Margins and margin recipes
Marcel van Herk
On behalf of the image guidance group
The Netherlands Cancer Institute Amsterdam, the Netherlands
Classic radiotherapy procedure Tattoo, align and scan patient
Draw target and plan treatment on RTP
Align patient on machine on tattoos and treat (many days)
In principle this procedure should be accurate but …
Things move: geometrical uncertainties
In the past large safety margins had to be used
Baseline shift: largest error in lung RT Organ motion: largest error in prostate RT
Example IGRT system: Elekta Synergy
• 1997: proposed by David Jaffray and John Wong
• 2004: prototype in clinical use at NKI
• 2005: Released for clinical use worldwide
• 6 at NKI, more than 500 world-wide
Over 100.000 scans made at NKI – 200 GByte scans per week
With such a system, this is no longer needed to precisely irradiate a brain tumor
We can use this instead: focus on patient stability, but let computer position the
patient with better than one mm precision
v Beek et al, in preparation
Accuracy registration: 0.1 mm SD Accuracy table: 0.5 mm SD {x, y, z} Intra-fraction motion: 0.3 mm SD
IGRT – The good, the bad, and the ugly
• Good: IGRT gives unprecedented precision of hitting any clearly defined point in the body
• Bad: This precision may give us overconfidence in the total chain accuracy: tumors are rarely clear
• Ugly: we may have to find this out from our clinical mistakes
Nomenclature • Gross error: mistakes, transcription errors, software
faults: • must be caught by QA
• Error: difference between planned value and its true
value during treatment, however small
• Uncertainty: the fact that unpredictable errors occur – quantified by standard deviations
• Variation: the fact that predictable or periodic errors occur
EPID dosimetry QA to catch gross errors: used for all curative patients at NKI
EPID movie
Reconstructed EPID dose (VMAT case)
per frame cumulative -140° 140°
Mans et al, 2010
Precision: within few %, enough to catch gross errors
Gross errors detected in NKI
0.4% of treatments show a gross error
(>10% dose)
9 out of 17 errors would not have
been detected pre-treatment !!
Mans et al, 2010
What happens in the other 99.6% ?
• There are many small unavoidable errors (mm size) in all steps of radiotherapy • In some cases many of these small errors point in the
same direction • I.e., in some patients large (cm) errors occur(ed)
• This is not a fault, this is purely statistics
• What effect does this have on treatment?
• We do not really know!
Motion counts? Prostate trial data (1996)
Risk+: initial full rectum, later diarrhea Heemsbergen et al, IJROBP 2007
N=185 (42 risk+) N=168 (52 risk+)
The major uncertainties not solved by IGRT
• Target volume definition • GTV consistency • GTV accuracy • CTV: microscopic spread
• Inadequacy of surrogate used for IGRT
• Motion that cannot be corrected
• Too fast • Too complex
CT (T2N2)
SD 7.5 mm
CT + PET (T2N1)
SD 3.5 mm
Delineation variation: CT versus CT + PET
Steenbakkers et al, IJROBP 2005 Consistency is imperative to gather clinical evidence!
Effect of training and peer collaboration on target volume definition
teacher
students groups
Material collected during ESTRO teaching course on target volume delineation
Glioma delineation variation (Beijing 2008)
SD (mm)
SD (mm) outliers removed
Margin (mm)
Homework 3.6 2.3 5.8
Groups 1.3 1.3 3.2
Validation 2.6 2.3 5.8
Delineation uncertainty is a systematic error that should be incorporated in the margin Consistency is imperative to gather clinical evidence
Other remaining uncertainties • Is the surrogate appropriate?
2.5 cm
Motion of tumor boundary relative to bony anatomy
Are prostate markers perfect ?
