IMRT

First page 1 Next page Last page

Animated publication

ESTRO Course Book IMRT and Other Conformal Techniques in Practice

4 - 8 October, 2015 Brussels, Belgium

NOTE TO THE PARTICIPANTS

The present slides are provided to you as a basis for taking notes during the course. In as many instances as practically possible, we have tried to indicate from which author these slides have been borrowed to illustrate this course. It should be realised that the present texts can only be considered as notes for a teaching course and should not in any way be copied or circulated. They are only for personal use. Please be very strict in this, as it is the only condition under which such services can be provided to the participants of the course.

Faculty

Marco Schwarz and Frank Lohr

Disclaimer

The faculty of the teachers for this event has disclosed any potential conflict of interest that the teachers may have.

Programme

Sunday 04 October 8.30 – 8.45 Introduction to the course - Marco Schwarz 8.45 - Group B : Going to UZ Brussel 9.30 – 10.00 Demo 1: Tomo patient QA - Katrien Leysen – Red Team 10.00 – 10.30 Demo 2: Tomo absolute calibration - Dirk Verellen – Orange Team 10.30 – 11.00 Coffee break 11.00-11.30 Demo 3: Real-time Tumour tracking - Jennifer Dhont – Green Team 11.30 – 12.00 Demo 4: Planning - Marlies Boussaert; Paul Bijderke - White Team Group A : Lectures at the Hotel Chair: Frank Lohr 9.00 – 9.30 Opening and welcome - Koen Tournel 9.30 – 10.00 Treatment of Rectal cancer using IMRT/IGRT at the UZ Brussel - Benedikt Engels 10.00 – 10.30 Coffee break 10.30 – 11.00 Treatment of lung and liver tumors using dynamic tracking - Thierry Gevaert

11.00 - 11.40 Treatment of oligometastases using an SBRT/IMRT/IGRT approach Robbe Van Den Begin 11.40 – 12.10 Radiosurgery at the UZ Brussel - Thierry Gevaert 12.30 - 13.30 Lunch 13.30 - Group A going to UZ Brussel 14.00 – 14.30 Demo 1: Tomo patient QA - Katrien Leysen – Red Team 14.30 – 15.00 Demo 2: Tomo absolute calibration - Dirk Verellen – Orange Team 15.30 – 16.00 Coffee break 16.00-16.30 Demo 3: Real-time Tumour tracking - Jennifer Dhont – Green Team 16.30 – 17.00 Demo 4: Planning - Marlies Boussaert; Paul Bijderke - White Team Group B - Lectures at the hotel: Chair: Matthias Söhn 14.00 – 14.30 Opening and welcome - Koen Tournel 14.30 – 15.00 Treatment of Rectal cancer using IMRT/IGRT at the UZ Brussel - Benedikt Engels

14.00 - 14.45 Image-guidance & Adaptive: concept and approaches – Matthias Söhn 14.45 - 15.30 Image-guidance & Adaptive: Clinical applications – Frank Lohr 15.30 - 16.00 Coffee break 16.00 - 16.45 IMRT in breast and risk of secondary cancer after IMRT – Frank Lohr 16.45 – 17.30 IMRT in Hodgkin's Lymphoma and secondary cancer risks– Andrea Filippi 17.30 – 22.30 Social Event @ Atomium Tuesday 6 Oct Chair: Matthias Söhn 9.00 - 9.45 ‘Patient specific’ QA – Marco Schwarz 9.45 – 10.30 Modeling adverse effects after 3DCRT and IMRT– Giovanna Gagliardi 10.30 - 11.00 Coffee break 11.00 -11.45 Review of Dose-volume relationships I: H&N - Giovanna Gagliardi 11.45 - 12.30 IMRT in Head and neck – Frank Lohr 12.30 - 14.00 Lunch

15.00 – 15.30 Coffee break 15.30 – 16.00 Treatment of lung and liver tumors using dynamic tracking - Thierry Gevaert 16.00 - 16.30 Treatment of oligometastases using an SBRT/IMRT/IGRT approach Robbe Van Den Begin 16.30 – 17.00 Radiosurgery at the UZ Brussel - Thierry Gevaert Monday 05 October Chair: Giovanna Gagliardi 9.00 - 9.30 Rational of IMRT. A clinician’s point of view - Frank Lohr 9.30 - 10.15 IMRT delivery techniques – Marco Schwarz 10.15 - 10.45 Coffee Break 10.45 - 11.30 Dosimetry issues in IMRT – Koen Tournel 11.30 - 12.00 TPS commissioning – M. Schwarz 12.00 – 12.45 IMRT optimization: algorithms and cost functions – Matthias Söhn

12.45 - 14.00 Lunch Chair: Koen Tournel

Clinical session 3 : local MD , Matthias Söhn – Prostate - LILLEHAMMER Room Group B: free

