BABE Ș-BOLYAI UNIVERSITY CLUJ -NAPOCA FACULTY OF ENVIRONMENTAL SCIENCE AND ENGINEERING Dosimetric and quality assurance procedure evaluation of the… [627283]
BABE Ș-BOLYAI UNIVERSITY CLUJ -NAPOCA
FACULTY OF ENVIRONMENTAL SCIENCE AND ENGINEERING
Dosimetric and quality assurance procedure
evaluation of the strut -adjusted SAVI hybrid
device used in accelerated partial breast
irradiation
DOCTORAL THESIS
SCIENTIFIC SUPERVISOR :
Prof. Univ.Dr. Constantin COSMA
Prof. Univ. Dr. Dumitru RISTOIU
DOCTORAL CANDIDATE:
Șerban MORCOVESCU
2017
ACKNOWLEDGMENTS
I enrolled into this doctoral degree program in 2006 strongly encouraged and
supported by the one who eventually had the bravery of taking on the responsibility of
being my supervisor and dissertation chair, Professor Dr. Constantin C OSMA, of blessed
memory. Unfortunately he is no longer with us and unable to partake of this special
moment, of seeing me cross ing the finish line. Time is never too generous. Prof. Dr.
Constantin COSMA has worked with me patiently during those years and helped me
eventually find and fine tune my area of research. He was a great mentor and invaluable
resource of wisdom and promoter of scientific debate and dialogue throughout the
program. Words are unable to express the loss of him, and the significance of his life in
my life. May the Lord grant him the peace and the rest he deserves and so much was
entitled to, after a life of hard work, done with dedication and unfailing passion.
I am therefore that much more grateful to Professor Dr. Dumi tru RISTOIU, for
being generous enough to adopt me, and to accept taking me under his wing as my
supervisior for the final stage of my doctoral program. His advice and leadership was in
fact vital and invaluable, and helped me correct, adjust, improve and finalize a thesis in
much need of trimming and format adjustments.
I owe a lot of gratitude to the special people and excellent professionals I started
working with and for in US A, Dr. Marius and Rodica Alecu from AROS LLC Texas , who
greatly helped me duri ng my first years of practice as a Therapeutical Medical Physicist in
Texas, always encouraging me to learn, study, settle, perfect and achieve my goals .
I am also grateful to Dr. Jeffery M orton , MD, from Texas Oncology Denton, my
practice Radiation Oncologist, and to all my colleagues and research partners , for the
constructive feedback , encouragement and opportunity to work on this c l i n i c a l
r e s e a r c h project , so well blended into my regular clinical schedu le.
L a s t , b u t n o t t h e l e a s t , h u g e t h a n k s t o m e w i f e A n c a a n d m y
e x t e n d e d f a m i l y for their unrelenting support , understanding and love they embraced
me with throughout the program.
Table of Contents
CHAPTER 1 INTRODUCTION ………………………….. ………………………….. ………………….. 5
CHAPTER 2 THEORETICAL ASPECTS ………………………….. ………………………….. …. 12
2.1. Ionizing Radiation. Fundamentals ………………………….. ………………………….. ……….. 12
2.2. Sources and types of ionizing radiation ………………………….. ………………………….. … 12
2.3 Energy transfer, absorption and attenuation ………………………….. ……………………….. 13
2.4. Interactions of photons with matter ………………………….. ………………………….. ………. 16
2.4.1. Photoelectric effect ………………………….. ………………………….. ………………………….. …….. 17
2.4.2. Compton effect ………………………….. ………………………….. ………………………….. ………….. 18
2.4.3 The Pair Production ………………………….. ………………………….. ………………………….. …….. 21
2.4.4 Interactions of charged particles with matter ………………………….. ………………………….. .. 22
2.5. Quantities describing the interaction of ionizing radiation with matter ………………. 23
2.5.1 Kerma ………………………….. ………………………….. ………………………….. ……………………….. 24
2.5.2 Exposure ………………………….. ………………………….. ………………………….. ……………………. 25
2.5.3. Absorbed Dose ………………………….. ………………………….. ………………………….. ………….. 26
2.6. Measurement of ionizing radiation ………………………….. ………………………….. ………. 28
CHAPTER 3 BRACHYTHERAPY ………………………….. ………………………….. ……………. 31
3.1. High Dose Rate Brachytherapy – General Aspects ………………………….. …………….. 31
3.1.1. Dose calculations in brachytherapy – TG43 formalism ………………………….. …………….. 31
3.1.2. Novel computational algorithms – ACUROS BV ………………………….. ……………………. 35
3.1.4 High Dose Rate unit description and source calibration ………………………….. …………….. 40
CHAPTER 4 OVERVIEW OF BREAST CANCER TREATMENT MODALITIES
………………………….. ………………………….. ………………………….. ………………………….. …………. 45
4.1. Breast Cancer and anatomy ………………………….. ………………………….. …………………. 45
4.2. Treatment modalities ………………………….. ………………………….. ………………………….. 46
4.3. Brachytherapy in the treatment of breast cancer. Developments. ………………………. 48
4.3.1 Partial breast irradiation. Brachytherapy devices and techniques. ………………………….. . 50
CHAPTER 5 DOSIMETRICAL EVALUATION OF A STRUT -ADJUSTED –
VOLUME -IMPLANT SAVI DEVICE USED FOR ACCELERATED PARTIAL
BREAST IRRADIATI ON ………………………….. ………………………….. ………………………….. 56
5.1 Device description ………………………….. ………………………….. ………………………….. ….. 56
5.2 Patient selection criteria ………………………….. ………………………….. ………………………. 58
5.3 Equipment ………………………….. ………………………….. ………………………….. …………….. 59
5.4 Structure Definitions and Nomenclature ………………………….. ………………………….. … 59
5.5 Treatment planning technique ………………………….. ………………………….. ………………. 60
5.6 Dose prescription and optimization ………………………….. ………………………….. ………. 65
5.7 Evaluation of dosimetric advantages of the SAVI device compared to balloon -type
APBI devices ………………………….. ………………………….. ………………………….. ………………. 66
5.7.1. Study Motivation ………………………….. ………………………….. ………………………….. ……….. 66
5.7.2. Materials and Methods ………………………….. ………………………….. ………………………….. .. 66
5.7.3. Results and Discussions ………………………….. ………………………….. ………………………….. . 69
CHAPTER 6 COMPREHENSIVE DOSIMETRIC ANALYSIS OF THE SAVI
DEVICE ………………………….. ………………………….. ………………………….. ……………………….. 73
6.1. Study Motivation ………………………….. ………………………….. ………………………….. …… 73
6.2. M aterials and Methods ………………………….. ………………………….. ……………………….. 73
6.3. Results and Discussions. Original contributions. ………………………….. ………………… 88
6.3.1 Multi -institutional study on all SAVI type devices ………………………….. …………………… 88
6.3.2 Single institution study results, on SAVI6 -1mini device – TG43 ………………………….. .. 89
6.3.3 Single institution study results, on SAVImini device – ACUROS ………………………….. . 92
CHAPTER 7 COMPREHE NSIVE EVALUATION OF A STRUT -ADJUSTED –
VOLUME -IMPLANT SAVI DEVICE QUALITY ASSURANCE PROGRAM …….. 94
7.1 Study motivation ………………………….. ………………………….. ………………………….. ……. 94
7.2 Pre -treatment Quality Assurance ………………………….. ………………………….. ………….. 95
7.2.1. Imaging and documentation for treatment planning ………………………….. …………………. 95
7.2.2 Treatment Time Nomogram for Strut -Based Accelerated Partial Breast Applicators.
Original Contributions. ………………………….. ………………………….. ………………………….. ………… 99
7.3 During and post treatment Quality Assuranc e – Interfractional Variance. Original
Contributions. ………………………….. ………………………….. ………………………….. ……………. 101
7.3.1 Materials and Methods ………………………….. ………………………….. ………………………….. . 101
7.3.2 Results and Discussion ………………………….. ………………………….. ………………………….. . 103
7.4 Overall Results and Discussion ………………………….. ………………………….. …………… 104
CHAPTER 8 CLINICAL RESULTS ………………………….. ………………………….. ………… 106
CONCLUSIONS ………………………….. ………………………….. ………………………….. …………. 110
BIBLIOGRAPHY ………………………….. ………………………….. ………………………….. ……….. 112
List of scientific papers presented at National and International Congresses and
Scientific Meetings ………………………….. ………………………….. ………………………….. ………. 128
List of Figures………………………………………………………………………..
List of Tables…………………………………………………………………………
ACKNOWLEDGEMENTS………………………………………………………….
REFERENCES……………………………………………………………………….
LIST OF PUBLICATIONS………………………………………………………………
LIST OF PRESENTATIONS………………………………………………………..
5
CHAPTER 1 INTRODUCTION
Ever since the discovery of X -rays and of radioactivity, slightly more than a
century ago, the study of th e nature and of the mechanisms of interaction of charged
particles became one of the most fertile research fields in the modern history of applied
science. This is mainly due to the fact that, like in Roentgen’s personal case, there was a
very strong conne ction and an immediate application of these newly discovered physical
phenomena to the field of medicine. The fact that x -ray diagnostic radiography was
adopted widely in both Western Europe and America within just one year after Roentgen’s
discovery speak s very clearly about how prominently the new technology impacted the
field of medicine and how quickly it was adopted and adapted to new practical
applications.
Radiological physics and radiological medicine are the two closely related
scientific fields t hat were born shortly after, the former dealing with the science of
ionizing radiation and the way it interacts with matter, and the second dealing with how
the findings of the first are integrated and ap plied in the field of medicine.
Cancer is considere d nowadays as being one of the most aggressive large group s of
diseases that can affect the normal functionality of any part or organ the human body, due
to its ability to grow and spread uncontrolled. When cancer cells develop and are able to
involve near by organs, metastasis occur, with cancer cells able to migrate in other parts of
the body, away from the original tumor site. External (exposure to alcohol , tobacco,
chemicals) or internal (genetic background, metabolical) factors can influence the
evoluti on of cancer, but there are various different main methods currently available for
the treatment of cancer: systemic chemotherapy, surgery, radiotherapy, hormone therapy,
and targeted therapy. For the purpose of our research project, this thesis focuses on the
aplication of radiation therapy in the treatment of cancer, more specifically, of breast
cancer.
Breast cancer is among the most common cancer in women in the world , and the
most common in women in USA. Breast tumors are usually diagnosed and found during
routine examinations, either through mammography, ultrasound examination or regular
physical examination. Advanced mammographic techniques are able to identify tumor s as
6
small as a few millimeters, but small tumor formations are easily identified through
regular checks performed by the educated patient.
Treatment of breast cancer can be a combination of local management and
systemic treatment. Most breast cancer pati ents are treated with both using local and
systemic treatment regimens. The selection of local management is made based on
diagnosis, patient and disease related characteristics, and it consists of surgery and, of
much interest for us, radiation therapy. C hemotherapy and hormonal therapy are the
systemic treatment options available.
The main purpose of the use of local therapies is the eradication of the primary
local disease. Mastectomy is many times involved, with or without reconstruction, but in
most ca ses, breast -conserving surgery combined with radiation therapy is used with much
success. When radiation therapy is the sole and primary treatment option, the excision of
the tumor is usually followed by a course of radiation therapy , which can be 1) exter nal or
2) internal ( brachytherapy), meant to remove residual microscopic disease. This is a
conservative treatment modality, since it conserves the breast, decreases the chance of
recurrence and eradicates the residual tumor. Various irradiation techniques , traditional or
novel, are available and can be employed (Devlin et al., 2016; Baglan et al., 2001; Benitez
et al., 2004; Mooij et al., 2014). The normal treatment regimen for external beam therapy
consists of daily treatments delivered for 4 to 5 weeks w ith the use of a linear accelerator
and a combination of photon and electron beams. The electron beams are usually used as a
boost field that only targets to original tumor bed. Another radiation treatment option
currently available is the use of brachythe rapy. Interstitial needles or balloon type devices
are implanted at the site of the primary tumor bed and radiation is delivered remotely,
using afterloader techniques, or using permanent implants, a more recent treatment option
developed in Canada in the last years.
The research objective of this thesis is to quantify the dosimetric performance and
to evaluate quality assurance procedures for a device specifically designed for the
treatment of early stage breast cancer patients by means of High Dose Rate brachytherapy.
There are many studies done on these particular areas of research, many focused on the
dosimetry aspect ( Edmunson et al ., 2002; Dou et al ., 2011; Gurdalli et al ., 2011;
Scanderbeg et al., 2009 ), others on the quality assurance program implementation (Dou et
7
al., 2010 ; Thomadsen, 2000; Ji et al ., 2016) , and several approaching both aspects
altogether (Gurdalli et al., 2007 ; Dou et al., 2012).
The numerous scientific papers published in the last two decade s clearly indicate
that o ne of the fastest growing medical procedures is breast brachytherapy. Over the last
two decades, breast conservation therapy (BCT) has been accepted as one of the standard
treatment regimens in patients with early -stage breast cancer. Especially since the early
2000s, Accelerated Partial Breast Irradiation (APBI) has been embraced with great
interest by both cancer care providers and breast cancer patients as a great and efficient
treatment alternative to conventional who le-breast irradiation. There is data that indicates
that APBI is an acceptable option of treatment for pr operly selected patients (Arthur et al.,
2003) .
Historically, partial breast irradiation was first performed with interstitial
implantation using multi -catheter brachytherapy that typically triggered the tumor bed plus
a generous margin of 2.0 -2.5 cm. Because of its procedural complexity, interstitial
brachytherapy was not widely adopted in the United States and eventually led to the
development of a number of other methods of delivering APBI, including intraoperative
radiotherapy with photons and electrons and conformal three -dimensional external -beam
approaches (Vaidya et. al , 2004) The MammoSite approach was eventually developed,
allowing for the delivery of hypofractionated high -dose rate brachytherapy to the tumor
bed in a relatively straightforward manner, which eliminated many of the technical
difficulties inherent to traditional double -plane interstitial implants.
Many approaches and technique s, all image -guided, have been developed and
improved since the beginning. The MammoSite® – Hologic, Bedford, MA – was designed
to be a simpler alternative to the interstitial implants, replacing the insertion and treatment
of many catheters with a single catheter centered in a balloon that fills the lumpectomy
cavity. With the simplicity of the procedure, many surgeons perform the insertion, which
helps bring them into the treatment team, increasing the number of cases treated. The
trade -off for the simpli city is a sacrifice of control over the dose distribution. For these
intracavitary treatments, the prescribed dose contour falls 1 cm outside of the balloon
surface. For spherical balloons, a single dwell position in the center serves as the source.
Due to the anisotropic dose distribution of the HDR source, the single dwell position
results in low doses near the source axis, leading some practitioners to add lightly
8
weighted dwell positions near where the catheter crosses the balloon surface. Oblong
balloo ns require multiple dwell positions. The dose decreases continually with distance
from the source center, so the dose near the surface of the balloon tends to be 1.7 to 2.25
times the prescription dose. Skin doses will exceed the prescription dose if the b alloon
surface lies closer than 1 cm to the skin, leading to a recommendation of a minimal
separation between the balloon and the skin of 6 mm. One common problem with these
applications occurs when air becomes trapped on the surface of the balloon during
insertion, pushing target tissue away from the prescribed dose contour.
After the launch and initial use of the single -lumen MammoSite, many other
treatment multicatheter d evices such as the multilumen Ma mmoSite (Hologic, Bedford,
MA), Contura Multi -Lumen Balloon (Contura MLB, SenoRx Inc.,Irvine, CA) , ClearPath
or Strut Adjusted Volume Implant (SAVI) were developed, in order to allow better
targeting of the primary tumor site and better sparing of the adjacent normal tissues and
organs. Among those, the SA VI device proves to be a unique solution for cases where
other APBI devices are not a fit (Morcovescu et al ., 2009). Because of its design, the
miniSAVI version of the SAVI applicator allowed excellent dosimetric conformance and
skin sparing for cases wher e the size of the breast and the location of the lumpectomy site
hindered the use of balloon -type devices, like MammoSite or Contura.
