Top Banner
1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements Team efforts from staff at Duke Radiation Oncology, especially to Dr. J Chang, Dr. J O’Daniel for providing slide information Technical and financial supports from Varian Medical Systems Clinical Implementation of VMAT Infrastructure/Installation Acceptance testing/commissioning Planning Delivery Quality assurance Fundamentals for RapidArc Fundamentals for VMAT Intensity Modulated arc therapy (VMAT) An arc-based approach to IMRT To be delivered on a conventional linear accelerator with a conventional MLC During an arc, the leaves of the MLC move and dose rate changes continuously as the gantry rotates RapidArc is one format of VMAT
22

Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

Oct 12, 2018

Download

Documents

vothuy
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

1

RapidArc: Clinical

Implementation

Fang-Fang Yin, PhD

Q. Jackie Wu, PhD

Duke University Medical Center

Acknowledgements

• Team efforts from staff at Duke Radiation

Oncology, especially to Dr. J Chang, Dr. J

O’Daniel for providing slide information

• Technical and financial supports from Varian

Medical Systems

Clinical Implementation of VMAT

• Fundamentals for RapidArc

• Infrastructure/Installation

• Acceptance testing/commissioning

• Planning

• Delivery

• Quality assurance

• Fundamentals for RapidArc

Fundamentals for VMAT

• Intensity Modulated arc therapy (VMAT)

– An arc-based approach to IMRT

– To be delivered on a conventional linear

accelerator with a conventional MLC

– During an arc, the leaves of the MLC

move and dose rate changes

continuously as the gantry rotates

• RapidArc is one format of VMAT

Page 2: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

2

The Principle of IMRT: Dose Painting

Beam ProfilePTV

OAR

Conventional 3-field RT Expected 3-field IMRT

PTV

OAR

Typical dose

distribution

Static and Rotational IMRT

One aperture

At each angle

VMAT

Multiple apertures

At each angle

Static gantry IMRT

One arc from 179o 181o

Count-clockwise

Rotational IMRTStatic Gantry IMRT

Field1 @180o

One of 7 fields

The Format of Cone-Beam IMRT Volumetric Rotational IMRT Options

• Existing Planning Systems• Eclipse (Varian) – Duke choice

• ERGO++/Monaco (Elekta)

• Pinnacle SmartArc (Philips)

• Prowess (Prowess)

• Existing Delivery Systems• RapidArc (Varian) – Duke choice

• VMAT (Elekta)

• Cone-beam Therapy (Siemens) (WIP)

• Existing QA Systems• Matrixx – Duke choice for routine QA

• Film – Duke choice for commission

• SunNuclear

• Delta 4 – Duke choice for future QA device

• Optical Scanner ……

Page 3: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

3

Where Are We (Duke)?

• Started investigation in June 2008

– A research RapidArc planning station from Varian

• Clinical installation in October

– Acceptance testing, commissioning, QA programs

– Single arc, no couch rotation, partial arc

• First patient treatment

– December 2008

• New versions in August 2009 and May 2011

– Allow multiple arcs, couch rotation, partial blocking, etc.

Clinical Implementation of RapidArc

• Fundamentals for RapidArc

• Infrastructure/Installation

• Acceptance testing/commissioning

• Planning

• Delivery

• Quality assurance

• Infrastructure/Installation

Infrastructure/Installation

• Staff (dedicated and trained)

• Existing machines:

– 21EX machine with 120-leaf millennium MLC

– NovalisTx with 120-leaf HD MLC (SRS, SRT,SBRT)

• ARIA version 8.6 or above (v10 now)

• Eclipse planning station (hardware and software)

• QA equipment

Clinical Implementation of RapidArc

• Fundamentals for RapidArc

• Infrastructure/Installation

• Acceptance testing/commissioning

• Planning

• Delivery

• Quality assurance

• Acceptance testing/commissioning

Page 4: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

4

Acceptance Testing

• Machine readiness

• Verification of installation against items included in the

purchase order

• Inspections of safety and quality of installation and

components

• VMAT performance

• Testing of functionality of each component and system

performance against specifications.