14.00 - 15.30 Group A: Clinical case discussion 1 (14.00-14.45) Clinical session 1 : Andrea Filippi , Koen Tournel Lymphoma - HARALD Room Clinical session 2 : local MD , Matthias Söhn – Prostate – STAVANGER Room Clinical session 3 : Frank Lohr , Giovanna Gagliardi– H&N - LILLEHAMMER Room Clinical case discussion 2 (14.50-15.30) Clinical session 1 : local MD , Matthias Söhn – Prostate - HARALD Room Clinical session 2 : Frank Lohr , Giovanna Gagliardi – H&N - STAVANGER Room Clinical session 3 : Andrea Filippi , Koen Tournel Lymphoma - LILLEHAMMER Room Group B: Vendor session Chair of the session: Marco Schwarz 15.30 - 16.00 Coffee break 16.00 – 16.45 Group A: Clinical case discussion 3 Clinical session 1 : Lymphoma Frank Lohr , Giovanna Gagliardi – H&N - HARALD Room Clinical session 2 : Andrea Filippi , Koen Tournel Lymphoma - STAVANGER Room

Wednesday 7 Oct Chair: Frank Lohr 9.00 - 09.45 Practical IMRT planning and ‘biological optimization’ – Marco Schwarz 9.45 - 10.30 Impact of geometrical uncertainties on IMRT dose distributions – Koen Tournel 10.30 - 11.00 Coffee break 11.00 - 11.45 Review of Dose-volume relationships II: Pelvis – Giovanna Gagliardi 11.45 – 12.30 IMRT of prostate cancer – F. Lohr 12.30-14.00 Lunch 14.00-15.30 Group B: Clinical case discussion 1 (14.00-14.45) Clinical session 1 : Andrea Filippi , Koen Tournel Lymphoma - HARALD Room Clinical session 2 : local MD , Matthias Söhn – Prostate – STAVANGER Room Clinical session 3 : Frank Lohr , Giovanna Gagliardi– H&N - LILLEHAMMER Room

Thursday 8 October Chair: Andrea Filippi 9.00 - 9.45 Dose calculations in static and rotational IMRT - Matthias Söhn 9.45 - 10.30 Potential and limitations of rotational IMRT – Koen Tournel 10.30-11.00 Coffee break 11.00 - 11.45 Highly conformal techniques in early stage lung cancer: indications, techniques, normal tissue constraints, results – Andrea Filippi 11.45 - 12.30 Highly conformal techniques in advanced stage lung cancer: indications, techniques, normal tissue constraints, results – Andrea Filippi 12.30 -13.00 Final discussion and closing of the course

Clinical case discussion 2 (14.50-15.30) Clinical session 1 : local MD , Matthias Söhn – Prostate - HARALD Room Clinical session 2 : Frank Lohr , Giovanna Gagliardi – H&N - STAVANGER Room Clinical session 3 : Andrea Filippi , Koen Tournel Lymphoma - LILLEHAMMER Room Group A: Vendor session / Chair of the session: Marco Schwarz 15.30-16.00 Coffee break 16.00 – 16.45 Group B: Clinical case discussion 3 Clinical session 1 : Lymphoma Frank Lohr , Giovanna Gagliardi – H&N - HARALD Room Clinical session 2 : Andrea Filippi , Koen Tournel Lymphoma - STAVANGER Room Clinical session 3 : local MD , Matthias Söhn – Prostate - LILLEHAMMER Room Group A: free

Faculty

Marco Schwarz

Protontherapy Centre Trento, Italy marco.schwarz@apss.tn.it University Medical Centre Mannheim, Germany f.lohr@gmx.de

Frank Lohr

Andrea Riccardo Filippi AOU Città della Salute e della Scienza Turin, Italy afilippi@unito.it

Giovanna Gagliardi

Karolinska University Hospital Stockholm, Sweden giovanna.gagliardi@karolinska.se

Matthias Söhn

LMU University Hospital Munich, Germany Matthias.Soehn@med.uni- muenchen.de UZ Brussel (VUB) Brussels, Belgium koen.tournel@uzbrussel.be

Koen Tournel

Welcome to Brussels Brussel/Bruxelles

NOT dr.Prof Mark De Ridder, head of department Koen Tournel, Medical Physicist

A rich history….

Bruocsella

est. 10 century AC

“Village by the swamp”

A rich history….

Pieter Brueghel the elder

Capital of Belgium

• 11.000.000 inhabitants • Official languages : Dutch (60%), French (39,3%), German (0,7%) (English widely spoken and understood) • Cities : Brussels, Antwerp, Liege, Gent, Namur, Brugge… Brussels: • 1.200.000 inhabitants • French (38%), Dutch (5%), Dutch and French ( 17%), French and others (23%), others (17%)

Language map of Belgium

Comm. of Flanders

Capital of Europe

Capital of surrealism

Rene Magritte 1898-1967

Belgium : country of surrealism

• Federal monarchy • 9 parliaments • 8 governments • 48 ministers + 10 secretaries of state • 10 provinces, 11 governors • 1 king, 2 queens

• world record holder in number of days without government : 541 • European and world champion in tax pressure on labor : 56% • Traffic jam capital of Europe

Belgium : country of surrealism

Sights and Sounds

The Atomium (expo 1958)

BCC Ferro crystal

Manneken Pis

Grote Markt / Grand Place

Mont des arts / kunstberg

Architecture

Galerie des Reines

Art nouveau

Home of Belgian beers….

