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