Quality Assurance In Stereotactic Radiosurgery And .

Quality Assurance in StereotacticRadiosurgery and Fractionated StereotacticRadiotherapyDavid Shepard,Shepard Ph.D.Ph DSwedish Cancer InstituteSeattle, WATimothy D. Solberg, Ph.D.University of Texas Southwestern Medical CenterDallas, TX

Quality Assurance in Linac SRS/SBRTOutline MechanicalM h i l aspects Linac Frames Beam data acquisitiong of TP systemy Commissioning End-to-end evaluation Imaging and Image Fusion FramelesslRadiosurgerydi References and Guidelines

Can we hit the target?C we putt thCanthe dosedwherehwe wantt it?

How accurate is radiosurgery?Stereotactic Radiosurgery, AAPM Report No. 54, 1995Other sources:MRI DistortionImage FusionRelocatable framesDosimetric

Frames & CTMaciunas et al, Neurosurgery 35:682-695, 1995

Isocentric Accuracy: TheWinston-Lutz Test

Isocentric AccuracyIs the projection of theball centered withinthe field?

Isocentric Accuracy:The Winston-LutzTestIs the projection of theball centered withinthe field?Good results 0.5 mm

A daily Lutz test is extremelyimportant because: It verifies the accuracy ofyour lasers The mechanical isocentercan shift over time The AMC board in Variancouches can fail The cone mount or MLC maynot be repositioned perfectlyafter serviceAfterBefore

A Lutz test with the MLC is alsoimportant because: The ConeCone-basedbased Lutz test does nottell you anything about themechanical isocenter of the MLC The MLC may not be repositionedperfectly after serviceLutz test with 12 x 12 mm2 MLC field

End-to-End Localization Accuracy

End-to-End Localization Accuracy(Surely(y myy vendor has checked this))Assess eeverythinger thing yourselfo rself

End-to-end localization evaluation

StructureC li dCylinderCubeConeSphereAP000.020.0-35.025.0Phantom SpecificationsLATVERT000.030 030.0-17.040.0-20.040.020.032.7iPlan Stereotactic CoordinatesAPLATVERT101.0040.430 830.820.8-17.142.4-34.6-19.740.825.520.233.5

Verification of MLCshapes and isocenter

System AccuracySimulate the entireprocedures: Scan,target, plan, deliverPhantom withfil holderfilmh ldPin denotesisocenter

Resulting film provides measure oftargeting accuracy Offset fromintended target

as well as falloff for a multiple arc delivery10.80.60.4020.20-1-0.500.5Off Ax is Distance (m m )1

RPC PhantomHidden Targetg Testscan,, pplan,,localize, assessLucy PhantomCourtesy Sam Hancock, PhD, Southeast Missouri Hospital

Imaging Uncertainties Use CT for geometric accuracy Use MR for target delineation“MRI contains distortions whichimpede direct correlation with CTdata at the level required for SRSSRS”Stereotactic Radiosurgery – AAPM Report No. 54Other ReferencesTS Sumanaweera,SJR Adler,AdlS Napel,Nl ett al.,l CharacterizationCht i tioffspatial distortion in magnet resonance imaging and its implicationsfor stereotactic surgery,” Neurosurgery 35: 696-704, 1994.

1.8 0.5 mm shift ofMR images relative toCT andd deliveredd lid dosedShifts occur in theffrequencyencodingdidirectionDue to susceptibilityartifacts between thephantom and fiducialmarkers of theLeksell localizationboxY Watanabe, GM Perera, RB Mooij, “Image distortion in MRI-basedpolymer gel dosimetry of Gamma Knife stereotactic radiosurgerysystems,” Med. Phys. 29: 797-802, 2001.

Frequency Encoding L/RFrequency Encoding A/PY Watanabe, GM Perera, RB Mooij, “Image distortion in MRI-basedpolymer gel dosimetry of Gamma Knife stereotactic radiosurgerysystems,” Med. Phys. 29: 797-802, 2001.