• End-to-end testing

• Dry-runs for a few test case from simulation to delivery

Acceptance Testing Sample

• Test 1.1: Gantry Angle Calibration

– Tolerance: + 0.5°

• Test 1.2: Isocenter Calibration

– Tolerance: + 1 mm

• Test 1.3: General Arc Dosimetry

– Range: 0.2 MU/° to 5.0 MU/°

– Tolerance: + 1%

Acceptance Testing Sample

• Test 1.4

• dMLC Dosimetry

• 0.5cm MLC slit sliding over 4 cm range

• Gantry:0°, 90°, 270°, 180°

• Tolerance: + 2% (over mean value)

Acceptance Testing Sample

• Test 2.1:

• Accuracy of dMLC position vs. gantry position

• Tolerance: + 1 mm

Page 5: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

5

Acceptance Testing Sample

• Test 2.2:

• Accuracy of dMLC position during arc

• Tolerance: + 1 mm

Acceptance Testing Sample

2.1: Picket Fence vs. Gantry Angle (static)

Acceptance Testing Sample

2.1: Picket Fence vs. Gantry Angle (static)

Acceptance Testing Sample

Page 6: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

6

Acceptance Testing Sample

• Test 2.3:

• Ability to accurately detect MLC position error

• Criteria: detect sub-millimeter error in position

Acceptance Testing Sample

2.3: Picket Fence with Errors

• Test 2.4:

• Accuracy of dose rate and gantry speed control during RapidArc

• Tolerance: + 2%

Acceptance Testing Sample

• Test 2.5:

• Ability to control leaf speed/position during RapidArc

• Tolerance: + 2%

Acceptance Testing Sample

Page 7: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

7

Acceptance Testing Sample

-200 -150 -100 -50 0 50 100 150 2001.00

1.02

1.04

1.06

1.08

1.10

1.12

1.14

1.16

1.18

1.20

Off x-axis Position (mm)

Rela

tive

Do

se

Off y-axis : -100 mmOff y-axis : 0 mmOff y-axis : 100 mm

Gantry speed

vs.

Dose rate

(Tolerance 2%)

Acceptance Testing Sample

-1.0

-0.8

-0.6

-0.4

-0.2

0.0

0.2

0.4

0.6

0.8

1.0

Perc

en

t D

evia

tio

n (

%)

-60 -40 -20 0 20 40 60

Off X-axis Position (mm)

Off y-axis: -100 mmOff y-axis: 0 mmOff y-axis: 100 mm

Variation of gantry speed and dose

rate (tolerance 2%)

