`
`Oncologya
`
`
`
`
`a
`
`Technologies
`Ii
`
`Radiation Oncology
`
`
`erm
`
`A.-L.Grosu
`
`eed a3
`
`ViewRay Exhibit 1011
`Page1 of 15
`
`ViewRay Exhibit 1011
`Page 1 of 15
`
`
`
`Wo trcanae ScHLece., PhD
`Professor, Abteilung Medizinische Physik
`in der Strahlentherapie
`Deutsches Krebsforschungszentrum
`Im Neuenheimer Feld 290
`
`69120 Heidelberg
`Germany
`
`THomas BorTFe.p, PhD
`Professor, Department of Radiation Oncology
`Massachusetts General Hospital
`30, Fruit Street
`Boston, MA O2114
`USA
`
`ANcA-LIGIA GRosuU,MD
`Privatdozent, Department of Radiation Oncology
`Klinikum rechts der Isar
`
`Technical University Munich
`Ismaningerstrasse 22
`#1675 Munchen
`
`Germany
`
`MEDICAL RaproLocy + Diagnostic Imaging and Radiation Oncology
`Series Editors: A. L. Baert: L. W. Brady + H.-P. Heilmann: M. Molls + K. Sartor
`
`Continuation of Handbuch der medizinischen Radiologie
`Encyclopedia of Medical Radiology
`
`Library of Congress Control Number: O04] 16501
`
`ISBN 3-540-00321-5 Springer Berlin Neidelberg New York
`ISBN 978-3-540-00321-2 Springer Berlin Heidelberg New York
`
`This work is sabject to copyright. All rights are reserved, whether the whole or part of the material is concerned,specifi-
`calky the: rights of tramslation, reprinting, ease of ilbastrarions, recitarions, broadcasting, reproduction on paiceotilin oo
`in any other way, and storage in data banks. Duplication of this publication or parts thereof is permitted only under the
`provisions of the German Copyright Law of September 9, 1965, in is current version, amd permissbon for use mast always
`be obtained from Springer-Verlag. Vielatioms are liable for prosecution under the German Copprighit Law.
`
`Springer |e part of Springer Sckenort Husiness Media
`hop)eww.epringeronline.com
`© Springer-Verlag Berlin Heidelberg 2006
`Printed in Germany
`The use of general descriptive names, trademarks, etc. in this publication does not inoply, even im the absence of a specific
`stabement, that such names are exempt fom the relevant protective laws and regulations and therefore free for general use.
`Prodect liability: The publishers canmot guarantee the accuracy ofany information about dosage and application contained
`in this book. In every case the user must check seach information by comsulting the relevant literature.
`Medical Editor: Drv. Ute Healmann, Heidelberg
`Teek Editar: Ursula M. Davia, Hetdhelbserg
`Prodection Editer: Kort ‘Teicknsann, Mauer
`Cover-Deesign and Typesetiing: Verlagsservice Teichmann, Manet
`
`Printed on acid-free paper=21/31$laq- 543210
`
`ViewRay Exhibit 1011
`Page 2 of 15
`
`ViewRay Exhibit 1011
`Page 2 of 15
`
`
`
`| New Technologies in 3D Conformal Radiation Therapy:
`Introduction and Overview
`WASCHDBGEDeb
`
`Basics of 3D Imaging ................5. bebe eee bee beeen ee
`
`7
`
`24D Reconstruction
`]. Hessen and D, STSEPANKOU. 0... cee eee eee ee ee eee eee es 9
`
`3 Processing and Segmentation of 3D Images
`G. Sakae and A. POMMERT ..2.ee beeen eee ee wel
`
`45D Visualization
`G. Savas and A.Pre 24
`
`5 Image Registration and Data Fusion for Radiotherapy Treatment Planning
`ML. Wessne and M. ROBB ARSON. 600 e cementite mene mhemem imeem 4d
`
`
`
`7 Clinical X-Ray Computed Tomography
`M. FAGHELRIBSS 00 seeeeee G7
`
`8 4D Imaging and Treatment Planning
`E. RIETZEL amd G. TY, CHEN 1. cceceeseeeeseeeeceteceneeseseseeeesosesees Bl
`
`9 Magnetic Resonance Imaging for Radiotherapy Planning
`LB. SCWAD oeoeee
`
`10 Potential of Magnetic Resonance Spectroscopy for Radiotherapy Planning
`
`A.PIBZRALL.ccsccccscsecasscsssscscssessssscscessssssosssssssssssssosesIG
`
`1] PET and PET/CT for Radiotherapy Planning
`SBS MUTIC . 4... c.ssecs babe eee dee bee ee ee beeen dee al
`
`12 Patient Positioning in Radiotherapy Using Optical Guided
`3D-Ultrasound Techniques
`W. A. Tomi, §. L. Meexs, N. P. Orton, L. G, BoucHet, and M. A. RITTER ...... 15]
`
`ViewRay Exhibit 1011
`Page 3 of 15
`
`ViewRay Exhibit 1011
`Page 3 of 15
`
`
`
`XIE
`
`Contents
`
`3D Treatment Planning for Conformal) Radiotherapy.......................... 165
`
`Definition of Target Volumeand Organs at Risk.
