasesThe role of surgery and SRS may be complementary forpatients with multiple metastases, particularly in cases wherethe largest lesion causes symptoms of mass effect and smalllesions are unresectable because of their small size or deeplocation. In this context, the ideal treatment may be surgicalresection of the larger or more symptomatic lesions combinedwith SRS for the surgically inaccessible lesions. Thiscombination approach allows for local treatment of all thebrain lesions, which may be the critical factor for a successfuloutcome.11 Since the University of Kentucky study clearlydemonstrated the need for adjuvant therapy after resectionof a brain metastasis, WBRT is required for these patients.Alternatively, some authors advocate the use of radiosurgeryin the resection cavity when WBRT is withheld,54 thoughthis is controversial.

Table 3: Randomized Trials of WBRT + SRS Versus WBRT Alone or SRS Alone

FirstAuthor

ManagementModalities

No. of Tumors

MaximumTumor

Size

PatientNo. Patient Eligibility

MedianSurvival(Months)

StatisticalSignificance

Andrews20042 WBRT 1–3 4 cm 94

Pts with prior surgery included, Pts with active disease excluded.

4.9 p = 0.0393

WBRT + SRS

92WBRT 37.5 Gy in 15 fractions; SRS 15–24 Gy, linear accelerator or GK.

6.5

Kondziolka199956 WBRT 2–4 2.5 cm 14

Pts with active disease included. WBRT 30 Gy in 12 fractions.

7.5 p = 0.22

WBRT + SRS 13 Study stopped at 60% accrual. 11

Chougule200023 SRS 1–3 3 cm 36

Pts with minimum life expectancy of 3 months were included.

7

WBRT + SRS 37 WBRT: 30 Gy in 10 fractions.SRS: 30 Gy to tumor margin.

5 NA

WBRT

31 SRS + WBRT: SRS 20 Gy to margin + WBRT 30 Gy in 10 fractions. 9

Aoyama20064 SRS 4 2 cm 60 30 Gy in 10 fractions over 2–2.5

weeks. 7.5 p = 0.42

WBRT + SRS

60 Pts with SCC, lymphoma, germinoma and multiple myeloma were excluded.

8

10

Radiosurgery in Addition to WBRT: Level I Evidence

Local Brain Tumor ControlThere have been three randomized trials examining the useof whole-brain radiation therapy plus radiosurgery boostcompared with whole-brain radiation therapy alone inselected patients with brain metastases. Two of these trialshave been reported in the peer-reviewed publishedliterature,2,56 and one has only been reported in abstract formwith the final report pending.23 In addition, one randomizedtrial has been published comparing radiosurgery with WBRTto radiosurgery alone.4

The RTOG 95-08 trial2 randomized 164 patients to WBRTand radiosurgery boost and 167 patients to WBRT alone.Patients with one to three newly diagnosed brain metastaseswere included. The arms of the trial were well balanced forbaseline characteristics known to affect survival such as age,Karnofsky Performance Status and status of extracranialmetastases. Brain metastases with the largest lesion up to amaximum diameter of 4 cm and additional tumors less than3 cm in size were included. Local brain control was definedas unchanged or improved on serial post-treatment MRIscans. Magnetic resonance imaging scans were judged incentral review by one neuroradiologist as complete response,partial response or stable disease. Progressive disease wasdefined as an increase in the size of any lesion, developmentof new intracranial lesions, or stable disease withdeterioration of the neurologic examination.

Kondziolka et al.56 randomized patients with two to four brainmetastases (all < 25 mm) to WBRT alone (30 Gy in 12fractions) or WBRT plus radiosurgery. The study wasdiscontinued at 60% accrual when only 27 patients wererandomized. The results were reported for 14 patients inthe WBRT group and 13 patients in the WBRT andradiosurgery boost group. The two groups were wellbalanced with respect to age, sex, tumor type, number oftumors and extent of extracranial disease. Local brain controlwas defined as no tumor growth based on MRI scans and noincrease in clinical symptoms associated with the lesion.Serial scans were read by an independent blinded observer.The Chougule et al. trial23 randomized patients with one tothree brain metastases to Gamma Knife® radiosurgery alone(30 Gy to the tumor margin), WBRT (30 Gy in 10 fractions)plus Gamma Knife® boost (20 Gy to the tumor margin), andWBRT alone (30 Gy in 10 fractions). Patients with tumorvolume < 30 cc and minimum life expectancy of > 3 monthswere included. Overall median survival was seven, five andnine months for the arms, respectively (not statisticallysignificant). The results of this trial were published only inabstract form. Local brain control was not defined in theabstract.23

In summary, three randomized trials2,23,55 detected animprovement in local brain control in patients treated withWBRT and radiosurgery boost compared with WBRT alone.Local brain control at one year ranged from 82–92% in theradiosurgery boost arm vs. 0–71% in the WBRT alone arm.The study by Kondziolka et al.56 also noted superiorintracranial control with the use of radiosurgery.

Aoyama et al. published the results of a prospectiverandomized trial of radiosurgery followed by WBRT versusradiosurgery alone.4 Local control rates with combinedmodality therapy were significantly better than withradiosurgery alone. The combination of the two therapeuticmodalities provided a local control benefit similar to thatseen in the combined modality arm of RTOG 95-08 (89%and 82%, respectively), demonstrating inter-studyconsistency.

Overall SurvivalNo statistically significant improvement in overall survivalwith the use of radiosurgery boost compared with WBRTalone was reported in these trials.2,23,55 However, RTOG 95-082 reported that WBRT and radiosurgery boost improvedsurvival in RPA Class I patients and in patients with favorablehistologic status, squamous cell or non-small cell lungtumors. Median survivals for the WBRT arm ranged from5.7–7.5 months and for the WBRT and radiosurgery boostarm ranged from 5–11 months.

The RTOG 95-08 trial2 reported that the overall meansurvival time was 6.5 months for WBRT alone and 5.7months for WBRT plus a radiosurgery boost (p = 0.1356).In Kondziolka et al.’s study,56 median survival was notstatistically different between the two groups (7.5 monthsfor WBRT alone vs. 11 months for WBRT and radiosurgeryboost [p = 0.22]). Survival was dependent on the extent ofextracranial disease (p = 0.02) but was not dependent onhistology or number of tumors.

The RTOG 95-082 used recursive partitioning analysis ofprognostic factors41 to analyze the relative contributions ofpretreatment variables on the survival of patients and toidentify subgroups of patients with homogeneous prognosticcharacteristics predictive of survival. The three classes were:Class I, patients with KPS > 70, < 60 years of age withcontrolled primary and no extracranial metastases; Class III,KPS < 70; and Class II, all others. From the historic database,the best survival was noted in Class I patients (median, 7.1months), intermediate in class II patients (median, 4.2months), and the worst in Class III patients (median, 2.3months). The RTOG 95-08 trial2 explored survival outcomesin certain subsets of patients. By multivariate analysis,WBRT and radiosurgery boost improved survival in RPAClass I patients (p < 0.0001) and in patients with favorablehistologic status, squamous cell or non-small cell lung tumors(p = 0.0121). In patients with single brain metastases(unresectable or inoperable) the median survival was 6.5months with radiosurgery boost compared with 4.9 monthswith WBRT alone (p = 0.0393).

Quality of LifeThe RTOG 95-08 trial2 reported that patients in theradiosurgery boost group were more likely to have a stableor improved KPS at six months follow-up than patients inthe WBRT alone group (43% vs. 27%, respectively, p =0.03). Steroid use six months after treatment decreased in41 of 76 patients treated with radiosurgery boost, comparedwith 25 of 75 patients treated with WBRT alone (p = 0.0158).Auchter et al., in a retrospective study,5 reported on quality

11

of life or symptom control outcomes. In this study,5 themedian time in which the KPS was sustained > 70 was 44weeks for patients treated with WBRT and radiosurgery.

ComplicationsThe Kondziolka et al. trial56 reported “no neurologic orsystemic morbidity related to [radiosurgery].” Andrews etal.2 reported a statistically nonsignificant increase in the riskof toxicity with radiosurgery boost, which was 3% acuteGrades 3 and 4 toxicities and 6% late Grades 3 and 4toxicities. However, overall quality of life using a validatedinstrument was not measured in any of these trials.

