MIONALJO
`
`Y•PHYSICS
`
`ISSN 0360-3016
`
`The Official Journal of the
`AMERICAN SOCIETY FOR THERAPEUTIC RADIOL.OGY AND ONCOLOGY
`
`Sponsored by the
`INTERNATIONAL SOCIETY CW RAOIATIO~
`CIKCUU) DE RADIOTERAPEUTAS IBERO-L\TI~
`
`Published by
`PERGAMON PRESS New York / Oxfor
`
`hd; ,,,r'l')-i11; .i0t·1;\J.
`.jour'l'"i,':l./. of
`r·,hi :;_,Jt: ii::.n i::incc.J.oqy
`::1h1/':i i C '.•;
`hi. C•.l i')IJ/''
`r::,.w Yr: M · i:·:t(H.11 1
`J:,t,11.
`UC ~i.:1n Oie1q,,
`Ht:·C'l:'.\ v,,•d , .. ,r,: 06·· l ;:, ... •:),:-:
`
`1
`
`Varian Exhibit No. 1015
`Page 0001
`
`

`

`INTERNATIONAL JOURNAL OF
`
`TION ONCOLOGY
`BIOWGY· PHYSICS
`
`VOLUME 23, NUMBER 2, 1992
`
`TABLE OF CONTENTS
`
`Editor's Note
`P. Rubin
`
`e CLINICAL ORIGINAL CONTRIBUTIONS
`
`Retrospective Analysis of 5037 Patients With Nasopharyngeal Carcinoma Treated During 1976-1985: Overall
`Survival and Patterns of Failure
`A. W. M. Lee, Y . F. Poon, W. Foo, S. C. K. Law, F. K. Cheung, D. K. K. Chan, S. Y. Tung, M. Thaw and
`J. H. C. Ho
`
`Carcinoma of the Nasopharynx: Factors Affecting Prognosis
`C. A. Perez, V. R. Devineni, V. Marcial-Vega, J. E. Marks, J. R . Simpson and N. Kucik
`
`The Predictive Role of Bioeffect Dose Models in Radiation-Induced Late Effects in Glottic Cancers
`S. M. Deare, S. J. Supe, V. Sharma and K. A. Dinshaw
`
`Survival FolJowing Locoregional Recurrence of Breast Cancer: Univariate and Multivariate Analysis
`K. J. Halverson, C. A. Perez, R.R. Kuske, D. M. Garcia, J. R. Simpson and B. Fineberg
`
`Comparison of Pathologic and Clinical Evaluation of Lymph Nodes in Prostate Cancer: Implications of RTOG
`Data for Patient Management and Trial Design and Stratification
`G. E. Hanks, J. M. Krall, M . V. Pilepich, S. 0. Asbell, C. A. Perez, P. Rubin, W . T . Sause and R . L. S. Doggett
`
`Tl and T2 Carcinoma of the Urinary Bladder: Long Term Results With External, Preoperative, or Interstitial
`Radiotherapy
`W. De Neve, M . L. M. Lybeert, C. G oar, M. A. Crommelin and J. G . Ribot
`
`Long-Term Results of Combined Modality Treatment With 1-125 Implantation for Carcinoma of the Pancreas
`M. Mohiuddin, F. Rosato, D . Barbot, A Schuricht, W . Biermann and R . Cantor
`
`lntracranial Ependymoma: Long-Term Results of a Policy of Surgery and Radiotherapy
`L. J. Vanuytsel, E. M. Bessell , s. E. Ashley, H . J. G . Bloom and M. Brada
`
`(Contents continued on page viii)
`
`259
`
`261
`
`27 1
`
`281
`
`285
`
`293
`
`299
`
`305
`
`3 13
`
`INDEXED IN Current Contents, B10S1S Oat.abase, Index Medicus, MEDLINE, Excerpta Medica, _Safety _Sci. Abstr., Energy Res. Abstr .. Energy Data Base, Toxicology
`Abstr .. Electronics & Commun. Abstr .. Compu ter & Info. Systems Abstr., C'ambndge Sci. Abstr., and CABS, PASCAL-CNRS Database
`
`ISSN 0360-3016
`(223)
`
`Varian Exhibit No. 1015
`Page 0002
`
`