Apex Base Sem. Vesicles +/-1 cm margin required
van der Wielen, IJROBP 2008 Smitsmans, IJROBP 2010
Best: combine markers with low dose CBCT
Intra-fraction motion: CBCT during VMAT
Intra-fraction motion: CBCT during VMAT
This amount of intra-fraction motion is rare for lung SBRT
Error distributionsCentral limit theorem:
the distribution of the sum of an increasing number of errors with arbitrary distribution will approach a Normal (Gaussian) distribution
Large errors happen sometimes if all or most of the small sub-errors are in the same direction
Normal distribution:
-3 0 30
200
400
600
800
1,000
1,200
1,400
mean = 0s.d. = 1N = 10000
-2..2 = 95%
Definitions (sloppy) • CTV: Clinical Target Volume
The region that needs to be treated (visible plus suspected tumor)
• PTV: Planning Target Volume The region that is given a high dose to allow for errors in the position of the CTV
• PTV margin: distance between CTV and PTV
• Don’t use ITV for external beam! (SD adds quadratically)
Time-scales for errors • Compare Xplanned with Xactual
• Xplanned – Xactual = εgroup + εpatient, group + εfraction, patient, group+ εtime, fraction, patient, group
• The appropriate average of each ε is zero
Xplanned – Xactual = Mg +/- σg +/- σp +/- σf
The nomenclature hell Proposed to ICRU Bel et al. Literature
Mg Mean group error M Mean group error
bias
(fraction)
Systematic error σg Intra-group
uncertainty Σ Inter-patient
uncertainty
σp Intra-patient uncertainty
σ Inter-fraction uncertainty
(fraction)
Random error
σf Intra-fraction uncertainty
Intra-fraction uncertainty
Analysis of uncertainties Keep the measurement sign!
mean =M
RMS = σ
SD = Σ
Intra-fraction
0.0
0.3
0.4
0.1
0.3
_________
Mean = 0.2 RMS of SD = σf
patient 1 patient 2 patient 3 patient 4fraction 1 0.5 0.0 0.2 0.7fraction 2 0.6 -0.5 0.3 0.2fraction 3 0.9 0.2 0.2 -0.4fraction 4 1.3 -1.1 0.3 -0.1
mean 0.8 -0.4 0.3 0.1sd 0.3 0.6 0.1 0.5
van Herk et al, Sem Rad Onc 2004
M = mean group error (equipment) Σ = standard deviation of the inter-patient error σ = standard deviation of the inter-fraction error σf = standard deviation of the intra-fraction motion {
Demonstration – errors in RT • Margin between CTV
and PTV: 10 mm
• Errors: • Setup error:
• 4 mm SD (x, y) • Organ motion:
• 3 mm SD (x, y) • 10 mm respiration
• Delineation error: optional
What is the effect of geometrical errors on the CTV dose ?
Treatment execution (random) errors blur the dose distribution
Preparation (systematic) errors shift the dose distribution
dose
CTV
Random: Breathing, intrafraction motion, IGRT inaccuracy
Systematic: delineation, intrafraction motion, IGRT inaccuracy
CTV
Analysis of CTV dose probability
• Blur planned dose distribution with all execution (random) errors to estimate the cumulative dose distribution
• For a given dose level:
– Find region of space where the cumulative dose exceeds the given level
– Compute probability that the CTV is in this region
Computation of the dose probability for a small CTV in 1D
x
x
..and compute the probability that the average CTV position is in this area
In the cumulative (blurred) dose, find where the dose > 95%
98%
95%
average CTV position
What should the margin be ?
0 100 minimum CTV Dose (%) 0
100
0 mm
6 mm
9 mm
12 mm
Typical prostate uncertainties with bone-based setup verification
Simplified PTV margin recipe for dose - probability
To cover the CTV for 90% of the patients with the 95% isodose (analytical solution) : PTV margin = 2.5 Σ + 0.7 σ
Σ = quadratic sum of SD of all preparation (systematic) errors σ = quadratic sum of SD of all execution (random) errors
(van Herk et al, IJROBP 47: 1121-1135, 2000)
