SRS SRT Dosimetry - Sonja Dieterich PDF
SRS SRT Dosimetry - Sonja Dieterich PDF
SRS SRT Dosimetry - Sonja Dieterich PDF
StanfordUniversityHospital
EllenWilcox,PhD
StFrancisHospital,CT
CarloCavedon,D.S.
Ospedale SanBartolo,Vicenza,Italy
WithslidesfromRemyDurand,Huaying Ji,JosephNovotny
ConflictofInterest
SonjaDieterich hasaconsultingagreementwith
Cyberheart Inc.
LearningObjectives
Describethebasicconceptsofsmallfieldreference
dosimetry
Knowwhichdetectorsareappropriatetouse
Learntherightquestionstoaskwhenyouseeaplan
Beawareofwholebodydose
0.3mm
Closertoclinicalreality:Endtoend(E2E)
ManufacturerE2Espec:<0.95mm
Myclinic:0.3mm 0.7mm
IstarttoinvestigateifE2E>0.7mm
OlderpaperbyYuetal (E2E
(E2E=1.1mm)
1.1mm)
had2mmslice thickness
Knowndependencyofspatialaccuracyvs.slicethickness
1mmor1.25mmiscurrentlystandardslicethickness
Thinkofdosimetricaccuracyas
y
shiftingisodoselines
WhyisDosimetryimportant?
Inthebrain:
PreSRSAVMembolizationwithOnyxorglue
Nearcavitiesorsuperficialtumors
SRShasmovedoutsidethebrain:
Lung
L
Tspine
OurSmallFieldsaregettingsmaller
WhatistheWBdosefor85(!)lesions?
T bl f C
TableofContents
ReferenceDosimetry
Whatisasmallfield?
ThenewIAEAframeworkforSRSfields
Relative(Patient)Dosimetry
Dosecalculationalgorithms
Invivodosimetry
v vo dos et y
Wholebodydose
WhatisaSmallField?
Smallchangeswithtime:
Nolaterale equilibrium(quick:whatisthatfor6MV?)
<10mm
<4mm?
Challenges:
Ch ll
Howtobuildacollimator
Focalspotsizelargecomparedtocollimator
Inversesquarelawbreakingdown
Detectorresolution
WhyMicrobeams?
Functionaltargets:
Epilepsy,Parkinsons(fornon
DBScandidates))
Neurology:Facetblocks,
hyperhydrosis
Intractablepain(palliative,
capsulotomy)
ObsessiveCompulsiveDisorder
Intractable,severedepression
Intractable severedepression
(fMRI:hyperactivityofBrodmann
25)
Smallanimalapplications
R f
d
f
d d fi ld
Referencedosefornonstandardfields
TG51presentations: Musthave10cmx10cmfield
Gammaknife?Cyberknife?Tomotherapy?
y
py
Thesefieldsarenotflatoverdetectorvolume
IAEAandAAPMtaskgroupsonsmallfielddosimetry
(TG51)60mmconeforCK
Standardbeamdataacquisition(TPR,OCR)forCK:
80cmSAD
Howtoobtain100cmSSD?
Easy:moverobotup(byhowmuch?)
HowtoobtainkQ?
Noflatteningfilter
N fl tt i filt noproblem(seeDWORogers)
bl ( DWOR
)
Howtohandlegradientcorrections?
TG51output=1.0153cGy/MUatdmax
TG21output=1.0162cGy/MUatd
p
y/
y
max sameday
15
EffectsofDetectorSizeon
ReferenceDosimetry
1.2
0.8
plan
0.6
BeamData
0.4
0.2
0
30
-20
-10
10
20
30
PracticalProcedure(1)
MoverobotuptoextendSSDto100cm.
Eqv
q square=0.9
q
9 Eqv
q circle*
Eqv square=0.9 6 100/80=6.75cm@100cmSSD
Measure%dd(10,6.75,100)withCyberKnife.
%dd(
)
h b
f
Compare%dd(10,6.75,100)withlocalorreferencedata
(e.g.BJRsup25
(e
g BJRsup25)
*
PracticalProcedure(2)
Takethecorresponding%dd(10,10,100)fromthe
referencedataset.
ObtainkQ with%dd(10,10,100).
ObtainPgr(10,10,100)fromreferencedata.
(10 10 100)fromreferencedata
ChangecalibrationconditiontoCyberKnife calibration
condition(SSDsetuptoSADsetup).
condition(SSDsetuptoSADsetup)
18
PracticalProcedure(3)
PerformmeasurementbaseduponstandardTG51
procedure.
MeasurePgr(CK)atthemeasurementpoint.