What do we do about MR spatial distortion?Use Image Fusion

Fusion Verification

MR Fusion – Lucy Phantom

Beam Data Acquisitionand DosimetryStandard DiodeSmall field measurementscan be challenging; Diodesand small ion chambersare well suited to SRSdosimetry, but theircharacteristics / responsemust be well understood.Stereotactic DiodePinpoint Chamber 0.015 cc

PTW DetectorsPinPoint0.015 ccCylindrical or SphericalMicroLion0.002 ccLiquid-filledi id fill d800 VDiode1 mm x 2.5 µmDiamond1-6 mm x 0.3 mm100 V

A140.016 ccA160.007 ccA14SL0.016 ccExradin Detectors

Wellhofer DetectorsCC040.04 ccCC010.01 cc

Diode Warnings!!!1) DiDioded response willill ddriftift over titimeRe-measure reference between each chance infield size2) Diodes exhibit enough energy dependence thatratiostibetweenb tlargelandd smallll fieldfi ld measurementstare inaccurate at the level required for radiosurgeryUse an intermediate reference field appropriate forboth diodes and ion chambers

Reference diode output to an intermediatefield sizeOutputFactorNO!YES! Reading (FS)diodeReading (Ref’)diodeXReading (Ref’)ICReading (Ref)ICReading (6 mm)diodeReading (100 mm)diodeReading (6 mm)diodedi dReading (24mm)diodeXReading (24 mm)ICReading (100 mm)IC

Radiosurgery beams exhibit a sharp decreasein output with decreasing field sizeSignificantuncertaintyDon’tt use high energyDonThis means that with small collimators,treatment times can be long

6X Output Factors – Circular ConesRelative Outpuut Factor10.90.80.70.6010.100246810Field Diameter (mm)121416

Circular Cones – Original NovalisInstitution 1Institution 2Circular Cones –Novalis TxInstitution 1Institution 2Institution 3Institution 3

HD-120Institution 1Institution 2

Small field depth dose show familiar trends

Novalis Tx /HD-120 XLow Standard ModeInstitution 1Institution 2

Off Axis Profiles – Cones

Penumbra: Cones versus MMLC3.5ConesMMLCPenumbbra, 80%-220% (mm)32.521.510.500204060Nominal Field Size (mm)80100

Need proof that beam dataacquisition for smallfields is difficult?Surveyed Beam Data from 40 identical radiosurgery units:Percent Depth DoseRelative Scatter FactorsAbsolute Dose-to-Monitor Unit CFReference ConditionApplied statistical methods to compare data

Relative Output Factor: 6 mm x 6 mm MLCOutpput Facctor 45%Institution

Even institutions in the U.S have difficultyCone size(mm)Original edOutput 0

A different institution in the U.S1.2Institution A1Institution B0.80.60.40.20050100150Depth (mm)200250300

And still another institution in the U.S11010090Institution AInstitution BPerceent Depth Dosse806.0 %7010.3 %6050403020100050100150200250Depth (mm)300350400450

Commissioning your system: Does calculationagree with measurement?

Phantom Plans

End-to-end testingDosimetric uncertaintyCalculationCalculationarc-step 10oarc-step 2oRelative Dosimetry

Start simple, and increase complexity1 isocenter4 field boxDDynamici ConformalC fl ArcsA2 isocenters – off axis2 isocenters – on axisIMRT

End-to-end testingDosimetricDoi et iuncertaintyAbsolute Dosimetry

Independent MUCalculations

End-to-end dosimetricevaluation

Absolute DosimetryLucy Phantom

Relative DosimetryLucy Phantom

Relative Dosimetryy PhantomLucyContours defined on MR

What about “Frameless Systems?”A “frameless” stereotactic system provideslocalization accuracy consistent with the safedelivery of a therapeutic dose of radiationgiven in one or few fractions, without the aidof an external reference frame, and in amanner that is non-invasive.Frameless stereotaxis is inherently image guidedAlAlsorequired:i dImmobilization – need not be linked to localizationAbility to periodically monitor / verify

(Stereo)photogrammetry the principle behindfframelesslttechnologiesh l iPhotogrammetry is ameasurement technologyin which the threedimensional coordinatesof points on an objectare determined bymeasurementst maded initwo or morephotographic imagest ktakenffrom differentdifftpositions

Stereophotogrammetry in RadiotherapySpatial Resolution: 0.05 mmTemporal Resolution: 00.0303 sLocalization Accuracy:OpticalPhotogrammetry0.2 mm

Stereophotogrammetry in RadiotherapyBova et al, IJROBP, 1999

Frameless Radiosurgery (X-ray Stereophotogrammetry)