*For a big CTV with smooth shape, penumbra 5 mm
2.5Σ + 0.7σ is a simplification • Dose gradients (‘penumbra’ = σp) very shallow in
lung smaller margins for random errors
• Number of fractions is small in hypofractionation • Residual mean of random error gives systematic error • Beam on time long respiration causes dose blurring
• If dose prescription is at 80% instead of 95%:
222 64.1)(64.15.2 ppM σσσ −++Σ=
222 84.0)(84.05.2 ppM σσσ −++Σ=
(van Herk et al, IJROBP 47: 1121-1135, 2000)
Practical examples
Prostate: 2.5 Σ + 0.7 σ
all in cm systematic errors squared random errors squared
delineation 0.25 0.0625 0 0 Rasch et al, Sem. RO 2005
organ motion 0.3 0.09 0.3 0.09 van Herk et al, IJROBP 1995
setup error 0.1 0.01 0.2 0.04 Bel et al,IJROBP 1995
intrafraction motion 0.1 0.01
total error 0.40 0.16 0.37 0.14
times 2.5 times 0.7
error margin 1.01 0.26
total error margin 1.27
Prostate: 2.5 Σ + 0.7 σ Now add IGRT
all in cm systematic errors squared random errors squared
delineation 0.25 0.0625 0 0 Rasch et al, Sem. RO 2005
organ motion 0 0 0 0 van Herk et al, IJROBP 1995
setup error 0 0 0 0 Bel et al,IJROBP 1995
intrafraction motion 0.1 0.01
total error 0.25 0.06 0.10 0.01
times 2.5 times 0.7
error margin 0.63 0.07
total error margin 0.70
Engels et al (Brussels, 2010) found 50% recurrences using 3 mm margin with marker IGRT
CNS: single fraction IGRT for brain metastasis
all in cm systematic errors squared random errors squared
delineation 0.1 0.01 0
organ motion 0 0 0
setup error 0.05 0.0025 0
intrafraction motion 0.03 0.0009
total error 0.11 0.01 0.03 0.0009
times 2.5 times 0.7
error margin 0.28 0.02
total error margin 0.30
Tightest margin achievable in EBRT ever due to very clear outline on MRI
Planning target volume concepts
GTV/ITV CTV PTV
Convention Free-breathing
CT scan Time-averaged mean position
Internal Target Volume
Motion
Gating @ exhale
Mid- Ventilation /Position
Crap Too large Margin ?
}
Image selection approaches to derive representative 3D data
4D CT
Mid-ventilation Exhale (for gating)
Vector distance to mean position (cm)
Very clear lung tumor: classic RT
all in cm systematic errors squared random errors squared
delineation 0.2 0.04 0
organ motion 0.3 0.09 0.3 0.09
setup error 0.2 0.04 0.4 0.16
Intra-fraction motion 0 0
respiration motion 0.1 0.01 0.3 0.111111 1(0.33A)
total error 0.42 0.18 0.60 0.361111
times 2.5 difficult equation(almost times 0.7)
error margin 1.06 0.41
total error margin 1.47
Using conventional fractionation, prescription at 95% isodose line in lung
Very clear lung tumor: IGRT hypo
all in cm systematic errors squared random errors squared
delineation 0.2 0.04 0
organ motion 0.1 0.01 0.1 0.01
setup error 0 0
Intra-fraction motion 0.15 0.0225 0.15 0.0225
respiration motion 0 0.7 0.444444 2(0.33A)
total error 0.27 0.07 0.69 0.476944
times 2.5 difficult equationnon-linear
error margin 0.67 0.22
total error margin 0.89
Using hypo-fractionation, prescription at 80% isodose line in lung
Planned dose distribution: hypofractionated lung treatment 3x18 Gy
Realized dose distribution with daily IGRT on tumor (no gating)
9 mm margin is adequate even with 2 cm intrafraction motion
2 cm
But what about the CTV ?
• By definition disease between the GTV and the CTV cannot be detected
• Instead, the CTV is defined by means of margin expansion of the GTV and/or anatomical boundaries
• Very little is known of margins in relation to the CTV • Very little clinical / pathology data • Models to be developed
Hard data: microscopic extensions in lung cancer
30% patients with low grade tumors (now treated with SBRT with few mm margins), have spread at 15 mm distance
Having dose there may be essential!