“gueuze” or “kriek”

Home of Belgian chocolates

Home of gastronomy

UZ Brussel

• Academic Hospital of the Vrije Universiteit Brussel • connected to the medical faculty and the Erasmus school for life sciences • Only native Dutch-speaking hospital in Brussels (in practice multilingual) • 3500 employees

Oncology Center

Radiotherapy

Oncology

Onc. Surgery

1600 patients/year

7 Linacs – 2 locations

Aalst

Jette (Brussels)

April 2014

• 12 Radiation oncologists (3 in training) • 7 physicists/2 dosimetrists • 5 engineers for linac maintenance • 25 nurses/therapists • 1 logistics

Multi-vendor environment

• 2 Elekta • 1 Varian • 2 tomotherapy • 1 Mitsubishi/brainlab • Brainscan/iPlan/Xio/Monaco/Tomo/ Eclipse/Raysearch

• Vero RV/Aria/Mosaiq • SN/IBA/PTW/Ashland • MIM

Milestones

• 1994 : Stereotactic frame • 1998 : IMRT using the mimic • 2000 : 1st European Novalis • 2003 : Stereoscopic Xray positioning for prostate • 2005 : 2nd European Tomotherapy device • 2013 : Dynamic tracking

Research interests • Rectal cancer : RectumSIB trial -B.Engels • SBRT T1,2 Lung (tracking, ITV) • NSCLC T3,4 : C.Collen • SBRT of metastatic disease (R.Van Den Begin) • SRS Brain : AVM, frameless : T.Gevaert • Markerless tracking : J.Dhondt

Preoperative RT of rectal cancer using IMRT/IGRT at the UZ Brussel

Benedikt Engels, MD PhD Department of Radiotherapy UZ Brussel, Vrije Universiteit Brussel

ESTRO course on IMRT and other conformal techniques 2015 Oct 4-8, Brussels, Belgium

Introduction

 823 patients with stage II/III rectal cancer  Randomly assigned  Preoperative 28 x 1.8 Gy + 5FU  Postoperative 31 x 1.8 Gy + 5FU ( 5 FU: 120 h CI, 1000 mg/m 2/day , week 1 & 5) Sauer et al, N Eng J Med 2004

Local recurrence rate (update)

Sauer et al, J Clin Oncol 2012

Preoperative chemoRT or RT alone?

 EORTC 22921 (1) and FFCD 9203 (2) : preoperative 5-FU chemoRT superior over RT alone with respect to pCR and LC, but no difference in:  Overall survival

 Rate of sphincter sparing surgery  Occurrence of distant metastases

(1) Bosset et al., N Engl J Med 2006 (2) Gérard et al., J Clin Oncol 2006

Bosset et al, Lancet Oncol 2014

Adverse effects of long-course chemoRT

 Radiation enteritis = primary acute/late side effect  German trial: preop chemoRT less toxic than postop (1)

 Acute grade ≥ 3 toxicity rate up to 5 times higher with the addition of 5-FU to preoperative RT as compared to preoperative RT alone (2)  Intensified chemotherapy + RT:

(1)Sauer et al., N Engl J Med 2004 (2)Gérard et al., J Clin Oncol 2006

Efforts to improve outcome

 Preoperatively:  Induction chemotherapy prior to 5-FU chemoRT  Concurrent multi-agent 5-FU based chemoRT:

 Biologic agents (cetuximab/bevacizumab)

 Oxaliplatin: phase III evidence

3D-Conformal RT: 3-field technique

postero-anterior beam (40% of the dose)

2 opposing latero-lateral beams (60% of the dose)

3D-conformal RT vs IMRT

Wolff et al, R&O 2012

PTV Organ at risk

+ Image-guided RT (IG-IMRT)

Planning CT CT prior to treatment co-registration => the dose distribution is delivered at the exact location within the patient by image guidance

UZ Brussel approach

 Alternative strategy UZ Brussel: IG-IMRT + simultaneous integrated boost (SIB) by helical tomotherapy

 To decrease the toxicity  by reducing the irradiated volumes of small bowel and bladder by IG-IMRT  by omitting concomitant 5-FU chemotherapy  To maintain the oncological safety  by delivery of a simultaneous integrated boost to bad T3 (CRM ≤ 2mm) and T4 tumors

 Phase II Study: primary endpoint 2-year local control

 ≤ 3/100 patients with local failure

De Ridder and Engels et al, Int J Radiat Oncol Biol Phys and Radiother Oncol 2008 - 2014