ConclusionThere is Level I evidence (three randomized trials) thatradiosurgery boost with WBRT, compared with WBRT alone,significantly improves local brain control rate for patientswith up to four metastases. There is Level I evidence toindicate that radiosurgery boost with WBRT improvessurvival in selected patients with a single brain metastasis,and there is Level I evidence that the ability to taper downsteroid dose and improvement of KPS was statistically betterin the radiosurgery arm at six months.2

Radiosurgery Alone as Initial Therapy: Level I Evidence

Local Tumor ControlIn a randomized trial reported in abstract form by Chouguleet al.,23 patients were randomized to Gamma Knife®

radiosurgery alone vs. WBRT and Gamma Knife®

radiosurgery vs. WBRT alone. The local brain control ratewas higher in the two radiosurgery arms: 87% for GammaKnife® radiosurgery alone and 91% for Gamma Knife®

radiosurgery and WBRT, compared with 62% in the WBRTonly arm.

Another randomized trial compared the use of radiosurgerywith WBRT plus radiosurgery as initial therapy in selectedpatients with brain metastases.4 Aoyama et al. reported theresults of a prospective, multi-institutional, randomizedcontrolled trial comparing WBRT plus SRS vs. SRS alonefor patients with limited (defined as < 4) brain metastaseswith a maximum diameter of 3 cm on contrast-enhanced MRIscan.4 Patients with metastases from small cell carcinoma,lymphoma, germinoma and multiple myeloma wereexcluded. Eligible patients had a KPS score of 70 or higher.The WBRT dosage schedule was 30 Gy in 10 fractions over2–2.5 weeks. Metastases with a maximum diameter of upto 2 cm were treated with SRS doses of 22–25 Gy and thoselarger than 2 cm were treated with doses of 18–20 Gy. Thedose was reduced by 30% when the treatment was combinedwith WBRT. Local tumor progression was defined as aradiographic increase of 25% or more in the size of ametastatic lesion. The primary end point of the study wasoverall survival. Secondary end points were cause of death,functional preservation, brain tumor recurrence, salvagetreatment and toxic effects of radiation. One hundred thirty-two patients were randomized (65 to WBRT + SRS and 67to SRS alone). The interim analysis was performed with122 patients (approximately 60 in each group). The JapaneseRadiation Oncology Study Group 99-1 trial4 reported an

actuarial one-year local tumor control rate of 88.7% in theWBRT + SRS group and 72.5% in the SRS-alone group (p= 0.002). The one-year actuarial rate of developing newbrain metastases was 41.5% in the WBRT + SRS group and63.7% in the SRS-alone group (p = 0.003).

A prospective, single arm, multi-institutional EasternCooperative Oncology Group (ECOG) Phase II study ofradiosurgery alone for “radioresistant” histologies(melanoma, sarcoma, renal cell carcinoma) in patients withone to three brain metastases has also been reported.69

Inclusion criteria were one to three newly diagnosed brainmetastases with a maximum diameter of 4 cm. In patientswith multiple lesions and any lesion > 3 cm, all remaininglesions were required to be < 3 cm. Of 36 patients accrued,31 were eligible and evaluable; 14 had melanoma, 14 hadrenal cell carcinoma and three had sarcoma. Three of thirty-one patients (10%) had partial response, 10 of 31 (32%)had stable disease, 14 of 31 (42%) had progressive disease,and 4 of 31 (14%) were not evaluable. At six months, 39.2%failed within the radiosurgery volume and 39.4% failedoutside the radiosurgery volume.

Several retrospective studies21,94,113,117,128 compared local braincontrol rates of those patients receiving initial radiosurgeryalone with those receiving whole-brain radiation therapy.Chidel et al.21 found a statistically significant improvementin two-year brain control with the use of WBRT in additionto radiosurgery boost: 80% vs. 52% in patients treated withradiosurgery alone (p = 0.034). Pirzkall et al.94 found one-year local control rates to be inferior with the radiosurgeryalone group: 89% vs. 92% in the WBRT and radiosurgeryboost group. Shehata et al.113 reported that patients who hadwhole-brain radiation therapy had superior local tumorcontrol rates (97%) compared with patients treated withradiosurgery alone (87%; p = 0.0001). Sneed et al.117

reported a statistically significant improvement in one-yearbrain freedom from progression rate in those patients treatedwith WBRT + SRS boost (69%) compared with those patientstreated with initial radiosurgery only (28%). It wascommented that the one-year brain control rate allowing forsalvage (using WBRT or serial SRS) at first failure was notstatistically different between those treated with initial WBRT+ SRS boost (73%) vs. those treated initially with SRS alone(62%). Wang et al.128 found that the local brain control rateof patients treated with SRS alone was 93.3%, comparedwith 95.6% in patients treated with WBRT + SRS boost.

SurvivalThe Japanese trial4 found no significant survival differencebetween the groups receiving WBRT + SRS and SRS alone.The median survival time was 7.5 months with WBRT +SRS and 8.0 months with SRS alone. In addition, nosignificant difference in the frequency of death due toneurologic causes was observed. Death was attributed toneurologic causes in 22.8% in the WBRT + SRS group andin 19.3% in the SRS alone group. In Chougule et al.’sabstract,23 median survivals were seven, five and nine monthsfor Gamma Knife® radiosurgery alone vs. WBRT and GammaKnife® radiosurgery vs. WBRT, respectively. Survival wasreported as not different among the three arms. The ECOG

12

Phase II trial69 of radiosurgery alone for radioresistanthistologies found median survival to be 8.2 months (95%CI, 7.4–12.2 months) in its cohort of patients.

Lutterbach performed a prospective study66 usingradiosurgery alone for the initial management of brainmetastases. However, no survival comparisons were madewith patients treated with WBRT. Several retrospectivestudies have reported on the use of radiosurgery alone asinitial management of selected patients with brainmetastases.15,21,39,49,53,105,109,113,115,117,118,124,128 Survival outcomesranged from 8–15 months. Chidel et al.21 reported the mediansurvival of patients treated with radiosurgery alone as 10.5months compared with 6.4 months in patients treated withradiosurgery boost and whole-brain radiation therapy (pvalue not stated). Sneed et al.117 reported that the mediansurvival of patients treated initially with radiosurgery alonewas 11.3 months, which was not statistically different fromthe survival of patients treated with WBRT + SRS boost(11.1 months). Wang et al.128 reported a median survival of15 months in patients treated with SRS alone vs. 20 monthsin patients treated with WBRT + SRS boost vs. 8.5 monthsfor patients treated with WBRT alone.

Pirzkall et al.94 found no difference in overall survival forpatients treated with radiosurgery alone or radiosurgery andWBRT; however, in the subset of patients withoutextracranial disease, omitting whole-brain radiation therapyresulted in a survival decrement from 15.4 to 8.3 months.Sneed et al.118 collected data from 10 institutions to comparethe survival probabilities of patients with newly diagnosedbrain metastases managed initially with SRS alone vs. SRSand WBRT. Of the 569 evaluable patients, 268 hadradiosurgery alone initially (24% of these ultimately neededsalvage WBRT) and 301 had radiosurgery and up-frontWBRT. The median survival times for patients treated withSRS initially vs. SRS + WBRT were 14.0 vs. 15.2 monthsfor RPA Class 1, 8.2 vs. 7.0 months for Class II, and 5.3 vs.5.5 months for Class III. With adjustment by RPA class,there was no survival difference comparing radiosurgeryalone initially with radiosurgery and up-front whole-brainradiation therapy.

There is Level I evidence from the recently publishedJapanese trial4 and Level II-3 evidence from literature thataddition of up-front WBRT does not improve survival inpatients treated with up-front radiosurgery. Thus patientswith newly diagnosed brain metastases can be treated withup-front SRS alone, reserving WBRT for salvage.

Quality of Life or Symptom ControlNo formal comparisons between radiosurgery alone vs.competing options of management such as WBRT have beenmade in terms of quality of life or symptom control amongthe studies. The only study that has reported KPS outcomeshas been the Japanese randomized trial.4 Actuarial one-yearKPS preservation rate (KPS > 70) was 25% in the SRS alonearm and 37% in the WBRT + SRS arm. No formal neurologicfunctional tests were prospectively performed. However,validated quality of life outcomes have not been reported inany of these studies examining the use of radiosurgery alone

(without whole-brain radiation therapy) as up-front treatmentfor brain metastases.