100%
50%
25%
Slide courtesy of Gilhuijs and Stroom, NKI
N=32
0
10
20
30
40
50
60
70
80
90
100
0 5 10 15 20 25 30 35 40 45
distance from GTV [mm]
% c
ases
with
ext
ensi
ons
Deformation corrected
Mapping of planned dose cubes to standard patient
prostate
Is dose outside the prostate related with outcome? detect disease spread in historical data
Dose differences due to: - randomization
- anatomy
- technique
Estimate pattern of spread from response to incidental dose in clinical trial data (high risk prostate patients)
Average dose no failures – average dose failures
≈ 7 Gy p = 0.02
< median (53.1 Gy)
Treatment group IV, Hospital A (n=67)
≥ median
p = 0.000
100%
0% 0 3 6 Y
80% 60%
40%
20%
- =
PSA controls PSA failures
Witte et al, IJROBP2009; Chen et al, ICCR2010
Conclusions
• We defined a margin recipe based on a given probability of covering the CTV with a given isodose line of the cumulative dose
• The margin with IGRT is dominated by delineation uncertainties
• Margins for random uncertainties and respiratory motion in lung can be very small because of the shallow dose falloff in the original plans
Conclusions
• In spite of IGRT there are still uncertainties that need to be covered by safety margins
• Important uncertainties relate to imaging and biology that are not corrected by IGRT
• Even though PTV margins are designed to cover geometrical uncertainties, they also cover microscopic disease
• Reducing margins after introducing IGRT may therefore lead to poorer outcome and should be done with utmost care (especially in higher stage disease)
Modern radiotherapy
Us
Clinical motion estimation (rigid body registration)
0% Tumor trajectory
10%
Redundant registration (10 x 9 times)
Roughly paint mask
Entire procedure takes about 1 minute: Estimates trajectory, automatically removes outliers Nijkamp et al., ICCR 2007
1 0 -1 -2
4D CT Mid V
1 2 3 4 5 6 7 8 9 10-10%
-5%
0%
5%
10%re
lativ
e di
ffere
nce
mea
n do
se G
TV
patient
mean = -0.9%
} } Conventional On-line Hypo
Mexner et al, ESTRO2007; submitted
Is full 4D planning necessary? Tumor peak-peak amplitudes range from 1.1 to 3.6 cm
The blurred mid-ventilation dose is almost the same as a full 4D dose calculation
Mid-position CT: in research 4DCT 4D
DVF
Deform 4DCT to local mean pos.
Mid-position CT
Average frames
Deformable registration
Wolthaus et al, Med Phys 2008 (in press)
Margins in lung hypo (3 x 18 Gy) Systematic Random
Delineation 2 mm SD -
Registration/couch shift 1.5 mm SD 1.5 mm SD
Intra-fraction motion 1.5 mm SD 1.5 mm SD
Total 3 mm SD 2.2 mm SD
Margin A=10 mm 7 mm + 0 mm
Margin A=20 mm 7 mm + 2 mm
222 84.0)(84.05.2 ppM σσσ −++Σ= σp ≈ 7.8 mm
Ensures 80% isodose encompasses GTV 90% of time in lung
Lung Treatment Baseline (off-line EPID, normal fractionation) Advanced (SBRT with on-line soft tissue based
IGRT)
Lung Systematic error Random error Systematic error Random error
Delineation 0.2 0.2
Baseline motion 0.3 0.3 0.1 0.1
Setup error 0.2 0.4
Intrafraction 0.15 0.15
Respiration 0.3 0.3
Total error 0.41 0.66 0.31 0.35
Margin 1.1 0.4 0.67 0.06
Total margin 1.5 0.7
56
Margins (cranio-caudal) GTV to PTV: classical
Clinical used margin: 1.5 cm cranio-caudal, 1 cm other directions
delineation 0.4 0.16 0 0.00 Steenbakkers et al, 2005
respiration 0.3 0.09 0.3 0.09 0.33 x A (1 cm); van Herk et al., 2004
baseline shift 0.3 0.09 0.3 0.09 Sonke et al, 2007
setup error 0.4 0.16 0.4 0.16 NO EPID protocol
total error 0.71 0.50 0.58 0.34 root of sum square
times 2.5 times 0.7 Simplification!
error margin 1.77 0.41 factors for random and systematic (2.5 and 0.7)
0.19 note: margin for SBRT
total error margin 2.18 sum
GTV-CTV margin 0.00
total margin 2.18 sum
57
Margins (cranio-caudal) GTV to PTV: add technology
Clinical used margin: 1.5 cm cranio-caudal, 1 cm other directions
delineation 0.4 0.16 0 0.00 Steenbakkers et al, 2005
respiration 0.3 0.09 0.3 0.09 0.33 x A (1 cm); van Herk et al., 2004
baseline shift 0.3 0.09 0.3 0.09 Sonke et al, 2007
setup error 0.4 0.16 0.4 0.16 NO EPID protocol
total error 0.71 0.50 0.58 0.34 root of sum square
times 2.5 times 0.7 Simplification!
error margin 1.77 0.41 factors for random and systematic (2.5 and 0.7)
0.19 note: margin for SBRT
total error margin 2.18 sum
GTV-CTV margin 0.00
total margin 2.18 sum
Uncertainty management: Conventional IMRT planning with margin
CTV PTV
Inverse optimization
Objective functions Poisson cell kill, EUD,
DVH points, ...