Phase II study 2005 – 2010

 108 patients with cT3-4 rectal cancer treated preoperatively

 Radiotherapy (RT): 23 x 2 Gy with IG-IMRT by helical tomotherapy with daily MVCT scan positioning and without concurrent chemotherapy

 SIB of 0.4 Gy/day on the primary tumor up to a total dose of 55.2 Gy for patients with a narrow CRM on MRI ( ≤ 2 mm) to maintain oncological safety

Radiotherapy

follow-up

Late toxicity Outcome

adjuvant chemotherapy

5-6 weeks

5 weeks

Dose-volume constraints for IMRT: small bowel  Baglan et al. (n=40, 3D-CRT)  V15 < 150cc: no grade 3+ acute GI toxicity  V15 > 150cc: 50% grade 3+ acute GI toxicity  Gunnlaugsson et al. (n=28, 3D-CRT)  V15 ≤ 150cc: 11% diarrhea  V15 > 150cc: 52% diarrhea  Tho et al. (n=41, 3D-CRT)  V5-V30 correlated with severity of diarrhea  Yang et al. (n=177, IMRT/3D-CRT)  V45 < 3%: no grade 2+ diarrhea  V45 ≥ 27%: 20% grade 2+ diarrhea  Reis et al. (n=45, 3D-CRT)  V5 < 292cc: 29% grade 2-3 diarrhea  V5 > 292cc: 82% grade 2-3 diarrhea  QUANTEC (Marks LB et al.): V15 < 120cc (< 10% risk of severe acute GI toxicity)

Dose-volume constraints for IMRT: bladder

 Less radiosensitive as compared to small bowel 

e.g. primary RT gynaecological/urological malignancies (often doses > 70Gy)

 Sauer et al: only 2% grade 3+ late GU toxicity  Appelt et al. (n=345, preoperative long-course chemoRT):  first report on dose-volume relationship for acute urinary toxicity in rectal cancer

 No clinically significant benefit is to be expected from IMRT with regard to GU toxicity

Dose-volume constraints for IMRT: pelvic bone + plexus

 Baxter et al: 

Increased risk of pelvic fractures in elderly women  Holm et al (follow-up data Stockholm trials):   No reports on dose-response relationship:  max 45-50 Gy

5.3% femoral neck/pelvic fractures + hospitalization (vs 2.4% no RT)

 Sacral insufficiency fractures:  Rare but significant morbidity  Incidence: 3-7% 

Independent risk factors: osteoporosis, female gender, age > 60 years  No data on dose-response relationship  Lumbosacral plexus:  Tunio et al (cervical cancer) 8% grade 2 or more plexopathy  Prospective studies in rectal cancer are warranted

Dose-volume constraints for IMRT: bone marrow  40% of total body bone marrow reserves located in pelvis:  up to 8% hematological toxicity with preop chemoRT  Mell et al (chemoRT anal cancer):  V10-20 significantly correlated with hematological toxicty  = low-dose treshold > myelosuppressive chemotherapy  Dosimetric benefit of IMRT (Mell et al, cervical cancer):  Reduction of lumbosacral BM irradiation all dose levels  Reduction of pelvic BM irradiation to high doses  Yang et al (rectal cancer):  Sacral BM: V45 < 51%  Coxal BM: V45 < 13%  Higher dose treshold (vs anal cancer)

Proposed dose-volume constraints IMRT rectal cancer

IG-IMRT by helical tomotherapy with a simultaneous integrated boost

SIB

Acute toxicity

Only 1% grade ≥ 3 acute toxicity

According to the NCI CTC AE v 3.0 scale

De Ridder and Engels et al, Int J Radiat Oncol Biol Phys and Radiother Oncol 2008 - 2014

Surgical characteristics and downstaging

8% pCR rate 40% of cT3-4 was downstaged to ypT0-2

De Ridder and Engels et al, Int J Radiat Oncol Biol Phys and Radiother Oncol 2008 - 2014

Late toxicity

According to the NCI CTC AE v 3.0 scale

9% grade ≥ 3 late gastrointestinal toxicity 4% grade ≥ 3 late urinary toxicity 13% any grade ≥ 3 late toxicity

De Ridder and Engels et al, Int J Radiat Oncol Biol Phys and Radiother Oncol 2008 - 2014

Follow-up (median 54 months; range: 27-79 months)

3 locoregional relapses

MVA: R1 resection (p=0.03) ypN2 disease (p=0.04) UVA: adjuvant chemotherapy (p=0.04)

5-year: 97% LC 57% PFS 68% OS

MVA: R1 resection (p=0.03) ypN2 disease (p=0.04) UVA: adjuvant chemotherapy (p=0.04) ypT3-4 tumor (p=0.01) Dworak grade 0-2 (p=0.03)

De Ridder and Engels et al, Int J Radiat Oncol Biol Phys and Radiother Oncol 2008 - 2014

2010: Multicentric randomized trial (NCT 01224392)