`Biological Target Volume
`AcL. Guoay, L. D, Spaacue. ad M, MOLLG pcp eet ieee terse ereeene ewe reeenes DEF
`
`Virtual Therapy Simulation
`BSNce ciieiedicece ie caerete toto
`
`Dose Caleulation Algorithms
`LL. ORLPKE ANd C.SCHOLE occ t ete tented EE
`
`13
`
`M 1
`
`5
`
`Monte Carlo Dose Calculationfor Treatment Planning
`Jo
`Iraaae
`
`17,
`
`Optimization of Treatment Plans, Inverse Planning
`TRcomeia eece de a cea
`
`it
`
`1g
`
`Biological Modelsin TreatmentPlanning
`MPAa hc ee nc il cehl
`
`2D and 3D Planning in Brachytherapy
`D.BALTAS and N. ZAMBOGLOU o.0 645s sce sersenennnes bch Aa add dhe 17
`
`New Treatment Techniques .. 2... 0. ee ee eee eee eee ees DIT
`
`20
`
`Zl
`
`a2
`
`24
`
`25
`
`26
`
`Beam Delivery in 3D Conformal Radiotherapy Using Multi-Leaf Collimators
`W. ScHioe.,K. H, Guossen ,PMAmiwo, and B.RHEIN 21.pesetpeteeee es 28F
`
`Stereotactic Radiotherapy/Radlosurgery
`A-L. Grosu, P. EwescHaumen, and W. SCHLEGEL ....2..0..0000ccecsea eases DOF
`
`Extracranial Stereotactic Radiation Therapy
`WRFeneee
`
`X-MRT
`5, Niet, R. Hinpemee and U. OBLFRE .. 1+sseetersrsterteseesteetercesrerses BMD
`
`Control of Breathing Motion:
`TechniquesandModels(GatedRadiotherapy)
`T..D.S0neenc, N. M. Wink, 5. E. Tews, 5. KernnsKn, GD,Huse,
`tNi
`
`Image-Guided/Adaptive Radiotherapy
`a Ai
`
`Predictive Compensation of Breathing Motion in
`Lung Cancer Radiosurgery
`A. SCHWEIEHARD aiid J.B. ADLER: ciccisccec ieee sees ieee ee aes aeaeed beans 337
`
`a7
`Proton Therapy
`H.PacanerttandT.Boareetnheeeeeeeeeeb ee be ee tee eeteeeeeM4
`
`ViewRay Exhibit 1011
`Page 4 of 15
`
`ViewRay Exhibit 1011
`Page 4 of 15
`
`
`
`(Coneabe
`
`28
`
`29
`
`30
`
`Heavy lon Radiotherapy
`O, JAMEL. ..,...
`
`XII
`
`-. 365
`
`Permanent-[mplant Brachytherapy in Prostate Cancer
`Mi ZAID ER. coe s cence eee eeee
`
`Vascular Brachytherapy
`B. PoRRAIAG, E. Mian. GC. Kinsitts and Ba POTTER as ese ceener cere eon eran OOP
`
`3) Partial Breast Brachytherapy After Conservative Surgery for
`EarlyBreastCancer: Techniques and Results
`¥. Be.cactma,|.-M. Hanmoun-Livi, and BE. LA@TIGAU,. 10: ccceee secs eeeee 497
`
`Verificationand QA.. SeeeeSeeeeeeeeeee +++409
`
`32_3D Quality Assurance (QA) Systems
`B, Ragin and P. HARING whey eee eaees pusertereranes peteraeerpererrrereeres 41)
`
`43 Quality Management in Radiotherapy
`GHAROMAeeeee eee!435
`
`Subject Index. usceeveserceceneaeevaees vueeveeneenaee Jeunseeee breed beuenons 449
`
`OUTike cod irraeta aiaee
`
`ViewRay Exhibit 1011
`Page 5 of 15
`
`ViewRay Exhibit 1011
`Page 5 of 15
`
`
`
`Beam Delivery in 3D Conformal Radiotherapy Using Multi-Leaf Collimators
`
`257
`
`20 Beam Delivery in 3D Conformal Radiotherapy
`Using Multi-Leaf Collimators
`
`W. Schlegel, K.H. Grosser, P. Häring, and B. Rhein
`
`CONTENTS
`
`Conformal Treatment Techniques 257
`20.1
`Multi-Leaf Collimators 258
`20.2
`20.2.1 Geometrical and Mechanical Properties 258
`Physical Properties 262
`20.2.2
`20.2.2.1 Focusing Properties and Penumbra 262
`20.2.2.2 Interleaf Leakage 263
`20.2.2.3 Leaf Transmission 263
`20.2.3 Operating Modes 265
`Commercial MLCs 265
`20.3
`Linac-Integrated MLCs 265
`20.3.1
`20.3.2 Accessory-Type MLCs 265
`20.4
`The Limits of Conventional Conformal Radiation
`Therapy 265
`References 265
`
`20.1
`Conformal Treatment Techniques
`
`Conformational radiation therapy (CRT) was intro-
`duced in the early 1960s by radiation oncologist S.