Commissioning

• Validate that VMAT is capable of delivering

radiation beams as good as SG-IMRT could

• Define the limitations of planning optimization,

gantry rotation, beam blocking, couch rotation,

and leaf speed, collimator settings

• Develop treatment process and

documentation

Workflow for RapidArc Treatment

Planning/prescription

Treatment/validation

Quality assurance

Target localization

Immobilization/simulation

Case selection

Page 8: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

8

Clinical Implementation of RapidArc

• Fundamentals for RapidArc

• Infrastructure/Installation

• Acceptance testing/commissioning

• Planning

• Delivery

• Quality assurance

• Planning

Planning - Process

SG-IMRT RapidArc

Planning - Preparation

• Site-specific, for each site

– 10 cases from previous IMRT

– Develop RapidArc plan with different options

– Comparison between IMRT vs. RapidArc

• Constraint/optimization

– Optimization algorithm

– Constraints sensitivity

Planning - Strategy

• Selected site-specific planning strategy

– 1 arc: prostate only, prostate bed, prostate+SV,

brain lesions

– 2 arcs: prostate+SV+LN, spinal, brain lesions

– Multiple arcs: Head and neck, brain lesions, anal-

rectal

– Partial arcs: partial breast, liver, …

Page 9: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

9

Planning - Optimization

• Key challenge

– interconnectivity of the beam shapes within consecutive

VMAT gantry positions

• Constraints

– Target and normal structures

– Mechanicals

• Optimization algorithms

– More parameters

– Aperture based objectives and algorithms

Planning – Quality and Efficiency

• MLC segments

– Number of segments: IMRT RapidArc

– Plan quality is comparable if same segments

– More segments: slow gantry rotation for RapidArc

– More segments: long treatment time

• Planner experience

– Understand algorithm

– Understand limitations

Planning - Logistics

• Training – multiple people involved in the

planning

• Gaining experience

• Develop planning protocols

• Last ~ 3 months before patients start

Clinical Implementation of RapidArc

• Fundamentals for RapidArc

• Infrastructure/Installation

• Acceptance testing/commissioning

• Planning

• Delivery

• Quality assurance

• Delivery

Page 10: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

10

Delivery - Tolerances

• MLC leaf motion

• Leaf motion limit: 2.5 cm/s, 5 mm/degree

• Dose rate and gantry rotation speed

• One arc = one field

• MU, doserate, gantry speed linked

• Larger MU max doserate varying gantry speed

• Smaller MU max gantry speed varying doserate

• Middle varying gantry speed and doserate

Delivery – Sample Parameters

RapidArc Delivery Limits

• Variable gantry speed

– 0.5 – 5.6 degree/sec

• Variable dose rate

– 0 – 600 MU/min (0 -1000 Novalis Tx)

• Variable dose per degree

– 0.2 – 20 MU/degree

• Variable MLC speed

– 0 –2.5 cm/s

Delivery - Efficiency

• Plan quality

• The quality of plan number of apertures

• Single arc – 177 apertures

• Complexity structures:

• more apertures/arcs

• multiple arcs

• Delivery time

• Single arc – less time (< 2 mins) but sometimes

inferior quality

• Multiple arcs – better quality but longer time

Page 11: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

11

Delivery – Treatment Time

• Treatment time = Patient set-up time + Delivery time

• Delivery time=beam-on time+between beam=on time

• Patient setup time: no reduction

• Beam-on time: 20-60% reduction (less MUs)

• Between beam-on time:

• 100% saving for single arc

• 20-60% saving for multiple arcs

Delivery – Partial Arc

• Mechanical collision

• If isocenter is close to

center

• full arc

• If isocenter is close to

peripheral

• partial arc

• Partial blocking is

also available

Partial RapidArc for Liver SBRT

RapidArc MU = 714

IMRT Total MU = 1572

Time: 1.3 min

Multiple Arcs vs. Single Arc

Single arc

Multiple arcs

Page 12: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

12

VMAT For Large Size PTV

• Large field size -> sometimes IMRT beam has triple

beam splits

• Multi-sections (parts) of the PTV, large variation of

PTV shape, OARs and their constraints

• Beam orientation selection is part of IMRT planning,

is often not used for RapidArc (i.e. full arc)

VMAT For Large PTV

17cm Field Size << PTV in some

directions

3 Arcs, 1000 degree rotation

VMAT For Large Size PTV

IMRT VMAT17cm

VMAT For Large PTV

IMRT VMAT17cm

Page 13: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

13

VMAT For Large PTV

IMRT VMAT17cm

VMAT For Large PTV

VMAT26CM VMAT 17cm

VMAT For Large PTV

VMAT17cm VMAT23cm

Field Size Effect On VMAT

Planning Quality

100

110

120

130

140

150

160

0.0 5.0 10.0 15.0 20.0 25.0 30.0 35.0 40.0 45.0

D1%

(%

)

Projected Field Size X (cm)

PTV Hot Spot (D1%)