 Preoperative IG-IMRT with a simultaneous integrated boost (SIB) compared to chemoradiotherapy for T3-4 rectal cancer

Collaborating Centres  National Cancer Institute Aviano, Italy  IRCC Candiolo, Italy  IRCCS San Martino-IST Genoa, Italy  Institute of Oncology Vilnius University, Lithuania  UZ Brussel, Belgium (PI)  University of Torino, Italy  Nantes, France (2014)



Study design

cT3-4 rectal cancer



Arm 2: RT-boost

Arm 1: chemoRT

IG-IMRT 23 x 2 Gy + capecitabine (825mg/m 2 )

IG-IMRT 23 x 2 Gy + SIB 0.4 Gy/day ( Σ 55.2 Gy)

TME surgery 6 weeks post RT + adjuvant capecitabine (x 6)

 Primary endpoint: reduction in metabolic tumor activity  SUV max on sequential 18FDG-PET imaging  Non-inferiority: difference in metabolic response ≤ 10%  Sample size: 78 in both arms ( α : 0.05, power: 80%)

Patient population (n = 169)

Acute toxicity

Surgical parameters

75%

68%

Pathology

49%

45%

51%

60%

Primary endpoint

 Mean decrease in SUV max on 18FDG-PET 5 weeks after completion of preoperative RT as compared to baseline:

 Difference: -2.9% (95% CI, -10.1% to 4.3%)

Outcome (median follow-up 14 months)

2-y 86% vs 82%

2-y 75% vs 82%

2-y 93% vs 98%

compared to chemoRT for T3-4 rectal cancer: a multicentric randomized trial

Conclusions:



 The use of image-guided and intensity-modulated radiotherapy (IG-IMRT) results in very low rates of acute grade 3 toxicity in both arms (6% vs 4%)

 IMRT improves significantly the tolerance of preoperative RT for rectal cancer, especially radiation enteritis

 Higher pCR rates are observed in the chemo-RT arm (24% vs 14% in the RT-SIB arm), no differences in major tumoral downstaging

 Difference in SUVmax reduction -2.9% in favor of chemoRT (95% CI, -10.1% to 4.3%, p = 0.06) => RT-SIB marginally failed to prove non-inferiority to chemoRT

Acknowledgements

Antonino De Paoli (Aviano), Gabriella Cattari (Candiolo), Fernando Munoz (Torino), Stefano Vagge (Genova), Darius Norkus (Lithuania), Gianna Tabaro (Aviano), Harijati Versmessen (Brussels), Mark De Ridder (Brussels), …

SBRT in UZ Brussel

Thierry Gevaert, PhD Radiotherapy, UZ Brussel

The VERO System

Gimbaled linac tilt/pan ampl.: ± 4.4 cm (at iso)

Dual O-ring Gantry rotation Horizontal axis ± 185°

Vertical axis: : ± 60°

Nicely flattened 6MV beam

MLC leaves : 60 x 5 mm single focus 50 mm/s leaf speed Max. field size : 15 x 15 cm

Transversal profiles @ 10cm depth

Diagonal profiles @ 10cm depth

The VERO positioning system

è Dual orthogonal kV X-ray system:

ExacTrac IR marker pre-positioning device

Stereoscopic X-rays (Imaging size : 15.89 x 21.18 cm) kV cone-beam CT imaging in 200°, FOV ∅ 20 x 15 cm

l

kV 1

kV 2

EPID MV portal imaging (0.2 mm pixel size) FPD 1 FPD 2

Fluoroscopy imaging for tracking (15 fr/sec)

Novalis Body 6DOF postioning

FPD 2

FPD 1

Image guidance accuracy

“ Hidden-target test on anthropomorphic phantom, pelvic region ”

kV planar, kV CBCT imaging

iso 2mm BB in isocenter

Plan 30x30mm open fields

Treatment plan delivery

Phantom positionning

EPID Image analysis

-EPID images (0.2mm pixels) -Automatic detection of field outline (Hough transform) and BB (Kernel based) -Detection estimated accuracy 0.1 mm

Image guidance accuracy

“ Hidden-target test on anthropomorphic phantom, pelvic region ”

kV 0°-90° bony

kV 45°-135° bony

kV CBCT CW bony kV 0°-90° marker

X (LAT) 0.23 (0.10) mm 0.08 (0.18) mm -0.63 (0.04) mm 0.17 (0.06) mm -0.18 (0.09) mm -0.20 (0.05) mm -0.08 (0.11) mm -0.18 (0.03) mm -0.13 (0.04) mm -0.58 (0.16) mm 0.32 (0.27) mm -0.71 (0.03) mm 3D VECTOR modulus 0.62 mm 0.67 mm 0.71 mm 0.75 mm Y (LONG) Z (VERT)