`

`Pergamon Press Offices:
`U.S.A.: Pergamon Press, 660 White Plains Road, Tarrytown, NY 10591-5153, USA, INTERNET
`"PPI@PERGAMON.COM"
`
`U.K.:
`
`Pergamon Press, Headington Hill Hall, Oxford, OX3 0BW, England
`
`KOREA: Pergamon Press, K.P.O. Box 315, Seoul 110-603, Korea
`JAPAN: Pergamon Press, Tsunashima Building Annex, 3-20-12 Yshima, Bunkyo-ku, Tokyo I 13, Japan
`
`Editorial Office: Division of Radiation Oncology, Strong Memorial Hospital, 601 Elmwood Avenue, RocheSter, NY 14642, USA,
`(716) 275-5175.
`
`Publishing, Subscription and Advertising Offices: Pergamon Press Inc., 660 White ~lains Road, Tarrytown, NY 10591-5153, USA,
`INTERNET "PPI@PERGAMON.COM"; and Pergamon Press Ltd., Headington Hill Hall, Oxford OX3 0BW, England.
`
`Published 15 per annum (vols. 22-24). Annual Institutional Subscription Rate ( 1992)_: ~S $1 ,29_0.00 (~710.00). T wo-year lnstitut!ona/
`Subscription Rate (1992/93): US $2,451.00 (£1,349.00). US dollar prices are defimt1ve. Sterhng pnc~s are quoted for_ convemence
`only, and are subject to exchange rate fluctuation. Prices include postage and insurance and ar~ subJec_t to change without notice.
`Back issues of all previously published volumes, in both hard copy and on microform , are available dlfect from Pergamon Press.
`Subscription rates for Japan are available on request.
`
`Reprints of all articles are available from Pergamon Press in lots of 50, minimum order I 00 copies. Call for details.
`
`Copyright © 1992 Pergamon Press Ltd.
`
`It is a condition of publication that manuscripts submitted to this journal have not been published and will not be simultaneously
`submitted or published elsewhere. By submitting a manuscript, the authors agree that the copyright for their article is transferred to
`the publisher if and when the article is accepted for publication. The copyright covers the exclusive rights to reproduce and distribute
`the article, including reprints, photographic reproductions, microform or any other reproductions of similar nature and translations.
`No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means. electronic.
`electrostatic, magnetic tape, mechanical, photocopying, recording or otherwise, without permission in writing from the copyright
`holder.
`
`While every effort is made by the publishers and editorial board to see that no inaccurate or misleading data, opinion or statement
`appears in this journal, they wish to make it clear that the data and opinions appearing in the articles and advertisements herein are
`the sole responsibility of the contributor or advertiser concerned. Accordingly, the publishers, the editorial board and editors and
`their respective employees, officers and agents accept no responsibility of liability whatsoever for the consequences of any such
`inaccurate or misleading data, opinions or statement.
`
`Drug and dosage selection: The authors have made every effort to ensure the accuracy of the information herein, particularly with
`regard to drug selection and dose. However, appropriate information sources should be consulted, especially for new or unfamiliar
`drugs or procedures. It is the responsibility of every practitioner to evaluate the appropriateness of a particular opinion in the context
`of actual clinical situations and with due consideration to new developments.
`
`U.S. COPYRIGHT LAW APPLICABLE TO USERS IN THE U.S.A.
`
`Photocopying information for users in the U.S.A. The Item-Fee Code for this publication indicates that authorization to photocopy
`items for internal or personal use is granted by the copyright holder for libraries and other users registered with the Copyright Clearance
`Center (CCC) Transactional Reporting Service provided the stated fee for copying, beyond that permitted by Section 107 or 108 of
`the United States Copyright Law, is paid. The appropriate remittance of $5.00 per copy per article is paid directly to the Copynght
`Clearance Center Inc., 27 Congress Street, Salem, MAO 1970.
`
`Permission for other use. The copyright owner's consent does not extend to copying for general distribution, for promotion, for
`creating new works, or for resale. Specific written permission must be obtained from the publisher for copying. Please contact the
`Subsidiary Rights Manager, Publishing Services Dept. at either Pergamon Press Ltd. or Pergamon Press Inc.
`The Item-Fee Code for this publication is: 0360-3016/92 $5.00 + .00.
`
`Discl~imer-Th_e Americ~n Society for Therapeutic ~adiolog~ and Oncology assumes no responsibility for the accuracy of any editfial
`matenal contained m this Journal, and such ed1tonal matenal does not represent official policy or recommendations of the Soc y
`r value
`· h' J
`· ·
`I d
`·
`·
`f d
`Th
`e appearance o a vert1smg m t ts ourna oes not constitute a guarantee or endorsement by the Society of the quality 0
`of any advertised products or services or of the claims made for them by the advertisers.
`
`IN_TERNATIONAL JOURNAL_OF RA~I_ATION ONCOLOGY, BIOLOGY, PHYSICS (ISSN 0360- 3016). Second-class~~~:::.
`paid at Elmsford, NY and additional mailing offices. Postmaster: send address changes to Int'!. J. Radiation Oncology, 8
`Physics, Subscription Dept. , Pergamon Press, 660 White Plains Road, Tarrytown, NY 1059 J-5153.
`
`Printed in the United States of America.
`
`Varian Exhibit No. 1015
`Page 0003
`
`

