Physics Aspects of SRS/SBRT Luke Rock Beacon Hospital 1
Physics Aspects of SRS/SBRT
Luke Rock
Beacon Hospital
1
2
Outline
• Programme Implementation
- equipment specification
- commissioning
- training
• Treatment Planning
- 4DCT
- treatment techniques
• Treatment Delivery
- respiratory management
- image guidance
- pre-treatment QA
• Quality Assurance
- end to end evaluation
- SOPs
- periodic QA
3
AAPM TG 42 1995 – SRS AAPM TG 76 2006 – Respiratory Motion
AAPM TG 142 2009 – Linac QA IPEM Report 103 2010 – Small Field Photon Dosimetry
ACR 2006
ACR 2009
AAPM 2010
ASTRO 2011
Guidance Documents
SABR: A Resource 2014
• Surgical analogy
• Full MDT support during treatment delivery
Multidisciplinary Team
4
TG 101 AAPM 2010; ASTRO 2011; ACR 2009
MDT
MDT
Radiation Oncologist
Medical Physicist
Radiation Therapist
Support
staff
• Standard Radiotherapy is generally safe
• May 2000 to Aug 2006: UK reported 181 incidents
40 per 100,000
{3 per 100,000 deemed clinically adverse}
• Introduction of a new technique challenges established
safety systems
5
SBRTSkills
Intensive
Complex
Technology
SBRT Programme Implementation
6
SBRT
Respiratory Motion
Treatment Planning
Treatment Delivery
Implementation – Equipment Acquisition
7
Excluding the pre-implementation planning - Total = 16-26 weeks
Commissioning – How long will it take?
8
• Treatment Planning System
- beam model creation and verification
- heterogeneity correction algorithm
• Image Guidance
- determine accuracy of IGRT system
• Respiratory management
- 4DCT commissioning
- gated treatment delivery
Commissioning – what’s needed?
• Small photon field data
• Very difficult measurement
- loss of lateral electronic equilibrium
- detector volume averaging
- detector positioning / orientation error
• High profile errors in Florida (77 pts),
France (145 pts), Minnesota (152 pts)
9
TG 101 AAPM 2010
Commissioning – Beam Data Acquisition
• Use appropriate detectors
• Beware of diode detector energy dependence when moving
from large to small fields
• <2cm diode; 2-4cm diode + chamber; >4cm chamber
• Compare your data to other centres with same equipment
• Large differences may be observed (up to 30%)
10
Measurement Detector
Depth Dose / TPR Diode /Ion chamber
Beam Profiles Diode /Ion chamber
Relative Output Diode /Ion chamber
Absolute Dose Ion chamber
Commissioning – Beam Data Acquisition
11
0.4
0.6
0.8
1
0 1 2 3 4
Ou
tpu
t F
acto
r
Cone Diameter (cm)
Cone Output Factor Comparison
SRS Diode
SHY
GBD
8% difference
Commissioning – Data Comparison Example
• Need to account for motion due to respiration – 4DCT
• Two types of 4DCT:
• Regardless of 4DCT technique, breathing cycle needs to be
regular and reproducible
Respiratory Motion Management – 4DCT
12
Prospective Retrospective
Image acquisition on selected part of
breathing cycle
Image acquisition across whole breathing
cycle
Standard size CT dataset Large CT dataset
No post-scan computation required CT data requires binning into phase bins
Limited breathing cycle information Full breathing cycle information
Prospective 4DCT
13
Respiration Waveform from RPM Respiratory Gating System
CT Scan
Axial scan trigger,
1st couch position
Axial scan trigger,
2nd couch position
Exhalation
Inhalation
Scan Scan Scan
Axial scan trigger,
3rd couch position
Prospective 4DCT
14
Conventional CT Image Gated CT Image
Images Courtesy Medical College of Virginia, Richmond VA
Tumor
Retrospective 4DCT
15
X-ray on
Exhalation
Inhalation
1st couch
position
2nd
couch
position
3rd couch
position
“Image acquired”
signal to RPM
system
(Ford 2003, Vedam 2003)
Respiration Waveform from RPM Respiratory Gating System
Retrospective 4DCT
16
4D Data and images courtesy
VUmc, Amsterdam, The Netherlands
80% isodose: volume 13 vs. 27 cc
20% isodose: volume 163 vs. 471 cc
4DCT Planning Volumes
17
Korreman et al 2012
4DCT Planning Volumes
18
Total Dose # Fractions Indications
25-34 Gy 1 Peripheral, small (<2cm) tumours,
>1cm from chest wall
45-60 Gy 3 Peripheral, >1cm from chest wall
48-50 Gy 4 Central, Peripheral, <4-5cm, <1cm
from chest wall
50-60 Gy 5 Central, Peripheral, <1cm from chest
