Clinical uncertainties: what is the magnitude?

Clinical uncertainties: what is the magnitude?

Clinical uncertainties: what is the
magnitude?
B h N t
BrachyNext
Miami 2014
Kari Tanderup
The 6 steps of 3D image guided brachytherapy
Applicator implant
3D imaging
Contouring
3D dose planning
Reconstruction of applicator
Dose delivery
Organ/applicator movement
2
1
Important publications!
 Radiotherapy and Oncology 107 (April 2013):

Special issue on uncertainties in brachytherapy (mainly
gyn)
16 original papers (contouring, inter-fraction, intra-fraction,
applicator, vagina dose reporting)
1 editorial


 Radiotherapy and Oncology

Kirisits C, Rivard MJ, Baltas D, Ballester F, De Brabandere M,
van der Laarse R, Niatsetski Y, Papagiannis P, Hellebust TP,
Perez-Calatayud J, Tanderup K, Venselaar JL, Siebert FA.
Review of clinical brachytherapy uncertainties: Analysis
guidelines of GEC-ESTRO and the AAPM. Radiother Oncol.
In press
3
Uncertainties in the high
gradient BT dose distribution
Radial dose profile
Dose profile at level of point A
Dose gradient per mm
300%
20%
6% pr mm at point A
250%
15%
Dose (%))
Dose (%
%)
200%
150%
100%
10%
5%
50%
0%
0
10
20
30
40
Distance from tandem (mm)
50
60
0%
0
10
20
30
40
50
60
Distance from tandem (mm)
4
2
Contouring uncertainties
HR-CTV on MRI
 HR CTV:

Mean deviation <4mm
4mm
 GTV, IR CTV:

Mean deviation <6-7mm
10
A
GTV
HR CTV
IDDSTAPLE (mm)
8
IR CTV
6
4
2
0
180
240
300
0
60
120
Angle (degrees)
Petric et al, R&O 2008
Petric et al, RO Vol 107, April 2013
5
Impact of contouring uncertainties
(MRI) on dose
10%
5%
8%
11%
Hellebust et al, Vol 107, April 2013
6
3
Contouring uncertainties on CT
(interstitial brachytherapy)
Organ
Coefficient of variance
Bladder
16% ± 10%
Rectum
7% ± 2%
Sigmoid
34% ± 19%
Damato et al, IJROBP in press
7
Reconstruction uncertainties
MRI based intracavitary/interstitial brachytherapy
Reconstruction uncertainties in mm (SD):
~ 0.5mm
0 5mm in lateral and ant
ant-post
post directions
~ 1-2mm in longitudinal directions
Longitudinal
SD (mm)
Lateral
SD (mm)
Ant-posterior
SD (mm)
Rotation
SD (mm)
Ring (plastic)
1.4
0.6
0.4
1.2
Tandem (plastic)
1.7
0.6
0.5
-
Titanium Needle
2.2
0.4
0.3
-
S Haack et al., Radiother Oncol 2009
8
4
Impact of reconstruction uncertainties on
DVH parameters
% pr. m
mm displacement
7
6
5
lateral
ant-post
longitudinal
rotation
4
3
2
1
0
GTV
HR-CTV
Rectum
Bladder
Sigmoid
K Tanderup et al., Radiother Oncol, 2008
9
Evaluation of needle displacement over 2 days
Prostate HDR implantation
Simnor, RO 2009
10
5
Impact of needle movement
 Needle movement has significant impact on
target coverage
Kolkman-Duerloo, RO 2011
11
Courtesy Nicole Nesvacil
12
6
Uncertainty budget GYN
Tanderup et al, RO Vol 107, 2013
13
Gradients in brachytherapy
 Longitudinal direction
 Gradients can be manipulated
 Dependent on application
 Radial direction
 Gradients cannot be manipulated
14
7
PTV margins in lateral direction?
EBRT
Dose
Dose
BT
l
PTV
Distance
PTV
CTV
Distance
CTV
15
Impact of uncertainties on total dose
OAR
6-8 Gy (SD)
Sigmoid
Bladder
Rectum
HR CTV
2-5 Gy (SD)
HR CTV
16
8
Examples total dose and uncertainty
 HR CTV:
D90 = 90 ± 4Gy
 Bladder:
D2cm3 = 85 ± 7Gy
 Rectum:
D2cm3 = 70 ± 4Gy
 Sigmoid:
D2cm3 = 70 ± 7Gy
17
Simulating dose‐response curves for HR CTV D90
HR CTV, D=10% per Fx
HR CTV,  D=10% per Fx
1
1
0.9
0.95
08
0.8
0.7
0.9
response
response
0.6
0.5
0.85
0.4
0.8
expected dose response
curve
0.3
0.2
mean response
(F(D,uncert))
01
0.1
simulated percentage of
responders
expected dose response
curve
mean response
(F(D,uncert))
0.75
simulated p
percentage
g of
responders
0.7
0
0
10
20
30
40
50
60
70
80
90
100
110
120
45
55
65
75
85
95
105
115
dose (Gy EQD2    =10Gy)
dose (Gy EQD2    =10Gy)
Simulated data with 10% uncertainty (SD per BT fraction, for EBRT+4xHDR BT).
Error bars: +/-1SD of response probabilities calculated for Individual patients
Negligible systematic differences between simulated „observed“ curve and assumed dose-response curve.
Ranges of response probability uncertainties around 85Gy EQD2 are around 2.5% for individual patients (1 SD of
all simulated values (n=10000)).
Courtesy Nicole Nesvacil
9
Simulating dose‐response curves for OAR
OAR,  D=20% per Fx
OAR,  D=20% per Fx
0.2
1
0.9
expected dose
response curve
0 18
0.18
expected
p
dose
response curve
0.8
mean response
(F(D,uncert))
0.16
mean response
(F(D,uncert))
0.7
simulated percentage
of responders
0.14
simulated percentage
of responders
response
response
0.6
0.5
0.4
0.12
0.1
0.08
0.3
0.06
0.2
0.04
0.1
0.02
0
0
40
60
80
100
120
140
160
180
45
dose (Gy EQD2    =3Gy)
Simulated data with 20% uncertainty (SD per BT fraction,
for EBRT+4xHDR BT).
Error bars: +/-1SD of response probabilities calculated for
Individual patients
50
55
60
65
70
75
80
85
90
dose (Gy EQD2    =3Gy)
-Small systematic offset around rectum dose constraint of 70-75Gy
-Range of response probability uncertainty for individual patients
~4-6%
Courtesy Nicole Nesvacil
Dosimetric margin
20
10
Interventions to decrease uncertainties
 Target

Contouring training

Re-imaging for assessment of catheter
position (interstitial)
•
ESTRO contouring workshops
 OARs


Re-imaging
Re-planning
 What do we need?


Kolkman-Duerloo, RO 2011
Tools for quick plan adaptation
Alternative imaging modalities (CBCT, US)
21
Conclusion
 Major clinical uncertainties are related to:



Contouring
Needle movement
Organ movement
 Geometrical PTV margins are not generally applicable
for brachytherapy
 In brachytherapy ”dosimetric margins” may be the
”secret of success” even in the presence of
considerable uncertainties
22
11