`Takahashi, who came up with many ideas of how to
`concentrate the dose to the target volume using vari-
`ous forms of axial transverse tomography and rotat-
`ing multi-leaf collimators (MLC; Takahashi 1965).
`Three-dimensional conformal radiotherapy (3D
`CRT) is an extension of CRT by the inclusion of 3D
`treatment planning and can be considered to be one
`of the most important advances in treating patients
`with malignant disease. It is performed in nearly all
`modern radiotherapy units.
`The goal of 3D CRT is the delivery of a high radia-
`tion dose which is precisely conformed to the target
`volume while keeping normal tissue complications at
`a minimum.
`
`W. Schlegel, PhD; P. Häring, PhD; B. Rhein, PhD
`Abteilung Medizinische Physik in der Strahlentherapie,
`Deutsches Krebsforschungszentrum, Im Neuenheimer Feld
`280, 69120 Heidelberg, Germany
`K. H. Grosser, PhD
`Radiologische Klinik der Universität Heidelberg,
`Abteilung Strahlentherapie, Im Neuenheimer Feld 400,
`59120 Heidelberg, Germany
`
`The preconditions which have to be fulfi lled in
`order to achieve conformal dose distributions are
`discussed in the preceding chapters (Chaps. 2–12)
`on imaging and treatment planning (Chaps. 13–19).
`In summary, it can be said that fi rst of all detailed
`diagnostic imaging information has to be available
`from a variety of sources including conventional X-
`ray imaging, CT, MRI and PET in order to be able to
`defi ne the target volume and the organs at risk with
`suffi cient accuracy. Furthermore, a 3D computerized
`treatment planning system and an exact and repro-
`ducible patient positioning system have to be used.
`If these boundary conditions are fulfi lled, conformal
`irradiation techniques can be used optimally.
`In general, the attainable dose conformity in con-
`ventional conformal radiation therapy depends on the
`boundary conditions described in Table 20.1. As is seen
`from this table, there are many approaches to confor-
`mal therapy using sophisticated irradiation techniques
`with multiple isocentric beam irradiations, irregularly
`shaped fi elds (either using cerrobend blocks or MLCs),
`and computer-controlled dynamic techniques).
`An important step in 3D CRT was the introduc-
`tion of irregularly shaped irradiation fi elds, made
`of metal blocks from alloys with low melting points
`(in radiotherapy often called “cerrobend” blocks).
`Individually shaped irregular fi elds realized by cer-
`robend blocking turned out to be time-consuming
`and expensive; therefore, great progress in conformal
`radiotherapy was achieved by the development and
`application of MLCs. The dose distributions achieved
`with MLCs turned out to be equivalent to conformal
`blocks; however, the cost of conformal radiation
`therapy could be minimized and the fl exibility sig-
`nifi cantly enhanced by using the new MLC technol-
`ogy (Adams et al. 1999; Foroudi et al. 2000).
`The MLCs are beam-shaping devices that consist
`of two opposing banks of attenuating leaves, each of
`which can be positioned independently. The leaves
`can either be moved manually or driven by motors to
`such positions that, seen from the “beam’s eye view”
`of the irradiation source, the collimator opening cor-
`responds to the shape of the tumor (Fig. 20.1). The
`leaf settings are usually defi ned within the virtual
`
`ViewRay Exhibit 1011
`Page 6 of 15
`
`
`
`258
`
`W. Schlegel et al.