ComplicationsIn the randomized Japanese trial,4 there was no statisticallysignificant difference in any grade of either acute or lateradiation toxicities between the radiosurgery alone arm andthe radiosurgery and whole-brain radiation therapy arm. Inthe ECOG trial,69 two Grade 3 events in 31 patients (oneseizure and one fatigue) were thought to be possibly relatedto radiosurgery. The prospective study by Lutterbach et al.66

reported 13% of patients experiencing complications withradiosurgery alone as initial treatment for brain metastases.Nine percent were acute toxicities and 4% were latetoxicities. Varying degrees of toxicities were reported inthe retrospective series examining the outcomes of patientstreated with radiosurgery alone.

ConclusionThere is Level I to Level II-3 evidence that addition of WBRTin patients treated with radiosurgery for 1–3 newly diagnosedbrain metastases does not improve survival, compared withradiosurgery alone with WBRT reserved for salvage therapy.There is Level I evidence that omission of WBRT results indecreased tumor control, both at the site of radiosurgery andalso in the remaining untreated brain. Level II-1 and LevelII-3 evidence further support this observation.

Other Management Strategies

ChemotherapyIn principle the use of chemotherapy for brain metastases isappealing because it can treat multiple brain tumors as wellas systemic cancer. The neurocognitive profile afterchemotherapy is better than with WBRT. However, thedevelopment of brain metastases while patients areundergoing systemic chemotherapy indicates that the blood-brain barrier makes the brain a sanctuary from manychemotherapeutic agents. The tight intercellular junctionsof the BBB prevent the free passage of drugs into brainparenchyma and CSF. However, some chemotherapeuticagents do cross the intact BBB fairly well. Such agentsinclude the nitrosoureas (e.g., BCNU and CCNU), thiotepaand temozolomide. Topotecan, irinotecan andhydroxyifosfamide are also probably effective. High-dosemethotrexate and cytarabine, etoposide and idarubicin donot cross the BBB well but may still achieve cytotoxic CSFconcentrations. Several studies have examined the use ofchemotherapy in brain metastases and have shown promisefor patients with brain metastases from small cell lungcancer,3,46,60,96 breast cancer38,175 and germ cell tumors. Brainmetastases from NSCLC are generally less chemosensitivethan those from SCLC.31 In summary, despite concerns aboutthe BBB, which plays a role in allowing development ofbrain metastases in patients undergoing systemicchemotherapy, there is good evidence that systemicchemotherapy is sometimes useful in managing brainmetastases. Its role will be limited to patients with multiplebrain metastases or active systemic cancer reasonably likelyto respond to chemotherapy. The optimal drugs, scheduleand route of administration remain to be determined.

13

Local Delivery of Radiation or Drug at the Time of SurgeryLocal radiation techniques, such as brachytherapy with I125

seeds and inflatable balloon catheters utilizing a liquid I125

(GliaSite® Radiation Therapy System), are being evaluatedfor resectable brain metastases. GliaSite® consists of aninflatable balloon catheter that is inserted into resectioncavities at the completion of surgery. Radiation is deliveredwith an aqueous solution of organically bound I125 (Iotrex®

[sodium 3-(125I)-iodo-4-hydroxybenzenesulfonate]), whichis temporarily introduced into the balloon portion of thedevice via a subcutaneous port approximately 5–14 daysfollowing surgery. The I125 is uniformly distributed withinthe treatment balloon and provides radiation to thesurrounding brain parenchyma. The Iotrex® is typicallyallowed to dwell in the device for 3–6 days, after which thedevice is explanted. Complications associated with GliaSite®

are reported to include pseudomeningocele development,wound infection and meningitis. The GliaSite® systemremains an option for the treatment of newly diagnosed orpotentially recurrent metastases when other forms oftreatment have failed.72 In addition, a Phase II study ofGliaSite® brachytherapy after resection of a single metastasishas demonstrated local control rate, median patient survivaltime, and duration of functional independence similar to thatachieved with resection plus whole-brain radiation therapy.101

GliaSite® can be particularly useful for tumors with volumestoo large to permit radiosurgery.

Gliadel® wafers (BCNU biodegradable polymer wafers) havebeen developed as a method of optimizing drug deliveryinto the CNS with satisfactory safety and efficacy withoutproducing undesirable systemic side effects. Thechemotherapy wafers are implanted in the tumor resectioncavity at the completion of surgery.13 Gliadel® was found tobe effective in breast cancer without subsequentadministration of radiotherapy.34,35 Local BCNUadministration was also shown to be effective.34 Localtherapies delivered at the time of surgical resection of brainmetastases hold promise but have yet to be rigorously provento provide benefit for patients with brain metastases.

Targeted Therapies for Brain MetastasesSignificant advances have been made in the understandingof the molecular and cellular biology of normal andcancerous cells. Molecular abnormalities includechromosomal aberrations, which can result in overexpressionof oncogenes or inhibition of tumor suppressor genes. On acellular level, abnormalities of growth factor regulation,signaling pathway dysregulation, and dedifferentiation haveall been implicated in malignant progression. The metastaticphenotype is also associated with abnormal angiogenesis andinvasion. All of these molecular and cytogeneticabnormalities are potential targets for therapy. A detaileddiscussion of targeted therapy for brain metastases iscomplex and beyond the scope of these guidelines. Severalagents have been developed to target pathways that arecritical for the ongoing growth and proliferation of cancer.Many of these signaling pathways are involved in criticalcellular events, including DNA repair, cell survival signals,invasion, angiogenesis, metastasis and apoptosis. In thefuture these pathways may play an important role in

mediating sensitivity to chemotherapy and radiation therapy.There is increasing evidence that the newer biologic agentstargeting cellular protein receptors or other components ofthe tumor microenvironment may work synergistically withconventional radiation and cytotoxic agents.

Repeat RadiosurgerySince tumor control rate after radiosurgery is 80–90%, othermanagement options after radiosurgery may be needed forpatients with documented tumor growth. Whole-brainradiation therapy, microsurgery, and in selected cases repeatradiosurgery, can be considered for patients with tumorgrowth despite radiosurgery. Very little data are availableon repeat radiosurgery for brain metastases.8

Indications for Radiosurgery• Newly diagnosed single or multiple brain metastases

without significant mass effect documented on imaging• Boost after WBRT for single or multiple brain metastases• Recurrent brain metastases after WBRT• Radiosurgery for residual tumor after resection

Clinical Algorithm

Several factors are considered in making a recommendation.These factors include:

1. Patient’s age2. Patient’s symptoms3. Status of systemic disease4. Patient’s current neurological status5. Patient’s medical condition6. Presence or absence of other organ metastases7. History of prior WBRT8. History of prior brain procedures9. Patient’s concern and risk tolerance for neuro-cognitive9. functions

10. Patient’s wishes

Tumor SizeRadiosurgery can be performed for tumors up to 4 cm inmaximum diameter. However, tumor volume, dose andlocation are more important variables.

Patient PreferencePatients’ preferences are also considered in selecting amanagement approach.

A broad outline of brain metastases diagnostic work-up andmanagement algorithms for single tumor, limited braindisease (2–4 tumors) and multiple metastases are shown.However, the final recommendation is usually influencedby the recommending surgeon’s, radiation oncologist’s andneuro-oncologist’s experiences along with patient preference.

14

Known Cancer No Known Cancer

Metastatic Work-up

Stereotactic Biopsy or Resection

Not Sure of Brain Met

Primary FoundNo Primary

Brain Lesion Suggestive of Metastasis on MRI

Metastatic Tumor Confirmed

1. Discuss roles of SRS, WBRT, Resection and Chemotherapy at differentstages in treatment.

2. Assess systemic disease (status of primary and metastases in otherorgan systems).

3. Address concerns regarding cognitive effects, local and distant tumorcontrol.

4. Help patient choose appropriate management option.5. Start treatment with patient’s first choice of management.

15

Single Brain Metastasis on MRI

Lobar, Resectable

Mass Effect No Mass Effect

Complete Resection

New Lesions

Resection or Repeat SRS

Tumor Recurrence

SRS + WBRT

Local

SRS or WBRT

Residual Tumor

Tumor Bed SRS or XRT

Non Lobar, Nonresectable

Resection

16

Limited (2–4) Brain Metastases on MRI

2nd SRS

Tumor Progression

SRS or WBRT

Local

PoorGood

SRS + WBRTSRS alone

Confirm limited number of brain metastases with high-resolution, thin slice (2 mm) double dose contrast enhanced MRI.