`

`( Contents continued)
`
`tactic Interstitial Brachytherapy With lod.'
`• St
`Patterns of Recurrence of Malignant Astrocytoma Following
`ereo
`ine-
`125 Implants
`C. B. Agbi. M. Bernstein, N. Laperriere, P. Leung and M. Lumley
`
`Prognostic Variables in Malignant Epithelial Tumors of the Parotid
`M. G. Poulsen, G. R. Pratt, B. Kynaston and L. B. Tripcony
`e BIOLOGY ORIGINAL CONTRIBUTIONS
`
`Subcutaneous or Topical Administration of 16,16 Dimethyl Prostaglandin E2 Protects From Radiation-Induced
`Alopecia in Mice
`W.R. Hanson, A. E. Pelka, A. K. Nelson and F. D. Malkinson
`
`Radiosensitivity Testing of Human Primary Brain Tumor Specimens
`M. J. Allalunis-Turner, G. M. Barron, R. S. Day, Ill, D.S. Fulton and R. C. UrtaSun
`
`Differences in Radiation Response Between Cells in S-Phase and Non-S-Phase Cells of the Granulocyte/Mac(cid:173)
`rophage Progenitor (GM-CFC) Compartment
`W. Nothdurft, K. Baltschukat and C. Selig
`
`Effects of 5'-lododeoxyuridine on the Repair of Radiation Induced Potentially Lethal Damage lnterphase Chro(cid:173)
`matin Breaks and DNA Double Strand Breaks in Chinese Hamster Ovary Cells
`Y. Wang and G. Iliakis
`
`Enhancement of Tumor Radiosensitivity and Reduced Hypoxia-Dependent Binding of a 2-Nitroimidazole With
`Normobaric Oxygen and Carbogen: A Therapeutic Comparison With Skin and Kidneys
`A. Rojas, M. C. Joiner, R. J. Hodgkiss, U. Carl, E. Kjellen and G.D. Wilson
`
`e PHYSICS ORIGINAL CONTRIBUTIONS
`
`32t
`
`327
`
`333
`
`339
`
`345
`
`353
`
`361
`
`Beam's Eye View Volumetrics: An Aid in Rapid Treatment Plan Development and Evaluation
`L. C. Myrianthopoulos, G. T. Y. Chen, S. Vijayakumar, H.J. Halpern, D. R. Spelbring and C. A. Pelizzari
`
`367
`
`e HYPERTHERMIA ORIGINAL CONTRIBUTIONS
`
`Use of Nitroprusside to Increase Tissue Temperature During Local Hyperthermia in Normal and Tumor-Bearing
`Dogs
`D. M. Prescott, T. V. Sarnulski, M. W. Dewhirst, R. L. Page, D. E. Thrall, R. K. Dodge and J. R. Oleson
`
`377
`
`e PHASE //II CLINICAL TRIALS
`
`Phase 1/11 Trial of Pre-Operative Radiation Therapy and Coloanal Anastomosis in Distal Invasive Resectable
`Rectal Cancer
`B. D. Minsky, A. M. Cohen, W. E. Enker and E. Sigurdson
`
`Hyperfractionation in Advanced Carcinoma of the Uterine Cervix: A Preliminary Report
`C. Varghese, F. Rangad , C. C. Jose, P. Raveendran, J. Subhashini, K. Ramadas, J. ldicula, A. Pais and R. K.
`Raul
`e BRIEF COMMUNICATIONS
`
`387
`
`393
`
`High Dose Endobronchial Irradiation in Recurrent Bronchogenic Carcinoma
`M. Gauwitz, N. Ellerbroek, R. Komaki, J. B. Putnam, Jr. , M. B. Ryan, L. DeCaro, M. Davis and J. Cundiff
`
`7
`
`39
`
`(Contents continued on page x)
`
`Varian Exhibit No. 1015
`Page 0004
`
`