Collimator 30

Collimator 45

Page 14: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

14

VMAT For Head-and-Neck

Field Thru

Shoulder Field Thru

Shoulder

VMAT For Head-and-Neck

VMATVMAT IMRT

VMAT For Head-and-Neck

VMAT IMRT

RapidArc For Head-and-Neck

PharynxLt Parotid

Rt Parotid

Oral CavityCord

Page 15: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

15

VMAT For Head-and-Neck

Segments Thru Shoulder

Clinical Implementation of RapidArc

• Fundamentals for RapidArc

• Infrastructure/Installation

• Acceptance testing/commissioning

• Planning

• Delivery

• Quality assurance• Quality assurance

Quality Assurance

• The QA program for the VMAT is similar to SG-IMRT

in principle but with different measurement

approaches due to its dynamic nature that, during

VMAT delivery,

– MLC leaves are moving

– Gantry is rotating

– Dose rate is changing

Quality Assurance

• The QA program: validate the functionality

and performance of accepted features

• For each planned delivery

– Patient specific QA

– Machine specific QA

Page 16: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

16

Quality Assurance

• Machine specific QA

– accuracy of the MLC leaf positions during VMAT

delivery

– ability of the system to accurately vary the dose

rate and gantry speed during VMAT delivery

– ability of the system to accurately vary the MLC

leaf speed during VMAT delivery

• Tolerances: Acceptance baselines

Machine QA Chart

• Daily– Standard linac QA

– Standard MLC QA

– Rotational delivery of dose to an ion chamber phantom

• Monthly– Leaf motion

– Gantry rotation

– Dose output

• leaf motion

• gantry motion

• dose rate changes

Patient Specific QA

• Hybrid QA technique

– Plan to phantom

– Dose measurement to phantom

• Rotational nature

– Not to single plan

• Phantoms

• Instruments

– Ion chamber

– 2-D array (ion chamber, diodes, film, …)

Patient Specific QA

• Data analysis

– Multiple planes (axial, coronal, sagittal)

– Profiles

– Points

– Gamma analysis

• Collision check

– Before patient on the couch

– When patient on the couch

Page 17: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

17

RapidArc QA vs. IMRT QA

• More complex treatment delivery

– Varying gantry angle, gantry speed, dose rate, and

MLC leaf motion

• ImSure does not calculate dose for RapidArc

delivery

• Current IMRT technique: Portal dosimetry

• RapidArc technique: Ion chamber, film, and

Matrixx

QA Measurements

• Ion chamber– Absolute dose: (meas – calc) / calc < 3%

• Film (coronal, sagittal, and axial planes)– Optional

• Matrixx (coronal and sagittal planes)

• 3-6 hrs/patient

“Gold Standard” QA Tools

Ion ChamberFilm

Ion Chamber

• Equipment

– 0.13cc or 0.01cc ion

chamber

• Calibration

– 10x10cm2, 100SSD, depth

= dmax, 200MU

– cGy/nC correction factor

• RA delivery

– Center of 30cm x 30cm x

20cm solid water

phantom

– Compare to Eclipse

calculation

• 39 VMAT plans

Page 18: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

18

Film

• Equipment

– Kodak EDR2 film

– OmniPro ImRT

• Calibration

– 12 2cm x 2cm squares

– 0 – 300cGy

• RA delivery

– MultiCube (IBA dosimetry)

– Coronal, sagittal, and axial planes

– Compare to Eclipse calculation

– Gamma analysis

• 8 MAT plans

Verification with “Gold Standard” QA

• Ion chamber:

– Median: +1.7%

– Range: -0.9% - 2.8%

• Film

– 24/24 > 93% passing rate

– 23/24 > 95% passing rate

– 20/24 > 97% passing rate

Ion chamber array vs. Eclipse

3%, 3mm DTA, 5% threshold

Ion Chamber vs. Eclipse Result: Axial Film vs. Eclipse

Film

Eclipse

Page 19: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

19

Film vs. Eclipse

60

65

70

75

80

85

90

95

100

1 2 3 4 5 6 7 8

Plan #

% P

ixels

Pass

ing

Ga

mm

a (

<1

)

Coronal

Sagittal

Axial

14 15 23 24 32

3%, 3mm DTA, 5% threshold

Film vs. Eclipse

3%, 3mm, axial thresholds up to 40%

Film vs. Eclipse

60

65

70

75

80

85

90

95

100

1 2 3 4 5 6 7 8

Plan #

% P

ixels

Pass

ing

Ga

mm

a (

<1

)