Strut -based applicat ors have been widely adopted in United States as an alternative to
balloon -type applicators in APBI, a nd were increasingly used at our practice since early
2008. The Strut Adjusted Volume Implant (SAVI) applicator studied (Cianna Medical,
Aliso Viejo, CA), also focuses on the smallest of its kind (6 -1mini), which has been
especially used on patients with reduced breast or/and lumpectomy cavity size.
Our research is a comprehensive dosimetric evaluation of various coverage
parameters, and doses to adjacent critical structures have been estimated in all patients
includ ed in a retrospective study encompassing more t han five years of cumulated clinical
data. Proposed improved guidelines for daily treatment clinical QA and workflow are also
presented and discussed.
The main body of our work consists on a comprehensive dos imetric analysis of
extensive clinical data, collected for all four different size SAVI devices (SAVI6 -1mini,
SAVI6 -1, SAVI8 -1 and SAVI10 -1). Our study is structured and focused on two subsets of
data: 1) a major pool of data collected at a multi -instituti onal level, that presents the
9
dosimetric analysis of the entire range of SAVI applicators, and a 2) minor pool, a subset
of the entire data, considering patients implanted with the smallest of the SAVI devices,
the SAVI6 -1mini device, in our clinic only . The total number of patients included in our
multi -institutional pool study is 817. There were 14 different participating institutions
involved in the multi -institutional study, each providing data for all four SAVI device
models. The subset study presented on the SAVImini device is a single -institution study of
plans created for 121 patients , treated over the span of 5 years, from 2009 to 2014. We
have also performed intercomparison studies among different APBI devices, which
allowed us to highlight the va rious dosimetrical advantages of the SAVI device over the
balloon type devices.
The dosimetric parameters reported in this study include: cavity volume, volume of
the defined treatment region (PTV_EVAL), V90(%), V95(%), V100(%), V150(cc),
V200(cc), skin di stance (minimum distance from the lumpectomy cavity wall to the skin),
chest wall and ipsilateral lung distances (mm), and the maximum doses to critical
structures (skin and chest -wall). Conformity Indexes (CI), related to reported air/seroma
and invaginat ion volumes, were also evaluated. Our dosimetric coverage criteria for this
study was V90>90%, V150<50 cm3, V200<20 cm3. Additional constraints are placed to
try limiting the chest wall and skin doses to 100%. All dosimetric data , both the major
pool and m inor subset, was stratified using 5 mm skin -bridge intervals, therefore
differentiating among cases with major or no PTV volume reduction.
The current thesis is structured in eight chapters, a short description of each being
presented in the following paragraphs.
Chapter 2 is meant to review the fundamental theoretical concepts and quantities
used to describe the interactions of ionizing radiation, both gamma and x -ray, with matter,
and the methods used to measure those quantities. We briefly surveyed the main types of
interactions of photons with matter, with an emphasis on the kinematics and probability of
the Compton interactions, as the Compton effect is the predominant mechanism of
interaction in the range of energies commonly used in radiotherapy.
Chapter 3 overviews general aspects of High Dose Rate Brachytherapy, with
special attention given to describing the standard TG43 dose calculation formalism
historically and widely used in USA for brachytherapy calculations, as well as to
addressing the n ovel computational algorithm commissioned and implemented in our
10
department, ACUROS BV, which allowed us to account for inhomogeneity corrections
and more realistically evaluated the extent to which these can become clinically relevant in
the setting of br achytherapy HDR planning and delivery.
Chapter 4 offers an overview of the treatment modalities currently available for
breast cancer, with an emphasis on the brachyth erapeutical options . We surveyed the
brachytherapy devices currently used clinically in t he USA, and highlight the device we
focused our research on, the SAVI device.
Chapter 5 presents the framework of our study, in which we describe the physical
characteristics of the SAVI device, the patient selection criteria, and we clarify the main
dosim etric parameters used in evaluating the performance of this device. We present our
initial results with the device, on the background of a preliminary comparison study we
performed on the dosimetric performance of balloon type devices, MammoSite versus
Contura. We are showing that the SAVI device is the device of choice when there is a
need to treat lumpectomy cavities of volumes of less than 35 cm3, which are usually
entirely filled by properly inflated balloon devices.
Chapter 6 is the main body of our work consisting of a comprehensive dosimetric
analysis of extensive clinical data, collected for all four different size SAVI devices
(SAVI6 -1mini, SAVI6 -1, SAVI8 -1 and SAVI10 -1). Our study is structured and focused
on two subsets o f data: 1) a major pool of data collected at a multi -institutional level, that
presents the dosimetric analysis of the entire range of SAVI app licators, and a 2) minor
pool, a subset of the entire data, considering patients implanted with the smallest of t he
SAVI devices, the SAVI6 -1mini device, in our clinic only . The total number of patients
included in our multi -institutional pool study is 817. There were 14 different participating
institutions involved in the multi -institutional study, each providing da ta for all four SAVI
device models. The subset study presented on the SAVImini device is a single -institution
study of plans created for 121 patients , treated over the span of 5 years, from 2009 to
2014.
In chapter 7 we present our own contribution to designing a comprehensive quality
assurance program that deals with all stages of an APBI treatment process in a busy
radiotherapy department . We bring to light all possible un -common clinical situations, we
highlight the common practices and the extra measur es we included into our customized
11
QA program, in an attempt to incorporate those into a comprehensive QA program
capable dealing with even the least frequent clinical situations .
The final results and conclusions of this multi -layered study are in fact
corroborated with the results of a 5 year long clinical study we were part of, that confirms
the validity of our dosimetrical study . This report also cofirmed outstanding target
coverage with excellent skin and rib sparing over the entire cohort of clinical data. We
concluded that t he SAVI applicators were designed to simplify brachytherapy APBI
compared to interstitial brachytherapy, allowing the advantages of brachytherapy over
other forms of accelerated partial breast ra diation therapy accessible to more women. The
strut op en architecture design and mul tiple catheter options allow dose sculpting to each
patient’s unique anatomy and cavity lo cation. This flexibility helps to overcome prior
concerns with skin spacing and tu mor beds positioned between the overlying skin and
chestwall that limited patient eligibilit y. The clinical report presented at the end of this
thesis confirms excellent tumor control comparable to other published APBI rates and
survival with low toxicity. Compared to external beam techniques for APBI,
brachytherapy seems to be as effective, with less toxicity.
12
CHAPTER 2 THEORETICAL ASPECTS
2.1. Ionizing Radiation. Fundamentals
The radiations of primary concern for us are the ones originating in atomic and
nuclear processes. Ionizing radiations are those radiations that can excite or ionize the
atoms of the material they interact with. They are usually identified with radiations that
surpass the minimal kinetic energy value of ~ 10 eV, the typical energy required by a
valence electron to escape an atom. The International Commission on Radiation Units and
Measurements (ICRU, 1971) makes a clear distinction between interactions of c harged
and uncharged particles, emphasizing the fact that there are two different mechanisms by
which the process of ionization can take place: directly and indirectly ionizing radiation.
2.2. Sources and types of ionizing radiation
It is therefore pro per and right to enumerate the most important sources and types
of radiations that can initiate and take part in an ionization process. An important source
consists of non -particulate radiations, 1) X-rays and 2) gamma -rays (γ-rays), which are
electromagne tic radiations of atomic or nuclear origin. They are created either due to the
interactions between energetic charged particles and atomic targets or due to nuclear
energetic instability (Betel, 1996) . The practical energy ranges of X -rays are stretching
from generating voltages of 0.1 KV all the way up to the megavoltage values, and are
normally grouped in voltage intervals, i.e. low -energy (0.1 -20 kV, diagnostic (20 -120kV),
orthovoltage (120 -300kV), intermediate (300kV -1 MV) and megavoltage (above 1 MV).
A graphic depiction of the mechanisms of X -ray production is shown in Figure 1.
Particulate radiations, 3) electrons (β -rays or δ -rays), 4) neutrons and other 5)
heavy charged particles (protons, alpha particles, etc), are either the products of
radioactivity or nuclear reactions or are obtained through the process of acceleration in
high energy generators. The range of energies of interest in clinical radiological physics is
from a few eV (for electrons) to hundreds of MeV (for protons or alpha -particles).
13
Figure 1 . Mechanisms of X -ray production
The main difference between the particulate and the non -particulate radiations has
to do with the process of energy deposition in matter, since this is a two -step process for
X-rays and γ -rays, a detail of much importance when describing and defining the basic
concepts of radiation dosimetry. We will further discuss the most important concepts
describing the mechanism of interaction of ionizing radiation with matter.
2.3 Energy transfer, absorptio n and attenuation
Ionizing x -ray or gamma photons indirectly deposit their energy by interacting
with the atoms of the material and producing high energy electrons. These electrons then
lose their energy by three major mechanisms: photoelectric effect, C ompton effect and
pair-production, and by two other minor ones, of not much interest for radiological
physics: coherent scattering and photonuclear reactions.
The energy absorption process is a complex phenomenon, of mutual interchange
and interaction bet ween particulate and non -particulate radiations (Cember, 1983) . Highly
14
energetic x -ray or gamma -ray beams have a wide array of applications in radiotherapy and
are the principal method of producing particulate radiation beams, mainly electron
fascicules, f or the same purpose. In fact, highly energetic photon beams are extremely
useful in clinical applications exactly because they are capable of transferring their energy
to the target materials or tissues by ejecting orbital electrons from their otherwise
relatively stable energetic states. These high -speed electrons have then the capability of
producing ionization and excitation of other atoms along their own paths of interaction.
This energy deposition and absorption process can be of great use when occurri ng inside
living tissues, since it has the potential of destroying or damaging the reproduction
capacity of tumor cells and tissues.
A photon beam consists of a very large number of photons, of various energies,
and in order to characterize the way this b eam of radiation interacts with the material it
transverse, some basic quantities of statistical nature were introduced:
1. The fluence (Φ) is the ratio between the number of photons dN that enter a finite
sphere surrounding the point of measurement and the cross -sectional area da of the
imaginary sphere:
dadN
(2.1)
usually expressed in units of m-2 or cm-2.
2. The fluence rate or flux density (φ) is the fluence dΦ per unit time dt,
dtd
(1.2)
and is expressed in units of m-2 s-1.
3. Energy fluence (Ψ) is the quotient of the total energy dE carried by all dN rays
entering the imaginary sphere and the cross -section da of that sphere,
dadE
(2.3)
usually expressed in units of J m-2 or erg cm-2. For the special case of a monoenergetic
beam,
15
dE = dN· hν (3.4)
where hν is the individual kinetic energy of any photon in the beam.
4. Energy fluence rate or energy flux density (ψ) is the energy fluence per unit time:
dtd
(4.5)
The quantities listed above are of much importance and useful in many practical
applications but they are not descriptive in terms of the energy and type of the photon
beams. These factors become extremely important especially when talking about radiation
detection and measurement.
Especially when discussing about uncharged ionizing radiations we have to
introduce a very important concept, i.e., exponential attenuation . Charged ionizing
radiations interact with the matter in a much more sophisticated way, through small
interactions, and their gradual attenuation is not descr ibed by an exponential function
(Meredith, 1972) . An uncharged particle can pass through matter without losing a
significant amount of energy, while a charged particle always loses some or all of its
energy.
A beam of uncharged particles has therefore no specific range and its energy
degradation mainly depends on its energetic profile. A monoenergetic beam is attenuated
exponentially, according to this equation:
dN = – μN dx or I(x) = I 0e-μx
(5.6)
where dN is the reduction in the number of incident photons, N is the initial number of
incident photons in the beam, I(x) is the intensity transmitted by a thickness x , I0 is the
incident energy on the target material and μ is the attenuation coefficient . When the beam
consists of a spectrum of photon energies, the attenuation is not exactly exponential, since
photons of different energies are attenuated differently in the medium, with the lower
energy photons being preferentially r emoved from the beam.
The mass attenuation coefficient , μ/ρ (cm2/g), is a more relevant indicator of the
attenuation of a photon beam in a medium, since it factors out the density of the material
16
and it brings into discussion its atomic composition. An even more relevant quantity is the
electronic attenuation coefficient μ e,
01
Ne cm2 / electron
(6.7)
since, ultimately, the attenuation of the photon beam and the inherent energy deposition in
the medium depends mainly on the electronic composition and characteristics of the target
material. Two other important processes take place when a photon interacts with the
electrons in a medium, described by two other coefficients:
i) energy transfer coefficient, μ tr:
hEavg
tr
tr (7.8)
where
avg
trE is the average energy transferred into kinetic energy of charged particles per
interaction and μ is the attenuation coefficient of the material, and
ii) energy absorption coefficient, μ en:
μen = μtr (1 – g)
(8.9)
where g is the fraction of the energy of secondary charged particles lost to bremsstrahlung
in the material. Energy transfer and energy absorption are the two processes by which the
energy of a photon beam is imparted to the surrounding electrons. A photon can have
multiple interactions in which its energy is converted into kinetic energy of the electrons it
interacts with, and then these high speed electrons impart their energy with other electrons
trough ionization or excitation. Part of their energy is lost in bremmstrahlung processes
and does not contribute to the energy absorbed in that specific volume. The energy
absorption coefficient is the q uantity of special interest for radiotherapy since it is used in
evaluations of the energy absorbed in tissues, therefore prone to producing biological
effects.
2.4. Interactions of photons with matter
As we mentioned earlier, there are five different me chanisms by which the photons
are interacting with matter. Each of these can be represented by its own attenuation
17
coefficient, characteristic for the energy of the incident photon and for the atomic number
of the absorbing material. The total attenuation coefficient can be expressed as the sum of
all these individual coefficients:
ph c coh (9.10)
where μcoh, τ, σ c, π, and μph are the attenuation coefficients for coherent scattering,
photoelectric effect, Compton effect, pair production and photo disintegration,
respectively. The coherent scattering, or Rayleigh scattering, involves no energy
absorption but only scattering of t he low energy incident photons at small angles relative
to their initial path, especially in high atomic number materials. It has no major relevance
in radiation dosimetry. The process of photo disintegration is, at the other end of the
energy interval, on ly occurring for photon energies above 10 MeV, and is, for that reason,
not taken into account when dealing with the dosimetry of X -rays in the range of clinically
relevant energies.
2.4.1. Photoelectric effect
The photoelectric effect is a phenomenon in which a photon interacts with and
imparts part of its energy to the orbital electrons of a target atom, ejecting them from the
atom. The ejection of the photoelectron results in the creation of a shell vacancy that can
be subsequently filled by an outer o rbital electron, with the emission of characteristic x –
rays. Auger electrons can also be produced by the re -absorption of these characteristic x –
rays inside the mother atom (Figure 2).
Figure 2. The photoelectric effect.
18
The probability of the photoelec tric effect depends on the energy of the incident
photon and on the atomic number of the absorbing material. This dependence is expressed
as
33
EZ (10.11)
and is at the foundation of many applications in diagnostic radiology. Low energy X -rays,
when used for therapeutical purposes, are highly absorbed in bone, due to its high Z. The
direction the photoelectrons are emitted is either close to 90o degree for low energy
photons or in a more forward direction for higher energy photons. The photoelectric effect
in water is predominant at energies around 30 keV but is almost nonexistent for photon
energies above 100 keV.
2.4.2. Compton effect
The Compton effect refers to the interaction between a photon and a stationary or
unbound electron, i.e. loosely bound electrons that can be considered “free” electrons
because of their weak and low energetic binding inside the atom, when compared with the
energy of the incident photon . The Compton effect is the dominant e ffect in the range of
energies regurarly dealt with in brachytherapy, therefore we will summarize important
theoretical aspects of this physical phenomenon.
2.4.2.1. The kinematics of the Compton effect
In terms of kinematics, the equations that describe the Compton effect are the
following:
) cos1)( / (1'2
cmhhh
o
(11.12)
T = hν – hν’ (12.13)
)2tan() 1( cos2
cmh
o (13.14)
19
where moc2 is the rest energy of the electron and T is its kinetic energy, ω and θ are the
scatter angle of the photon and o f the electron, respectively . Figure 3. represents the
collision between a photon of energy hν with an unbounded electron that has no kinetic
energy (being in a stationary state). The forward momemntum of the incident photon is
transferred to the electron in the form of kinetic energy T, which then is scattered at angle
θ, with both kinetic energy a nd momentum conserved during the process.
Figure 3. The Compton effect.