`

`( Contents continued)
`
`Interstitial Pneumonitis After Hyperfractionated Total Body Irradiation in HLA-Matched T-Depleted Bone
`Marrow Transplantation
`P. Latini, C. Aristei, F. A versa, F. Checcaglini, E. Maranzano, B. M. Panizza, E. Perrucci , A. Carotti and M. F.
`Martelli
`
`Radiotherapy During Pregnancy for Clinical Stages IA-IIA Hodgkin's Disease
`S. Y. Woo, L. M . Fuller, J. H. Cundiff, M. L. Bondy, F. B. Hagemeister, P. McLaughlin, W. S. Velasquez,
`F. Swan, Jr., M. A. Rodriguez, F. Cabanillas, P. K. Allen and R. J. Carpenter, Jr.
`
`• TECHNICAL INNOVATIONS AND NOTES
`
`Stereotaxic Radiosurgery for Brain Metastases: The Importance of Adjuvant Whole Brain Irradiation
`B. G. Fuller, I. D . Kaplan, J . Adler, R. S. Cox and M.A. Bagshaw
`
`The Objective Evaluation of Alternative Treatment Plans III: The Quantitative Analysis of Dose Volume His(cid:173)
`tograms
`D. A. Viggars, S. Shalev, M. Stewart and P. Hahn
`
`Vertex Field Verification in the Treatment of Central Nervous System Neoplasms
`S. A. Reisinger, J. Palta and L. Tupchong
`
`Tangential Breast Irradiation: Simple Improvements
`P. Cross, D. J. Joseph, J. Cant, S. G. Cooper and J. W. Denham
`
`Mantle Irradiation in the Upright Position: A Technique to Reduce the Volume of Lung Irradiated in Patients
`With Bulky Mediastinal Hodgkin's Disease
`K. C. Marcus, G. Svensson, L. P. Rhodes and P. M. Mauch
`
`Sagittal Magnetic Resonance Imaging in the Design of Lateral Radiation Treatment Portals for Patients With
`Locally Advanced Squamous Cancer of the Cervix
`A. H . Russell, J.P. Walter, M. W. Anderson and C. L. Zukowski
`
`e ONCOLOGY IN TELLIGENCE
`
`Loss of Local Control With Prolongation in Radiotherapy
`J. F. Fowler and M. J . Lindstrom
`
`e EDITORIALS
`
`Trends in the Clinical Management of Nasopharyngeal Carcinoma
`L. Taifu
`
`The Treatment of Local-Regional Recurrence of Carcinoma of the Breast After Mastectomy
`L. J. Solin
`
`(Contents continued on par,e xii)
`
`401
`
`407
`
`413
`
`419
`
`429
`
`433
`
`443
`
`449
`
`457
`
`469
`
`473
`
`Varian Exhibit No. 1015
`Page 0005
`
`

`

`(Contents continued)
`
`e CORRESPONDENCE
`
`One Versus Two or More Brachytherapy Applications in Cervical Cancer
`B. J. Smit
`
`High Energy Photon Irradiation of the Olfactory Mucosa in Humans
`S. A. Costello, C. J. Wynne, A. Faid and M. J. Gray
`
`Interstitial Implantation for Base of Tongue Carcinoma
`B. Esche, J. Crook, L. Grimard and P. Genest
`e MEETINGS
`
`477
`
`477
`
`477
`
`479
`
`Varian Exhibit No. 1015
`Page 0006
`
`