`
`Table 20.1 Irradiation techniques for conformal radiotherapy. MLC multi-leaf collimator
`
`Physical parameter
`
`Impact on conformity
`
`Scorecard
`
`Drawbacks
`
`No. of beam incidents
`
`Optimization of beam
`directions
`
`Optimization of beam
`energy (photons)
`Application of an MLC
`Width of the MLC leaves
`
`More than one target
`point at the same time
`
`Moving beam irradiations;
`computer-controlled
`dynamic radiotherapy
`
`Better conformity can be achieved
`with a higher number of irradiating
`directions
`Higher conformity possible
`
`Higher conformity possible
`
`Higher conformity possible
`Small leaf width enables a better fi eld
`adjustment and therefore better con-
`formity
`In cases where more than one target
`volume is treated simultaneously, con-
`formity might be higher
`Depends on shape of target volume
`
`++
`
`++
`
`+
`
`+++
`++
`
`+
`
`+
`
`Higher complexity, longer planning
`time, normal tissue dose becomes
`possibly larger, longer irradiation time
`Higher complexity, longer planning
`time; in case of non-coplanar beam
`directions: longer treatment time
`
`Higher complexity, longer planning
`time
`
`Higher complexity, longer planning
`time, sometimes a lower homogeneity
`
`Longer irradiation time, complex
`verifi cation and quality assurance
`
`20.2
`Multi-Leaf Collimators
`
`Multi-leaf collimators permit the quick and fl exible
`adjustment of the irradiation fi elds to the tumor shape
`and the shape of the organs at risk. Though already
`proposed by Takahashi in 1960, it took nearly 25 years
`before the fi rst commercial computer controlled MLCs
`appeared on the market. This was due to the fact that
`MLCs are mechanical devices with high mechanical
`complexity, and they have to fulfi ll very rigid technical,
`dosimetric, and safety constraints. Detailed reviews of
`the history and performance of MLCs for 3D CRT are
`given by Webb (1993, 1997, 2000). The use of MLCs for
`static or dynamic IMRT is discussed in more detail in
`another work by Webb (2005).
`This chapter describes briefl y the general design
`and performance of the MLCs as they are currently
`being used in routine applications for 3D CRT.
`
`20.2.1
`Geometrical and Mechanical Properties
`
`Fig. 20.1 Beams eye view of a planning target volume (PTV) in
`the brain, together with organs at risk (green: brain stem; red:
`eyes; blue: optic nerves) and MLC setting (yellow)
`
`therapy simulation program of 3D treatment plan-
`ning (see Chap. 14; Boesecke et al. 1988, 1991; Ésik
`et al. 1991).
`There were different other designs of MLCs with
`up to six leaf banks (Topolnjak et al. 2004), which,
`however, up to now have not played an important role
`in the clinical practice of 3D CRT.
`
`The most important technical parameters (Fig. 20.2)
`which characterize the performance of an MLC are
`mechanical and geometrical properties such as:
`1. The maximum fi eld size
`2. The leaf width
`3. Maximum overtravel
`
`ViewRay Exhibit 1011
`Page 7 of 15
`
`
`
`Beam Delivery in 3D Conformal Radiotherapy Using Multi-Leaf Collimators
`
`259
`
`4. Interdigitation
`5. Confi guration of the MLC with respect to the col-
`limator jaws
`
`For MLCs which are used for IMRT, other impor-
`tant parameters are also the minimum and maximum
`leaf speed and the precision of leaf positioning. Other
`aspects are of course the complexity of calibration
`and the overall efforts for maintenance.
`
`Fig. 20.2 Multi-leaf collimator with the most relevant mechani-
`cal parameters
`
`Maximum Field Size
`
`Two kinds of MLCs are employed presently: those for
`medium-sized and large fi elds of up to 40×40 cm2,
`
`Table 20.2 Commercial integrated MLCs
`
`which are implemented in the gantry of linacs; and
`“add-on-MLCs” for small fi eld sizes (often called
`mini- or micro-MLCs) which can be attached to the
`accessory holder of the treatment head, and, for ex-
`ample, used in conjunction with stereotactic confor-
`mal radiotherapy. Mini- and micro-MLCs have char-
`acteristic maximum fi eld sizes of about 10×10 cm2.
`Maximum fi eld sizes depend for some MLCs (see
`Tables 20.2, 20.3) from the maximum overtravel:
`when the maximum overtravel is used, maximum
`fi eld size will become smaller, because the whole leaf
`banks have to be shifted in order to achieve complete
`overtravel.
`
`Leaf Width
`
`MLCs integrated into the linac head. Computer-con-
`trolled MLCs integrated into the head of the accelera-
`tor usually have a spatial resolution of 0.5–1 cm in
`the isocenter plane, perpendicular to the leaf-motion
`direction, and a positioning accuracy in the range of
`1 mm in the direction of the motion.
`The leaf width (measured in the isocenter plane)
`should be adapted to the size and complexity of the
`target volumes. Maybe an effective leaf width of
`10 mm is completely suffi cient in case of prostate
`cancer; however, in the case of a small target volume
`located around the spinal cord, 10 mm is too large!
`A leaf width of 5 mm is presently considered to be a
`good compromise.