Compton scattering is nearly elastic for low energy photons, and is able to transfer
almost its entire energy to the recoiling electron. Special cases of the Compton effect are
a) the direct hit , when the energy transferred to the electron is maximum and the scattered
photon travels backward at minimum energy, b) 90-degree photon scatter , when the
photon is scattered at right angles ( ω=90o) and the photon and, c) grazing hit , when the
photoelectron is deflected at 90o and the scattered photon continues on the forward
direction ( ω=0o). When the energy of the incident photon is high, if the radiation is
scattered at right angles it attains a maximum energy of 0.511 MeV, and if the radiation is
scattered backwards it attains the energy of 0.255 MeV. These are special situations of
much importance in calculating shield barrier and wall thicknesses against scattered
radiation.
Because the Compton interaction practically involves elect rons weakly bound in
the atom, it is independent of the atomic number Z of the material. Even more, since the
number of electrons per gram decreases slowly but steadily with Z, the Compton mass
attenuation coefficient is practically the same for all materi als. Therefore, σ c/μ ∞ 1/E,
20
since the Compton effect decreases with increasing the photon energy. In the range of
photon energies commonly used in radiation therapy, the Compton effect is the most
relevant mode of interaction of incident photons with the a bsorbing medium.
2.4.2.2 Probability of Compton Interactions
J.J. Thompson was the one who first introduced the concept of the probability of
Compton interactions, theory which assumed that the free electron oscillates under the
influence of the electri c vector of an electromagnetic incident wave. The electron is not
transferred any of the kinetic energy of the process, but it releases a photon of the same
energy as of the incident one in an elastic scattering.
The differential cross section per electro n for a photon scattered at angle ω, per
unit solid angle, is expressed as:
𝑑 𝜎0𝑒
𝑑𝛺𝜑= (𝑟02
2) (1+ 𝑐𝑜𝑠2𝜔) (14.15)
in units of cm2 sr-1 per electron. The quantity r o = e2/moc2 = 2.818 x 10-13 is the “classical
electron radius”.
The total Thomson scattering cross section per electron, 𝜎0𝑒, can be obtained by
integrating the last equation, over all directions of scatteri ng. This is simplified by
assuming cylindrical symmetry and integrating over 0≤ ω ≤π, with the annular element of
solid angle given by dΩ φ = 2π sin ω dω, in terms of angle ω:
𝜎0𝑒= ∫𝑑 𝜎0𝑒𝜋
𝜑=0= 𝜋𝑟02∫(1+ 𝑐𝑜𝑠2𝜔)𝜋
𝜑=0sin𝜔𝑑𝜔=
= 8𝜋𝑟02
3=6.65 𝑥 10−25 𝑐𝑚2/𝑒𝑙𝑒𝑐𝑡𝑟𝑜𝑛 (15.16)
Because this cross section value predicted in J.J. Thomson’s theoretical description
of the Compton process is too large for photon energies ℎ𝜈 > 0.01 MeV, further theoretical
work was done by Dirac and Nishina in 1928, by applying Dirac’s relativistic th eory to
describe the Compton interaction process (Attix, 1986) . They were able to better predict
and more comprehensively express the differential cross section for photon scattering, as
21
𝑑 𝜎0𝑒
𝑑𝛺𝜑= (𝑟02
2) (ℎ𝜈′
ℎ𝜈)2
(ℎ𝜈′
ℎ𝜈+ ℎ𝜈
ℎ𝜈′− sin2𝜔) (16.17)
with ℎ𝜈′ given by Eq. (2.12) . Since for low energies ℎ𝜈′ ≈ ℎ𝜈, Eq. (2.17) becomes
𝑑 𝜎0𝑒
𝑑𝛺𝜑= (𝑟02
2) (2− 𝑠𝑖𝑛2𝜔)= (𝑟02
2) (1+ 𝑐𝑜𝑠2𝜔) (17.18)
form identical to Eq. (2.15) , proving that the Klein Nishina differential cross section is
simply the Thompson cross section for the special case of low photon energies.
If we integr ate Eq ( 2.17) over all photons scattering angles ω, the total Klein
Nishina cross section per electron becomes:
𝜎𝑒=2𝜋∫𝑑 𝜈𝑒
𝑑𝛺𝜑𝜋
𝜑=0 sin𝜔·𝑑𝜔=
= 𝜋·𝑟02∫(ℎ𝜈′
ℎ𝜈)2
(ℎ𝜈′
ℎ𝜈+ ℎ𝜈
ℎ𝜈′− sin2𝜔)𝜋
0 sin𝜔·𝑑𝜔=
= 2𝜋·𝑟02{1+𝛼
𝛼2 [2(1+𝛼)
1+2𝛼− 𝑙𝑛(1+2𝛼)
𝛼]+ 𝑙𝑛(1+2𝛼)
2𝛼− 1+3𝛼
(1+2𝛼)2} (18.19)
where α = hν/m0c2, with m0c2 = 0.511 MeV and hν also expressed in MeV.
2.4.3 The Pair Production
This mechanism of interaction takes place when the energy of the incident photon
is greater than the threshold energy of 1.02 MeV. In this process there is a strong
interaction between the electromagnetic field of the atomic nucleus and the incident
photon in which the photon releases and imparts its entire energy to a newly created pair
of particles, a negative electron (e-) and a positron (e+) (see Figure 4).
22
Figure 4 . Pair production
Because the pair production is a process of interaction with the energetic field of
the nucleus, it greatly depends on the atomic number, its probability increasing with higher
Z. The pair production mass attenuation coefficient π/ρ is proportional to Z2. This
mechanism predominates for energies above 25 MeV.
In conclusion, the total mass attenuation coefficient is large at low energies (<
100keV) due to the predominance of the photoelectric effect, then slightly decreases with
energy up to 25 MeV, in the C ompton effect predominance region (and remains relatively
independent of Z), and increases slightly above 25 MeV due to the predominance of pair
production.
2.4.4 Interactions of charged particles with matter
Charged particles interaction and exchange of energy with matter is very different
then the one specific to photon beams. This is due to the Coulombian nature of interaction
between a moving charged particle and the bound electrons or with the nucleus. Any
charged particle is losing its energy through a process of successive interactions and
energy losses with the surrounding medium. The heavier the particle is, the smaller the
number of interactions will be. In the case of electrons , a Bragg peak is not ob served
because they exchange energy in multiple successive scattering interactions. Neutrons
have a very complex nature of interaction, and their dosimetry very intricate exactly
because they create a multitude of other types of radiations (heavy particles , neutrons,
gamma rays) when they interact with matter. Heavy particles lose their energy at a very
23
high rate, proportional with the square of the particle charge and inversely proportional
with its velocity. The Bragg peak indicates a high energy loss related to the slowing down
process that takes place.
Figure 5 . Spread Out Bragg peak for a proton beam
It is possible to manipulate and use weighted superposition methods in order to
obtain a much flatter and broader Bragg peak, as shown in Figure 5 above, for a proton
beam. The red the dose curve represents the Bragg peak of a monoenergetic, thin pencil
beam of protons. By grouping multiple Bragg peaks of different proton ranges and
energies together, it is possible to deposit a homogenous dose in th e target region (Perez,
1992) . The resulting (range -modulated) proton beam distribution is called Spread Out
Bragg Peak (SOBP, indicated in blue). The dose degradation with depth curve of an X -ray
beam is shown in green. The picture shows that protons depo sit a much smaller dose than
X-rays outside plateau region, which can be of great applicability when a uniform
deposition of energy in a medium is desired.
2.5. Quantities describing the interaction of ionizing radiation with matter
The quantities describing the interaction of radiation with matter are 1) Kerma, K ,
a quantity that describes the energy transfer from indirectly ionizing radiations (photons)
to charged particles, 2) the Exposure, X , a quantity that describes gamma or X -ray beams
24
in terms of their ability to ionize a volume of air, and 3) the Absorbed Dose, D , a quantity
that describes the energy transfer from directly ionizing radiations to matter.
2.5.1 Kerma
The Kerma K (kinetic energy released in a medium) can be def ined by introducing
another quantity, the energy transfer, E tr, in a volume V, and the radiant energy, R :
Etr = R i – Ro + Q (19.20)
where
Ri is the radiant energy of uncharged particles entering volume V,
Ro is the radiant energy of uncharged particles exiting volume V, except that
created through radiative losses of kinetic energy by charged particles inside
volume V, and
Q is the net energy derived from rest mass in V.
All processes, in which the kinetic en ergy of a charged particle is converted into
kinetic energy of a photon, either by bremsstrahlung or by in -flight annihilation of
positrons, are considered radiative losses.
The radiant energy R is defined as the energy of particles (excluding the rest
energy), emitted, received or transferred to a medium. Etr does not reflect how the energy
imparted to charged particles is spent inside the specified volume V, and does not count
for the energies by charged particles in volume V.
Kerma K is defined simply as
dmdEKtr (20.21)
the total energy transferred to charged particles by uncharged particles, per unit mass,
including the radiative loss energy but excluding the energy axchenged among charged
particles in volume V. The unit of kerma is gray (Gy), which equals 1 J/kg. The special
unit for kerma is rad. There is a strong relation between kerma and enegy fluence for
25
photons, and for monoenergetic x -ray beams of energy E, that transverse a material of
atomic number Z, this can be expressed as
gK
ZEen
ZEtr
11
, ,
(21.22)
where μ tr/ρ is the mass energy transfer coefficient, μ en/ρ is the mass energy absorption
coefficient, Ψ the energy fluence at the point A of measurement and K is the kerma at
point A in that volume. The energy gained by an electron after an interaction with an x -ray
photon can be partially lost as ionization energy in collision interactions with the atomic
electrons of the absorbent, or by radiative interactions, whit the emission of secondary x –
ray photons. We can therefore make a distinction between two components of kerma,
K = K col + K rad =
gg
ZEen
ZEen
1, ,
(22.23)
where K col is the collision kerma and K rad is the radiation kerma. The collision component
of kerma, K col, is a very important quantity used when discussing the concept of charged –
particle equilibrium (CPE) in a given volume V, and can be viewed as the net energy
transferred to the charged particl es, since it does not incorporate the radiant energy losses
Rrad of the charged particles originated in V,
Enet = R i – Ro – Rrad + Q = E tr – Rrad (23.24)
2.5.2 Exposure
Exposure is another important quantity for radiological physics, one that has
historical precedence over the concepts of kerma and absorbed dose. It is a measure of
ionization produced in air by photons and is defined as the quotient of the absolute value
of the total charge, dQ, of the ions of one sign produce d in air when all the electrons
liberated by photons in air mass dm are completely stopped in air,
26
dmdQX (24.25)
its SI unit is coulomb per kilogram (C/kg), and its special unit is roentgen (1R). Exposure
can be expressed in relationship with the collision component of kerma by knowing that
the mean energy required to produce an ion pair in air is almost constant for all electron
energies, and has a value of W = 33.97 eV/ion pair. Therefore X equates to K col as follows:
WeKXair
col =
air airen
airWe
(25.26)
where e/W is the average energy required per unit charged of ionization produced.
Exposure is a quantity that applies only to uncharged radiations below 3MeV, and is a
measure of ionization in air only.
2.5.3. Absorbed Dose
The absorbed dose is a quantity that is relevant to all types of ionizing radiation
fields, whether indirectly or directly ionizing. The absorbed dose is the ratio between the
mean energy dE imparted by ionizing radiation to a material of mass dm,
dmdED (26.27)
its SI unit is gray (Gy) and its special unit is rad ( radiation absorbed dose), like for kerma.
There is a strong relationship between absorbed dose and collision kerma and it brings into
discussion the concept of electronic equilibrium in a given absorbing volume. Kerma is
maximum at surface and decreases with depth and the dose builds up to a maximum and
then decreases with depth, proportionally with kerma. If we introduce a new quantity, β,
defined as
colKD
(27.28)
we can differentiate between two different electronic equilibrium regions: for β < 1, we
are in the electronic build -up region , when β = 1, electronic equilibrium is achieved, and
when β > 1, we are referring to the region of transient electronic region. This is graphically
lustrated in Figure 6 below.
27
Figure 6. Absorbed dose and collision kerma relationship, for a megavoltage photon beam
In the transient electronic equilibrium region, the absorbed dose is greater than
kerma because of the combined effect of photon attenuation and the predominantly
forward motion of electrons in the field. β is known for various photon energies in
different medium like air and water and it varies with en ergy.
The relationship between absorbed dose and photon energy fluence Ψ at a point
where transient electronic equilibrium exists is
D = β • (μ/ρ) • Ψ (28.29)
At the foundation of radiation dosimetry stays the relationship that can be
established between absorbed dose, kerma and exposure, in conditions of electron
equilibrium. Because for energetic photon beams, in the order of a few MeV, the photon
fluence at a certain point in the medium is related to and depends on the photo n fluence
some distance upstream of the point of interest, the process of photon attenuation is taken
into account and energy -dependant corrections need to be considered. A rigorous
determination of absorbed dose form exposure is limited to energies up to ~ 1 MeV, when
charged particle equilibrium ca be achieved. In these conditions,
Dair = (K col)air = X • W/e = 0.876 (rad/R) • X(R) (29.30)
28
where 0.876 is the roentgen -to-rad conversion factor for air, under electronic equilibrium
conditions.
2.6. Measurement of ionizing radiation
The measurement of ionizing radiation effects was performed in a somehow non –
quantitative way at the beginning of the 20 -th century, when the science of radiology was
born. Biological and chemical effects, like the amount of reddening of human skin, where
considered. But in late 1920’s, the ICRU introduced the unit of exposure, the roentgen, as
a unit of measuring x -ray radiation exposure.
For the first time, the concept of electronic equilibrium was introduced and applied
to the ionization produced by x -rays in free-air ion chamber , which consisted of a
coupling of ion -collection plates connected to a voltage.
The free -air ionization chamber is design for the measurement of roentgen, and it
becam e the primary standard used for the calibration of other exposure and dose
measuring devices. A fee -air chamber is a parallel plate camber; a voltage of ~ 100 V/cm
is applied across the plates and because of this voltage, the electrons created when any x –
ray beam crosses and ionizes the chamber’s air volume are accelerated between and
collected on the plates. The ionization is then measured for a length defined by the
limiting electrical field lines between the plates, defined by the adjacent guard electrod es
(see Figure 7 below). The collected charge is then converted in units of exposure.
Figure 7. Free-air ionization chamber
29
It was for practical reasons that other devices where designed for the purpose of measuring
dose and exposure. Thimble chambers (Figure 8) are ionization
Figure 8. Thimble ionization chamber
chambers of small volumes that function on the same principle as free -air ionization
devices. Special theories were developed in order to describe the process of charge
collection and measurement for small cavities of air enclosed in a very thin solid material,
the walls of the chamber, since conditions of electronic equilibrium must be achieved. The
effective Z of the chamber walls and central electrode need to carefully selected as mo st of
the electrons producing ionization inside such a cavity originate in the immediate
surroundings of the small air cavity. The exposure measured with a thimble chamber is
obtained from
AQX1 (30.31)
where Q is the ionization charge in the cavity volume of denity ρ and volume υ; A is a the
fraction of the energy fluence transmitted through the wall.
Other types of chambers are used for measuring x -rays if higher energies. In 1955,
Farmer designed a chamber tha t became the standard measurement device for all energies
in the therapeutical range. Its collecting volume is 0.6 cm3, its walls are made of graphite,
its central electrode of aluminum and the insulator of polytrichlorofluorethylene. The
charged collected is measured by another device, the electrometer, which holds a constant
bias voltage of ~ 300 V on the collector. The electrometer charge reading is converted in
units of exposure by applying various correction factors like pressure and temperature
correc tion, ion chamber calibration factor for a given energy, stem correction, etc.
30
The dosimetry of ionizing radiation is a very complex topic in the field of medical
physics, because of its high degree of applicability in radiotherapy (Knoll, 2000) . Major
concepts, quantities and units are to be thoroughly understood, since they are the
foundation of all the computational systems and models that were and are still used in the
evaluation and measurement of dose deposited in tissues by ionizing radiation, the
fundamental process that contributed to the emergence of novel fields of medicine, like
radiology, radiation oncology, nuclear medicine.
The calculation of dose in both external beam therapy and brachytherapy is a very
complex process because of the complexi ty of factors that are to be taken into account.
