Point A - AMOS Onlineamos3.aapm.org/abstracts/pdf/68-19734-230352-86926.pdf · 82 Mean Values 93 81...

1

AAPM 2012– IGABT for Cervical Carcinoma – C Kirisits

Image guided Adaptive Brachytherapy

for Cervical Cancer

Christian Kirisits


Acknowledgements

The financial support by the Federal Ministry of Economy, Family

and Youth and the National Foundation for Research, Technology

and Development is gratefully acknowledged.


The Department of Radiotherapy at the Medical University of

Vienna received/receives financial and/or equipment support for

research and educational purposes from Nucletron, an Elekta

Company and Varian Medical Systems, Inc.

Christian Kirisits is a consultant to Nucletron, an Elekta Company.

This research was supported by the Austrian Science Fund FWF

grant No L562.

2


Point A

2cm2cm

2cm 2cm

2


Point A / target dose

84 Gy

84 Gy

60 Gy

D90 = 65 Gy EQD2


84 Gy

93 Gy

D90 = 75 Gy EQD2




84 Gy

84 Gy

84 Gy

D90 = 90 Gy EQD2

~ 500 Gy

3



79 Gy

84 Gy

84 Gy

D90 = 85 Gy EQD2

~ 500 Gy


Recommendations

• GYN GEC ESTRO recommendations I (Haie-Meder et al.) - contouring

• GYN GEC ESTRO recommendations II (Pötter et al.) - dose parameters

• GYN GEC ESTRO recommendations III (Hellebust et al.) - reconstruction

• GYN GEC ESTRO recommendations IV (Dimopoulos et al.) - imaging

• ABS recommendations on GYN general (Viswanathan and Thomadsen, ABS

Cervical Cancer Recommendations Committee) -general

• ABS recommendations on GYN HDR (Viswanathan et al.)

• ABS recommendations on GYN PDR (Lee et al.)

• ICRU 38 revision (coordinators: R. Pötter and C. Kirisits,

committee members: B. Erickson, C. Haie-Meder, J. Lindegaard, E. van

Limbergen, J. Rownd, K. Tanderup, B. Thomadsen)

8


DOSE AND VOLUME PARAMETERS FOR

PRESCRIBING, RECORDING, AND REPORTING

TARGET VOLUMES AND ORGANS AT RISK

Target parameters

TRAK, point A

D98, D90, D50 for HR CTV, IR CTV* (*if used for prescription)

D98, D90 for GTV

D98, D90 for PTV (if applicable, only for research!!)

D98?, D50?, D90? for pathological lymph nodes

(state of the art parameters in bold)

4


DVH for target volumes

10


DOSE AND VOLUME PARAMETERS FOR

PRESCRIBING, RECORDING, AND REPORTING

TARGET VOLUMES AND ORGANS AT RISK

OAR parameters

D2cm³, D0.1cm³ for high doses to bladder, rectum,

sigmoid and bowel (if applicable)

Vaginal points at level of vaginal sources

Intermediate dose by e.g. V50Gy, D50, and points and lengths

AAPM 2012– IGABT for Cervical Carcinoma – C Kirisits 12

DVH for OAR

5


Dose

V

Typical DVH curve for OAR from intracavitary BT


Dose

V

An increase of TRAK will result in an increase of the dose to the OAR. The shape of the new curve is strechted, the ratio of the two upper points on the DVH curve becomes larger. The impact on the low dose regions is smaller. The same ratio applies. The blue symbols are from the previous figure with lower dose to the OAR. The third point in the low dose region is not needed necessarily to describe the shape.


Dose

V

The orange curve could be the result of different external beam component or other treatment technique. The shape of the curve is completely changed. All three points on the DVH curve are needed to describe the difference to the two curves shown before.