wall
60-70 Gy 8-10 Central tumours
Treatment Planning – Fractionation Schemes
19
• Multiple series for Lung SABR – no consensus
OAR 1 Fraction 3 Fractions 4 Fractions 5 Fractions
Spinal Cord 14 Gy 18 Gy
(6 Gy/fx)
26 Gy
(6.5 Gy/fx)
30 Gy
(6 Gy/fx)
Esophagus 15.4 Gy 30 Gy
(10 Gy/fx)
30 Gy
(7.5 Gy/fx)
32.5 Gy
(6.5 Gy/fx)
Brachial Plexus 17.5 Gy 21 Gy
(7 Gy/fx)
27.2 Gy
(6.8 Gy/fx)
30 Gy
(6 Gy/fx)
Heart 22 Gy 30 Gy
(10 Gy/fx)
34 Gy
(8.5 Gy/fx)
35 Gy
(7 Gy/fx)
Great Vessels 37 Gy 39 Gy
(13 Gy/fx)
49 Gy
(12.3 Gy/fx)
55 Gy
(11 Gy/fx)
Airways 20.2 Gy 30 Gy
(10 Gy/fx)
34.8 Gy
(8.7 Gy/fx)
32.5 Gy
(6.5 Gy/fx)
Chest Wall 30 Gy 30 Gy
(10 Gy/fx)
30 Gy
(7.5 Gy/fx)
32.5 Gy
(6.5 Gy/fx)
Skin 26 Gy 30 Gy
(10 Gy/fx)
36 Gy
(9 Gy/fx)
40 Gy
(8 Gy/fx)
Treatment Planning – OAR Dose Constraints
20
NCCN Guidelines v3 (2012) based on RTOG 0618; 0813; 0915
Treatment Planning – Beam Arrangements
21
• Use multiple beams from many directions
• Typically 8/9 fields for 3DCRT/IMRT
• 2/3 partial arcs for VMAT
• Skin dose <30% of prescribed dose TG 101 AAPM 2010
Treatment Planning – Coplanar or Non-coplanar?
22
Non-coplanar Coplanar
Treatment Planning – Coplanar or Non-coplanar?
23
• Advantages
- lower lung dose
- improved distribution in axial plane
• Disadvantages
- complicated treatment (increase risk of error)
- longer treatment time
- potential collisions (gantry/couch/patient)
Treatment Planning – Calculation Algorithm
24
• Dose calculation in a challenging environment
– small field size
– target surrounded by low density tissue
• Heterogeneity correction is required- Pencil beam algorithms not sufficient
- Minimum requirement is an algorithm that accounts for 3D
scatter integration (e.g. convolution/superposition)
- Ideal algorithm will account for photon and electron transport
(e.g. Monte Carlo; LBTE)
TG 65 AAPM 2004
TG 101 AAPM 2010
TG 101 AAPM 2010
Treatment Planning – Calculation Algorithm
25
• Reported Inaccuracies:
- 1D correction up to 30%
- 2D (e.g. Pencil Beam) up to 10-15%
- 3D (e.g. Convolution/Superposition) up to 5%
- Transport equations (e.g. Monte Carlo, LTBE) up to 2%
• Know limitations of your algorithm!
Treatment Planning - Interplay effect
Li et al JACMP 2013
• Effect increases with tumour motion
• Predominant in high dose gradients
• Use gating or sufficient margins to reduce effect
Chen et al 2009;
Li et al 2013
Treatment Delivery - Immobilisation
27
BodyFix by Medical
Intelligence
Body Pro-Lok
by CIVCO
Stradivarius by
Q Fix
Treatment Delivery - Immobilisation
28
• IGRT is essential for SBRT
Treatment Delivery - Localisation
29
TG 101 AAPM 2010
TG 101 AAPM 2010; ASTRO 2011
• Determine the accuracy of your IGRT system
TG 101 AAPM 2010; ASTRO 2011; ACR 2009
• Physician led image review recommended
TG 101 AAPM 2010
• Patient localisation must be verified before each
treatment fraction
SRS/SBRT - Quality Assurance
30
• Two types of QA
- QA of individual components of system (Linac, IGRT, respiratory
gating, MLC, etc); usually performed frequently (e.g. daily, weekly)
- QA of whole treatment system (End-to-End evaluation);
usually performed annually
Quality Assurance – Winston-Lutz Test
31
• Measure of accelerator isocentre accuracy
• TG142 tolerance is avg 0.75mm (max 1mm)
• Commercial software available
Quality Assurance – Winston-Lutz Test
32
• Can the isocentre move?
• Yes
• Noted drift in couch
isocentre over 6 month
period
• Adjust accelerator position
by ~1.0mm
Imaging
Planning
LocalisationTreatment Delivery
Analysis
Quality Assurance – End to End Evaluation
33
Quality Assurance – End to End Evaluation
34
• Annually
• Radiological Physics Center, MD Anderson
• External, independent audit of treatment chain
Quality Assurance – TG 142 Summary
35
Conclusions
36
• SRS/SBRT are well established techniques
• Both present significant technical challenges
• Requires multiple complex technologies
- Image Guided RT
- Respiratory management
- Immobilisation
- Treatment planning
- Precise treatment delivery
• QA programme requires more resources than standard RT
• Comprehensive practice guidelines available