Assess systemic disease control and functional status.

New Lesions

WBRT + SRS Boost

Radiosensitive Tumors Radioinsensitive Tumors

17

Multiple (>4) Brain Metastases on MRI

Conventional

Management

Emerging

Strategies

WBRT

Progression

SRS Limited WBRT Boost

SRS + WBRT

Progression

Repeat SRS

Limited WBRT Boost

18

References

1. Amendola BE, Wolf AL, Coy SR, Amendola M,Bloch L: Gamma knife radiosurgery in the treatmentof patients with single and multiple brain metastasesfrom carcinoma of the breast. Cancer J 6:88-92,2000

2. Andrews DW, Scott CB, Sperduto PW, Flanders AE,Gaspar LE, Schell MC, et al: Whole brain radiationtherapy with or without stereotactic radiosurgeryboost for patients with one to three brain metastases:phase III results of the RTOG 9508 randomised trial.Lancet 363:1665-1672, 2004

3. Antonadou D, Paraskevaidis M, Sarris G, ColiarakisN, Economou I, Karageorgis P, et al: Phase IIrandomized trial of temozolomide and concurrentradiotherapy in patients with brain metastases. J ClinOncol 20:3644-3650, 2002

4. Aoyama H, Shirato H, Tago M, Nakagawa K, ToyodaT, Hatano K, et al: Stereotactic radiosurgery pluswhole-brain radiation therapy vs stereotacticradiosurgery alone for treatment of brain metastases:a randomized controlled trial. Jama 295:2483-2491,2006

5. Auchter RM, Lamond JP, Alexander E, Buatti JM,Chappell R, Friedman WA, et al: A multiinstitutionaloutcome and prognostic factor analysis ofradiosurgery for resectable single brain metastasis.Int J Radiat Oncol Biol Phys 35:27-35, 1996

6. Barnholtz-Sloan JS, Sloan AE, Davis FG, VigneauFD, Lai P, Sawaya RE: Incidence proportions of brainmetastases in patients diagnosed (1973 to 2001) inthe Metropolitan Detroit Cancer SurveillanceSystem. J Clin Oncol 22:2865-2872, 2004

7. Bendell JC, Domchek SM, Burstein HJ, Harris L,Younger J, Kuter I, et al: Central nervous systemmetastases in women who receive trastuzumab-basedtherapy for metastatic breast carcinoma. Cancer97:2972-2977, 2003

8. Bhatnagar A, Heron DE, Kondziolka D, LunsfordLD, Flickinger JC: Analysis of repeat stereotacticradiosurgery for progressive primary and metastaticCNS tumors. Int J Radiat Oncol Biol Phys 53:527-532, 2002

9. Bhatnagar AK, Flickinger JC, Kondziolka D,Lunsford LD: Stereotactic radiosurgery for four ormore intracranial metastases. Int J Radiat OncolBiol Phys 64:898-903, 2006

10. Bhatnagar AK, Kondziolka D, Lunsford LD,Flickinger JC: Recursive partitioning analysis ofprognostic factors for patients with four or moreintracranial metastases treated with radiosurgery.Technol Cancer Res Treat 6:153-160, 2007

11. Bindal RK, Sawaya R, Leavens ME, Lee JJ: Surgicaltreatment of multiple brain metastases. J Neurosurg79:210-216, 1993

12. Borgelt B, Gelber R, Kramer S, Brady LW, ChangCH, Davis LW, et al: The palliation of brainmetastases: final results of the first two studies bythe Radiation Therapy Oncology Group. Int J RadiatOncol Biol Phys 6:1-9, 1980

13. Brem H, Piantadosi S, Burger PC, Walker M, SelkerR, Vick NA, et al: Placebo-controlled trial of safetyand efficacy of intraoperative controlled delivery by

biodegradable polymers of chemotherapy forrecurrent gliomas. The Polymer-brain TumorTreatment Group. Lancet 345:1008-1012, 1995

14. Chambers AF, MacDonald IC, Schmidt EE, MorrisVL, Groom AC: Clinical targets for anti-metastasistherapy. Adv Cancer Res 79:91-121, 2000

15. Chang EL, Hassenbusch SJ 3rd, Shiu AS, Lang FF,Allen PK, Sawaya R, et al: The role of tumor size inthe radiosurgical management of patients withambiguous brain metastases. Neurosurgery 53:272-280; discussion 280-281, 2003

16. Chang EL, Selek U, Hassenbusch SJ 3rd, Maor MH,Allen PK, Mahajan A, et al: Outcome variationamong “radioresistant” brain metastases treated withstereotactic radiosurgery. Neurosurgery 56:936-945; discussion 936-945, 2005

17. Chang EL, Wefel JS, Maor MH, Hassenbusch SJ 3rd,Mahajan A, Lang FF, et al: A pilot study ofneurocognitive function in patients with one to threenew brain metastases initially treated with stereotacticradiosurgery alone. Neurosurgery 60:277-283;discussion 283-284, 2007

18. Chang JE, Robins HI, Mehta MP: Therapeuticadvances in the treatment of brain metastases. ClinAdv Hematol Oncol 5:54-64, 2007

19. Chao ST, Suh JH, Raja S, Lee SY, Barnett G: Thesensitivity and specificity of FDG PET indistinguishing recurrent brain tumor fromradionecrosis in patients treated with stereotacticradiosurgery. Int J Cancer 96:191-197, 2001

20. Chernov MF, Nakaya K, Izawa M, Hayashi M, UsubaY, Kato K, et al: Outcome after radiosurgery for brainmetastases in patients with low Karnofskyperformance scale (KPS) scores. Int J Radiat OncolBiol Phys 67:1492-1498, 2007

21. Chidel MA, Suh JH, Reddy CA, Chao ST, LundbeckMF, Barnett GH: Application of recursivepartitioning analysis and evaluation of the use ofwhole brain radiation among patients treated withstereotactic radiosurgery for newly diagnosed brainmetastases. Int J Radiat Oncol Biol Phys 47:993-999, 2000

22. Chitapanarux I, Goss B, Vongtama R, Frighetto L,De Salles A, Selch M, et al: Prospective study ofstereotactic radiosurgery without whole brainradiotherapy in patients with four or less brainmetastases: incidence of intracranial progression andsalvage radiotherapy. J Neurooncol 61:143-149,2003

23. Chougule PB, Burton-Williams M, Saris S, ZhengZ, Ponte B, Norén G, et al: Randomized treatment ofbrain metastasis with gamma knife radiosurgery,whole brain radiotherapy or both. Int J Radiat OncolBiol Phys 48(Suppl 1):114, 2000

24. Coffey RJ, Flickinger JC, Bissonette DJ, LunsfordLD: Radiosurgery for solitary brain metastases usingthe cobalt-60 gamma unit: methods and results in 24patients. Int J Radiat Oncol Biol Phys 20:1287-1295, 1991

25. Cormio G, Maneo A, Colamaria A, Loverro G, LissoniA, Selvaggi L: Surgical resection of solitary brainmetastasis from ovarian carcinoma: an analysis of22 cases. Gynecol Oncol 89:116-119, 2003

19

26. DeAngelis LM, Delattre JY, Posner JB: Radiation-induced dementia in patients cured of brainmetastases. Neurology 39:789-796, 1989

27. Decker DA, Decker VL, Herskovic A, CummingsGD: Brain metastases in patients with renal cellcarcinoma: prognosis and treatment. J Clin Oncol2:169-173, 1984

28. Delattre JY, Krol G, Thaler HT, Posner JB:Distribution of brain metastases. Arch Neurol45:741-744, 1988

29. DeYoung BR, Wick MR: Immunohistologicevaluation of metastatic carcinomas of unknownorigin: an algorithmic approach. Semin DiagnPathol 17:184-193, 2000