6


Rectum:D2cc 63%

D0.1cc 80%

Typical dose distribution


Rectum:D2cc 63%

D0.1cc 100%

Optimized only based on limited parameter set


Standard loading

7


Inverse optimizationwithout thinking


0.1 cm³

2 cm³

> 5 cm³

Reference Points

Bladder

Rectum

Va

gin

a

Small Bowel


upper

mid

lower

ventral

dorsal

right

center

x c

m

Vag

inal le

ng

th

cm

cm

Vaginal dose –

DVH, Dose surface histogram, points, dimensions

8


FROM PLANNING AIMS TO PRESCRIPTION

Traditional concepts:

“when prescribing to a target, the prescription

dose is the planned dose to cover this target as

completely as possible.”

or

prescription to a 100% isodose which is “to

cover” the target volume”


Need for common terminology according to ICRU

reports on proton treatment and IMRT

• Planning aim dose

� Set of dose and dose/volume constraints for a treatment

• Prescribed dose

� Finally accepted treatment plan (which is assumed to be delivered

to an individual patient)

• Delivered dose

� Actually delivered dose to the individual patient

23


Need for common terminology according to ICRU

reports on proton treatment and IMRT

Example:

Previously: 4x7 Gy ~ 84 Gy EQD2 prescribed, D90 was mean 93 Gy

Planning aim was to deliver 4 x 7 Gy ~ 84 Gy, D2cm³ for rectum,

sigmoid < 70 Gy EQD2, bladder < 90 Gy EQD2

Prescribed dose was mean 93 Gy ± 13 Gy (1SD) EQD2 to D90 HR CTV

Delivered dose ? Depending on variations and uncertainties – on

average no systematic deviation from prescribed dose

24

9


LINKING DVH PARAMETERS TO CLINICAL OUTCOME

for TARGET/TUMOUR

1

Pro

bab

ilit

y o

f lo

ca

l co

ntr

ol

0 10 20 30 5040 90 1007060 80 110 120 130 140

0

0.1

0.2

0.3

0.5

0.4

0.9

0.7

0.6

0.8

D90 (HR CTV)

Entire population (n=141)

Tumours > 5cm (n=76)

Dimopoulos et al. IJROBP 2009, Strahlentherapie 2009

141 patientsFIGO: IB-IVA, median follow-up:51 monthsD90 for the HR-CTV and probability of local control

20% less D90 per fraction ~ 10 Gy less for total dose


Incid

ence V

RS

> 3

0

10

20

30

40

50

60

70

80

90

100

Dose [Gy]

30 40 50 60 70 80 90 100 110 120 130 140

Incid

en

ce L

EN

T/S

OM

A >

2

0

10

20

30

40

50

60

70

80

90

100

D2ccD1ccD0.1ccDICRU

P. Georg IJROBP 2010

Koom et al. IJROBP 2007

Rectum dose

N=141Clinical

Symptoms

0,00

0,10

0,20

0,30

0,40

0,50

0,60

0,70

0,80

0,90

1,00

1,10

30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

D2cc (Gy)

p

21

N=35Changes visible

with rectoscopy

Georg et al. 2009


www.embracestudy.dk

> 650 patients enrolled!

10


82

Mean Values

93 81 85 89 94 90 81 89 91 92 81 94 86

HR CTV D90 by Center Preliminary - 3D QA pending!


HR-CTV dose and technique

Median HR CTV volume: 30cc HR CTV >30cc Mean ± SD

HRCTV D90IC 81.9 ± 9.5 Gy

IC/IS 88.2 ± 7.7 Gy

Bladder D2ccIC 79.1 ± 10.3 Gy

IC/IS 78.7 ± 13.4 Gy

Rectum D2ccIC 65.6 ± 8.2 Gy

IC/IS 64.6 ± 6.1 Gy

Sigmoid D2ccIC 64.4 ± 6.4 Gy

IC/IS 63.7 ± 7.6 Gy

Large tumours (>30cc): � IC/IS increases target dose by 6Gy

� OAR doses not increased

Small tumours (≤30cc)� No difference for target or OARs

IC/IS Combination of intracavitaryand interstitial application


Application technique and

patterns of tumor regression

11


IIIB, 8 cm width, insufficient response (11/99)

no adaptation of application technique

intracavitary approach only

Brachytherapyrecurrence

9 months after treatment

DiagnosisOptimization with

Tandem/ringonly

HR CTVD90: 69 Gy


Improvement of local control

MRI based brachytherapy Vienna 98-03

Late side effects at 3 years

Total G3/4: 98-03: 7/145; 98-00: 6/73;