Only the fundamental factors were presented in this paper but many other are included in
current dose calculation formalisms, especially because the evolvement of new
technologies that were adopted into the clinical environment nowadays. Special treatment
procedures like intensity modulated radiation therapy IMRT, stereotactic radiosurgery
SRS, tomotherapy are in use today, and specific factors are to included and accounted for
into dose computations for thes e new technologies.
31
CHAPTER 3 BRACHYTHERAPY
3.1. High Dose Rate Brachytherapy – General Aspects
As previously discussed, the use of remote afterloaders provides the ability to
irradiate tumors at a variety of dose rates, from high dose rate to conventional low dose
rate. Since the method of testing used in this project utilized High Dose Rate
Brachytherapy (HDR), only the characteristics related to this method of treatment are
presented in the following pages.
Since the introduction of remot e afterloading technology in the 1950’s, numerous
authors published studies regarding the advantages and disadvantages of HDR
brachytherapy over the traditional Low Dose Rate (LDR) brachytherapy (Kubo et al ,
1998).
One major concern in the era of LDR brachytherapy was the staff’s exposure to the
harmful effects of ionizing radiation. This led to the development of remote afterloaders,
where the empty applicators are inserted in the patient, then the radiation is delivered
remotely from the control room , thus reducing the staff’s exposure to radiation.
3.1.1. Dose calculations in brachytherapy – TG43 formalism
The most popular and almost universally used dose calculation formalism today is
TG43 (Raviner et al ., 1994) , a formalism used to establish the 2-D dose distribution
around cylindrically symmetric sources. Cylindrical type sources are the most common
shape of clinically used sources and their dose distribution can be described in terms of a
polar coordinate system with its origin at the source cen tre, where r is the distance from
the origin to the point of interest P and θ is the angle with respect to the long axis of the
source, as shown in Figure 9. Point P( r0, θ0) is the reference point that lies on the
transverse bisector of the source at a distance of 1 cm from the origin (i.e. at r0 = 1 cm and
θ 0 = π/2).
32
Figure 9. TG43 dose calculation formalism geometry for a linear source
The dose rate
),(rD at point P in a medium from the center of the source of air kerma strenght Sk
is expressed as:
)( ),()2/,1(),(),( rg rFGrGS rDk (3.1)
Where:
Sk is the air kerma strength (U) of the source [45,54,55];
G(r,θ) is the geometry function (cm-2) that describes inverse square falloff ,
and accounts for the distribution of the radioactive material (see
following paragraphs);
F(r,θ) is the anisotropy function that accounts for angular dependence of
dose due to absorption and scatter by the encapsulation and the
medium, is dimensionless, and is equal to unity on the transverse
axis;
g(r) is the radial dose function that accounts for radial dependence of
dose on the transverse axis due to photon absorption and scatter in
the medium, is dimensionless, and is equal to unity at 1 cm on the
transverse axis.
33
Λ is the dose rate constant (cGy h-1 U-1) and is described in more
detail below;
r is the radial distance (cm) of a point of interest from the source
center;
θ is the polar angle (radian) formed by the longitudinal axis of the
source and the ray from the source center to the point of interest;
G(r,θ)
As already indicated, Λ is the dose rate constant, defined as the dose rate per air
kerma strength S k at 1 cm along the transverse axis of the source, and is given by:
Λ = D (1,π/2)/ Sk (3.2)
It depends on the source type, its construction and material the source is
encapsulated in. At this point we have to clarify that the quantity air kerma strength is
recommended by the AAPM as the standard indicator of a source strength; this quantity, in
turn, relates to another one, the air kerma rate constant Γ δ, which is the quantity ICRU
recommended to replace a former one, namely exposure rate constant (which is the
exposure rate in R/h at a point 1 cm from a 1 mCi point source). Its SI unit is m2Jkg-1h-1Ci-
1. The air kerma rate constant at a distance l from the source is given by
dtdk
Alair2 (3.3)
and it relates the activity A of a radionuclide emitting photons of energy greater than δ to
the exposure rate. The relationship between the air kerma strength Sk and air kerma rate is
given by
2lK Sl k (3.4)
The geometry function , G(r, θ), accounts for the variation of the relative dose with
the distance form the source, due to the spatial distribution of activity within the source.
G(r, θ) reduces to 1/ r2 for point source approximation and to β/( Lr sin θ) for a line source
approximati on with β and L as defined in Figure 12. For idealized source geometry,
AAPM TG 43 recommends the use of the following equation to express G(r, θ):
34
G(r,θ) = r –2 for a point source approximation
G(r,θ)= 𝜃2 −𝜃1
𝐿𝑟 𝑠𝑖𝑛𝜃 for a line source approximation of active length L
L, r and θ are defined in Figure xxx. For points along the transverse axis, the geometry
factor for a line source is given by :
G(r,π
2)=2 𝑡𝑎𝑛−1(𝐿/2𝑟 𝑠𝑖𝑛𝜃)
𝐿𝑟 𝑠𝑖𝑛𝜃 (3.5)
The radial dose function g(r) accounts for the effects of attenuation and scatter in
the medium on the transverse plane of the source (θ = π/2), excluding falloff, which is
included by G(r, θ). It is also influenced by filtration of photons, by the encapsulation
materials and source type and is given by :
g(r)=𝐷̇ (𝑟,𝜋 / 2) ⋅ 𝐺(1,𝜋 / 2)
𝐷̇ (1,𝜋 / 2) ⋅ 𝐺(𝑟,𝜋 / 2) (3.6)
For a point source, this reduces to:
g(r)=𝐷̇(𝑟)𝑟2
𝐷̇ 𝑟02 (3.7)
The anisotropy function , F(r, θ), accounts for the anisotropy of dose distribution
around the source, including the effects of absor ption and scatter in the medium:
),( )2/,()2/,( ),(),( rG rDrG rDrF (3.8)
F(r, θ) is defined as unity on the transfer plane, for θ = π/2 ; however, its value off
the transfer plane decreases: (1) as r decreases; (2) as θ approaches 0ș or 180ș; (3) as the
source encapsulation thickness increases; and (4) as the photon energy decreases.
The dose distributions around brachytherapy sources are calculated assuming
photon interactions only, and are influenced by the emitted radiation and the surrounding
media. The dose at any point from a single finite source can be considered as a summation
35
of doses from multiple point sources. When the source is in free space, no absorption or
scattering effects are present; however, when the source is placed in water, absorption and
scatter will influence the dose rate at any point away from the source.
If the source is approximated by a point, Eq. (3.1) can be simplified to provide the
approximate value of dose rate from an actual point source:
𝐷̇ (r)≈ΛSk𝑔 (𝑟)
r2 Φ an(r) (3.9)
where Φan(r) is a distance dependent average anisotropy factor, which is defined as the
ratio of 4 π averaged dose rate at a given radial distance divided by the dose rate at the
same distance along the transverse axis of the source (Rivard et al., 2004)
3.1.2. Novel computational algorithms – ACUROS BV
Acuros BV is an algorithm that allows dose -to-medium distribution calculations,
in addition to the standard dose -to-water calculations available in moden treatment
planning systems using just the TG43 formalissm. Advantages also include to possibility
of assigning C T values of different materials for structures identified and reconstructed in
a series of CT images, and accounting for heterogeneity effects introduced by these
structures, thus allowing for a more realistic and accurate rendition of the dose
distributio n in inhomogeneous media.
Specific to the Acuros BV is that it calculates dose distri butions through solving
the Linear Boltzmann Transport Equation (LBTE). The LBTE is the equation that
describes the macroscopic behavior of radiation particles (neutrons, gamma -rays,
electrons, etc.) as they travel through and interact with matter. For a g iven volumetric
region , subject to a radiation source, the solution to the LBTE would render a more
accurate dose distribution within that particular region of interest . Though, analytic
solutions to the LBTE can only be obtained for a few simplified situa tions , therefore the
LBTE must be solved in a non -analytic manner.
Of the two general approaches to obtaining such solutions to the LBTE, the first is
the widely known Monte Carlo method, which stochastically predicts particle transport
through media by tr acking a statistically significant number of particles through successive
36
random interactions , and t he second is to explicitly solve the LBTE using mathematical
numerical methods. Monte Carlo methods only obtain an indirect solution to this equation,
while some methods explicitly solve the LBTE equation, as it is the case with Acuros BV.
Acuros BV is a strong alternative to Monte Carlo computational methods in the
field of brachytherapy, with even predecessors of this solver being validated on a broad
range of applications (Gifford et. al , 2006).
Acruos BV is a computational method that solves the time -independent form of
LBTE. For a computational volume, V, with surface, δV, the LBTE is given by:
Ω̂ ·Δ⃗⃗ Ψ+ 𝜎𝑡Ψ= qscat+ ∑qp
4πp
p=1δ(r −r p) (3.10)
Ψ=0,𝑟 ∈ 𝛿 𝑉,Ω̂· 𝑛⃗ <0 (3.11)
where
𝚿 is the angular photon fluence (or flux if not time integrated),
Ψ(r ,E,Ω̂), as a function of position, r = (x,y,z), E, and direction, Ω̂ = (μ, η, ζ);
qscat is the photon scattering source qscat(r ,E,Ω̂), as a function of
position, r = (x,y,z), energy, E and direction, Ω̂ = (μ, η, ζ);
qp is the photon point source, qp(E,Ω̂), for brachytherapy point
source p, at position r p, of which there are P point sources in total;
𝛅(𝐫 −𝐫 𝐩) is the Dirac delta function between the point source location, r p,
and position, r ;
𝝈𝒕 is the macroscopic total cross section, 𝜎𝑡(r ,E ), and
𝒏⃗⃗ is the normal vector to surface 𝛿 𝑉.
The first term on the left -hand side of the above equation is the streaming operator
and the second is the collission operator. The first term on the right -hand side of the
equation is the scattering source, and the second is the source from the prescribed
brachytherapy point sources. The scattering source is also a function of the angular fluence
and is de fined as
37
qscat(r ,E,Ω̂)= ∫∫𝜎𝑠
4𝜋∞
0(𝑟 ,𝐸′ →𝐸,Ω̂ ⋅ Ω̂′)Ψ(r ,E,Ω̂′)𝑑Ω̂′ 𝑑𝐸′ (3.12)
where 𝜎𝑠(𝑟 ,𝐸′ →𝐸,Ω̂ ⋅ Ω̂′) is the macroscopic differential scatter cross section, whose
angular dependence is expressed in terms of Legendre polynomials in μ o = Ω̂ ⋅ Ω̂′.
The LBTE equation is a six -variable integro -differential equation which is solved
by Acuros BV by discretizing three variables in space, (x,y,z), two variables in angle, (μ,
η, ζ = √1−η2 − μ2 ), and one variable in energy, E. By solving the discretized equat ion,
Acuros BV calculates the angular and energy dependent photon fluence at every spatial
degree of freedom in the computational domain.
Because the radioactive sources in HDR are geometrically small compared to the
spatial region in which the dose calcu lation is done, they can be aproximated as point
sources. In this representiation, the collided, scattered particles are transported and
accounted for differently than the uncollided, non -scattered particles.
Therefore, three independent processes of discretization are operated through
mutiple methods: 1) energy discretization, through a multigroup method that divides the
particle energy range of interest, E max ≥ E ≥ E min, into a finite number of intevals, 2)
angular discretization of the collided com ponent, through a discrete ordinates method
using Triangular -Tchebyshev quadrature sets of various orders, ranging from N=4 (24
discrete angles) to N=36 (1368 discrete angles), and 3) spatial discretization, namely the
linear discontinous Galerkin finite -element method (DFEM) (Lewis et al., 1984), in which
the computational volume domain, V, is subdivided in variable sized Cartesian elements,
and the local element size is adapted based on the anatomic al and applicator material
properties and gradients in th e scattered photon fluence.
In brachytherapy radioactive sources low energy ranges, specifically for a 192Ir
mean weighted energy of roughly 350 keV, electron equilibrium can be assumed to exist,
therefore the kinetic energy released per unit mass, KERMA, approximation can be
employed, which in turn eliminates the need to account for secondary electrons. When
KERMA approximation is used, dose calculation is done using the following formula:
38
𝐷(𝑟 )= 1
𝜌(𝑟 ) ∑𝜎𝐾𝐸𝑅𝑀𝐴,𝑔(𝑟 ){∑𝛷𝑝,𝑔𝑢𝑛𝑐(𝑟 )+𝑃
𝑝=1 𝛷𝑔𝑐𝑜𝑙𝑙(𝑟 )}𝐺
𝑔=1 (3.13)
where
𝐷(𝑟 ) is the dose at position 𝑟
𝜌(𝑟 ) is the density at position 𝑟
𝜎𝐾𝐸𝑅𝑀𝐴,𝑔(𝑟 ) is the macroscopic KERMA cross section
∑𝛷𝑝,𝑔𝑢𝑛𝑐(𝑟 )+𝑃
𝑝=1 𝛷𝑔𝑐𝑜𝑙𝑙(𝑟 ) is the total scalar fluence, that includes uncollided
and collided components, at position 𝑟
The fundamental data used by Acuros BV are the macroscopic atomic cross
sections accounted for all four types of photon interactions of photons with matter:
Comp ton and Rayleigh scatter (incoherent and coherent, respectively), photoelectric
effecta dn pair production. Macroscopic cross sections are obtained from two physical
quantities: the microscopic cross section for a given reaction, and the mass density of th e
material. In order to perform a dose calculation, Acuros BV has each voxel of the grid a
macroscopic section assigned, with their correspondent densities derived from HU values.
3.1.3 Radiobiological models
Although there are several advantages of HDR over LDR [41,43,57], the biggest
controversies over the two techniques of dose delivery, were related to the radiobiological
effects. For LDR the most significant advantage in terms of radiobiology is the dose -rate
effect, where for low doses and low dose -rates the repair of sublethal damage takes place.
It was demonstrated, that the dose -rate effect is greater for normal cells than for tumor
cells and this is the reason why the fractionation and dose -rate are playing major roles in
both external beam radia tion therapy and brachytherapy [59 -63]. Based on this theory, in
order to obtain comparable clinical results with HDR as with LDR, the dose per fraction
and fractionation needed to be increased.
39
3.1.3.1 The linear quadratic model
Historically, the comparison between LDR and HDR was done by the application
of the Linear Quadratic Model (L -Q model). In this model the biologically effective dose
(Fowler et al, 1992) (BED) is expressed as:
BED = – (ln S.F.) / α = NRt [ 1 + G x Rt / (α/β)] – kT (3.14)
where:
S.F. is the cell surviving fraction
N is the number of fractions
R is the dose rate expressed in Gy/h
t is the time for each fraction
T is the overall time, expressed in days, available for repopulation
α, β are tissue specific parameters, with α relating to the initial slope of the cell
survival curve, and β defining its curviness
α/β is the dose in Gy for which α and β are equal [61]
G is a function of the irradiation time, the dose rate, the cellular repair and
time between fractions
40
The calculation of G is different for LDR vs. HDR treatments. For conventional
LDR treatments G can be expressed as:
G = 2/µt [1 – (1 – e –µt) / µt] (3.15)
where µ is the repair -rate constant [59].
For HDR treatments, where the time between fractions is long compared to the
half-time for repair, and the fraction is short, time t is approaching zero, so in this case G
equals unity.
Since each HDR treatment is delivered in a time that is short compared to the half –
time for repair, it is obvious that the difference between LDR and HDR is that significant
repair occurs during LDR exposure, while only negligible repair takes places during a
short HDR procedure [59].
Several publications contrasted and compared the effects of LDR and HD R with
respect to tumor control and late effects [60, 65 -67], and it was concluded that at least
from radiobiological standpoints HDR can safely replace LDR if enough fractions are
delivered and the total doses are reduced accordingly.