`

`Vol. 23. PP. 419-427
`Onrok~~y Bwl Phw
`In!. J Rndmrwn
`Printed I” the U.S.A. All rghts resewed
`
`.oO
`$5.00 +
`0360-3016/92
`Copyright 0 1992 Pergamon Press Lid.
`
`??Technical Innovations and Notes
`
`TREATMENT PLANS III:
`THE OBJECTIVE EVALUATION OF ALTERNATIVE
`THE QUANTITATIVE
`ANALYSIS OF DOSE VOLUME HISTOGRAMS
`
`PH.D.,‘,~
`SHALEV,
`PH.D.,‘%~ SHLOMO
`DAVID A. VIGGARS,
`MARGARET
`STEWART,
`R.T.R.
`AND PER HAHN, M.D.’
`
`100 Olivia Street, Winnipeg, R3E 0V9, Manitoba, Canada; and
`Manitoba Cancer Treatment and Research Foundation,
`Departments
`of ‘Radiology and 2Physics, University
`of Manitoba, Winnipeg, Manitoba, Canada
`
`in a consistent way for use in 3-dimensional
`The computer program OSCAR evaluates dose-volume histograms
`treatment planning. Based on a dose prescription specified by a radiation oncologist,
`the technique provides a
`quantitative and easily understood visual analysis of a proposed dose distribution. Rapid, reliable, and consistent
`choices can be made between alternative treatment plans, and if necessary the results of OSCAR calculations can
`be used to guide the design of a plan that will be closer to the required prescription. The method is well suited to
`use in the definition of treatment protocols. The use of OSCAR
`is demonstrated by applying
`it to the evaluation
`of alternative volumetric treatment plans for ca lung. The results demonstrate the importance of using corrections
`for inhomogeneous
`tissue density in the calculation of 3-dimensional dose distributions.
`
`Radiation treatment planning, Dose volume histograms, Score functions, Density correction, Normalization, Protocol,
`Dose prescription.
`
`INTRODUCTION
`
`in three
`plans
`treatment
`of alternative
`The comparison
`is a for-
`isodose charts
`dimensions
`using only
`traditional
`superim-
`midable
`task. By examining
`isodose contours
`posed on CT images,
`the radiation
`oncologist must decide
`whether
`the target will receive an adequate dose through-
`out
`its volume,
`and whether
`unacceptable
`volumes
`of
`healthy
`tissue will receive high radiation
`doses. Since each
`plan will have many CT slices and
`there will be several
`plans
`to compare,
`this is clearly a time consuming
`task
`and one that
`is very difficult
`to perform
`objectively
`and
`consistently.
`(DVH) are a convenient way
`Dose volume histograms
`of summarizing
`the information
`in a 3-dimensional
`dose
`distribution.
`Their convenience
`is achieved by excluding
`detailed positional
`information
`about
`the location of dose
`levels within
`the region under consideration,
`and conse-
`quently
`they cannot
`entirely
`replace other means of dis-
`playing
`the dose distribution
`such as isodose charts and
`images of regret (22) which retain positional
`information.
`However, DVH’s are extremely
`useful
`in the initial stages
`of comparing
`and evaluating
`alternative
`plans and are
`increasingly
`being used
`in external
`beam
`radiotherapy
`planning
`(2-6,
`14, 20, 24, 26-28,
`30). They have also
`
`dose
`been used to correlate outcome with the treatment
`distribution
`( 1, 9, 18, 19). When calculated
`on a single
`CT-slice
`they are
`referred
`to as dose-area
`histograms
`(7, 11).
`for a volume of tissue are in
`Two forms of the DVH
`use. The cumulative
`dose volume
`histogram
`(CDVH)
`shows V(D) plotted against D, where V(D) is the volume
`of tissue
`in which
`the dose is greater
`than or equal
`to D.
`The differential
`dose volume
`histogram
`(DDVH)
`shows
`v(D) plotted against D, where v(D) is the volume of tissue
`in which
`the dose is between D and D + AD, and AD is
`the “bin width” of the histogram.
`In constructing
`such
`DDVHs
`care is needed
`in the choice of bin width. Too
`large a value will mask details of the dose distribution,
`in
`while
`too small a value may cause
`large fluctuations
`the DDVH
`because of the finite grid used for the dose
`calculations.
`benefit from the use of DVH’s
`To realize the maximum
`a technique
`is needed
`for comparing
`and evaluating
`them
`objectively
`and consistently.
`Such a technique would also
`enable
`them
`to be used in defining and ensuring adherence
`to a treatment
`protocol. At present most comparisons
`of
`DVH’s for sensitive
`organs are made by assuming
`that
`smaller volumes
`at high dose mean better plans. Such a
`comparison
`is simple
`to make when
`the CDVH
`for one
`
`to: D. A. Viggars, Manitoba Cancer Treat-
`requests
`Reprint
`Foundation,
`100 Olivia St., Winnipeg,
`ment
`and Research
`Manitoba, R3E 0V9 Canada.
`
`419
`
`Acknowledgement-This
`work was supported
`Ltd.
`atronics
`International
`Accepted
`for publication
`
`5 December
`
`in part by Ther-
`
`199 1.
`
`Varian Exhibit No. 1015
`Page 0007
`
`