How do we know the system is targeting properly?End-to-end evaluation that mimics a patient procedureX-rayIdentify target & planDRRSet upp in treatment roomIrradiateEvaluate

Results of Phantom Data((mm))L tLat.LLong.V tVert.3D .560.320.820.42 Sample size 50 trials (justified to95% confidence level,, /- 0.12mm))

Comparison in 35 SRS patientsand 565 SRT fractions

Difference Between conventional and frameless localization1.2Multiple FractionSingle FractionSuperior / Inferior10.8060.60.41.01 0.54 mm2.36 1.32 mm0.20-8.0-6.0-4.0-2.00.02.04.06.0 Frameless localization is equivalent to frame-based rigid fixation Frameless localization improves accuracy of relocatable frames8.0

End-to-end evaluation:Extracranial3D error 1.2 0.4 mm

End-to-end evaluation: CyberKnife3D error 1.1 0.3 mmChang et al, Neurosurgery 2003

Localization usingimplanted fiducialsCourtesy Sam Hancock, PhD, Southeast Missouri Hospital

Localization using implanted fiducialsCourtesy Sam Hancock, PhD, Southeast Missouri Hospital

Radiosurgery Guidelines ACR / ASTRO Practice Guidelines Whath ddo theyhcover?Personnel Qualifications / ResponsibilitiesProcedure SpecificationsQuality Control / Verification / ValidationFollow-upp

Radiosurgery Guidelines Task Group ReportsAAPM Report #54 – Stereotactic RadiosurgeryAAPM Report #91 – The Management of Motion in Radiation Oncology(TG 76)TG 68 – Intracranial Stereotactic Positioning SystemsTG 101 – Stereotactic Body RadiotherapyTG 104 – kV Localization in TherapyTG 117 – Use of MRI in Treatment Planning and Stereotactic ProceduresTG 132 – UseUoff IImage RegistrationR i t tianddDDatat FFusioniAlgorithmsAlithanddTechniques in Radiotherapy Treatment PlanningTG 135 – QA for Robotic RadiosurgeryTG 147 – QA for Non-Radiographic Radiotherapy Localization andPositioning SystemsTG 155 – Small Fields and Non-Equilibrium Condition Photon BeamDosimetryTG 176 – Task Group on Dosimetric Effects of Immobilization DevicesTG 178 – Gamma Stereotactic Radiosurgery Dosimetry and QATG 179 – QA for Image-Guided Radiation Therapy Utilizing CT-BasedTechnologies

Radiosurgery Guidelines RTOG ProtocolsRTOG 9005 – Single Dose Radiosurgical Treatment of Recurrent PreviouslyIIrradiateddi t d PrimaryP iBrainB i TTumors andd BrainB i MetastasesM t tRTOG 9305 – Randomized Prospective Comparison Of StereotacticRadiosurgery (SRS) Followed By Conventional Radiotherapy (RT) WithBCNU TTo RT With BCNU AlAlone FFor SSelectedl t dPPatientsti t With SSupratentorialt t i lGlioblastoma Multiforme (GBM)RTOG 9508 – A Phase III Trial Comparing Whole Brain Irradiation AloneVVersusWholeWh l BrainB i IrradiationIdi tiPlusPlStereotacticStt ti RadiosurgeryR diforfPatients with Two or Three Unresected Brain MetastasesRTOG 0236 – A phase II trial of SBRT in the treatment of patients withmedicallydi ll inoperableibl stagetI/II non-smallll cellll lunglcancerRTOG 0618 – A phase II trial of SBRT in the treatment of patients withoperable stage I/II non-small cell lung cancerRTOG 0813 – Seamless phase I/II study of SBRT for early stage, centrallylocated, non-small cell lung cancer in medically operable patients

Other DocumentsASTRO/AANS Consensus Statement on stereotacticradiosurgery quality improvement, 1993RTOG Radiosurgery QA Guidelines, 1993European Quality Assurance Program on StereotacticRadiosurgery, 1995DIN 68756875-11 (Germany) Quality Assurance inStereotactic Radiosurgery/Radiotherapy, 2004 and read the literature andd ttalklk withith your colleaguesll

Quality Assurance in Stereotactic Radiosurgery and Fractionated Stereotactic Radiotherapy David Shepard Ph DDavid Shepard, Ph.D. Swedish Cancer Institute Seattle, WA Timothy D. Solberg, Ph.D. U