3.1.4 High Dose Rate unit description and source calibration
The high-dose rate remote afterloader used in our department (Figure 10) is a
VariSource iX HDR afterloader (VARIAN Medical Systems, Inc., Palo Alto, CA) that
uses a patented active wire of 0.59 mm in diameter, with two built -in Iridium -192 (192Ir)
source s, of 5 Ci (185 Gbq) nominal activity each . Both sources are encapsulated in a
homogeneous nickel titanium alloy wire so that there are no welds to break, that would
potentially create stiff spots in the wire itself. 192Ir is the radioisotope of preference for
HDR afterloaders, because of its small dimensions, high specific activity, and low photon
energy. Due to its short half -life, the 192Ir source requires replacement at least every 4
months , or wh en the control system warns about approaching the cycle limit of 1000
cycles for the active wire. When not in use, the active source wire is dwelling in a
shielded safe inside the HDR unit and is only remotely extended, and automatically sent
41
from the sto rage place and into the applicators during treatment s or quality assurance tests
peformed by the Medical Physicist . The design of the HDR unit includes 20 individual
channels, the source can be moved into sequentially via the turret of the unit . The source
wire travels from the safe, through the unit’s channels into the transferring tubes and
finally in the applicators that are placed interstitially or intracavitary into the tumor.
Before the i nitiation of each treatment, an inactive, dummy wire tests all c onnections
before an actual treatment run is permitted, to verify that they are free of obstructions. If
the path is clear the source wire is engaged, and dwells at predetermined positions for the
prescribed and planned treatment times.
Figure 10. Depiction of a VariSource iX unit, with its main components
One of the most important tasks a Medical P hysicist is required to perform is
patient specifi c quality assurance (QA) (Khan, 2003) . A robust quality assurance program
is mandatory for the smooth operation of an HDR department . Radiation oncology
facilities licensed for the use of HDR are following well established safety and emergency
procedures, addressing patient and facility pre- and post – treatment surveys , redundant
checks of the patient s’ identity, dose prescription, treatment times and location of the
42
tumor. The HDR unit and console operation are tested prior to delivery of each treatment
for safety interlocks, source decay and treatment time calculation (Kutcher, 1995).
The 192Ir source has a half -life of 73.82 days with a rate of decay of roughly
1%/day. In order to ensure an optimum rate of dose delivery , the active wire is replaced
every 12 to 14 weeks, process that implicitly requires the new installed source to be
comprehensively tested for mechanical and safety interlocks functionality, positioning
accuracy and reproducibility. One of the most important tests performed is the source
calibration, procedure summarize d below.
As already mentioned, the strength of a brachytherapy source can be specified in
terms of activity, apparent activity, exposure rate at a specified distance or air kerma
strength. The American Association of Physicists in Medicine recommends the use air
kerma strength (S k) to specify the source activity. S k can be determined from the
measurement of the exposure rate X measured in free ai r at a distance of 1m from the
source following the formalism (Beaulieu, 2012) ,
Sk = Kl x l2 (3.16)
where K l is the kerma rate in free space, and l is the distance at which is specified, usually
1m. The relation between kerma and exposure is:
K = X (W/e) avg x [ (µ tr/ρ )avg / (µen/ρ)avg] (3.17)
where K is kerma, X is exposure, (W/e) avg is the average energy absorbed per unit charge
of ionization in air, (µ tr/ρ )avg and (µ en/ρ)avg are the average values of the mass transfer
coefficient and the mass energy absorption coefficient of air for the photons. In the energy
range of brachytherapy photons µ en/ρ ≈ µ tr/ρ, so Eq. 3.17 becomes :
K = X (W/e) avg (3.18)
We can then express the air kerma strength as:
Sk = Xl (W/e) avg l2 (3.19)
43
When the exposure rate is measured in R/h at l = 1m, and with the value of (W/e) avg =
0.876 cGy/R for dry air, air kerma strength for a brachytherapy source to be used in high
dose rate brachytherapy is expressed as:
Sk = X(R/h) (8.76 x 103 m2 µGy/R) (3.20)
The calibration of 192Ir sources is routinely performed using a well -type re -entrant
ionization chamber. One type of a well -type ionization chamber is filled with air and
communicates to the outside air through a vent hole. The chamber has an outer shell of
conductive material with walls that are forming an inner well electrically connected to and
disposed within the outer shell (Goetsch, 199 1). The active volume of the chamber is
usually 245 cm3, which give an optimum ionization current that is measured with an
electrometer. The chamber’s sensitive volume is defined between the inner well and the
outer shell.
The value of air kerma strength can be determined from the measurement of the
ionization current produced by the 192Ir source in the ionization chamber, corrected for ion
recombination , temperature and pressure at the time of source and chamber calibration
(Khan, 2003).
Sk = I x C T,P x N el x N c x A ion x P ion (3.21)
where:
I is the current reading, expressed in nA ,
CT,P is the correction for temperature and pressure ,
Nel is the electrometer calibration factor ,
Nc is the chamber calibration factor ,
Aion is the ion recombination correction factor at the time of chamber
calibration ,
Pion is the ion recombination correction at the time of source calibration.
44
Traditionally, orthogonal films were used for the purpose of treatment planning for
HDR b rachytherapy but nowadays the most frequent imaging modality employed is
Computer Tomography (CT) , which allow s for a full 3D anatomy reconstruction, dose
calculation and isodose dis tribution (Williamson, 199 6). As described in detail earlier in
the chapter, the recommended dose calculation formalism for point and linear sources is
AAPM’s TG -43 (Nath, 1995).
Dose distribution analysis in the irradiated volume is most efficiently done using
dose-volume histogram curves, since they allow individual patient plan evalutation and
provide an excellent comparison tool for subsequent plans performed on the same CT data
set (Gurdalli, 2008) . DVHs allow for dose uniformity assesment, evaluation of the extent
of hot spots in the irradiated volume, and greatly help in the plan optimization process.
45
CHAPTER 4 OVERVIEW OF BREAST CANCER TREATMENT
MODALITIES
4.1. Breast Cancer and anatomy
Breast anatomy is very important in order to understand the methods used in
radiotherapy. The breast is made up of the mammary gland, fat, blood vessels, lymphatics
and nerves ( see Figure 1 1). The surface of the breast has deep attachments of fibrous septa
which runs between the superficial fascia attached to the skin and the deep fascia covering
the pectoralis major and the other muscles of the chest wall.
The mamma consists of glandular tissue arranged in multiple lobes composed of
lobules connected in ducts, blood vessels an d areolar tissue. The smallest lobules are
arranged in clusters of alveoli that open into the small branches of the lactiferous ducts,
which in turn form larger ducts that converge into single canals in the nipple,
corresponding to each lobe of the gland. A network of lymph vessels encompasses the
entire surface of the chest, neck and abdomen and become very dense under the alveola.
Figure 11. Breast anatomy : A ducts B lobules C dilated section of duct to hold milk D
nipple E fat F pectoralis major muscl e G chest wall/rib cage
46
The breast can be virtually divided into four separate quadrants ( see Figure 12).
The most common site of origin of breast cancer is the upper outer quadrant (UOQ)
(~38%), and the incidence decreases in the following order: central area (~30%), upper
inner quadrant (UIQ), lower outer quadrant (LOQ) and lower
Figure 12. Quandrants of a left breast
inner quadrant. The rates of incidence correlate very well with the amount of breast tissue
in the respective quadrants. The tumors usually travel along the ducts, breaking then
through the basement membrane of the duct, invading adjacent lobules, ducts and
mammary fat, spreading through the lymphatics and eventua lly into the peripheral
lymphatics.
The staging of breast cancer is actually related to the amount of tumor
dissemination. Staging helps determining the best treatment available and estimates
prognosis. The most widely used staging system for breast can cer, based on tumor size,
number of lymph nodes involved and evidence of distant metastasis, is the TNM system
developed by the American Joint Committee on Cancer. The most common histological
type of breast cancer is infiltrating ductal carcinoma (~75%), followed by infiltrating
lobular carcinoma (~10%), the rest consisting of other histological types.
4.2. Treatm ent modalities
As it was already mentioned in the introduction, the t reatment of breast cancer can
be a combination of local management and sys temic treatment. When radiation therapy is
employed, as a breast -conservation therapy alternative to mastectomy alone, early trials
47
indicated that whole breast irradiation significantly improved the manangement of risk
recurrence after surgery, and whole b reast irradiation became a standard component of
breast cancer therapy (Wazer et al., 2006) . The standard radiotherapeutical options
available for breast cancer are external beam therapy and internal radioisotope therapy, or
brachytherapy. We will breafly survey both techniques, with an emphasis on the latter.
In external beam therapy, immobilization and positioning of the patient is of major
importance and reproducibility of the treatment set -up very important in order to warrant
the consistency o f treatment delivery. Traditionally, the patients are positioned in a supine
position, with the arm on the involved side elevated above the shoulder level (see Figure
13).
Figure 13. Common set -up for external radiation for breast: A breast tissue, B radiation
beam, C radiation field LINAC collimation, D arm support system
different directions in order to spare as much normal tissue as possible. The breast and
chest wall are normally treated with opposed tangential fields, and the supraclavicular and
axillary nodes with a combination of anterior and posterior fields adjacent to the tangential
fields. Matching the field margins is extremely difficult and requires a lot of attention, as
overlapping can result in unreasonably high doses at depth. Various treatment accessories
are used in order to achieve uniform distribution of dose in the breast tissue. Physical
wedges of different angles can tilt the dose profiles as to make up for the oblique
incidence on the breast and consequent air gaps involved. A common dose distribution in
external beam therapy is shown below, in Figure 14.
48
Figure 14. External beam therapy dose distribution
While external beam therapy provides excellent tumor coverage and control, it also
comes with high dose deposition in underlying tissues, especially the lung, heart (for felt
breast treatments), an d skin. Due to scatter form the wedges, when high energy photon
beams are used, the opposed breast can receive undesirable amounts of radiation,
sometimes linked to and promoting secondary breast cancer incidence in the opposing
breast.
Certain disadvant ages of external beam radiation therapy for breast cancer, mostly
related to i ts relative complexity and relat ed expenses, led to investigations done in the
direction of accelerated radiation courses that would eventually alleviate those problems.
Even though these many different strategies of delivering APBI, including low -dose-rate
interstitial brachytherapy, HDR brachytherapy, orHDR interstitial brachytherapy differ
with respect to certain variables, they all have in common the shortening of the treat ment
schedule from 6 to 7 weeks to a course extended over 1 week or less. The following
subchapter is offering an historical overview of brachytherapy options for breast cancer.
4.3. Brachytherapy in the treatment of breast cancer. Developments.
The hist ory of brachytherapy can be traced all the way back to the beginning of the
20th century, with the use of radium. The radium and radon gas where used in tumor
therapy in the years between the two world wars, and techniques were developed either in
49
Sweden o r France. Low intensity radium tubes were used in the early days, and soon,
Patterson and Parker developed special techniques for radium clinical applications in
England, still in use today. Quimby studied the dose distributions of radium packs in USA
and published dosimetry tables for individual sources, establishing simple rules to follow
in clinical practice.
The years of 1950 to 1960 brought into picture a lot of new isotopes, due to the
flourishing of research in the atomic field, some of them rapidly integrated into the field of
medicine, because of their properties and features. Afterloading techniques were
developed during the same years, because of increasing concern and knowledge about the
effects of radiation exposure on the radiation workers, mos tly practicing physicians.
The 1990s were the booming years of brachytherapy because of the discovery of
new isotopes, of the more refined dosimetry and because of the new treatment
brachytherapy methods, most of them employing remote afterloading techniq ues.
Brachytherapy (sometimes referred to as curietherapy or endocurie therapy) is
nowadays a term used for the short distance treatment of cancer with radiation from small,
encapsulated radionuclide sources. Brachytherapy is administered by placing source s
directly into or near the volume to be treated. The dose is then delivered continuously,
either over a short period of time (temporary implants) or over the lifetime of the source to
a complete decay (permanent implants). Most common brachytherapy source s emit
photons but, in certain situations, sources emitting beta or neutron radiations are used.
There are two main types of brachytherapy treatments :
● Intracavitary, in which the sources are placed in body cavities close to the
tumour volume;
● Interstit ial, in which the sources are implanted within the tumour volume
Historically, partial breast irradiation was first performed with interstitial
implantation using multi -catheter brachytherapy that typically triggered the tumor bed plus
a generous margin of 2.0-2.5 cm . While in the 1970s and 1980s LDR iridium interstitial
implants of the breast were sometimes used to boost the dose to the site of the tumor
following whole breast irradiation, this most recent wave of applications uses
brachytherapy to implement 1 week long accelerated partial breast irradiation as an
alternative to 6 weeks of whole breast external beam radiation therapy. In the early 1990s,
50
Kuske and colleagues developed techniques for brachytherap y implants following
tylectomy – segmen tal mastectomy and subsequently lumpectomy – as the sole
radiotherapy. Initially, the implants consisted of two planes of catheters laid in place
during the surgical procedure , which did not produce good implant geometries upon
closing the operative site , but soon the implants were performed after healing of the
surgical site, guided by digital mammograms, ultrasound, or computer tomography.
In attempts to combine the simplicity of the single catheter insertion and the ability
of the multi -catheter implant s to tailor the dose distribution to the target, a new set of
applicators were developed , the MammoSite balloon type devices, that contained several
cathe ters – anywhere from seven (7) to eleven (11) – connected at the tip and back,
bunched together for in sertion into the cavity, that then spread into a more spherical shape
to fill the cavity. A hyb rid design, ConturaTM Multilumen Balloon (MLB) -SenoRx, Alis o
Viejo, CA, has several catheters inside of a balloon. Another novel device, SAVI – Cianna
Medical, Aliso Viejo, CA – uses the same catheters to spread the cavity and provide paths
for the source.
In concept, having multiple catheters in the breast cavity allows reducing the dose
in some directions, such as toward the skin, while continuing to push the d ose to the
prescription distance in other directions, covering the target with increased uniformity. It
is the purpose of this study to dosimetrically evaluate the SAVI device and to establish a
sound procedural scheme for the clinical implementation of th is device.
4.3.1 Partial breast irradiation . Brachytherapy devices and techn iques.
With evolving brachytherapy techniques in recent years, a lot of research was done
in the attempt to find better treatment options for patients diagnosed with early stages
breast cancer. It is for this reason why the National Surgical Adjuvant Breast Proje ct
(NSABP) was initiated in USA, which tried to produce general and standard treatment
guidelines for novel treatment techniques emerging from the brachytherapy side mostly.
Breast conserving therapy (BCT) has become an accepted option in the treatment
of most patients with early stage breast cancer (stage I and II). The major advantages of
BCT are superior cosmetic results and reduced psychological and emotional trauma
51
compared to mastectomy. However, BCT also has relative disadvantages. The technique is
a more complex and prolonged treatment regimen requiring approximately 5 to 7 weeks of
treatment. For patients who are elderly or who live a significant distance from treatment
centers, who can not be accommodated or do not have access to in -patient hospita ls,
logistical problems can be overwhelming. Thus, despite the obvious cosmetic and
potential emotional advantages of BCT, less than half of the patients who are candidates
for breast conservation actually receive it. Most of the logistical problems associ ated with
BCT relate to the protracted course of external beam RT delivered to the whole breast. As
we mentioned earlier, standard therapy regimens after tumor excision usually take up to 5
(25 fractions) weeks of external beam RT to the whole breast (45 -50 Gy) followed by a
boost to the tumor bed with either an additional 8 to 10 fractions (1 fraction/day) of
external beam RT or a 2 to 3 day interstitial implant. The rationale for this approach is
based upon two principles.
First, higher doses of RT are g iven to the “tumor bed” in an attempt to control
small foci of cancer which may be left behind after excision alone. Second, WBI is used to
eliminate possible areas of occult multicentric in situ or infiltrating cancer in remote areas
of the breast. That s uch remote, multicentric areas of cancer exist has long been
established. However, the biological significance of these areas of occult cancer is
unknown, and the necessity to prophylactically treat the entire breast has recently been
questioned. For insta nce, there are recent trials that indicate that the majority of
recurrences in the breasts of patients who did not receive RT occurred at or in the area of
the tumor bed. In addition, the rate of development of new cancers in remote areas of the
breast (un related to the original lesion location) was similar whether or not WBI was
administered.
It is therefore not clear that the inclusion of these safe margins is in fact helping
with the tumor remission. In a similar pattern to invasive breast cancer, in no n-invasive
breast cancer (DCIS), 75% of recurrences developed in or adjacent to the original tumor
site following breast conservation surgery with or without the use of radiation. Previous
work has shown that less than 10% of DCIS cases have satellite lesi ons more than one
centimeter from the original tumor. A trial from Ochsner Clinic involving patients with
DCIS treated by brachytherapy resulted in a substantially low local recurrence rate.