`
`Manufacturer Product
`name
`
`Leaf width at
`isocenter (mm)
`
`Midline over-
`travel (cm)
`
`No. of leaves Maximum fi eld
`size (cm2)
`
`Focusing prop-
`erties
`
`Remarks
`
`Elekta-1
`
`Elekta-2
`
`Siemens-1
`Siemens-2
`Siemens-3
`
`Varian-1
`
`Varian-2
`
`Varian-3
`
`Integrated
`MLC
`Beam modu-
`lator
`3D MLC
`Optifocus
`160 MLC
`
`Millennium
`MLC-52
`Millennium
`MLC-80
`Millennium
`MLC-120
`
`10
`
`4
`
`10
`10
`5
`
`10
`
`10
`
`Central 20 cm
`of fi eld: 5 mm;
`outer 20 cm of
`fi eld: 10 mm
`
`12.5
`
`11
`
`10
`10
`20a
`
`20a
`
`20a
`
`20a
`
`40×2
`
`40×2
`
`29×2
`41×2
`80×2
`
`26×2
`
`40×2
`
`60×2
`
`40×40
`
`16×22
`
`40×40b
`40×40
`40×40
`
`26×40
`
`40×40
`
`40×40
`
`Single focusing
`
`Single focusing
`
`Double focusing
`Double focusing
`Single focusing Announced
`for 2006
`
`Single focusing
`
`Single focusing
`
`Single focusing
`
`aRequires movement of the complete leaf bank and leads to reduced maximum fi eld sizes
`bThe Siemens 3D MLC consists of 2×27 inner leaves with 1-cm leaf width and two outer leaves with 6.5-cm leaf width
`
`ViewRay Exhibit 1011
`Page 8 of 15
`
`
`
`260
`
`W. Schlegel et al.
`
`Table 20.3 Commercial add-on MLCs (mini- and micro-MLCs)
`
`Company
`
`BrainLAB (m3) Radionics
`
`3D Line
`(Wellhöfer)
`
`Direx AccuLeaf
`
`Siemens
`(MRC)
`(cid:43)-MLC
`
`Siemens
`(MRC)
`Moduleaf
`
`40
`7.3×6.4
`1.4
`1.6
`<1
`1.5
`30
`
`40
`12×10
`5.5
`2.5
`<1
`3
`30
`
`24
`11×10
`2.5
`4.5
`0.5
`1
`30
`
`26
`10×10
`5
`3.0–5.5
`<4
`1.5
`31
`
`31
`10×12
`No data
`4.0
`<2
`2.5
`35
`
`No. of leaf pairs
`Field size (cm2)
`Overcenter travel (cm)
`Leaf width (mm)
`Leaf transmission (%)
`Maximum speed (cm/s)
`Clearance to isocenter
`(cm)
`Total weight (kg)
`Geometric design
`
`31
`Single focused
`
`38
`35
`Single focused Parallel
`
`35
`39.7
`Single focused Double focused
`
`36
`11×10
`No data
`No data
`<2
`1.5
`31
`
`27
`Two sets of leaf
`pairs at 90°
`
`a
`
`b
`
`Some of the commercially available MLCs have
`sections with various leaf widths. The central section
`of the leaf bank has a higher spatial resolution than
`the outer sections.
`Examples of integrated MLCs are shown in
`Fig. 20.3.
`Accessory-Type MLCs. The need for conformal,
`homogeneous dose distributions in connection with
`stereotactic radiotherapy and radiosurgery treat-
`ments was the motivation for the development of
`high-resolution MLCs for small fi eld sizes. These col-
`limators are attachable to the accessory holder of the
`linac (Schlegel et al. 1993, 1997; Debus et al. 1997).
`The leaf resolution is in the range of 1.5–4 mm (see
`Table 20.2).
`As examples for accessory MLCs used in stereo-
`tactic treatments, Fig. 20.4 shows the manual and
`the computer- controlled micro-MLCs developed at
`DKFZ (Heidelberg, Germany; Schlegel et al. 1992,
`1993, 1997).
`The Optimum Leaf Width of a MLC. In general,
`it seems evident, that the higher the spatial resolu-
`tion of the MLC, the better the quality of the result-
`ing dose distributions formed with such a MLC. This
`has empirically been shown using clinical treatment
`planning examples for irregularly shaped target vol-
`umes (Föller et al. 1998; Nill 2002). There is, how-
`ever, a defi nitive limit given as a physical constraint
`in principle: for a MLC with the penumbra p (=dis-
`tance between the 20 and 80% isodose produced by
`the leaf edge) a leaf width fi ner than p/2 does not lead
`to further improvement in the dose distribution. This
`was concluded by Bortfeld et al. (2000) according
`to sampling considerations. For a 6-mV beam, for in-
`stance, the optimum leaf width of a stereotactic add-
`
`Fig. 20.3 a Integrated MLC with a leaf width of 1 cm at the
`isocenter (the 3D-MLC from Siemens; see Table 20.2). b New-
`generation MLC: the 160 MLC from Siemens (see Table 20.2)
`
`on MLC with a penumbra of approximately 3 mm
`therefore is in the range of 1.5–2 mm. An integrated
`MLC is much closer positioned to the target and has
`a penumbra of 8–10 mm. The optimum leaf width is
`thus in the range of about 5 mm.