30. DiLuna ML, King JT, Jr., Knisely JP, Chiang VL:Prognostic factors for survival after stereotacticradiosurgery vary with the number of cerebralmetastases. Cancer 109:135-145, 2007

31. Dziadziuszko R, Ardizzoni A, Postmus PE, Smit EF,Price A, Debruyne C, et al: Temozolomide in patientswith advanced non-small cell lung cancer with andwithout brain metastases. a phase II study of theEORTC Lung Cancer Group (08965). Eur J Cancer39:1271-1276, 2003

32. Engh JA, Flickinger JC, Niranjan A, Amin DV,Kondziolka DS, Lunsford LD: OptimizingIntracranial Metastasis Detection for StereotacticRadiosurgery. Stereotact Funct Neurosurg 85:162-168, 2007

33. Epstein BE, Scott CB, Sause WT, Rotman M, SneedPK, Janjan NA, et al: Improved survival duration inpatients with unresected solitary brain metastasisusing accelerated hyperfractionated radiation therapyat total doses of 54.4 gray and greater. Results ofRadiation Therapy Oncology Group 85-28. Cancer71:1362-1367, 1993

34. Ewend MG, Sampath P, Williams JA, Tyler BM,Brem H: Local delivery of chemotherapy prolongssurvival in experimental brain metastases from breastcarcinoma. Neurosurgery 43:1185-1193, 1998

35. Ewend MG, Williams JA, Tabassi K, Tyler BM, BabelKM, Anderson RC, et al: Local delivery ofchemotherapy and concurrent external beamradiotherapy prolongs survival in metastatic braintumor models. Cancer Res 56:5217-5223, 1996

36. Farnell GF, Buckner JC, Cascino TL, O’Connell MJ,Schomberg PJ, Suman V: Brain metastases fromcolorectal carcinoma. The long term survivors.Cancer 78:711-716, 1996

37. Flickinger JC, Kondziolka D, Lunsford LD, CoffeyRJ, Goodman ML, Shaw EG, et al: A multi-institutional experience with stereotactic radiosurgeryfor solitary brain metastasis. Int J Radiat Oncol BiolPhys 28:797-802, 1994

38. Franciosi V, Cocconi G, Michiara M, Di Costanzo F,Fosser V, Tonato M, et al: Front-line chemotherapywith cisplatin and etoposide for patients with brainmetastases from breast carcinoma, nonsmall cell lungcarcinoma, or malignant melanoma: a prospectivestudy. Cancer 85:1599-1605, 1999

39. Fukuoka S, Seo Y, Takanashi M, Takahashi S,Suematsu K, Nakamura J: Radiosurgery of brainmetastases with the Gamma Knife. Stereotact FunctNeurosurg 66 Suppl 1:193-200, 1996

40. Galanaud D, Nicoli F, Le Fur Y, Guye M, RanjevaJP, Confort-Gouny S, et al: Multimodal magneticresonance imaging of the central nervous system.Biochimie 85:905-914, 2003

41. Gaspar L, Scott C, Rotman M, Asbell S, Phillips T,Wasserman T, et al: Recursive partitioning analysis(RPA) of prognostic factors in three RadiationTherapy Oncology Group (RTOG) brain metastasestrials. Int J Radiat Oncol Biol Phys 37:745-751,1997

42. Gaudy-Marqueste C, Regis JM, Muracciole X,Laurans R, Richard MA, Bonerandi JJ, et al: Gamma-Knife radiosurgery in the management of melanomapatients with brain metastases: a series of 106 patientswithout whole-brain radiotherapy. Int J RadiatOncol Biol Phys 65:809-816, 2006

43. Gerosa M, Nicolato A, Foroni R, Tomazzoli L,Bricolo A: Analysis of long-term outcomes andprognostic factors in patients with non-small cell lungcancer brain metastases treated by gamma kniferadiosurgery. J Neurosurg 102 Suppl:75-80, 2005

44. Glantz MJ, Cole BF, Forsyth PA, Recht LD, WenPY, Chamberlain MC, et al: Practice parameter:anticonvulsant prophylaxis in patients with newlydiagnosed brain tumors. Report of the QualityStandards Subcommittee of the American Academyof Neurology. Neurology 54:1886-1893, 2000

45. Graus F, Walker RW, Allen JC: Brain metastases inchildren. J Pediatr 103:558-561, 1983

46. Grossi F, Scolaro T, Tixi L, Loprevite M, ArdizzoniA: The role of systemic chemotherapy in thetreatment of brain metastases from small-cell lungcancer. Crit Rev Oncol Hematol 37:61-67, 2001

47. Hammoud MA, McCutcheon IE, Elsouki R, SchoppaD, Patt YZ: Colorectal carcinoma and brainmetastasis: distribution, treatment, and survival. AnnSurg Oncol 3:453-463, 1996

48. Hankins JR, Miller JE, Salcman M, Ferraro F, GreenDC, Attar S, et al: Surgical management of lungcancer with solitary cerebral metastasis. Ann ThoracSurg 46:24-28, 1988

49. Hasegawa T, Kondziolka D, Flickinger JC,Germanwala A, Lunsford LD: Brain metastasestreated with radiosurgery alone: an alternative towhole brain radiotherapy? Neurosurgery 52:1318-1326; discussion 1326, 2003

50. Healy ME, Hesselink JR, Press GA, Middleton MS:Increased detection of intracranial metastases withintravenous Gd-DTPA. Radiology 165:619-624,1987

51. Hussain A, Brown PD, Stafford SL, Pollock BE:Stereotactic radiosurgery for brainstem metastases:Survival, tumor control, and patient outcomes. Int JRadiat Oncol Biol Phys 67:521-524, 2007

52. Iwadate Y, Namba H, Yamaura A: Significance ofsurgical resection for the treatment of multiple brainmetastases. Anticancer Res 20:573-577, 2000

53. Joseph J, Adler JR, Cox RS, Hancock SL: Linearaccelerator-based stereotaxic radiosurgery for brainmetastases: the influence of number of lesions onsurvival. J Clin Oncol 14:1085-1092, 1996

54. Kim PK, Ellis TL, Stieber VW, McMullen KP, ShawEG, McCoy TP, et al: Gamma Knife surgery targetingthe resection cavity of brain metastasis that has

20

progressed after whole-brain radiotherapy. JNeurosurg (Suppl) 105:75-78, 2006

55. Komarnicky LT, Phillips TL, Martz K, Asbell S,Isaacson S, Urtasun R: A randomized phase IIIprotocol for the evaluation of misonidazole combinedwith radiation in the treatment of patients with brainmetastases (RTOG-7916). Int J Radiat Oncol BiolPhys 20:53-58, 1991

56. Kondziolka D, Patel A, Lunsford LD, Kassam A,Flickinger JC: Stereotactic radiosurgery plus wholebrain radiotherapy versus radiotherapy alone forpatients with multiple brain metastases. Int J RadiatOncol Biol Phys 45:427-434, 1999

57. Kurtz JM, Gelber R, Brady LW, Carella RJ, CooperJS: The palliation of brain metastases in a favorablepatient population: a randomized clinical trial by theRadiation Therapy Oncology Group. Int J RadiatOncol Biol Phys 7:891-895, 1981

58. Lagerwaard FJ, Levendag PC, Nowak PJ,Eijkenboom WM, Hanssens PE, Schmitz PI:Identification of prognostic factors in patients withbrain metastases: a review of 1292 patients. Int JRadiat Oncol Biol Phys 43:795-803, 1999

59. Landis SH, Murray T, Bolden S, Wingo PA: Cancerstatistics, 1998. CA Cancer J Clin 48:6-29, 1998

60. Landonio G, Sartore-Bianchi A, Giannetta L, RengaM, Riva M, Siena S: Controversies in themanagement of brain metastases: the role ofchemotherapy. Forum (Genova) 11:59-74, 2001

61. Lassman AB, DeAngelis LM: Brain metastases.Neurol Clin 21:1-23, vii, 2003

62. Lavine SD, Petrovich Z, Cohen-Gadol AA, MasriLS, Morton DL, O’Day SJ, et al: Gamma kniferadiosurgery for metastatic melanoma: an analysisof survival, outcome, and complications.Neurosurgery 44:59-64; discussion 64-56, 1999