O-ring stability

BrainLab/MHI Vero

Elekta Infinity

BrainLab/Varian Novalis

Elekta Compact

Varian Truebeam

Tomotherapy Hi Art

Brussel, Aalst, BE

Munich, GE

Zurich, CH

Milano, IT Como, IT

Dallas, TX

USA

Europe

Phys. Med. Biol. 57 (2012) 2997–3011)

O-ring stability

è “ visual ” inspection of star shot film è Quality indicated by multi-axial symmetry of the star pattern l Radiosensitive film placed in plane of rotations l Sequence of narrow fields at different angles l Intersection of these fields forms a star-shaped pattern

Central “ Star ” pattern

Depuydt et al. 2012 (Phys. Med. Biol. 57 (2012) 2997–3011)

O-ring stability

Quality indicated by radius of smallest intersecting circle

1 Digitization of film 2 Detection of beam axis 3 Representation of beam axis as linear equations 4 Calculate smallest intersecting circle radius

Depuydt et al. 2012 (Phys. Med. Biol. 57 (2012) 2997–3011)

O-ring stability

Collimator rotation

Gantry rotation

Vero

Tomotherapy

Truebeam

Infinity

Couch rotation

O-ring rotation

Gantry Wobble

Truebeam

Vero

Novalis

All fields collimated with MLC

Phys. Med. Biol. 57 (2012) 2997–3011)

O-ring stability

- Best O-ring and best C-arm system show comparable results ...

Depuydt et al. 2012 (Phys. Med. Biol. 57 (2012) 2997–3011)

iPlan treatment techniques

TECHNIQUES IN DEVELOPMENT??

Multiple Modality Treatment Delivery

CONFORMAL BEAM

INVERS ARC

DYNAMIC CONFORMAL ARC

IMRT

Accomodation moving tumors l 4D CT: respiratory correlation, motion measurement

ITV GTV

Reliability 4D CT

è Repeated 4D CT before treatment planning: l Four 4D CT in ten minutes interval (Guckenberger IJROBP ‘07)

Ø No systematic changes of motion pattern Ø Increased variability for lower lobe tumors

l Two successive 4D CT (van der Geld radiat oncol ‘06) Ø Volume of the PTV not systematically different Ø Motion range variability < 2mm in 81 % Ø Coverage not compromised

No benefit of repeated 4D CT in one session

Reliability 4D CT

è Repeated 4D CT during treatment course: l Second 4D CT after > 2 fractions (Haasbeck IJROBP ‘07) Ø No systematic changes of motion pattern and PTV Ø Target coverage compromised in one patient (atelectasis) l Repeated 4D CBCT scans during RT (Sonke IJROBP ‘08) Ø Stable trajectory with variability (1SD) less than 1mm Ø Significant base-line shifts l Tumor tracking in EPID images (Richter IJROBP ’10) Ø Stable tumor trajectory, both intra- and inter fractional No benefit of replanning because of motion variability

Accomodation moving tumors

l Movement < 7mm : ITV concept

ITV GTV

l GTV contoured on 10 phase CT (propagation) = ITV

PTV = ITV + 5 mm

l

Reliable day to day treatments: ITV

Planning CT

Vero CBCT

Cone beam CT positioning with ITV approach

PTV ITV

Online Epid imaging

EPID%

ITV%

MLC%

Accomodation moving tumors

l Movement > 7mm : tracking concept

ITV GTV

l Placement of internal marker + dynamic tumor tracking

Visicoil TM

è Effectiveness of Tracking relies on a stable relationship between the marker and tumor l Changes in tumor geometry (long-course RT) l Migration between CT and radiation delivery (bronchofiberscopy)

è Ideal location Visicoil:

in the center of the lesion

l

l not in the same axial-CT image plane (e.g. 45 degree angle)

è Time interval between marker placement and CT for ‘scarring in’

è Lung and liver: percutaneous insertion under CT- fluoroscopic guidance

Surrogate value: COM Single Visicoil marker vs. COM GTV Fiducial marker surrogate value for GTV position

“ Max deviation ”

Ex. 1

LAT

1.2 mm in CC

(CT slice 2 mm)

(x,y,z) COM , Visicoil =(x,y,z) COM,GTV +Cte

Ex. 2

LAT

+ Regular verification of marker surrogate value on CT imaging

Acitve versus passive approach

Active approach: Tracking approach Passive approach: ITV approach

Example ITV vs. Tracking approach

PTV definition: Tracking: CTV=GTV, PTV TR =CTV + 5mm isotropically ITV backup plan: ITV= sum CTV all phases + 5 mm isotropically (for CBCT guidance) PTV ITV =ITV+5mm isotropically PTV TR = 18.3 ccm , PTV ITV =30.2 ccm (39% PTV volume reduction)

Patient History: Male, 74 years old: 7/2009: primary disease: cT4N3M0, treated induction chemo , consolidation RT 8/2010: metastatic disease: Tarceva 5/2012: progression of solitary metastasis lower left lobe , RT 2x20 Gy Delineation: On 60% phase in stable exhale ( volume 4.30 ccm ) Size: AP: 2.6 cm, CC: 2.2 cm, LR: 2.2 cm Motion: Visicoil marker 0.75 mm diameter, 10 mm length Amplitude CC of 14mm p2p in 4D CT

Goal of dynamic tracking as compared to the ITV approach

1) Reduction of volume receiving high-doses (PTV 100cc vs PTV 50cc)

=> potential advantage in terms of NTCP still needs to be determined!