Adding additional strength to this argument is the observation th at in patients undergoing
standard BCT and treated with conventional WBI, the development of new cancers in the
52
ipsilateral breast (remote from the primary lesion) is similar to that observed in the
contralateral breast. Certain trials indicate that electi ve treatment with RT beyond the
lumpectomy bed provided minimal additional benefit. If this were the case, radiation
therapy could be delivered in 1 to 2 weeks, thus significantly shortening treatment time
and potentially reducing health care costs. A shor tened treatment schedule would decrease
the burden of care for patients undergoing BCT, thus making available the conservation
option for more women.
By reducing the length of time required to deliver RT, the logistical problems
associated with integratin g local and systemic therapies would also be eliminated.
Additionally, toxicity to adjacent normal structures (i.e., heart, underlying chest wall,
contralateral breast) should be reduced significantly by decreasing the volume of irradiated
tissue.
Several recent studies on the use of RT for breast cancer patients clearly document
a reduction in cancer -specific mortality during the first 5 -10 years after treatment that is
partially off -set by late effects of radiation on adjacent tissues. Since it remains u ncertain
if the additional volume of normal tissue that is irradiated (in order to encompass the entire
breast for presumed occult disease) provides any additional benefit in reducing breast
cancer recurrence, the potential detrimental effects of this addi tional RT would be
eliminated. There currently is a large body of mature Phase I and II data (and some
preliminary Phase III findings) that have investigated the replacement of WBI after
lumpectomy with an accelerated course of radiation therapy delivered in only 4 -5 days and
restricted to the region of the tumor bed. Five -year results from the majority of these trials
have demonstrated local control rates in the breast comparable to those observed after
conventional WBI. Most of these data have been genera ted using interstitial breast
brachytherapy . This is an invasive procedure consisting of temporarily placing a series of
10-20 needles or catheters in the breast to encompass the lumpectomy cavity (Figure 15)
53
Figure 15. Interstitial brachytherapy for breast
These catheters are then loaded with a radiation source, which delivers a
tumoricidal dose of RT to the lumpectomy cavity region alone in 4 -5 consecutive days,
generally as an outpatient procedure. With the patient's entire RT course completed, the
catheters are then removed.
This technique has not yet gained widespread popularity because of the relative
complexity associated with performing an interstitial implant and the lack of significant
patient interest in an ad ditional invasive procedure. As a consequence, PBI as a potential
treatment option has been limited to only a handful of institutions with experience in
interstitial breast brachytherapy. Several recent developments have occurred that are
rapidly increasin g interest in the use of PBI as a treatment option. First, newer “user
friendly” interstitial breast brachytherapy techniques have been developed that are more
easily taught and performed and that are more comfortable for patients.
Figure 16. The Mammo Site device
54
In addition, two interstitial breast brachytherapy schools (one conducted by the
American Brachytherapy Society) have also been developed and are conducted on a
regular basis. Second, single -entry, single – and multi -lumen intracavitary FDA appr oved
devices are now available simplifying the brachytherapy technique and providing a more
reproducible method to perform breast brachytherapy, allowing many more physicians and
institutions the opportunity to deliver high quality PBI. The single -lumen in tracavitary
balloon was the first available device that was soon followed by three multi -lumen FDA
approved intracavitary devices. Proxima Therapeutics was the first to develop and achieve
FDA approval for a new breast brachytherapy catheter (MammoSite® – Figure 16 on the
previous page). Since the device consists of only one catheter temporarily positioned in
the breast (as opposed to 10 -20 needles), patient comfort is potentially improved.
Mammosite gained a lot of popularity after its promotion and is now widely used in USA.
Since the initiation of NSABP B -39/RTOG 0413 in 2005, there have been three
FDA (Federal Drug Administration – the entity that regulates the use of medical devices in
USA) approved single -entry multi -lumen intracavitary devices that h ave been introduced
to the market: MammoSite® ML (Multi -Lumen), Contura MLB (Multi -Lumen Balloon),
and SAVI® (Strut -Adjusted Volume Implant) (Figure 17)
Figure 17. The Contura Multi -Lumen Balloon and the SAVI device
55
These multi -lumen devices offer a more sophisticated and better optimized
radiation delivery approach as compared to the original single -lumen MammoSite®
balloon catheter , even though minimal dose optimization can be obtained with the single
catheter Mammo Site as well (Dickler et. al , 2004) . All three of these intracavitary devices
preserve the single -entry concept while providing the radiation oncologist with the ability
to improve dosimetric coverage of the target and reduce dose to nearby rib and skin wh en
needed. Through an increased ability to achieve the dosimetric goals outlined in this
protocol, these devices provide the potential to expand the number of patients who can be
appropriately treated with an intracavitary brachytherapy approach. Third, th ree-
dimensional (3D) conformal external beam radiation therapy techniques have also been
developed and successfully used to treat patients with PBI using a similar, shortened
treatment schedule.
This 3D technology is readily available in the majority of r adiation facilities
allowing many more radiation oncologist groups that do not perform brachytherapy to
deliver PBI. Perhaps the greatest advantage of this method of PBI is the fact no additional
invasive procedure is required. Collectively, these three de velopments have generated
tremendous interest in PBI since there are now several comparable and reproducible
techniques available along with 5 -year interstitial breast brachytherapy data demonstrating
efficacy. Despite these findings, mature Phase III data documenting the long -term efficacy
of this treatment approach and the group of patients most suitable for its application are
not yet available. Since the majority of patients treated in the Phase I/II PBI studies
discussed above were highly selected and treated at only a handful of institutions, it
remains to be determined if the excellent results observed to date are a reflection of the
true efficacy of PBI or to other confounding factors. The only scientifically valid approach
to resolve this concern is in the completion of a well -designed, sufficiently powered Phase
III study comparing outcome in similarly staged and selected patients randomized between
standard WBI versus accelerated PBI. Therefore, NSABP protocol proposes that selected
patients underg oing BCT for stages 0, I, and II breast cancer be randomly assigned after
lumpectomy to either standard WBI or PBI. Patients randomized to the PBI arm will be
treated with one of three different types of PBI depending upon the experience and
credentialing of the participating institution, technical considerations, and patient
preference.
56
CHAPTER 5 DOSIMETRICAL EVALUATION OF A STRUT –
ADJUSTED -VOLUME -IMPLANT SAVI DEVICE USED FOR
ACCELERATED PARTIAL BREAST IRRADIATION
5.1 Device description
The SAVI (Strut -Adjusted -Volume Implant) device is a multi -catheter, single entry
device manufactured by Cianna Medical, Aliso Viejo, CA, USA, that received 510(k)
clearance in July 2006. The device was slowly but gradually introduced and gained
popularity on the U S market, after the first patient cases reportedly being performed as
early as the late 2008. The first preliminary scientific papers eventually including the
SAVI device as a treatment option on their studies concerning accelerated partial breast
irradiat ion were eventually first accepted and published no sooner th an early of 2009
(Yashar et al., 2009), with publication of first clinical follow -up studies shortly thereafter
(Yashar et al. 2009; Yashar et. al, 2011) . The Texas Cancer Center clinic (one of the over
175 practice locations of one of the largest private oncology companies in US), where the
candidate functions as a solo Physicist since late 2003, was elected to be one of the beta
testing sites for the SAV I device, and therefore our first planned and delivered patient
cases were the first ones done in Texas, and among the firsts in the US , in November of
the year 2008.
The device is shown in Figure 18 in both expanded (post -insertion) and collapsed
(pre-insertion) format. The SAVI device comes in four different sizes, as displayed in
Figure 1 9. It consists of a central strut surrounded by 6, 8, or 10 peripheral struts,
depending on the size of the device. The configuration of the struts allows for a
differen tial radioactive source dwell -time loading, which translates in optimal dose
modulation around the lumpectomy cavity and sparing of adjacent normal tissues.
57
Figure 18. The SAVI device: A) collapsed and B) fully expanded
The major advantage of the SAVI device is patient -specific dose optimization from
the multiple dwell positions in each strut to minimize dose to normal tissues, including
skin, chest wall, and lung. A fixed hub located near the base of the implant and an
expansion tool that slides over the central strut allows for the expansion of the device once
inside the lumpectomy cavity, and for the collapsing of the device upon removal. The
device is inserted through the incision site in the collapsed position and is expanded when
fully inserted in the lumpectomy cavity . The fully expanded peripheral struts anchor the
device against the cavity walls securing the struts in place and a very stable position, if the
SAVI device size elected for the implant properly fits th e lumpectomy cavity size and
shape.
Figure 19. SAVI applicator sizes: 6 -1Mini (top), 6 -1, 8-1 and 10 -1 (bottom)
A
B
58
The expansion tool of the SAVI device is used at the time of the surgical placement
in the lumpectomy cavity, and then reinserted and engaged prior to each treatment, for
quick collapse and removal of the device in case of an emergency. Radio -opaque markers
are built -in on three of the peripheral struts (numbers 2, 4, and 6) for identification during
the reconstruction process, the one on numbe r 2 strut (M2) being located distally, closest
to the tip of the implant, the one on number 4 (M4) being located midway along the length
of the device and the number 6 marker (M6) being located most proximally. The different
components of the SAVI device a re demonstrated in Figure 20.
Of note, an experimental design version of the SAVI device, a double helix
prototype, having only three struts, one central to two peripheral ones, was tested in 2010,
and proved to dosimetrically feasible, but it never made it in the market, for clinical use
(White et al., 2010) .
Figure 20. The different components of the SAVI device
5.2 Patient selection criteria
Patient eligibility criteria for a SAVI implant verify those currently and widely
accepted and employed in the industry , and in compliance with American Brachytherapy
Society (ABS) and American Society of Breast Surgeons (ASBS) guidelines and the
59
ASTRO consensus statement. As a rule, only patients with invasive breast cancer or ductal
carcinoma in situ , stage 0, I or II breast cancer resected by lumpectomy, up to 3 cm tumor
size, and excised with negative lumph nodes involvement, age ≥ 45 years, were accepted.
Device selection and appropriateness is assessed and done by the breast surgeon during
surgery, after the tumor is excised and the lumpectomy cavity is created.
5.3 Equipment
In order to evaluate the dosimetric features of the four (4) different SAVI devices
subtypes described in the previous section, the following equipment was also used in this
study designed to explore and demonstrate the SAVI device applicability to a diverse array
of clinical situation and scenarios often encountered in APBI.
The CT scanner used in this study was an 8 slice GE LightSpeed helical multislice
CT Scanner. All treatm ent plans were created and optimized using a Varian BrachyVision
Treatment Planning System – Brachytherapy Planning Software (Version 13.6).
5.4 Structure Definitions and Nomenclature
As emphasized in many publications on breast brachytherapy (Arthur et al., 2003;
Patel et al., 2007; Hong et al ., 2012; Mooney et al., 2015) , dose volume histogram
evaluation is the standard method of dosimetric plan characterization, usually used to
determine the reliability of an implant, as eventually reflected in skin an d normal tissue
toxicity. Therefore we used the following prescribing, optimizing and reporting indices in
this study:
Rx Dose – prescribed dose, expressed in cGy.
PTV – Ideal Planning Target Volume. This is the intended treatment volume (cubic
centimeters).
PTV_EVAL – Adjusted Planning Target Volume. This is the actually treatment
volume used for optimization and coverage evaluation (cubic centimeters).
Isodose Curve – A geometric curve graphically documenting all the points that
receive an equ al radiation dose.
60
DVH – Dose -Volume Histogram. A plot of a cumulative dose -volume frequency
distribution that graphically summarizes the simulated radiation distribution within
a volume of interest of a patient that would result from a proposed radiation
treatment plan.
Coverage index (CI) – a measure of the fraction of the target volume receiving a
dose equal to or greater to the prescribed dose, i.e. V100 expressed as a percent.
V100 – Volume of tissue receiving at least 100% of the prescribed dose, e xpressed
in absolute terms (cubic centimeters) .
V95 – Volume of tissue receiving at least 95% of the prescribed dose, expressed as
a percentage of the total target volume.
V90 – Volume of tissue receiving at least 90% of the prescribed dose, as a
percent age of the total target volume.
D90 – the percentage of the prescribed dose delivered to 90% of the PTV.
V150 – the volume of tissue receiving at least 150% of the prescribed dose,
expressed in absolute terms (cc’s).
V200 – the volume of tissue receivin g at least 200% of the prescribed dose,
expressed in absolute terms (cc’s).
Max Skin Dose – the maximum dose as calculated by the DVH in the Max Skin
structure, expressed in absolute terms (cGy) or in relative terms (as a percentage of
the prescribed dose ).
Max Chest -Wall/Rib Dose – the maximum dose as calculated by the DVH in the
Max Chest -Wall or Rib structure, expressed in absolute terms (cGy) or in relative
terms (as a percentage of the prescribed dose).
DHI – dose homogeneity index, which is the equivalent to the fraction of the total
treatment volume which receives a dose between 100% and 150% of the Rx dose
5.5 Treatment planning technique
The CT images are sent to a BrachyVision treatment planning computer (Varian
Medical Systems, Inc., Palo Alto, California) where a full three -dimensional
reconstruction of the SAVI device is performed. Planning is performed only once, as re –
61
planning is only employed and necessary if intra -fractional in -out or rotational motion of
the device is assessed and c onfirmed.
Treatment planning is usually more time consuming than for a typical MammoSite
balloon applicator, but planning times are not prohibitive as standard template plans are
created for each of the SAVI device types. Standardization of created APBI S AVI cavities
was later adopted by others (Dahl et al ., 2014), since it improves consistency of SAVI
cavity reconstruction. This allows for quick digitization and reconstruction of the multiple
struts. The lumpectomy or SAVI cavity is defined by the physici an, as this becomes the
reference structure from which all planning target volumes are eventually obtained.
CT images for all 4 SAVI devices subtypes were exported to the BrachyVision
Treatment Planning . A carefully designed selection procedure is employed, in order to
select and model each device type based on published dimensional data. Each device type
was evaluated in axial and multiplanar reconstruction views for long axial and saginal
lenghts, symmetry and conformance of the applicator, in or der to closely match the
manufacturer’s specifications, see Table 1 below.
Table 1. Summary of criteria and recommendations for SAVI size selection based on
lumpectomy cavity dimensions (Courtesy of Cianna Medical)
Long axis of cavity
(cm) Diameter of cavi ty (cm)
2-3 3-4 4-5
2-3 SAVImini n/a n/a
3-4 SAVImini SAVImini n/a
4-5 SAVImini SAVImini n/a
5-6 SAVI6 -1 SAVI6 -1 SAVI6 -1
6-7 SAVI6 -1 SAVI 8-1 SAVI 8-1
7-8 SAVI 8-1 SAVI 10-1 SAVI10 -1
The Planning Target Volume (PTV) is generated by a 1 cm uniform expansion of
the lumpectomy cavity volume, and it is defined as the difference between the expanded
volume and the cavity volume. The Planning Target Volume for Evaluation (PTV_EVAL)
is, according to the definition giv en by the National Surgical Adjuvant Breast and Bowel
Project (NSABP) B -39/Radiation Therapy Oncology Group (RTOG) 0413 Protocol [17],
62
the same as the PTV but limited to 5 mm from the skin surface and by the posterior breast
tissue extent (chest wall and p ectoralis muscles not included).
Other structures are created, including but not limited to these: Air/Seroma , Heart ,
Normal Tissue , Chest Wall , Ipsilateral Lung and Skin Surface .
The library of created structures was created based and following closely on the
recommendations of the NSABP B -39/RTOG 0413 protocol. The Body structure is
automatically reconstructed by the BrachyVision software upon the import of any CT
studies into the system. The device is then accurately reconstructed, following a minutious
ste-by-step procedure, and all other structures associated with the device and with the
(hypothetical ) surrounding healthy tissue are created for prescription and optimization and
described below.
For each device studied the SAVI cavity is reconstructed by using the met al frame of
the device, easily detectable in the CT axial, sagital or coronal views . 2.5 mm CT slices
are always used, with no gaps between the slices, in order to provide excellent image
resolution. Patients are always scanned with arms up and with a breath hold technique that
allows contiguous and smooth data acquisition, free of any m otion artifacts.