Dose distribution
90% prob. of D ≥ 95% Dprescribed
in CTV
M = 2.5Σ+0.7σ
OAR
Uncertainty management: Probabilistic biological IMRT planning without margin
CTV
Inverse optimization
Objective functions with simulated errors
⟨TCP⟩, ⟨NTCP⟩
Dose distribution
Maximum ⟨TCP⟩ for given
OAR ⟨NTCP⟩
OAR
Σ, σ
no PTV margin!
87 Gy
80 Gy
74 Gy
65 Gy
39 Gy
Points to add
Effect of geometric uncertainty on TCP (1mm = x%)
Vs effect of dosimetric unc on TCP (1% = 1%)
Kill ITV Kill gating Volume of underdosage effect Read mans paper Do not treat OAR
Radiotherapy in the past (25 years ago)
Advanced imaging: 4D PET/CT
Shows correct tumor shape and its components
Shows range of respiratory motion
Allows optimal tumor targeting taking motion into account
Main problem now: target definition
- 11 observers from 5 institutions, 22 patients - newly developed delineation software - delineation on CT + (one year later) CT+PET
Steenbakkers et al, IJROBP 2005
Consistency is good
• Our clinical goal is to irradiate a target volume – this volume should be well defined
• This does not mean that the well-defined target volume is correct!
• However, if all physicians (e.g., in a trial) agree on target volume, clinical experience allows us to learn whether it is right or not
• The alternative is pathology validation – but this can only be done in surgery patients
Pathology validation (NKI/ MAASTRO/OLVG) Gilhuijs, Stroom and Boersma
34 surgical lung cancer patients
Pre-op : CT, PET
CT 18F-FDG PET
Post-op : Pathology
Macroscopical
Microscopical Analysis : 1361 slides
Accuracy CT/PET for gross tumor
Effective diameter GTV pathology (mm)
Effe
ctiv
e di
amet
er G
TV P
ET
(mm
)
GTV PET: 42% Max SUV contours
0 20 40 60 100 80
20
40
60
80
100
0
Effective diameter GTV pathology (mm)
Effe
ctiv
e di
amet
er G
TV C
T (m
m)
0 20 40 60 100 80
20
40
60
80
100
0
Slide courtesy of Gilhuijs and Stroom, NKI
Was this patient perfectly aligned after shifting the couch with IGRT?
Multiple clipbox registration
purple = planning CT green = cone beam CT
Conventional single ROI registration (used for global patient setup)
multiple ROI (mROI) registration (local misalignments)
Clinical in NKI, not commercially available
Pitfalls of IGRT • Overconfidence in precision of delineation
• we use 5 mm margin for the clearest of tumors
• Intra-fraction motion • generally requires only small margins
• Uncorrectable errors such as deformations and large rotations
• Anatomical changes warrant adaptation – but for which patients and when?
• Too much focus on anatomy, not on treatment
Repetitive 4D CT: treatment response
Differential motion
No couch correction can solve this problem
Seeds allow low dose imaging 0.35 mm visicoils
Seeds and soft tissue (seminal vesicles) visualized in low dose 1 minute scan
0.4 cGy 3 cGy
Hypofractionated lung treatment (3 x 18 Gy)
100x real speed
Danger case: target close to spinal cord
SBRT – soft tissue guidance • Alignment to room lasers • 4D-CBCT • Registration 4D-CBCT – Planning CT
• Bone • Tumour
• Couch shift • 4D-CBCT for verification • First arc + 4D-CBCT • Second arc + 4D-CBCT
Image respiration on cone beam CT ?