63. Lippitz BE, Kraepelien T, Hautanen K, Ritzling M,Rah T, Ulfarsson E, et al: Gamma knife radiosurgeryfor patients with multiple cerebral metastases. ActaNeurochir Suppl 91:79-87, 2004

64. Loeffler JS, Kooy HM, Wen PY, Fine HA, ChengCW, Mannarino EG, et al: The treatment of recurrentbrain metastases with stereotactic radiosurgery. JClin Oncol 8:576-582, 1990

65. Long DM: Capillary ultrastructure in humanmetastatic brain tumors. J Neurosurg 51:53-58,1979

66. Lutterbach J, Cyron D, Henne K, Ostertag CB:Radiosurgery followed by planned observation inpatients with one to three brain metastases.Neurosurgery 52:1066-1073; discussion 1073-1074, 2003

67. Magilligan DJ Jr., Duvernoy C, Malik G, Lewis JW,Jr., Knighton R, Ausman JI: Surgical approach to lungcancer with solitary cerebral metastasis: twenty-fiveyears’ experience. Ann Thorac Surg 42:360-364,1986

68. Mamelak AN, Withers GJ, Wang X:Choriocarcinoma brain metastasis in a patient withviable intrauterine pregnancy. Case report. JNeurosurg 97:477-481, 2002

69. Manon R, O’Neill A, Knisely J, Werner-Wasik M,Lazarus HM, Wagner H, et al: Phase II trial ofradiosurgery for one to three newly diagnosed brain

metastases from renal cell carcinoma, melanoma, andsarcoma: an Eastern Cooperative Oncology Groupstudy (E 6397). J Clin Oncol 23:8870-8876, 2005

70. Mathieu D, Kondziolka D, Cooper PB, FlickingerJC, Niranjan A, Agarwala S, et al: Gamma kniferadiosurgery in the management of malignantmelanoma brain metastases. Neurosurgery 60:471-481; discussion 481-482, 2007

71. Mavrakis AN, Halpern EF, Barker FG 2nd, GonzalezRG, Henson JW: Diagnostic evaluation of patientswith a brain mass as the presenting manifestation ofcancer. Neurology 65:908-911, 2005

72. McDermott MW, Cosgrove GR, Larson DA, SneedPK, Gutin PH: Interstitial brachytherapy forintracranial metastases. Neurosurg Clin N Am7:485-495, 1996

73. Mehta M, Noyes W, Craig B, Lamond J, Auchter R,French M, et al: A cost-effectiveness and cost-utilityanalysis of radiosurgery vs. resection for single-brainmetastases. Int J Radiat Oncol Biol Phys 39:445-454, 1997

74. Mehta MP, Gervais R, Chabot P: Motexafingadolinium (MGd) combined with prompt wholebrain radiation therapy (RT) prolongs time toneurologic progression in non-small cell lung cancer(NSCLC) patients with brain metastases: results of aphase III trial (Abstract). J Clin Oncol 24:7014,2006

75. Mehta MP, Rodrigus P, Terhaard CH, Rao A, Suh J,Roa W, et al: Survival and neurologic outcomes in arandomized trial of motexafin gadolinium and whole-brain radiation therapy in brain metastases. J ClinOncol 21:2529-2536, 2003

76. Mehta MP, Rozental JM, Levin AB, Mackie TR,Kubsad SS, Gehring MA, et al: Defining the role ofradiosurgery in the management of brain metastases.Int J Radiat Oncol Biol Phys 24:619-625, 1992

77. Merchut MP: Brain metastases from undiagnosedsystemic neoplasms. Arch Intern Med 149:1076-1080, 1989

78. Mintz A, Perry J, Spithoff K, Chambers A, LaperriereN: Management of single brain metastasis: a practiceguideline. Curr Oncol 14:131-143, 2007

79. Mintz AH, Kestle J, Rathbone MP, Gaspar L,Hugenholtz H, Fisher B, et al: A randomized trial toassess the efficacy of surgery in addition toradiotherapy in patients with a single cerebralmetastasis. Cancer 78:1470-1476, 1996

80. Muacevic A, Kreth FW, Horstmann GA, Schmid-Elsaesser R, Wowra B, Steiger HJ, et al: Surgery andradiotherapy compared with gamma kniferadiosurgery in the treatment of solitary cerebralmetastases of small diameter. J Neurosurg 91:35-43, 1999

81. Murray KJ, Scott C, Greenberg HM, Emami B,Seider M, Vora NL, et al: A randomized phase IIIstudy of accelerated hyperfractionation versusstandard in patients with unresected brain metastases:a report of the Radiation Therapy Oncology Group(RTOG) 9104. Int J Radiat Oncol Biol Phys39:571-574, 1997

82. Nam TK, Lee JI, Jung YJ, Im YS, An HY, Nam DH,et al: Gamma knife surgery for brain metastases inpatients harboring four or more lesions: survival and

21

prognostic factors. J Neurosurg 102 Suppl:147-150,2005

83. Nieder C, Nestle U, Motaref B, Walter K, NiewaldM, Schnabel K: Prognostic factors in brainmetastases: should patients be selected for aggressivetreatment according to recursive partitioning analysis(RPA) classes? Int J Radiat Oncol Biol Phys46:297-302, 2000

84. Nutt SH, Patchell RA: Intracranial hemorrhageassociated with primary and secondary tumors.Neurosurg Clin N Am 3:591-599, 1992

85. Pan HC, Sheehan J, Stroila M, Steiner M, Steiner L:Gamma knife surgery for brain metastases from lungcancer. J Neurosurg (Suppl) 102:128-133, 2005

86. Patchell RA: Brain metastases. Neurol Clin 9:817-824, 1991

87. Patchell RA: Metastatic brain tumors. Neurol Clin13:915-925, 1995

88. Patchell RA, Posner JB: Neurologic complicationsof systemic cancer. Neurol Clin 3:729-750, 1985

89. Patchell RA, Tibbs PA, Regine WF, Dempsey RJ,Mohiuddin M, Kryscio RJ, et al: Postoperativeradiotherapy in the treatment of single metastases tothe brain: a randomized trial. Jama 280:1485-1489,1998

90. Patchell RA, Tibbs PA, Walsh JW, Dempsey RJ,Maruyama Y, Kryscio RJ, et al: A randomized trialof surgery in the treatment of single metastases tothe brain. N Engl J Med 322:494-500, 1990

91. Percy AK, Elveback LR, Okazaki H, Kurland LT:Neoplasms of the central nervous system.Epidemiologic considerations. Neurology 22:40-48,1972

92. Petrovich Z, Yu C, Giannotta SL, O’Day S, ApuzzoML: Survival and pattern of failure in brain metastasistreated with stereotactic gamma knife radiosurgery.J Neurosurg 97:499-506, 2002

93. Phillips TL, Scott CB, Leibel SA, Rotman M,Weigensberg IJ: Results of a randomized comparisonof radiotherapy and bromodeoxyuridine withradiotherapy alone for brain metastases: report ofRTOG trial 89-05. Int J Radiat Oncol Biol Phys33:339-348, 1995

94. Pirzkall A, Debus J, Lohr F, Fuss M, Rhein B,Engenhart-Cabillic R, et al: Radiosurgery alone orin combination with whole-brain radiotherapy forbrain metastases. J Clin Oncol 16:3563-3569, 1998

95. Posner JB: Brain metastases: 1995. A brief review. JNeurooncol 27:287-293, 1996

96. Postmus PE, Smit EF: Chemotherapy for brainmetastases of lung cancer: a review. Ann Oncol10:753-759, 1999

97. Quinn JA, DeAngelis LM: Neurologic emergenciesin the cancer patient. Semin Oncol 27:311-321, 2000

98. Rades D, Bohlen G, Pluemer A, Veninga T, HanssensP, Dunst J, et al: Stereotactic radiosurgery aloneversus resection plus whole-brain radiotherapy for 1or 2 brain metastases in recursive partitioninganalysis class 1 and 2 patients. Cancer 109:2515-2521, 2007