`

`420
`
`I. J. Radiation Oncology 0 Biology 0 Physics
`
`Volume 23, Number 2, 1992
`
`the
`throughout
`volumes
`cumulative
`plan has smaller
`whole dose range. However, when
`the CDVH’s for alter-
`native plans cross,
`it is more difficult
`to decide which
`is
`preferable.
`in a treat-
`An example of this type of difficulty occurred
`ment
`for cancer of the uterine
`cervix
`that had extended
`to the pelvic and para-aortic
`lymph nodes (3). A dynamic
`conformal
`technique was compared with treatment
`using
`more conventional
`techniques.
`The dose distributions
`in
`healthy
`liver tissue generated
`by the dynamic
`conformal
`technique
`and by one of the conventional
`plans gave sim-
`ilar average doses, but
`the CDVH’s
`for the
`two plans
`crossed. The conformal
`plan was judged
`to be superior
`on the grounds
`that
`it restricted
`high dose to a smaller
`volume
`than
`in the conventional
`plan.
`et al. (1) evaluated
`the
`In contrast, Austin-Seymour
`radiation
`tolerance
`of healthy
`liver
`tissue during
`heavy
`ion treatments
`of carcinoma
`of the pancreas
`and biliary
`system. They concluded
`that a plan with a smaller volume
`of liver above 30-35 GyE would be tolerated better
`than
`a plan with a higher volume
`above 30-35 GyE, even
`though
`the former plan might expose
`larger volumes
`of
`liver to doses near 60 GyE
`than
`the latter. The criterion
`used by Chin et al. (3) would have
`led to the opposite
`conclusion.
`To resolve such ambiguities
`it is necessary
`to
`characterize DVH’s quantitatively,
`preferably
`in terms of
`parameters
`that
`can be
`related
`to
`the outcome
`of
`treatment.
`to characterize DVH’s objectively have
`Several attempts
`been based on calculations
`of the risk of complication
`in
`a sensitive organ
`(15-17, 21, 29). These calculations
`de-
`pend on assumptions
`about how
`to allow
`for non-uni-
`formity of dose within
`the organ and on parameters
`de-
`scribing
`the dose response of the tissue, which are usually
`very poorly known. They have therefore not been widely
`used.
`In this paper we describe
`a convenient
`objective
`technique
`for characterizing,
`comparing
`and evaluating
`DVH’s which uses a simple dose prescription
`provided
`by a radiation
`oncologist based on clinical experience
`and
`dose response
`data. The
`technique
`provides
`visual and
`quantitative
`tools for the consistent
`evaluation
`and com-
`parison
`of alternative
`treatment
`plans. As an example,
`conventional
`and conformal
`treatment
`plans
`for a case
`of ca lung are compared.
`
`METHODS
`
`AND MATERIALS
`
`treatment
`3-dimensional
`for evaluating
`The scheme
`plans which we describe here has been developed
`on a
`commercial
`treatment
`planning
`system*
`and has been
`fully integrated with the conventional
`software
`so that it
`can be used easily on a routine
`basis. The scheme has a
`number
`of components.
`First,
`it uses a dose prescription
`which summarizes
`the radiation
`oncologist’s perception
`
`* Theraplan, Theratronics
`
`International, Kanata, Canada.
`
`or group of
`for a patient
`requirements
`of the treatment
`patients. The dose prescription, which
`is fully described
`below, can also be used as part of the definition
`of a treat-
`ment protocol. Once a dose prescription
`has been prepared
`the other components
`of the scheme can be used. They
`are a) images of regret on multiple CT slices, which are
`completely
`analogous
`to the 2-dimensional
`images of re-
`gret on single slices described
`in an earlier paper
`in this
`series (22), b) a visual display of the prescribed
`dose-vol-
`ume
`limits on the CDVH,
`c) objective
`score functions
`which quantify
`the deviation of the dose distribution
`from
`the dose prescription,
`d) histograms
`of regret
`in either
`cumulative
`or differential
`form, which provide a striking
`and easily assimilated
`visual comparison
`of the CDVH
`or DDVH with the dose prescription.
`the evaluation
`of
`The
`components
`and programs
`to collectively
`as OS-
`scheme described here are referred
`CAR (Objective Scoring with Colored Areas of Regret).
`
`Dose prescription
`in the target
`We assume
`that the ideal dose distribution
`is uniform
`at 100% of the prescribed
`dose and zero in all
`other
`tissues as shown
`in Figure 1 by the solid
`line his-
`tograms. Dose distributions
`which can be achieved
`in
`practice
`are less uniform
`in the target and are non-zero
`in normal
`tissue as shown by the dashed curves
`in Figure
`1. The quality of a proposed plan may therefore be judged
`by how far its CDVH departs
`from
`the ideal histograms,
`and a dose prescription
`can be defined by specifying
`the
`maximum
`acceptable deviations
`from the ideal shape. We
`refer to such deviations
`as “regret”. A complete descrip-
`tion of the acceptable
`limits of the shape of the CDVH
`would
`require a specification
`of the maximum
`and min-
`imum permissible
`cumulative
`volumes
`at all doses, but
`we have found
`that a much
`simpler prescription
`is ade-
`quate for practical purposes. A typical prescription
`is given
`in Table 1 which
`is a generalization
`to three dimensions
`of the constraints we have used in earlier papers (10, 11,
`22,23). Similar constraints
`have been used by Langer and
`Leong ( 13).
`levels
`at two dose
`limits are defined
`Target overdose
`(optionally
`only one), each corresponding
`to a prescribed
`maximum
`partial
`target volume which may exceed
`the
`dose
`limit. They are represented
`on the CDVH by two
`triangles,
`one shaded and one open. The
`triangles point
`downward,
`indicating
`that
`the CDVH
`is constrained
`to
`pass below
`them
`if the target overdose
`limits are not to
`be violated. Accordingly,
`the solid and dotted histograms
`in Figure 2a are within
`the overdose
`limits, but the dashed
`histogram
`violates both of them. For convenience,
`the
`lower and higher overdose
`limits are referred
`to as the
`mild and severe
`limits,
`respectively.
`in the target
`Similar constraints
`are placed on underdose
`by specifying
`two dose
`limits
`and
`two corresponding
`
`Varian Exhibit No. 1015
`Page 0008
`
`