[93-97: 14/189] 01-03: 1/72Pötter et al. Radiother Oncol 2007

Depending on treatment period(experience, modification of application)

D90: 81 Gy (98-00) – 90 Gy (01-03)


Pötter et al. 2011

Current local control rates (Vienna experience)

12


Uncertainties?

Where to improve?


Example for HDR intracavitary Cervix brachytherapy – per fraction

Category Optimum level Assumptions

Source strength 2% PSDL traceable calibrations

Treatment planning 3% Reference data with the appropriate bin width is used

Medium dosimetric Applicator without shielding and CTVcorrections 1% inside pelvis (concerning for scatter)

Dose delivery including Accurate QA concept for commissioning and registration of applicator constancy checks, especially for sourcegeometry to anatomy 4% positioning and applicator/source path

geometry, appropriate imaging techniques, applicator libraries

Interfraction/Intrafraction For one treatment plan per applicatorchanges 12% insertion but several subsequent fractions –

check for major deviations in subsequent

fractions

Total dosimetric uncertainty 13%

for one single fraction


Example for HDR intracavitary Cervix brachytherapy – total dose 4 fractions

Category Optimum level Assumptions

Source strength 2% PSDL traceable calibrations

Treatment planning 3% Reference data with the appropriate bin width is used

Medium dosimetric Applicator without shielding and CTVcorrections 1% inside pelvis (concerning for scatter)

Dose delivery including Accurate QA concept for commissioning and registration of applicator constancy checks, especially for sourcegeometry to anatomy 4% 2% positioning and applicator/source path

geometry, appropriate imaging techniques, applicator libraries

Interfraction/Intrafraction For one treatment plan per applicatorchanges 12% 6% insertion but several subsequent fractions –

Total dosimetric uncertainty 7%for entire BT

1 / √N

1 / √N

We would need interfraction value < 3% to have total uncertainy < 5 %

13


Visualization of the “real” source positions in relation to the

outer dimensions and holes of the Vienna ring applicator

r26


Applicator surface


Source path

14


Applicator + Source path


Reconstruction


Reconstruction

15


Reconstruction


Reconstruction


Reconstruction

16


Reconstruction


Reconstruction


MRI for each fraction?

17


3D applicator reconstruction

2nd application: CT


2nd application: CT3D applicator reconstruction

Target transfer

Targets from first application MRI


Target transfer from MRI to CT with applicator as reference system

2nd application: CT

18


2nd application: CT


2nd application: CTContouring OAR on CT

Target contours from

1st appliction MRI

OAR contours from

2nd application CT


2nd application: CTDose planning and optimization based on copied target and individual OAR

contours. All dose constraints for targets and OAR have to be achieved.

19


Levels

• X-ray

• CT-based for one application

• CT-based for all applications

• MRI for first, CT for subsequent applications

• MRI for each application

• MRI for each fraction


Ultrasoundtechnology


Outlook

• ICRU 38 revision

• EMBRACE studies to increase insight into dose-effect

relationships

• New cost effective methods keeping accuracy reasonable

Point A - AMOS Onlineamos3.aapm.org/abstracts/pdf/68-19734-230352-86926.pdf · 82 Mean Values 93 81...

Documents

Transcript of Point A - AMOS Onlineamos3.aapm.org/abstracts/pdf/68-19734-230352-86926.pdf · 82 Mean Values 93 81...