When reconstructing the cavity, we always aim for replicating the physical
dimensions of the SAVI applicator (Figures 21 -24), according to the manufacturer’s
specifications indicated below (the length refers to the distance between the pr oximal hub
and the base of the distal tip), Table 2:
Table 2. SAVI devices physical reference dimensions
Type of Device Length of central shaft Width/Max Expansion
mm mm
SAVImini 50 24
SAVI6+1 61 30
SAVI8+1 67 40
SAVI10+1 75 50
63
Figure 21. SAVI6+1 reconstructed cavity and applicators
Figure 22. SAVI8+1 reconstructed cavity and applicators
64
Figure 23. SAVI10+1 (the largest SAVI device) reconstructed cavity and
applicators
Figure 24. SAVI 6-1mini (the smallest SAVI device) reconstructed cavity and
applicators
65
Once the cavity is reconstructed, we use a reconstruction wizard tool to reconstruct
the ipsilateral lung. The CHEST_WALL structure is then created by applying a 5.0 mm
margin exterior extension to the LUNG, and the SKIN structure is created by applying a
5.0 mm margin interior extension to the BODY contour. The Planning Target Volume
(PTV) is generated as a uniform layer of 1cm around the SAVI structure. Using Boolean
tools, we create d the PTV_EVAL structure by substracting the CHEST_WALL and SKIN
volumes from the PTV volume. The SKIN structure is then reset, and its final rendition
will be obtained by applying a 3.0mm exterior margin to the BODY contour. No
subqutaneous region is included in the SKIN, so all reported Max Skin values reflect the
dose values at the very surface of the skin edge. The process is shown in Figure 25 below:
Figure 25. Reconstruction of device and creation of PTV volume
5.6 Dose prescription and optimization
The standard fractional dose is of 340 cGy to the outer surface of PTV_EVAL. The
total dose of a full course of treatment is of 3400 cGy, delivered in 10 fractions, twice
daily, with daily pair -fractions at least six hours apart. The planning criteria used for
planning are matching the once recommended by the NSABP B -39/ RTOG 0413 protocol
guidelines for APBI irradiation with respect to D90, V100, V150, V200 and conformity
indexes, as well as to Maximum Skin Dose (MSD). V100, V150 and V200 represent the
volumes (in cm3) covered by the respective (IL) isodose line (in %).
D90 represents the percentage of the prescribed dose delivered to 90% of the PTV.
MSD is the Maximum Skin Dose, MCWD is the Maximum Chest Wall Dose, V150 and
66
V200 represent the volumes (cc) covered by the percentage (%) of the dose. I n preliminary
papers on this topic (Wazer et al., 2006), it is indicated that escalated values for these
entities are linked to the increasing risk of developing fat tissue necrosis. Lately, different
authors or groups (Cuttino et al., 2009) published early tolerances and late toxicities
related to th e dosimetric parameters indicated above .
5.7 Evaluation of dosimetric advantages of the SAVI device compared to
balloon-type APBI devices
5.7.1. Study Motivation
As stated before, the main precursors of the SAVI device were two balloon -type
devices, the later being the ConturaTM MLB applicato r, a five (5) lumen balloon -type
device , clinically launched in January 2008 by SenoRX, Inc, believed to be a better
solution for certain cases unfit for PBI (partial breast irradiation) using a similar device,
the MammoSite (Hologic) single -lumen balloon, due to either minimal balloon -to-skin
distance, balloon symmetry or tissue -balloon conformance. Multiple studies (Arthur et. al,
2003; Foo et al., 2008; Brown et al., 2009; Kim et al., 2011 ; Sherman et al., 2011 ) show
that balloon type devices with multiple lumen entries, like the Contura device, allow for
better PTV_EVAL coverage, reduction of maximum doses on skin and chest -wall
compared to single lumen devices, like MammoSite , the use of which can also lead to late
chest wall or skin toxicities (Brashears et al ., 2008) . Also, the SAVI applicator use is
proved to result in lower doses to the contraletral breast, compared to external beam
radiation (Robinson et al., 2015). In this section we present our dosimetri cal evaluation of
both balloon -type applicators against each other and relative to the SAVI device, in a
comparative study (Morcovescu et al., 2009) meant to highlight the relevant advantages of
one balloon type device over the other, and of the SAVI devic e over both balloon type
devices, in view of the RTOG 0413 protocol.
5.7.2. Materials and Methods
We considered ten (10) cases elected to be treated with the ConturaTM balloon. The
treatment plans were designed following the RTOG 0413 planning guidelines. All five (5)
67
available channels were used when creating plans with the ConturaTM balloon, as shown in
Figure 26.
Figure 26. ConturaTM balloon , all 5 channels fully loaded
Figure 2 7. ConturaTM balloon, central lumen, single position
Surrogate plans were then designed for each of the ten cases, using only the
centered lumen of the ConturaTM balloon (see Figure 27) , in order to mimic the use of a
68
MammoSite -contura (MM c) applicator. Single and multiple dwell position surrogate plans
were created. For each patient, the outcome of the three plans were then compared and
analyzed both from a dosimetrical perspective and from the treatment delivery hands -on
experience perspe ctive as well. 3D treatment planning and DVH analysis was employed in
order to evaluate geometric and dosimetric parameters.
Figure 28. Body contour optimization
The SAVI device was the applicator of choice in our first two cases presented in
the next study , because of the small lumpectomy cavity sizes, 9.6cc and 8.5cc
69
respectively. The SAVI mini catheters were differentially loaded following a PTV
optimization process, in order to allow V95>95%, V150<50cc, V200<20cc and acceptable
rib and skin maximum doses. These two cases were evaluated in a comparison with the
other two commonly used APBI treatment balloon -type devices, the MammoSite and the
Contura balloons . A single source dwell position was placed in the central lumen of the
SAVImini devi ce in order to create a virtual MammoSite plan.
The prescription dose of 34 Gy was delivered at 3 cm radial distance from this
central single dwell position. This is the equivalent radius of the PTV spherical volume
created for a regularly filled 35.0 cc MammoSite balloon.
A Contura plan template was also superimposed on the CT image set and fit into
the cavity volume. A plan was then created as to deliver the prescription dose to the same
PTV surface as the one considered for the MammoSite virtual plan. A Body -Deformed
Contour was created in order to mimic the shape of the skin (Figure 28) when a balloon –
type applicator was virtually placed inside the cavity. More realistic maximum skin doses
were than assessed and reported .
5.7.3. Results and Discussions
When comparing the balloon type applicators, t he intrinsic design of the ConturaTM
MLB applicator corroborated with an automated plan optimization process that considered
the use of all its five (5) channels, contributed to a generally better dosimetrical
characterization of the target volumes, when trying to limit the Maximum Skin Dose
(MSD) to less then 145% of the Prescription Dose (PD). Coverage Indexes for both PTV
and PTV_EVAL structures, as defined in the RTOG 0413 protocol, were evaluated. With
Contura, the PTV Coverage Index – CI1 is better by an additional 1.5 –3.5% compared to
the MammoSite ( MM) balloon. An increase of V150 and V200 values were record ed for
Contura compared with MM , in the range of 1.0 -2.0 cc (V150) and 0.1 -1.0 cc (V200),
while mantaining acceptable limits for both. The most common limiting factor for the use
of interstitial PBI is the MSD. While improving the coverage of both PTV and
PTV_EVAL with the Contura, the MSD was reduced by 8.0 -12.0% of PD compared to the
MM. Close and thorough monitoring of the balloon positioning during the treatment
course is also es sential, as rotation of the balloon may occur.
70
The symmetry of the balloon no longer plays a crucial role with the Contura
balloon, since adequate shaping of the dose cloud can be achieved by diferential loading of
off-centered cathethers. Better conformity can also be achieved with the Contura balloon
due to availability of a suction/vacuum chanel that allows immediate action at the time of
the initial CT simulation, when air or/and seroma trapped around the balloon can be
removed. The better dosi metrical outcome that the ConturaTM MLB applicator potentially
offers comes with a price though: it involves a more time -consuming planning process and
more extensive quality assurance program, that includes prior to each fraction balloon
position monitori ng and adjustments, incision site retaping and rebandaging.
Overall, compared to the simulated MammoSite -Contura balloon, better target
coverage was possible with the new ConturaTM MLB applicator while being able to reduce
the MSD values and to achieve be tter conformity.
When comparing the cases where the SAVI device was elected against the MM
and Contura balloon devices, the following results are reported. Lumpectomy cavity
volumes of less than 10 cm3 were easily accommodated by the S AVImini. Because of
possible differential loading of up to 7 catheters, even cases that would normally not meet
the criteria outlined in NSABP B -39 can were successfully treated without any clinical and
dosimetrical compromises. In general though, studies show that plans done with both
balloon -type and SAVI devices conform well with guidelines specified in national
protocols (Scanderbeg et al., 2009; Sehgal et al., 2011
The 95% isodose line coverage in all three situations is very similar, in the 96.5% –
99.8% range. Though, th e maximum skin dose and the maximum rib dose vary greatly,
especially for the case where the recorded SAVI -to-skin distance was 1.3 mm.
Balloon -type applicators can not accommodate volumes of less than 15cc without
causing extreme patient discomfort, skin overstretching and prohibitive skin doses, up to
almost 600% of the prescription dose, as the study shows. Body contour modeling or
editing should be employed in order to realistically account for skin stretching and shaping
caused by the use of a balloon applicator in a lumpectomy volume less than 15cc, when
assessing the maximum dose to skin for comparative studies.
Maximum skin doses of 244.3% (MM) and 249.4 % (Contura) were estimated
when the body surface was conformed to the shape of a balloon applic ator ( Figure 28 ).
71
These are still unacceptable values, but realistically lower values than the ones reported
for case #2 in Table 2.
Table 2. Comparison data for Contura, MM and SAVImini cases
Because of the size o f the cavity in the case of SAVI 6-1minis, V150/V100 volume
ratios tend to exceed the 0.5 value, therefore further investigation and reevaluation of the
DHI acceptance criteria , and relevance, for APBI for small cavity volumes is necessary .
APBI has become a very efficient and attractive method of treatment for women
diagnosed with early stage breast cancers. Balloon -type applicators like MammoSite or
Contura can accommodate a large range of clinical situations but fail to address the ones
where the lumpectomy cavity volumes are below 15cc. Prope rly inflated balloons can fill
up cavities of at least 30cc, normally larg er than 35cc. The SAVI 6-1mini applicator proves
to be the only implant solution for small lumpectomy volumes, below 15cc. Its design
allows a proper dose optimization, with excellent PTV coverage and acceptable skin
sparing.
New devices are developed for the treatment of breast cancer and better results are
observed. An intricate and exhaustive study of the advantages of one over the others needs
to be done, in order to improve the current treatment options and allow for further # Applicator
type V100
(cc) V95
(%) V90
(%) V150
(cc) V200
(cc) DHI Max
skin
dose
(%) Max
CW
dose
(%) PTV
volum
e
(cc) Min
Skin
Distance
(mm)
1 miniSAVI 47.2 98.6 99.6 27.4 17.1 0.419 72.5 170.3 48.92
13.9 MM1dw 73.3 96.5 98.3 25.1 5.1 0.658 147.2 355.7
82.0
Contura 77.0 99.1 99.8 28.7 7.8 0.627 143.0 325.3
2 miniSAVI 24.8 97.0 98.7 16.0 10.2 0.355 105.7 163.9 26.3
1.3 MM1dw 78.4 98.4 99.0 30.0 8.2 0.617 582.7 323.1
82.0
Contura 79.4 99.3 99.6 31.6 9.5 0.602 592.8 280.1
72
developments. Our studies indicate that the SAVI device proves to be a desirable option,
with excellent coverage results and minimal skin reactions and late secondary effects.
73
CHAPTER 6 COMPREHENSIVE DOSIMETRIC ANALYSIS OF THE
SAVI DEVICE
6.1. Study Motivation
Appropriateness of the accelerated partial breast irradiation option in the case of
breast cancer patients was and still is one of the most important questions to be answered
when determining whether this should be offered to a woman candidate, and is historically
understood in view of many aspects, including pathology, anatomic patterns, etc., which
are mainly seen as clinical patient selection criteria. With the evolvement and development
of different technologies designed specifically for the treatment of breast cancer alone,
another important additional aspect that is crucial to the implementation of a
comprehensive accelerated partial breast irradiation (APBI) program is a robust quality
assurance program that confirms that the treatment target is appropriately defined and
dosimetrically covered within the intended prescription dose ( Wazer et al., 2006), and that
the organs at risk are not irradiated above widely accepted and protocol impos ed tolerance
values (NSABP Protocol B-39/RTOG 0413 Protocol, 2007), when also confronted and
confirmed with values obtained by the application of novel computational algorithms such
as Acuros or Monte -Carlo (Graf et al., 2011).
Our preliminary studies (Mor covescu et al ., 2009; Morcovescu & Morton, 2009)
indicated that the SAVI device’s dosimetric performance is superior to that of balloon type
APBI devices, so we further attempted to comprehensively evaluate the dosimetric
performance of SAVI type devices.
6.2. Materials and Methods
The main body of our work consists on a comprehensive dosimetric analysis of
extensive clinical data, collected for all four different size SAVI devices (SAVI6 -1mini,
SAVI6 -1, SAVI8 -1 and SAVI10 -1). Our study is structured and focused on two subsets of
data: 1) a major pool of data collected at a multi -instututional level, that presents the
dosimetric analysis of the entire range of SAVI applicators, and a 2) minor pool, a subset
of the entire data, consideri ng patients implanted with the smallest of the SAVI devices,
the SAVI6 -1mini device , in our clinic only . The total number of patients included in our
multi -institutional pool study is 817 . There were 1 4 different participating institutions
74
involved in the multi -institutional study , each providing data for all four SAVI device
models. The subset study presented on the SAVImini device is a single -institution study of
plans created for 121 patients , treated over the span of 5 years, from 2009 to 2014.
The dos imetric parameters reported in this study include: cavity volume, volume of
the defined treatment region (PTV_EVAL), V90(%), V95(%), V100(%), V150(cc),
V200(cc), skin distance (minimum distance from the lumpectomy cavity wall to the skin),
chest wall and i psilateral lung distances (mm), and the maximum doses to critical
structures (skin and chest -wall). Conformity Indexes (CI), related to reported air/seroma
and invagination volumes, were also evaluated. Our dosimetric coverage criteria for this
study was V 90>90%, V150<50 cm3, V200<20 cm3. Additional constraints are placed to
try limiting the chest wall and skin doses to 100%. All dosimetric data , both the major
pool and minor subset, was stratified using 5 mm skin -distance (SD) intervals, therefore
differen tiating among cases with major or no PTV volume reduction.
We have also evaluated and reported the volumes of the cavity , of the PTV and of
the PTV_EVAL, and evaluated the volume reduction of the PTV. For this we proposed
and used this equation for calcula ting PTV -VR:
PTV -VR (%) = (PTV volume – PTV_EVAL volume) / PTV volume (6.1)
The Task Group TG43 formalism was employe d on all dosimetric evaluations, and
both TG43 and ACUROS formalisms used on the SAVI6 -1mini device study only.