(CK)atthemeasurementpoint
Calculatedosetodmax:
DwQ
M N DQ, w
%dd
or TPR
M M raw P
Q
D,w
Co 60
D,w
kQ
Pgr (CK )
Pgr (10,10,100)
19
Example
1.
2.
3
3.
4.
5.
6.
7.
8.
9.
100cmSSD60mmcone,%dd(10,6.75,100)=64.02%.
SSD6
%dd( 6
) 6
%
%dd(10,10,100)=65.4%*.
Withthis%dd(10 10 100) kQ=0.9928.
Withthis%dd(10,10,100),k
=0 9928
Pgr(10,10,100)=0.9897*.
78.5cmSSD,60mmcone,%dd(10,5.3,78.5)=59.43%.
78.5
c SS , 60
co e, %dd( 0,5.3,78.5) 59.43%.
Pgr(10,5.3,78.5)=0.9878
CyberKnife newkQ =0.9909
TG51output=1.0153cGy/MUatdmax
TG21output=1.0162cGy/MUatdmax sameday
* Calculated from BJR data.
20
Measuringeachfieldnotpossible
EvenwithPerfexion,8segments
Planclass:Isocentric,allbeams,defined
,
,
collimator
GKPlanClassSpecificReference
p
Calibration
VerificationofDoseCalculation
SuitableDetectors?
Plus:
Plus
BANGgels
Diamond
OLD
UncertaintiesforOutputFactors
Errorbarsgettinglarger
below20mm
MCofbeam
MC f b
MCofdetectors
Fi ld
D
Si
Fieldvs.DetectorSize
1.000
1.5cm
1 5cm depth
0.800
0.600
Widthofa
Wid h f
5c
7.5c
10c
12.5c
15c
site 5c
site 7.5c
site 10c
site 12.5c
site 15c
0.400
0.200
0 000
0.000
0.0
2.0
4.0
6.0
8.0
10.0
12.0
14.0
PTW60012diode
comparedtosmall
collimatorOARs
Al ith i ( ld ) C i l SRS
Algorithmsin(older)CranialSRS
y
Aheadisverysimilartoa
sphere(surface
corrections!)
Fairlyhomogeneous
Nasopharynx
AVMembolizations
Densitycanbe
approximatedbyH
i t db H2O
(hencenoCTusedon
GammaKnife)
NeedforbetteralgorithmsinSRS
Mostsystemsusedpathlengthcorrection
Narrowbeams:
fielddimensionssmallerthenmaximumrangeof
secondaryelectrons
Steepdosegradients
Getsexacerbatedbytissueheterogeneity
Betteralternative:
Collapsedconeconvolutionsuperposition
MonteCarlo
Example:
Example:
6MVPhotonMonteCarlo
UseMCinalllungcases,Tspine,Head&Neck
Brain:supraorbital,pituitary,<1.5cmtoskin,embolized
AVM
Recalculationorreoptimization
Older(e.g.Raytracing)algorithms:
+8%14%offforRPClungphantom(RTOG0236)
Dosedifferencemaybemuchhigherforsmalllesions
D diff
b
hhi h f
lll i
( t %!)
(upto40%!)
Dosedifferenceofvarying%knownissueforall olderTx planning
algorithmsincombinationwithsmallbeams,not limitedto
C b k if
Cyberknife
ExampleI:DosimetryforSBRT
a p e : os et y o S
ExampleII:DosimetryforSBRT
a p e : os et y o S
Independent verification of MC
IndependentverificationofMC
RecentlygotapprovedforRTOG0618
RTOG0618=RTOG0236+inhomogeneity
3
g
y
corrections
Doneonamotionplatform
June 11, 2009
WhyInVivoDosimetry?
Why?
Challenges?
FramelessSRS
Fieldsize:noexitfieldfor
di d l
diodeplacement
SBRT:
Imaging:NospaceforEPID
Gating/ABC
(GammaKnife)
Motion
Motionadaptivewith
adaptivewith
Synchrony
Nonisocentricity
(Cyberknife)
Motionadaptivewithmoving
MLC
C
Fieldmatchingfor
retreatments
Justoneideaforinvivo
Antennaisverysimilartoagoldseed
Invasiveproceduretoimplantmarker
Implantable Dosimeter
11.25
25 mm CT slice thickness
Looks like two closely placed
fiducials!
Not uncommon in patients
01/26/2007
39
Whyisitimportant?
Wearetreatingbenign&functionalcases
Longsurvival
Pediatrics(protons!)
MUperdelivereddose
Theissues:
Whatistheerrorbaronsecondarycancerrisk?
Absolutevs.relativerisk
Absolutevs relativerisk
Delayedradiationresponse
Riskoftreatingvs.notreatment/othertreatments
Differenceinrisktolerance(e.g.RadOnc
Diff
i i k l
( R dO vs.Neurosurgeon)
N
)