99. Ransohoff J: Surgical management of metastatictumors. Semin Oncol 2:21-27, 1975

100. Richards GM, Khuntia D, Mehta MP: Therapeuticmanagement of metastatic brain tumors. Crit RevOncol Hematol 61:70-78, 2007

101. Rogers LR, Rock JP, Sills AK, Vogelbaum MA, SuhJH, Ellis TL, et al: Results of a phase II trial of theGliaSite radiation therapy system for the treatmentof newly diagnosed, resected single brain metastases.J Neurosurg 105:375-384, 2006

102. Sabatier J, Ibarrola D, Malet-Martino M, Berry I:[Brain tumors: interest of magnetic resonancespectroscopy for the diagnosis and the prognosis].Rev Neurol (Paris) 157:858-862, 2001

103. Samlowski WE, Watson GA, Wang M, Rao G, KlimoP, Jr., Boucher K, et al: Multimodality treatment ofmelanoma brain metastases incorporating stereotacticradiosurgery (SRS). Cancer 109:1855-1862, 2007

104. Sanghavi SN, Miranpuri SS, Chappell R, Buatti JM,Sneed PK, Suh JH, et al: Radiosurgery for patientswith brain metastases: a multi-institutional analysis,stratified by the RTOG recursive partitioning analysismethod. Int J Radiat Oncol Biol Phys 51:426-434,2001

105. Sansur CA, Chin LS, Ames JW, Banegura AT,Aggarwal S, Ballesteros M, et al: Gamma kniferadiosurgery for the treatment of brain metastases.Stereotact Funct Neurosurg 74:37-51, 2000

106. Sawaya R, Bindal RK, Lang FF, Abi-said D:Metastatic Brain Tumors, ed 2nd ed. New York:Churchill Livingstone, 2001

107. Sawaya R, Ligon BL, Bindal RK: Management ofmetastatic brain tumors. Ann Surg Oncol 1:169-178,1994

108. Schouten LJ, Rutten J, Huveneers HA, Twijnstra A:Incidence of brain metastases in a cohort of patientswith carcinoma of the breast, colon, kidney, and lungand melanoma. Cancer 94:2698-2705, 2002

109. Serizawa T, Ono J, Iichi T, Matsuda S, Sato M, OdakiM, et al: Gamma knife radiosurgery for metastaticbrain tumors from lung cancer: a comparison betweensmall cell and non-small cell carcinoma. JNeurosurg 97:484-488, 2002

110. Shapiro WR, Mehta MP, Patchell RA: Motexafingadolinium (MGd) combined with whole brainradiation therapy prolongs time to neurologicprogression in non-small cell lung cancer (NSCLC)patients with brain metastases: pooled analysis of tworandomized phase 3 trials. Neurooncology 8:489,2006

111. Shaw E, Scott C, Souhami L, Dinapoli R, Kline R,Loeffler J, et al: Single dose radiosurgical treatmentof recurrent previously irradiated primary braintumors and brain metastases: final report of RTOGprotocol 90-05. Int J Radiat Oncol Biol Phys47:291-298, 2000

112. Sheehan JP, Sun MH, Kondziolka D, Flickinger J,Lunsford LD: Radiosurgery for non-small cell lungcarcinoma metastatic to the brain: long-termoutcomes and prognostic factors influencing patientsurvival time and local tumor control. J Neurosurg97:1276-1281, 2002

113. Shehata MK, Young B, Reid B, Patchell RA, St ClairW, Sims J, et al: Stereotactic radiosurgery of 468

22

brain metastases < or =2 cm: implications for SRSdose and whole brain radiation therapy. Int J RadiatOncol Biol Phys 59:87-93, 2004

114. Siker ML, Mehta MP: Resection versus radiosurgeryfor patients with brain metastases. Future Oncol3:95-102, 2007

115. Simonova G, Liscak R, Novotny J Jr., Novotny J:Solitary brain metastases treated with the Leksellgamma knife: prognostic factors for patients.Radiother Oncol 57:207-213, 2000

116. Smalley SR, Laws ER Jr., O’Fallon JR, Shaw EG,Schray MF: Resection for solitary brain metastasis.Role of adjuvant radiation and prognostic variablesin 229 patients. J Neurosurg 77:531-540, 1992

117. Sneed PK, Lamborn KR, Forstner JM, McDermottMW, Chang S, Park E, et al: Radiosurgery for brainmetastases: is whole brain radiotherapy necessary?Int J Radiat Oncol Biol Phys 43:549-558, 1999

118. Sneed PK, Suh JH, Goetsch SJ, Sanghavi SN,Chappell R, Buatti JM, et al: A multi-institutionalreview of radiosurgery alone vs. radiosurgery withwhole brain radiotherapy as the initial managementof brain metastases. Int J Radiat Oncol Biol Phys53:519-526, 2002

119. Soffietti R, Ruda R, Mutani R: Management of brainmetastases. J Neurol 249:1357-1369, 2002

120. Sturm V, Kober B, Hover KH, Schlegel W, BoeseckeR, Pastyr O, et al: Stereotactic percutaneous singledose irradiation of brain metastases with a linearaccelerator. Int J Radiat Oncol Biol Phys 13:279-282, 1987

121. Szeifert GT, Massager N, DeVriendt D, David P, DeSmedt F, Rorive S, et al: Observations of intracranialneoplasms treated with gamma knife radiosurgery. JNeurosurg 97:623-626, 2002

122. Tajran D, Berek JS: Surgical resection of solitarybrain metastasis from cervical cancer. Int J GynecolCancer 13:368-370, 2003

123. van de Pol M, van Aalst VC, Wilmink JT, TwijnstraA: Brain metastases from an unknown primarytumour: which diagnostic procedures are indicated?J Neurol Neurosurg Psychiatry 61:321-323, 1996

124. Varlotto JM, Flickinger JC, Niranjan A, BhatnagarAK, Kondziolka D, Lunsford LD: Analysis of tumorcontrol and toxicity in patients who have survived atleast one year after radiosurgery for brain metastases.Int J Radiat Oncol Biol Phys 57:452-464, 2003

125. Vecht CJ, Haaxma-Reiche H, Noordijk EM, PadbergGW, Voormolen JH, Hoekstra FH, et al: Treatmentof single brain metastasis: radiotherapy alone orcombined with neurosurgery? Ann Neurol 33:583-590, 1993

126. Verger E, Gil M, Yaya R, Vinolas N, Villa S, Pujol T,et al: Temozolomide and concomitant whole brainradiotherapy in patients with brain metastases: aphase II randomized trial. Int J Radiat Oncol BiolPhys 61:185-191, 2005

127. Walker AE, Robins M, Weinfeld FD: Epidemiologyof brain tumors: the national survey of intracranialneoplasms. Neurology 35:219-226, 1985

128. Wang LG, Guo Y, Zhang X, Song SJ, Xia JL, FanFY, et al: Brain metastasis: experience of the Xi-Jinghospital. Stereotact Funct Neurosurg 78:70-83,2002

129. Weiss L, Grundmann E, Torhorst J, Hartveit F,Moberg I, Eder M, et al: Haematogenous metastaticpatterns in colonic carcinoma: an analysis of 1541necropsies. J Pathol 150:195-203, 1986

130. Wen PY, Loeffler JS: Management of brainmetastases. Oncology (Williston Park) 13:941-954,957-961; discussion 961-962, 969, 1999

131. Wronski M, Arbit E: Resection of brain metastasesfrom colorectal carcinoma in 73 patients. Cancer85:1677-1685, 1999

132. Wronski M, Arbit E, Burt M, Galicich JH: Survivalafter surgical treatment of brain metastases from lungcancer: a follow-up study of 231 patients treatedbetween 1976 and 1991. J Neurosurg 83:605-616,1995

133. Yamamoto M: Radiosurgery for metastatic braintumors. Prog Neurol Surg 20:106-128, 2007

134. Yamamoto M, Ide M, Nishio S, Urakawa Y: GammaKnife radiosurgery for numerous brain metastases:is this a safe treatment? Int J Radiat Oncol BiolPhys 53:1279-1283, 2002

135. York JE, Stringer J, Ajani JA, Wildrick DM,Gokaslan ZL: Gastric cancer and metastasis to thebrain. Ann Surg Oncol 6:771-776, 1999

COMPLETE SUMMARY

TITLE:Stereotactic Radiosurgery for Patients with Metastatic BrainTumors

RELEASE DATE:May 2008

DEVELOPER AND FUNDING SOURCE:International RadioSurgery Association (IRSA)

DEVELOPER COMMENT:International RadioSurgery Association is an independententity dedicated to promoting the development ofscientifically relevant practice guidelines for stereotacticradiosurgery. IRSA is a professional organization that worksto educate and provide support for physicians, hospitals,insurers and patients.