Goal of dynamic tracking as compared to the ITV approach

DRIFT

2) Real-time adaptation of periodic breathing motion and systematic drifts in patients with a variable and unpredictable magnitude of motion ( lower lobe tumors and patients with poor pulmonary function*)

*Guckenberger et al, IJROBP 2007

3) Increased efficiency of respiratory correlated treatments (as compared to gating)

Translation tracking versus BEV tracking

Accuracy CyberKnife

BrainLab/MHI Vero

XYZ Translation Tracking

Pan/Tilt Rotation Beams-eye-view Tracking

Behavior symmetry of Pan/Tilt, tracking/MLC decoupling

Gimbals Tracking

DMLC Tracking

Motion

MLC

Moving No tracking

Not Moving No tracking

Moving Tracking

EPID images (3fps)

Courtesy Andreas Krauß, DKFZ

Dynamic tracking concept

Tumor with implanted visicoil marker

Moving infrared marker on skin (surrogate)

“Mathema#cal model to predict mo#on of the on the basis of the moving infrared on skin”

Pa#ent

X-‐ray images

Internal moving tumor

Workflow dynamic tracking

1. X-ray guided setup based on bony landmarks CBCT s ft tissue matching

Patient Positioning

Acquiring breathing signal

20-40 sec kV fluoro sequence + infra- red marker motion

Visicoil detection + building correlation model

VERO BREATH-HOLD CONE BEAM CT

kV imaging monitoring

Beam on

Treatment time aspects

patient setup

modelling Treatment delivery

modelling

Treatment delivery

...

modelling

Treatment delivery

Average breathing amplitude [mm]

Netto treatment time

Time in room MIN

Time in room AVERAGE

Time in room MAX

Building of Cor. Model

Usage of Cor. Model

Site

2m 59s 5m 59s 20m 58s 34m 31s 34m 38s 34m 44s 3m 3s 5m 28s 14m 29s 19m 9s 22m 41s 37m 44s 2m 23s 8m 48s 20m 0s 32m 12s 33m 49s 36m 15s

Patient 1 lung (17Gy/ fr)

10.8

lung (5Gy/fr)

Patient 2

6.8

Patient 7 lung (12Gy/ fr)

23.7

2m 50s 6m 24s 18m 49s

31m 26s

Average

15.3

Dynamic tracking new version

Analyze of deviation between predictive and real time position marker Based on these information, update of correlation model

Updating correlation model

è Investigation of potential benefit of updating the Vero correlation model using X-ray verification images during real-time tumor tracking: è Gain in time!! l 30% reduction in overall treatment time l 5.5 min. ± 4 min. è Occurred in 80% of fractions

Poels et al.

EBT3 film Sinusoidal tumor motion : Tracking for different bpm

25 bpm

15 bpm

static

“ The 20%-80% isodose distance was not significant increased for the 15 bpm, and only 0.2 mm for the 25 bpm signal. ”

Tracking for tumor motion from patients

CTV-PTV margin calculation

Prediction error tolerance level of 3 mm

σ = 1.5 mm (dose blurring) Σ= 1.5 mm (shift)

3 mm

1.5 mm

R=3 mm

rebuild CM

Mechanical tracking errors

Penumbra: σ p =4.05mm β=0.73 (for 80% isodose)

marker-TV surrogate uncertainty

Gimbals systematic error

Correlation model Fit errors ( σ = 1 mm)

σ = 0.5 mm

Σ=CT-slice/2=1 mm

Σ= 0.4 mm

M=2.5*√(( 1mm ) 2 +( 1.5mm ) 2 +( 0.4mm ) 2 )+0.73*√(( 1.5mm ) 2 +( 0.5mm ) 2 +( 1mm ) 2 +(4.05mm) 2 )-0.73*4.05mm

-surrogate vs TV relative rotation in “ relative ITV ”

Based on Van Herk et al. 2000

CTV-PTV margin calculation

Prediction error tolerance level of 3 mm

σ = 1.5 mm (dose blurring) Σ= 1.5 mm (shift)

3 mm

1.5 mm

R=3 mm

rebuild CM

Mechanical tracking errors

Penumbra: σ p =4.05mm β=0.73 (for 80% isodose)

marker-TV surrogate uncertainty

Gimbals systematic error

Correlation model Fit errors ( σ = 1 mm)

σ = 0.5 mm

Σ=CT-slice/2=1 mm

Σ= 0.4 mm

M=2.5*√(( 1mm ) 2 +( 1.5mm ) 2 +( 0.4mm ) 2 )+0.73*√(( 1.5mm ) 2 +( 0.5mm ) 2 +( 1mm ) 2 +(4.05mm) 2 )-0.73*4.05mm =4.9 mm => 5 mm

Based on Van Herk et al. 2000

Patient specific intra/post-fraction QA

Gimbals position logging

“Redundant sources of tracking performance information”

Per fraction QA through combination of different information sources

Vero Tracking

kV Monitoring Imaging

EPID MV Imaging

Poels et al.