`
`ViewRay Exhibit 1011
`Page 9 of 15
`
`
`
`Beam Delivery in 3D Conformal Radiotherapy Using Multi-Leaf Collimators
`
`261
`
`Fig. 20.5 The problem of leaf interdigitation
`
`not be delivered with such an MLC. This is gener-
`ally not such an important issue for conventional
`conformal radiotherapy, but for IMRT applications,
`where many small and often complexly shaped seg-
`ments have to be delivered, such leaf “interdigitation”
`is often required.
`
`MLC confi guration in the treatment head.
`
`MLC confi gurations with respect to the rectangu-
`lar collimator jaws may be the following:
`1. Total replacement of the upper jaws
`2. Total replacement of the lower jaws
`3. Tertiary collimator confi guration
`
`The three main vendors of integrated MLCs haven
`chosen different confi gurations leading to different
`performances, especially for dosimetric properties as
`leakage and penumbra. The confi gurations are illus-
`trated in Fig. 20.6. (Accessory-type MLCs are always
`used in the tertiary confi guration, of course.)
`
`a
`
`b
`
`Fig. 20.4 a Accessory-type manual MLC with a leaf width of
`1.6 mm at the isocenter (DKFZ and Leibinger, GmbH; for de-
`tails see Schlegel et al.1992, 1993). b Accessory-type com-
`puter-controlled micro-MLC with a leaf width of 1.6 mm at
`the isocenter (DKFZ and SMS/OCS; see Table 20.3)
`
`Maximum Overtravel
`
`The overtravel characterizes how far a leaf can be
`moved over the midline of the MLC (Fig. 20.2). A large
`overtravel is important for very complexly shaped
`target volumes, but even more for the production of
`intensity-modulated fi elds in IMRT. Large overtravel
`is a mechanical challenge, because very long leaves
`are needed, which may lead to big weights and me-
`chanical guiding problems. Overtravel distances for
`commercial collimators are listed in Tables 20.2 and
`20.3. It has to be recognized that complete overtravel
`can only be realized in some collimators by moving
`a whole leaf bank (which is the case in all Varian
`MLCs and in the 160-MLC from Siemens). Moving
`the whole leaf bank of course reduces the maximum
`fi eld size.
`
`Interdigitation
`
`In some cases one leaf cannot pass an adjacent op-
`posing leaf without collision (Fig. 20.5).; thus, fi elds
`designed without considering such constraints can-
`
`Fig. 20.6 Principle of MLC confi gurations with respect to the
`upper and lower jaws for Elekta, Varian, and Siemens MLCs
`
`ViewRay Exhibit 1011
`Page 10 of 15
`
`
`
`262
`
`20.2.2
`Physical Properties
`
`20.2.2.1
`Focusing Properties and Penumbra
`
`The penumbra is an important design feature of a
`beam defi ning device. In order to obtain a steep dose
`gradient between the target volume and healthy tis-
`sue, the penumbra has to be as small as possible.
`First of all, penumbra depends on the position of
`the collimator relative to the source and the patient’s
`surface and on the diameter of the source. As a rule,
`in order to obtain a small penumbra, the source di-
`ameter has to be as small as possible (2–3 mm in
`modern linacs) and the distance between the source
`and the collimator as large as possible. On the other
`hand, the clearance between the patient and the ir-
`radiation head should be as large as possible in order
`to have the full fl exibility both for using accessories
`(block trays, wedges, or accessory MLCs) and to ap-
`ply non-coplanar beams. In that sense, a compromise
`has to be made between penumbra and clearance.
`Secondly, the penumbra also depends on the col-
`limator edges. In an MLC, in order to produce a small
`
`W. Schlegel et al.
`
`penumbra, the edges of the leaves must always be
`directed towards the source, independent of the leaf
`position. This property is called “focusing.”
`
`Focusing Perpendicular to the Leaf Motion Direction
`
`Good focusing properties are reached by trapezoid
`leaf cross sections which causes focusing in the direc-
`tion perpendicular to the leaf motion (Fig. 20.7a).
`
`Focusing in the Direction of the Leaf Motion
`
`Focusing in the leaf direction can be obtained by
`moving the leaves on a circular path or by rotating
`the leaf edges (see Fig. 20.7c, d; Pastyr et al. 2001).
`Both solutions are connected with engineering
`problems. That is why in most modern MLCs curved
`edges are being used which also give a reasonable
`penumbra (Fig. 20.7b). In the case of curved leaves
`penumbra is, however, not completely independent
`of the position of a leaf (Butson et al. 2003). The
`penumbra variation has to be implemented into the
`treatment planning systems. It also may complicate
`the delivery especially of small off-center segments
`in IMRT.