Coronal

Sagittal

Axial

14 15 23 24 32

3%, 3mm DTA, 5% threshold Sagittal and Coronal, 40% threshold Axial

Film vs. Eclipse

Validation of 2D Ion Chamber Array

• MatriXX Evolution (IBA Dosimetry)

• 1020 ionization chambers– 0.07 cm3 sensitive volume

– 0.4cm diameter

• 24 x 24 cm2 grid

• 7.6mm spacing

• Automatic temperature-pressure correction

Herzen et al., PMB 2007; 52: 1197-1208

Angular Dependency:

-1.7% 15x

-1.8% 15x

+2.0% 15x

+2.2% 15x

+3.2% 15x-2.0% 6x

-3.5% 6x

+2.8% 6x

+3.0% 6x

+3.7% 6x

Fro

nt

Back

Measured/Calculated (%)

Future solution from IBA:

Gantry angle sensor

Apply correction

factor to each ion

chamber based on

angularity

Page 20: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

20

Matrixx QAMatrixx vs Eclipse

88

90

92

94

96

98

100

0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34

RapidArc (Plan #)

%P

ixels

Passin

g (

Gam

ma <

= 1

)

MatrixxCoronal

MatrixxSagittal

Results Patient Specific QA

3%, 3mm, axial thresholds up to 5%

Fig. 1: Ion Chamber vs. Eclipse and Matrixx vs. Eclipse

-6%

-4%

-2%

0%

2%

4%

6%

1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 39

Plan #

%D

iffe

ren

ce f

rom

Ecl

ipse

Ca

lcu

latio

n

IC

IC Array

BrainProstate

BedProstate

Prostate

+ SV Spine

Ion Chamber vs. Eclipse & ICA vs. Eclipse Film vs. Ion Chamber Array

88

90

92

94

96

98

100

0 1 2 3 4 5 6 7 8 9

Plan #

%P

ixe

ls P

ass

ing

Ga

mm

a (

<1

)

Coronal

Sagittal Gamma = 3%, 3mm, 5% threshold

Film vs. Ion Chamber Array

1 2 3 14 15 23 24 32

Page 21: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

21

Effective vs. Efficient

• Stage 1: Intensive QA

– Ion chamber, film in 3 planes, ion chamber array in

2 planes, 3D polymer gel dosimetry

• Stage 2: Rigorous QA

– Ion chamber, ion chamber array in 2 planes

• Stage 3: Effective and Efficient QA

– Ion chamber, ion chamber array in 1 plane

– Ion chamber array only

Preparation Delivery Analysis

Ion chamber – 1st 15 15 5

Film – 1st 15 20 20

Ion chamber array – 1st 15 15 5

Ion chamber – additional 15 7 5

Film – additional 15 10 10

Ion chamber array –

additional

15 7 5

IC + 3 Film + 2 ICA 90 77 55

IC + 2 ICA 45 37 15

IC + ICA 30 30 10

Effective vs. Efficient

• When implementing a new technology

– Perform intensive patient-specific QA for the first

group of patients

– Rely on QA “gold standards”

– 3D QA very useful

– Use “gold standard” QA technology to transition to

newer QA devices

• Goal: Effective QA

Effective vs. Efficient Conclusion

• RapidArc is one format of rotational IMRT for dose

painting

• Implementation of RapidArc requires careful

planning, testing, and verifications.

• Thoroughly testing and commissioning are

necessary prior to patient treatment

• QA is a critical step, always compare with static field

IMRT plan in the early phase

• RapidArc should be judged by its

accuracy, safety, efficiency, applicability, integration

, and adaptation

Page 22: Acknowledgements RapidArc: Clinical Implementation · 1 RapidArc: Clinical Implementation Fang-Fang Yin, PhD Q. Jackie Wu, PhD Duke University Medical Center Acknowledgements •Team

22

Thank you for your attention