2 x real time speed 3D reconstruction
Software demo of 4D lung SBRT
about 200 patients done this way in NKI – solution commercial since march 2009
Scattered Fluence
+
kV only MV scatter only
kV only
kV only vs kV+MV
kV only vs kv+MV-<MV>
Registration protocols in
XVI4.5 Protocols:
Clipbox only Mask only Dual registration
Registration tools
(for all workflows) Bone Seed Grey value 4D Grey value
Workflow
Registration Correction Overview
Incorrect deformable registration
Prior After: looks OK Only edges are correc
Simulation: effect of microscopic spread on margin requirement
Microscopic spread is partly covered by the PTV margin and beam penumbra
0 2 4 6 8 10 12 14 1
10
100
1000
10000
100000
1000000
10000000
Margin
Cell ratio GTV/Shell
Solid tumor
Tumor with microscopic spread
Decreasing TCP
(5% steps)
Iso-TCP
Planning target concepts
GTV/ITV CTV PTV
Convention Free-breathing
CT scan Time-averaged mean position
Internal Target Volume
Motion
Gating @ exhale
Mid- Ventilation
Bad Too large Margin ?
}
Modes of Tumor Regression
‘elastic’ ‘erosion’
Opportunities of IGRT • mm precision • Soft-tissue guidance • Visualize normal structures • Detect anatomical changes • Novel correction protocols • Novel fractionation schemes • Monitor during treatment (VMAT) • Daily 4D imaging • Detailed knowledge of delivered dose
4D CT (PET): less artifacts + motion data
Allows determination of correct shape, SUV, mean position and trajectory of tumor
Fused 4DCT and 4DPET: Wolthaus et al, PMB 2005
What to do with this data ? • Full 4D planning and plan optimization:
• Not in this talk
• Motion estimation • Image registration
• Derive representative 3D CT from 4D data
• Image processing • Image selection
• Motion management
• Gating/tracking • Motion inclusive planning
Image processing approaches to derive representative 3D data
4D CT Mean (‘Slow CT’) Max (‘MIP’)
Free breathing-like Low resolution Density correct
ITV-like Density over-estimated
Mid-ventilation is very simple (used clinically on hundreds of patients)
Mid-ventilation CT 4D CT Eliminates systematic error due to imaging (except hysteresis)
Geometrically and dosimetric very close to full 4D plan!
Wolthaus et al, IJROBP 2006; Nijkamp et al ICCR 2007
Mid-position CT: deform all anatomy to its mean position and average over all frames
Mid-ventilation image Mid-position image
Wolthaus et al, Med Phys 2008 (in press)
Reduces noise and artifacts
mean correction strategy: correcting the average error
Possible correction strategies: • Mean corrections • Minimizarion of
maximum error • …
Adaptive replanning on average anatomy
Planning CT
daily CBCTs deformation vector fields
systematic deformations Average anatomy
Kranen et al, ESTRO2009
N
Inter-observer variation in delineation
0 – 1 1 – 2 2 – 3 3 – 4 4 – 5 5 – 6 6 – 7 7 – 8 8 – 9 9 – 10 10 – 11 11 – 12 12 – 13 13 – 14 14 – 15 > 15
LOCAL SD (mm)
SD is 2 mm at best: this is what we use for our hypo patients Steenbakkers et al, IJROBP 2005
Concurrent VMAT – CBCT acquisition
No MV-Beam With MV- Beam
Error distributions & central limit theorem
Average of N uniform distributed numbers This works for any distribution
Soren Bentzen:
Accuracy is quality not quantity priority of resources, assymptotic diminishing returns
cost benefit level of QA for trials precision reproducibikity
Q: rmse = s^2 + bias^2?; P (x>X) hangt niet af van RMSE Q: do we know the target, even the dose!
Gamma DR for indentical mice about 4 (H.Suit), is this true if you irradiate the tumor only?
Double trouble: Dose unc affects total dose & dose/fraction: GammaN > GammaD
variation in dose delivery do not cancel when dE2/dD2 not zero for 5% tcp loss: precision between 3 and 5% needed
1% dose accuracy: about 1% TCP, 1 mm about 5% loss importance of accuracy dose/geometry depends on your baseline
Ellen Yorke look at mellin 2011 pigs 1 fx d50 20 GY
Jeraj Dark ages
bio im is qual imaging even in wisc did not pass national QA bowen et al NMB: tracer retention mechanism
Q: tracer transfer function depends on goal due to non-uniformity!
pH (7.1-7.3) and other pars strongly affects Cu-ATSM response
FLT short term gives perfusion mostly not prolif Q: can we fix shift effect on PET images
Demonstration – margins in lung
• Margin between CTV and PTV: 7 mm
• Errors: • Delineation error:
• 2 mm SD • Registration error
• 2 mm SD • Intra-fraction motion
• 2 mm SD • Organ motion
• 10-30 mm respiration