`

`Evaluation
`
`of dose volume histograms
`
`0 D. A. VIGGARS et al.
`
`421
`
`I
`
`--.
`
`*\
`
`r
`
`100
`
`s
`5 T
`
`50
`
`-
`
`a) Target
`
`?
`.-
`s
`i
`E
`u’
`
`0
`
`0
`
`100 -.
`
`2
`E
`2
`p
`P
`.-
`ij
`i
`E
`a
`
`50.
`
`0
`
`0
`
`\
`
`‘.
`
`50
`Dose (%)
`
`100
`
`b) Organ
`
`.
`
`‘.
`
`‘\
`
`\
`
`\
`
`‘.
`
`-_
`
`50
`Dose (%)
`
`100
`
`4
`P
`‘2
`0
`z
`2
`
`50.
`
`a) Target
`
`0
`
`0
`
`I I
`
`ql-__
`
`50
`
`100
`
`Dose (46)
`
`Dose (%)
`
`dose volume histograms
`Fig. 1. Cumulative
`and realistic
`(dashed
`line) dose distributions.
`
`for ideal (solid line)
`
`dose volume histograms
`Fig. 2. Cumulative
`showing dose-volume
`limits from Table 1.
`
`for target and lung
`
`a
`
`receive
`which may
`volumes
`target
`partial
`maximum
`dose less than these limits. In this case the upper underdose
`limit
`is referred
`to as mild, and the lower limit as severe.
`As in the overdose
`case a single underdose
`limit may be
`used. To represent
`the
`target underdose
`constraints
`on
`the CDVH, use is made of the fact that
`if a volume V%
`of the tissue
`is below a certain dose then a volume
`(loo-
`V)% of the tissue
`is above
`that dose. Hence, an underdose
`limit that permits a maximum
`of 10% of the target volume
`to be below 95% dose
`is equivalent
`to requiring
`that at
`
`Table 1. Dose urescription
`
`for treatment
`
`of ca luna
`
`Type of regret
`
`Target overdose
`(severe)
`Target overdose
`(mild)
`Target under overdose
`(severe)
`Target under overdose
`(mild)
`Non-target
`Left lung
`Right
`lung
`Spinal cord
`
`overdose
`
`Dose limit
`@)
`
`volume*
`Maximum
`@)
`
`110
`
`105
`
`90
`
`95
`
`95
`50
`50
`75
`
`20
`
`50
`
`5
`
`50
`
`100
`30
`30
`0
`
`dose: 60 Gy at the isocenter.
`Prescribed
`as a
`for target and non-target
`tissue are expressed
`* Volumes
`percentage
`of target volume. Volumes
`for specific organs are
`expressed
`as a percentage
`of the organ volume.
`
`be above 95% dose. Therefore,
`least 90% of the volume
`if an underdose
`limit specifies
`that no more
`than V% of
`the target volume
`should be below D% dose, an upward
`pointing
`triangle
`is placed at dose D% and volume
`( lOO-
`V)% to indicate
`that
`the CDVH
`should pass above
`this
`point. This is illustrated
`in Figure 2a, where the solid and
`dashed histograms
`are within
`the underdose
`limits but
`the dotted histogram
`is not.
`For all non-target
`tissue and for specific sensitive organs
`single dose volume
`limits are specified
`to limit the volumes
`above chosen doses. This is illustrated
`in Figure 2b, where
`the solid histogram
`is acceptable but the dashed histogram
`violates
`the dose volume
`limit
`indicated by the downward
`pointing
`triangle.
`
`Histograms of regret
`of how well a dose distri-
`A clear visual
`representation
`bution
`conforms
`to a dose prescription
`can be given by
`shading
`areas of the DDVH plot that are in violation
`of
`the dose limits. Figure 3a illustrates a DDVH
`for the target
`in which any volume
`above
`the severe or mild overdose
`limits are shaded dark or light orange
`respectively.
`Sim-
`ilarly, any target volume below
`the severe or mild under-
`dose limits is shaded dark or light blue, respectively. Purple
`shading
`is used
`to show overdosed
`volumes
`of sensitive
`organs as illustrated
`in Figure 3b. This representation
`of
`the dose distribution
`is referred
`to as a “histogram
`of re-
`gret” since
`the sizes of the shaded areas are proportional
`to the volumes of tissue
`in violation
`of the corresponding
`dose limits. A similar approach
`is used for the CDVH as
`
`Varian Exhibit No. 1015
`Page 0009
`
`