`
`a
`
`b
`
`c
`
`Fig. 20.7a-d. Focusing properties of MLCs: the leaves have trapezoid cross sections to perform focusing perpendicular to the
`direction of the leaf motion (a). Focusing in the direction of leaf motion can either be realized by leaves traveling on a circular
`path (b), rounded leaf edges (c), or rotating leaf edges (d)
`
`d
`
`ViewRay Exhibit 1011
`Page 11 of 15
`
`
`
`Beam Delivery in 3D Conformal Radiotherapy Using Multi-Leaf Collimators
`
`263
`
`The three main vendors of linear accelerators
`(Siemens, Elekta, and Varian) have implemented their
`MLCs at different distances from the source, and they
`also have different edge designs of the MLC leaves
`(Table 20.2). This leads to different performances and
`has to be considered when purchasing a new linear
`accelerator.
`From a dosimetric point of view, penumbra is nor-
`mally specifi ed as the distance between the 20 and 80%
`isodose line. If, in the worst case, the leaf movement
`has a direction of 45° to the isodose lines, the pen-
`umbra will become larger for leaves with bigger leaf
`width, because of the ribble which superimposes to the
`isodose lines (Fig. 20.8). Figure 20.9 shows dosimetric
`fi lm measurements for this 45° situation and a physical
`leaf width of 1, 2, and 3 mm. It can easily be recognized
`that penumbra is increasing with leaf width.
`
`P
`
`80%
`
`50%
`
`20%
`
`Fig. 20.8 Penumbra in the edge region of an MLC with leaf
`motion under 45° to the PTV boundary
`
`20.2.2.2
`Interleaf Leakage
`
`In order to avoid friction, there has to be small gap
`of about 0.1 mm between the leaves. This gap causes
`leakage radiation, which has to be minimized below a
`level of about 4%. This is especially a problem when
`the leaves have a trapezoid cross section for beam fo-
`cusing (Fig. 20.10b). To suppress interleaf leakage, the
`leaves are manufactured using a tongue-and-groove
`design (Fig. 20.10c). Another trick to reduce inter-
`leaf leakage is to slant the whole arrangement of the
`leaves with respect to the direction of the divergent
`rays (Fig. 20.10d).
`Interleaf leakage cannot be avoided completely by
`any of the abovementioned leaf designs. Figure 20.11
`shows dosimetric fi lm measurements with the typical
`spikes caused by interleaf leakage.
`
`20.2.2.3
`Leaf Transmission
`
`When high energy X-rays have to collimated, there is
`always a small fraction of X-rays which will penetrate
`through the jaws or leaves (Fig. 20.12). That is why
`high-Z material such as tungsten has to be used for
`the jaws or leaves. For tungsten, the thickness of the
`material has still to be in the range of 8–10 cm in
`order to reduce transmission below 1%.
`In general, the fraction of intensity transmitted
`through the collimators is higher in IMRT step-and-
`shoot treatments than in conventional treatments be-
`cause the treatment volume is irradiated with more
`fi eld components to reach the prescribed dose level.
`
`Fig. 20.9 Film dosimetry of quadratic
`fi elds (3×3 cm2) generated with 1-, 2-,
`and 3-mm leaves under 45° to the PTV
`boundary. The increase of penumbra
`with increasing leaf width can be rec-
`ognized
`
`ViewRay Exhibit 1011
`Page 12 of 15
`
`
`
`W. Schlegel et al.
`
`264
`
`a
`
`c
`
`b
`
`d
`
`Fig. 20.10a-d. Various MLC concepts to prevent leakage a unfocused without correction, b focused without correction, c with
`tongue and groove, and d with fl ipped collimator
`
`e
`
`f
`
`x
`
`I(0)
`
`I(d)=I(0) e-(cid:43)x
`
`Fig. 20.11 Leaf transmission for leaves with height X and trans-
`mission coeffi cient
`
`Fig. 20.12 Dose profi le measured with dosimetric fi lm under a
`completely closed accessory-type MLC. Leaf transmission and
`interleaf leakage can be detected
`
`At a rough guess, the transmitted intensity is
`twice the original physical level. If the transmission
`is, for example, 2%, the maximum of the transmit-
`ted intensity mounts up to 4% at some positions.
`Especially for MLCs used in IMRT, transmission as
`well as interleaf leakage should therefore be kept as
`low as possible.
`
`Restrictions for leakage radiation of MLCs are
`given in IEC (1998): If the MLC is covered by rectan-
`gular jaws, which are automatically adjusted to the
`MLC shape, leakage radiation must be below 5% of
`an open 10×10-cm2 fi eld; otherwise, maximum leak-
`age should be less than 2% and average leakage less
`than 0.5%.
`
`ViewRay Exhibit 1011
`Page 13 of 15
`
`
`
`Beam Delivery in 3D Conformal Radiotherapy Using Multi-Leaf Collimators
`
`265
`
`20.2.3
`Operating Modes
`
`There are in principle two different operating modes
`for MLCs for 3D CRT, depending on whether the
`leaves are moving when the beam is on (dynamic
`mode) or the beam is shut off (static mode).