Patient specific intra/post-fraction QA

Poels et al.

Patient specific intra/post-fraction QA Beams-Eye-View Tracking error

Prob. [%] BEV Tracking error< 5mm

Average breathing amplitude [mm]

Site

99%

Patient 1

lung (3x18Gy/fr) lung (10x5Gy/fr) liver(10x5Gy/fr) liver (10x5Gy/fr) liver (10x5Gy/fr) lung (10x5Gy/fr)

10.8

100%

Patient 2

6.8

97%

Patient 3

11.5

100% 100%

Patient 4

6.3

Patient 5

14.1

93%

Patient 6

19.7

100%

Patient 7

lung (4x12Gy/fr)

23.7

93% 98%

Patient 8

lung (4x12Gy/fr)

17.9

Average

13.9

Poels et al, Radiother Oncol 2013

Impact of motion management

VERO SBRT Tracking

Korreman et al, Int J Radiat Oncol Biol Phys 2012

Conclusion

è Depending on tumor location, we have to correct for tumor motion è Different treatment strategies l Passive ITV or mid ventilation approach o All cases l Gating / Tracking o Fewer amount of patients will benefit of this approach o Depending on philosophy of department, tumor location

Conclusion

è Designer machines: Gadget or useful tool? l Latest machine incorporate the knowledge of the past l More robust way of treating patients

o Online verification on board o Patient specific treatment

Treatment strategy UZB

4D-CT

< 7 mm => PTV = ITV + 5 mm* ITV approach ≥ 7 mm => internal marker if no contra-indication PTV = GTV + 5mm Dynamic tracking •lesions < 1cm: 8mm PTV margin •lesions < 1cm: no Tracking

Courtesy of B. Engels, MD PhD

Acknowledgments

Motion compensation techniques

Classic

Motion encompassing (passive)

Active compensation

Coventional free-breathing

ITV

Mid-position

Gated at exhale

Tracking

Max. exhale

Time-weighted average position

Max. inhale

Courtesy of Wolfhaus et al IJROBP 2008

Motion compensation techniques

Prescription isodose

ITV

è The radiation beam does not necessarily need encompass the complete breathing amplitude l Broad beam penumbra in the lung tissue

Time spend at edges of “ITV” is short

l

l Dose loss at edges can be compensated for by higher doses at the centre

Radiother Oncol. 2011 Mar;98(3):317-22

A moving treatment beam

Medical linac full beam line

“ What parts of the beam line should move to create a moving beam? ” Dynamics of breathing/tracking: -Frequencies up to 30 Hz -Amplitudes of a few centimeters -Sub-milimeter accuracy Too heavy !!! (>>1000kg)

Electron Gun

RF Power

Accelerator

Target Primary collimator

Flattening filter

Jaws (X,Y)

“ Loose some of that weight ” -by compact design? -by omitting some heavy components?

MLC

A moving treatment beam… Robot

Medical linac “ reduced ” beam line

Accuray CyberKnife -Light and compact linac -Mounted on an robot ” -Both rotation and translations

Electron Gun

RF Power

Accelerator

Target Primary collimator

Cones or Circular Diaphragm

A moving treatment beam

Medical linac full beam line

Electron Gun

RF Power

Accelerator

Target Primary collimator

Flattening filter

Jaws (X,Y)

“ Loose some of that weight ” -by compact design? -by omitting some heavy components?

MLC

A moving treatment beam … Gimbaled Linac

Keeping the MLC/FF = Still very heavy (700kg)

Vero “ reduced ” beam line

Electron Gun

RF Power

Astronaut training: “ the gimbals chair ”

Accelerator

Target Primary collimator

Flattening filter

-Allows rotation of the linac/MLC assembly around the “ Center of mass ” -Only 2 DOF dynamic ( Pan/Tilt rotation ) -MLC and flattening filter remain

MLC

Conclusion

Linac

“ Safety margins incorporating motion ”

“ Gating ”

Linac

“ Exposure of large volumes of healthy tissue ”

-static beam -static couch -small beam -20-30% duty cycle “ Very long treatment ti es (duty cycle 20-30%) ”

-static beam -static couch -wide beam -100% duty cycle

Linac

“ Dynamic couch compensation ”

“ Tracking/Pusuit ”

-static beam “ Some parts of the achine have to move actively, responding real-time to tumor motion for counteracting or pursuit ” -dynamic couch -small beam ->90% duty cycle -dynamic beam -static couch -small beam ->90% duty cycle