`

`422
`
`1. J. Radiation Oncology 0 Biology 0 Physics
`
`Volume 23. Number 2, 1992
`
`(a)
`
`(b)
`(d)
`Fig. 3. Histograms of regret for an arbitrary dose distribution. DDVH for (a) the target and (b) a specific organ:
`and CDVH for (c) the target and (d) the same specific organ.
`
`the regions enclosed
`in Figures 3c and 3d, where
`shown
`by the dose limits,
`the ideal CDVH and the actual CDVH
`are shaded. For
`the CDVH,
`it is the maximum
`height
`rather
`than
`the area of the colored
`regions which has
`quantitative
`significance.
`The colors used and
`their
`in-
`terpretation
`correspond
`to the colors of the areas of regret
`described
`in an earlier paper
`in this series (2 1).
`
`Score,functions
`To provide
`a quantitative measure of how well a pro-
`posed
`treatment
`plan conforms
`to the dose prescription,
`we define a set of score functions which compare
`the actual
`deviations
`of a plan from the ideal CDVH with the max-
`imum
`deviations
`allowed by the dose prescription.
`For
`each dose volume
`limit
`[Di,Ri(max)]
`in the prescription,
`the score function
`is derived
`from a ratio ri defined as:
`
`c = Ri(D;)/Ri(max).
`
`is the actual volume of tissue,
`limits Ri(Di)
`For overdose
`the dose limit Di, and Ri(max)
`is the max-
`Vi(DI), above
`imum
`permitted
`or “tolerance”
`volume, Ti, above
`the
`dose limit Di, as shown
`in Figure 4a so that:
`
`r, = Vi(Di)/Ti.
`
`limits Ri(Di) and Ri(max) are volumes be-
`For underdose
`low the dose limit Di and
`therefore,
`as shown
`in Figure
`4b:
`
`ri = [ 100 - Vi(Di)]/[ 100 - Ti],
`
`where V(Di) is the volume above the dose limit Di, which
`can be read directly
`from
`the CDVH. For the target and
`for non-target
`tissue T,, Ri(max), Ri(Di), and Vi(Di) are
`expressed as percentages
`of the target volume. For specific
`organs at risk these volumes
`are expressed as percentages
`of the volume of the relevant organ. For a particular
`dose
`
`Varian Exhibit No. 1015
`Page 0010
`
`

`

`Evaluation
`
`of dose volume histograms ??D. A. VICGARS et al.
`
`423
`
`a) Overdose
`
`limit
`
`-0
`
`Di
`
`50
`
`100
`
`Dose (X)
`
`I
`
`1DD
`
`55
`
`V(Di)
`
`Ti
`
`50
`
`0
`
`i
`0
`
`2
`0
`.h
`z
`E
`(;
`
`1 100
`
`b) Underdose
`
`limit
`
`50
`
`Dose
`
`(X)
`
`used in calculating score functions:
`Fig. 4. Definition ofquantities
`limits, and (b) for underdose
`limits.
`(a) for overdose
`
`the range of ri is from 0 for an ideal distri-
`limit,
`volume
`to 1 for a distribution
`at the limit of acceptability.
`bution
`For a distribution
`which violates
`the dose volume
`limit,
`ri is greater
`than 1.
`To obtain a score function with more convenient
`erties we define
`
`prop-
`
`S, = lO[ 1 - ri],
`
`zero at the limit of
`is 10 for an ideal distribution,
`which
`acceptability,
`and negative when
`the dose-volume
`limit
`is violated. A single score for target overdose
`is calculated
`by averaging
`the two Si for the two target overdose
`limits,
`and similarly
`the scores for the two target underdose
`limits
`are averaged
`to give a single target underdose
`score. This,
`admittedly
`somewhat
`arbitrary,
`procedure
`allows some
`“trading-off’
`within
`each pair of constraints.
`An alter-
`native approach,
`if only one constraint
`of the pair is re-
`quired
`to be satisfied, would be to concentrate
`on
`the
`larger member
`of the pair of over- or underdose
`scores.
`If the scores
`for the
`two constraints
`in a pair differ by
`more than 5, the program
`issues a warning
`recommending
`that the two scores be examined
`individually.
`allows
`A special case occurs when
`the dose prescription
`no part of an organ
`to be above
`the dose limit, as happens
`for the spinal cord
`in Table
`1. Then
`the corresponding
`score function
`is set to 10 if this condition
`is satisfied and
`is set to zero otherwise.
`on a single slice
`are calculated
`If the score functions
`using area
`instead
`of volume,
`they are identical
`to the
`
`in an earlier paper
`defined
`score functions
`evaluation
`of 2-dimensional
`dose distributions.
`
`(23) on the
`
`RESULTS
`Application to planning a case of ca lung
`five
`above,
`To demonstrate
`the procedures
`described
`alternative
`plans were compared
`for a case of squamous
`cell carcinoma
`of the left lung. A CT scan of the thoracic
`region of the patient was acquired
`from lung apex to dia-
`phragm, using 29 slices 1 cm apart. The tumor was visible
`on 10 slices and a target
`region consisting
`of the tumor
`plus a 1.5 cm margin was outlined
`on each of these. The
`lungs and spinal cord were also outlined
`on all available
`slices.
`equally weighted
`All the plans used two perpendicular
`beams of photons
`from a 25 MV linear accelerator. The
`gantry angles were 70” and -20”
`as shown
`in Figure 5.
`The prescription
`required a dose of 60 Gy at the isocenter.
`A 55” wedge was used on both beams
`in plans A, B and
`C, a 27” wedge on both beams
`in plan D, and no wedges
`in plan E. In plan A rectangular
`beams were used with
`widths selected so that the 90% isodose covered
`the target
`area on the central slice through
`the tumor. No corrections
`for inhomogeneous
`tissue density were used
`in this cal-
`culation. The cumulative
`histogram of regret for the target
`in plan A is shown
`in Figure 6a and
`indicates
`that
`there
`are subs