`Although modern MLCs in principle have dynamic
`properties, they are still commonly applied in static
`treatment techniques. The real potential of MLCs in
`3D CRT will be demonstrated in the future, when,
`for example, dynamic arc treatment techniques with
`MLCs will be more widely accepted and implemented
`to perform dynamic fi eld shaping.
`The two modes (static mode=step-and-shoot
`technique; dynamic mode=sliding-window tech-
`nique) currently play a bigger role in IMRT delivery
`(see Chap. 23). In IMRT dynamic delivery is in many
`cases more time effi cient than step and shoot, but the
`step-and-shoot technique is less complex and qual-
`ity assurance may be easier to perform. Chui et al.
`(2001) compared the delivery of IMRT by dynamic
`and static techniques.
`
`20.3
`Commercial MLCs
`
`20.3.1
`Linac-Integrated MLCs
`
`The major manufacturers of commercial MLCs inte-
`grated into the irradiation head of a linear accelera-
`tor are the companies Elekta, Siemens, and Varian.
`Galvin (1999) provides a useful review with tables
`of MLC properties. Huq et al. (2002) compared the
`MLCs of all three manufacturers using precisely the
`same criteria and experimental methods. The result
`of this investigation was that there is no clear su-
`periority of one vendor compared with the others:
`the different designs and confi gurations of the MLCs
`are leading to a balance of advantages and disadvan-
`tages and there was no clear superiority of one MLC
`compared with the others. An updated list of MLC
`characteristics is given in Table 20.2.
`
`20.3.2
`Accessory-Type MLCs
`
`There are a couple of companies manufacturing and
`distributing accessory MLCs, which are especially
`
`suited to treat small target volumes in conjunction
`with stereotactic irradiation techniques. Bortfeld et
`al. (1999) have provided an overview on the charac-
`teristics and performances of these mini- and micro-
`MLCs. The specifi cations of these add-on collimators
`are summarized in Table 20.3. In summary, MLCs of
`this type are closer positioned to the patient’s surface
`and have smaller leaf widths. As a consequence, mini-
`and micro-MLCs have much smaller penumbras and
`produce dose distributions with higher conformity;
`however, they are restricted to the treatment of small
`target volumes.
`
`20.4
`The Limits of Conventional Conformal
`Radiation Therapy
`
`Complex-shaped target volumes close to radio-sensi-
`tive organs remain a challenge for the radiotherapist.
`With MLCs and conventional irradiation techniques
`conformal and homogeneous dose distributions can-
`not be obtained in all cases. This is particularly true
`for concave-shaped target volumes. In Chaps. 17 and
`23 (Inverse planning and IMRT) of this book it is
`shown that as the result of a superposition of several
`irradiation segments with homogeneous intensity, a
`concave-shaped dose distribution is produced. The
`MLCs to produce these IMRT fi eld segments have
`to fulfi ll very special criteria concerning leaf speed,
`accuracy, and reproducibility of leaf positioning,
`overtravel, transmission, and leakage. The specifi c
`requirements of MLCs in IMRT are discussed and
`analyzed in much more detail by Schlegel and
`Mahr (2000) and Webb (2005).
`
`References
`
`Adams EJ, Cosgrove VP, Shepherd SF, Warrington AP, Bedford
`JL, Mubata CD, Bidmead AM, Brada M (1999) Comparison
`of a multi-leaf collimator with conformal blocks for the
`delivery of stereotactically guided conformal radiotherapy.
`Radiother Oncol 51:205–209
`Boesecke R, Doll J, Bauer B, Schlegel W, Pastyr O, Lorenz WJ
`(1988) Treatment planning for conformation therapy using
`a multi-leaf collimator. Strahlenther Onkol 164:151–154
`Boesecke R, Becker G, Alandt K, Pastyr O, Doll J, Schlegel
`W, Lorenz WJ (1991) Modifi cation of three-dimensional
`treatment planning system for the use of multi-leaf col-
`limators in conformation radiotherapy. Radiother Oncol
`21:261–268
`Bortfeld T, Schlegel W, Höver KH (1999) Mini- and micro-mul-
`tileaf collimators. Med Phys 26:1094
`
`ViewRay Exhibit 1011
`Page 14 of 15
`
`
`
`266
`
`W. Schlegel et al.
`
`Bortfeld T, Oelfke U, Nill S (2000) What is the optimum leaf
`width of a multileaf collimator? Med Phys 27:2494–2502
`Butson MJ, Yu PK, Cheung T (2003) Rounded end multi-leaf
`penumbral measurements with radiochromic fi lm. Phys
`Med Biol 48:N247–N252
`Chui CS, Chan MF, Yorke E, Spirou S, Ling CC (2001) Delivery
`of intensity