COMMITTEE:The IRSA Medical Advisory Board Guidelines Committeeand representatives in the industry.

GROUP COMPOSITION:The radiosurgery guidelines group is comprised ofneurosurgeons, neuro-oncologists, radiation and medicaloncologists and physicists.

Names of Group Members: Ajay Niranjan, M.B.B.S.,M.Ch., Neurosurgeon, Chair; L. Dade Lunsford, M.D.,Neurosurgeon; Richard L. Weiner, M.D., Neurosurgeon; GailL. Rosseau, M.D., Neurosurgeon; Gene H. Barnett, M.D.,F.A.C.S., Neurosurgeon; Massaki Yamamoto, M.D.

23

Neurosurgeon; Lawrence S. Chin, M.D., F.A.C.S.,Neurosurgeon; Paul J. Miller, M.D., Radiation Oncologist;Andrew E. Sloan, M.D., Neurosurgeon; Burton L. Speiser,M.D., Radiation Oncologist; Sandra S. Vermeulen, M.D.,Radiation Oncologist; Harish Thakrar, M.D., RadiationOncologist; Frank Lieberman, M.D., Neuro-Oncologist;David Schiff, M.D., Neuro-Oncologist; Sammie R. Coy,Ph.D., Medical Physicist; Tonya K. Ledbetter, M.S., M.F.S.,Editor; Rebecca L. Emerick, M.S., M.B.A., C.P.A., exofficio.

DISEASE/CONDITION:Brain metastases

NUMBER OF REFERENCES:135

CATEGORY:Treatment, proposed management, radiosurgery

CLINICAL SPECIALTY:Neurological surgeryNeurologyMedical oncologyRadiation oncology

INTENDED USERS:PhysiciansHealth Care ProvidersHospitalsManaged Care OrganizationsMedical PhysicistsNursesUtilization Management

OBJECTIVES:To develop an evidence and consensus-based stereotacticradiosurgery practice guideline for radiosurgery treatmentrecommendations to be used by medical and public healthprofessionals following the diagnosis of brain metastaticdisease.

TARGET POPULATION:Patients diagnosed with metastatic brain disease.

INTERVENTIONS AND PRACTICES:The management options for brain metastatic disease includecareful serial observation, surgical removal, radiosurgery,whole-brain radiation therapy and chemotherapy.Stereotactic radiosurgery is the preferred managementapproach for patients with small to moderate tumor size.

Follow-up after stereotactic radiosurgery of the brainmetastases is performed using the following schedules:

• Assess MRI scans for tumor response every 2–3months for one year, then every 4–6 months.

OUTCOMES CONSIDERED:Tumor growth control, overall survival, new tumors,functional improvement, adverse events, quality of life andoverall patient satisfaction.

METHODS TO COLLECT EVIDENCE:Hand searches of published literature (primary sources); handsearches of published literature (secondary sources); searchesof electronic databases

DESCRIPTION OF METHODS TO COLLECTEVIDENCE:MEDLINE and PUBMED searches were completed for theyears 1966 to May 2008. Search terms included: brainmetastases, metastatic brain tumor, stereotactic radiosurgery,Gamma Knife,® linear accelerator, irradiation, clinical trials,research design, practice guidelines and meta-analysis.Bibliographies from recent published reviews were reviewedand relevant articles were retrieved.

METHODS TO ASSESS THE QUALITY ANDSTRENGTH OF THE EVIDENCE:Expert consensus (committee)

METHODS TO ANALYZE EVIDENCE:Review of published meta-analysis

REVIEW METHODS:External peer review; internal peer review

DESCRIPTION OF REVIEW METHODS:The recommendations were a synthesis of research obtainedin the evidence gathering process by a core group of twomembers (AN and LDL). These recommendations weremailed to all committee members. Feedback was obtainedthrough this mailed survey consisting of proposed guidelinesasking for comments on the guidelines and whether therecommendation should serve as a practice guideline. Nosignificant disagreements existed. The final statementincorporates all relevant evidence obtained by the literaturesearch in conjunction with the final consensusrecommendations supported by all working group members.

MAJOR RECOMMENDATIONS:• Patients with brain metastases, defined by modern

neurodiagnostic imaging (CT, MRI scan) constitutethe study group. Such patients typically present withseizures or symptoms of mass effects such as headache,nausea, vomiting, weakness, numbness of limbs orspeech problems. Many patients’ tumors are detecteddue to MRI surveillance before they develop anysymptoms.

• Stereotactic radiosurgery is a minimally invasive,single session, high-dose, closed skull strategy that maybe especially suitable for patients who have limitedmetastatic brain disease and have controlled systemicdisease with good functional status.

24

• Stereotactic radiosurgery is typically employed aloneor as a boost after WBRT for patients with metastaticbrain tumors.

• A high resolution double dose contrast-enhanced MRimaging is usually necessary to determine the numberof metastatic tumors. For radiosurgery dose planning,double dose contrast-enhanced volumetric gradientrecalled MR stereotactic images are ideal.

• Current radiation delivery technologies for volumetricstereotactic conformal single session radiosurgeryinclude Gamma Knife,® proton beam using Bragg peakeffect, and specially modified or dedicated linearaccelerators like Novalis Tx™ and Axesse.™

• The optimal dose range for volumetric conformalstereotactic brain metastases radiosurgery has beenlargely established based on tumor anatomy (proximityto eloquent brain regions), tumor volume, priorradiation therapy and estimated adverse radiation risks.Minimum doses to the margin typically range from14–24 Gy in a single session.

• Patients may receive a single stress dose ofcorticosteroids at the conclusion of the radiosurgeryprocedure. Patients can continue to take othermedications (antiseizure or antiedema drugs) asrecommended by their physicians.

• Post-radiosurgical clinical examinations and MRstudies are requested by referring physicians at 2–3month intervals or earlier if the patient develops a newsymptom suggestive of a new tumor, brain edema orhemorrhage.

• Patients with large tumors causing symptomatic masseffect may need surgical decompression of the tumor.Residual tumor or tumor bed can be treated byradiosurgery or radiation therapy.

• Causes for local failure of stereotactic radiosurgeryinclude inadequate visualization of the tumor, lack ofintraoperative stereotactic 3-D (volumetric) imaging,new metastatic deposits and insufficient dose (due tolarge tumor volume and proximity to eloquent brainlocations) to achieve the growth control response.

TYPE OF EVIDENCE:Types I, II and III evidence exist in support of stereotacticradiosurgery for brain metastases.

POTENTIAL BENEFITS:Minimally invasive approachHigh rates of tumor growth control (80–90%)

SUBGROUP(S) MOST LIKELY TO BENEFIT:Patients diagnosed with small to medium size brainmetastases without symptomatic brain compression. Patientswith residual or recurrent brain metastases after resection.Patients with residual or recurrent brain metastases afterWBRT.

POTENTIAL HARMS:Major adverse effects of radiosurgery are based on location,volume, and dose, and these risks can be estimated based onpublished data and experience. Individual risks are relatedto the anatomic location of tumors.

SUBGROUP(S) LIKELY TO BE HARMED:Patients with large volume tumors causing symptomatic masseffect on the brain.

GUIDELINE STATUS:This is the full current release of the guideline.

GUIDELINE AVAILABILITY:Electronic copies: available in Portable Document Format(PDF) from www.IRSA.org/guidelines.html

PATIENT RESOURCES:Patient resources are available online at www.IRSA.org, bye-mail at office1@IRSA.org, or by calling +717-260-9808.

See “publications” for patient resources for metastaticbrain tumors: www.IRSA.org/publications.htmlBrain Talk® Volume 5, No. 2

COPYRIGHT STATEMENT:Copyright IRSA 2008