75
TG43 V90 V95 V100 V150 V200 V cav V PTV V PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 762.0 637.0 768.0 665.0 758.0 514.0 532.0 693.0 687.0 470.0 419.0 317.0
MEAN 96.5 93.7 89.9 28.6 14.0 28.7 90.9 66.9 2884.3 12.1 2565.0 15.8
MEDIAN 97.5 94.8 90.9 26.7 13.8 24.6 78.8 62.0 2954.0 10.0 2620.0 13.0
MINIMUM 75.2 70.9 65.7 8.2 3.7 2.8 14.6 14.2 120.0 0.5 217.6 0.0
MAXIMUM 100.0 100.0 100.0 70.1 38.7 78.4 202.8 160.1 7854.0 76.4 6730.0 93.5
STANDARD
DEV 3.2 4.5 5.7 9.1 4.1 18.0 39.5 24.7 942.8 9.5 1152.5 13.8
Table 3. Full data, for all S AVI devices ( above) Table4. Sorted Data for the SAVI10 -1 device ( below)
TG43 V90 V95 V100 V150 V200 V cav V PTV V PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 115.0 80.0 123.0 82.0 109.0 96.0 89.0 101.0 103.0 84.0 81.0 68.0
MEAN 97.0 94.6 90.7 41.7 17.5 57.6 148.3 105.5 2786.3 12.9 2538.0 15.8
MEDIAN 97.8 95.2 91.1 42.4 17.6 58.6 155.0 107.6 2850.0 10.4 2689.4 13.2
MINIMUM 90.3 82.8 75.6 19.2 6.3 29.4 49.7 49.7 500.0 1.0 250.0 1.0
MAXIMUM 100.0 100.0 99.3 70.1 36.5 78.3 202.8 160.1 6543.0 50.0 4284.0 93.5
STANDARD
DEV 2.6 3.8 4.8 6.9 3.8 9.4 31.8 18.1 886.0 10.7 979.7 15.1
76
TG43 V90 V95 V100 V150 V200 V cav V PTV V PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 193.0 148.0 189.0 158.0 185.0 152.0 146.0 172.0 169.0 123.0 119.0 84.0
MEAN 96.6 93.7 89.5 33.4 15.2 34.5 107.6 79.9 2853.3 11.5 2720.4 14.9
MEDIAN 97.4 94.9 90.9 34.1 15.4 33.2 110.9 82.4 2850.0 9.2 2740.0 12.0
MINIMUM 79.3 79.2 68.2 19.0 5.2 7.0 30.2 30.2 315.3 0.5 390.0 0.0
MAXIMUM 100.0 99.8 99.3 49.9 26.8 78.4 160.8 121.5 5694.0 76.4 6400.0 66.2
STANDARD
DEV 3.1 4.7 6.0 6.6 3.4 7.7 25.3 15.8 819.6 9.7 1104.4 12.6
Table 5. Sorted Dat a for the SAVI8 -1 device ( above) Table 6. Sorted Dat a for the SAVI6 -1 device ( below)
TG43 V90 V95 V100 V150 V200 V cav V PTV V PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 321.0 279.0 322.0 297.0 324.0 181.0 187.0 290.0 291.0 159.0 151.0 111.0
MEAN 95.9 92.9 89.1 24.6 12.8 16.7 71.2 55.0 2951.8 12.3 2551.4 15.0
MEDIAN 97.1 94.0 90.0 23.9 12.3 15.9 70.8 55.5 2992.0 10.0 2550.0 11.0
MINIMUM 75.2 70.9 65.7 8.2 3.7 2.8 14.6 22.5 450.0 2.0 400.0 0.0
MAXIMUM 100.0 100.0 100.0 58.1 38.7 50.4 167.0 113.7 7854.0 56.3 6730.0 79.6
STANDARD
DEV 3.5 4.7 5.7 6.8 4.0 6.9 19.7 13.5 992.8 8.9 1253.0 13.0
77
TG43 V90 V95 V100 V150 V200 V cav V PTV V PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 128.0 126.0 129.0 123.0 135.0 81.0 107.0 126.0 121.0 104.0 68.0 54.0
MEAN 97.1 94.9 91.6 23.2 12.5 10.5 53.9 45.1 2860.4 11.7 2355.6 18.9
MEDIAN 98.2 95.7 92.7 22.7 12.2 10.0 52.9 46.3 3033.0 10.0 2459.0 14.0
MINIMUM 85.3 79.6 74.0 10.5 4.5 5.0 25.4 14.2 329.8 1.0 217.6 0.0
MAXIMUM 100.0 100.0 99.8 52.6 23.0 24.1 99.1 75.1 5359.0 53.4 5650.0 71.2
STANDARD
DEV 2.7 3.9 5.2 5.7 3.0 3.3 10.1 9.3 960.1 9.3 1178.8 15.3
Table 7. Sorted Data for the SAVI6 -1mini device (above) Table 8. Sorted Dat a for all devices, SD < 3mm ( below)
TG43 V90 V95 V100 V150 V200 V cav V PTV V PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 38.0 33.0 40.0 26.0 37.0 28.0 40.0 36.0 43.0 44.0 31.0 32.0
MEAN 94.6 92.7 88.3 25.3 12.7 28.7 80.5 55.8 3221.4 2.3 2916.0 8.2
MEDIAN 94.2 91.5 88.1 23.9 12.0 29.3 64.7 47.4 3200.0 2.6 2856.0 7.1
MINIMUM 77.0 79.2 65.7 14.4 5.2 5.3 15.9 20.8 2274.6 0.5 1170.0 0.0
MAXIMUM 100.0 98.7 97.9 42.5 19.9 68.5 160.1 105.1 4617.2 3.0 6730.0 18.0
STANDARD
DEV 4.3 4.3 6.6 7.8 3.9 18.9 38.4 21.9 517.5 0.7 1253.6 5.5
78
TG43 V90 V95 V100 V150 V200 V cav V PTV V PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 56.0 51.0 61.0 45.0 55.0 39.0 51.0 48.0 58.0 62.0 38.0 35.0
MEAN 96.6 93.8 90.7 28.4 14.3 28.5 86.3 65.8 3083.7 4.4 2758.9 13.5
MEDIAN 97.0 94.1 90.9 26.5 13.6 24.1 76.2 58.9 3128.0 4.1 3018.0 11.0
MINIMUM 89.8 85.2 79.0 12.4 4.5 5.8 25.4 25.4 329.8 3.3 873.0 0.0
MAXIMUM 99.9 99.3 99.3 48.5 23.0 68.9 177.2 119.5 5359.0 5.0 4570.0 58.0
STANDARD
DEV 2.2 3.1 4.6 8.5 3.8 17.0 36.8 24.9 651.5 0.5 933.6 12.1
Table 9. Sorted data, all devices, 3mm < SD < 5mm ( above) Table 10. Sorted data, all devices, 5mm< SD < 7mm (below)
TG43 V90 V95 V100 V150 V200 V cav V PTV V PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 50.0 44.0 47.0 36.0 49.0 33.0 40.0 47.0 48.0 51.0 36.0 35.0
MEAN 96.4 94.4 90.0 28.0 13.8 28.5 83.2 64.0 3058.1 6.4 2456.8 16.3
MEDIAN 97.5 95.9 90.7 26.0 13.7 18.7 68.8 57.2 3055.0 6.4 2285.0 14.5
MINIMUM 90.0 84.4 78.8 8.2 3.7 8.1 31.8 23.2 1990.0 5.2 445.3 1.9
MAXIMUM 99.9 99.4 98.2 42.8 19.9 77.9 182.9 123.7 4134.0 7.0 5000.0 53.2
STANDARD
DEV 2.9 3.8 5.4 8.7 3.4 19.6 37.4 24.4 432.9 0.5 1079.7 12.4
79
TG43 V90 V95 V100 V150 V200 V cav V PTV V PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 295.0 245.0 303.0 237.0 293.0 216.0 218.0 265.0 260.0 313.0 200.0 212.0
MEAN 96.9 94.9 91.1 30.1 14.9 29.5 86.5 70.2 2465.2 15.9 2569.4 17.4
MEDIAN 97.9 95.8 91.6 29.2 14.9 24.3 78.6 64.7 2600.3 13.0 2616.1 14.0
MINIMUM 79.3 73.0 68.2 11.6 4.6 6.6 14.6 14.2 340.0 7.1 217.6 0.0
MAXIMUM 100.0 100.0 99.8 52.6 35.7 78.4 190.3 160.1 4180.0 76.4 6400.0 93.5
STANDARD
DEV 3.1 4.3 5.5 8.0 3.6 18.6 35.9 24.9 743.3 9.5 1199.7 14.8
Table 11. Sorted Data, all devices, SD > 7mm ( above ) Table12. SAVImini full data TG 43 (below)
TG43 V90 V95 V100 V150 V200 V cav V PTV V
PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 72 72 72 72 72 72 72 72 72 72 72 71
MEAN 98.53 97.38 95.34 24.44 14.1 8.3 49.5 43.2 2595.4 11.9 3220.1 15.7
MEDIAN 99.45 98.30 96.00 25.85 14.7 8.3 49.6 47.1 2893.5 10.4 3187.0 12.0
MINIMUM 91.60 88.10 84.80 10.90 6.5 6.2 25.7 14.2 329.8 1.0 438.0 0.3
MAXIMUM 100.00 100.00 99.80 31.30 20.0 10.4 57.9 53.9 3594.0 53.4 9064.0 61.0
STANDARD
DEV 1.92 2.76 3.42 4.72 2.91 0.90 3.82 8.96 714.93 8.54 1684.5 12.04
80
ACUROS V90 V95 V100 V150 V200 V
air/seroma V
invag VR Ipsi
Lung d Max
Skin D Max
CW D
% % % cc cc cc cc cc mm cGy cGy
n = 71 71 71 71 71 71 71 72 71 71 71
MEAN 98.03 96.53 94.11 23.88 13.80 2.02 0.40 0.13 20.73 2489.2 3129.4
MEDIAN 98.90 97.10 94.60 24.80 14.20 1.10 0.00 0.07 17.00 2730.0 3080.0
MINIMUM 89.90 86.30 82.20 10.40 6.20 0.00 0.00 0.00 5.30 905.0 403.0
MAXIMUM 100.00 100.00 99.70 49.50 19.00 19.10 4.00 0.59 66.00 3371.0 9697.0
STANDARD
DEV 2.28 3.10 3.86 5.43 2.78 3.26 0.95 0.16 12.04 643.2 1745.1
Table13. SAVImini full data Acuros (above) Table 14. SAVImini data , SD < 5mm TG43 (below)
TG43 V90 V95 V100 V150 V200 V cav V PTV V
PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 13 13 13 13 13 13 13 13 13 13 13 12
MEAN 96.12 93.97 91.48 17.45 10.0 8.1 47.7 29.7 3262.3 2.9 3290.0 18.3
MEDIAN 97.10 95.60 93.60 18.10 10.2 8.1 48.6 28.6 3298.0 3.0 3958.0 10.1
MINIMUM 91.60 88.10 84.80 10.90 6.5 7.1 41.5 20.8 2910.0 1.0 438.0 2.0
MAXIMUM 98.70 97.20 95.70 24.80 12.3 9.6 51.0 38.5 3594.0 4.3 6200.0 61.0
STANDARD
DEV 2.39 3.05 3.63 3.82 1.81 0.72 2.66 5.67 167.92 1.13 1953.8 17.58
81
ACUROS V90 V95 V100 V150 V200 V
air/seroma V
invag VR Ipsi
Lung d Max
Skin D Max
CW D
% % % cc cc cc cc cc mm cGy cGy
n = 12 12 12 12 12 12 12 13 12 12 12
MEAN 95.27 92.86 89.96 16.60 9.90 1.46 0.08 0.37 23.31 3119.3 3216.5
MEDIAN 96.60 94.30 91.40 16.85 10.40 0.70 0.00 0.37 15.05 3173.0 3656.5
MINIMUM 89.90 86.30 82.20 10.40 6.20 0.00 0.00 0.21 7.00 2754.0 403.0
MAXIMUM 98.20 96.60 94.90 21.30 12.00 4.60 1.00 0.59 66.00 3371.0 6200.0
STANDARD
DEV 2.84 3.48 4.07 3.50 1.95 1.71 0.29 0.12 17.58 177.4 2059.2
Table15. SAVImini data , SD < 5mm Acuros (above) Table16. SAVImini data , 5mm<SD < 10mm TG43 (below)
TG43 V90 V95 V100 V150 V200 V cav V PTV V
PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 20 20 20 20 20 20 20 20 20 20 20 20
MEAN 98.28 96.98 95.03 23.65 13.7 8.4 50.5 42.2 3052.0 7.0 3523.4 12.6
MEDIAN 98.40 96.90 95.10 23.40 13.4 8.2 49.7 42.8 3115.0 7.0 3723.0 11.2
MINIMUM 96.20 94.30 92.30 17.10 10.2 6.4 45.7 30.8 1816.0 5.0 1145.0 3.1
MAXIMUM 100.00 99.90 98.20 28.40 18.3 10.1 57.9 50.7 3325.0 9.5 7807.0 30.0
STANDARD
DEV 1.09 1.37 1.43 2.85 2.00 0.93 2.67 4.93 328.27 1.27 1531.7 8.22
82
ACUROS V90 V95 V100 V150 V200 V
air/seroma V
invag VR Ipsi
Lung d Max
Skin D Max
CW D
% % % cc cc cc cc cc mm cGy cGy
n = 20 20 20 20 20 20 20 20 20 20 20
MEAN 97.59 95.89 93.29 22.60 13.18 1.72 0.50 0.17 17.64 2867.7 3402.0
MEDIAN 97.40 95.70 93.45 22.15 12.90 0.80 0.00 0.16 16.20 2949.5 3492.0
MINIMUM 95.60 93.80 90.10 16.60 9.80 0.00 0.00 0.00 8.10 1726.0 1083.0
MAXIMUM 100.00 99.70 98.40 27.60 17.80 6.90 3.60 0.33 35.00 3177.0 7689.0
STANDARD
DEV 1.18 1.43 1.93 2.71 2.01 2.07 1.13 0.08 8.22 312.5 1609.4
Table 17. SAVImini data for 5mm<SD<10mm ACUROS (above) Table 18. SAVImini data for 10mm<SD<15mm TG43 (below)
TG43 V90 V95 V100 V150 V200 V cav V PTV V
PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 20 20 20 20 20 20 20 20 20 20 20 20
MEAN 99.28 98.47 96.70 26.98 15.9 8.1 50.1 48.0 2655.4 12.5 3161.2 16.5
MEDIAN 99.80 99.10 97.45 26.95 15.6 8.3 50.0 48.0 2694.5 12.8 3080.0 12.3
MINIMUM 94.00 91.00 87.80 16.90 9.6 6.2 45.5 35.1 1639.0 10.0 634.0 0.3
MAXIMUM 100.00 100.00 99.00 31.30 20.0 10.4 57.6 53.9 3108.0 14.3 9064.0 40.0
STANDARD
DEV 1.37 2.00 2.52 2.95 2.18 1.07 2.51 3.59 357.39 1.35 1965.3 11.27
83
ACUROS V90 V95 V100 V150 V200 V
air/seroma V
invag VR Ipsi
Lung d Max
Skin D Max
CW D
% % % cc cc cc cc cc mm cGy cGy
n = 20 20 20 20 20 20 20 20 20 20 20
MEAN 98.80 97.55 95.27 25.88 15.41 2.05 0.79 0.04 21.49 2530.7 3102.4
MEDIAN 99.40 98.10 95.85 26.10 15.10 0.45 0.00 0.02 17.25 2548.5 2947.0
MINIMUM 91.70 88.40 84.80 16.10 9.80 0.00 0.00 0.00 5.30 1546.0 591.0
MAXIMUM 100.00 100.00 99.70 29.50 19.00 19.10 4.00 0.29 45.00 3058.0 9697.0
STANDARD
DEV 1.83 2.52 3.21 2.83 2.01 4.48 1.28 0.07 11.27 372.7 2070.2
Table 19. SAVImini data , 10mm<SD<15mm ACUROS ( above) Table20. SAVImini data , 15mm<SD<20mm TG43 (below)
TG43 V90 V95 V100 V150 V200 V cav V PTV V
PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 8 8 8 8 8 8 8 8 8 8 8 8
MEAN 99.95 99.51 97.96 27.79 15.9 8.7 47.9 46.1 2085.4 17.3 2802.4 19.5
MEDIAN 100.00 99.60 97.55 28.10 15.9 8.7 50.3 49.8 2024.5 17.6 3176.0 12.2
MINIMUM 99.80 99.00 97.30 25.70 14.2 7.1 25.7 14.2 1724.0 15.2 499.0 4.5
MAXIMUM 100.00 99.90 99.00 29.50 18.0 10.1 53.1 52.9 2666.0 19.7 4740.0 53.3
STANDARD
DEV 0.08 0.31 0.70 1.52 1.17 0.96 9.07 12.96 277.54 1.69 1601.0 17.91
84
ACUROS V90 V95 V100 V150 V200 V
air/seroma V
invag VR Ipsi
Lung d Max
Skin D Max
CW D
% % % cc cc cc cc cc mm cGy cGy
n = 8 8 8 8 8 8 8 8 8 8 8
MEAN 99.78 98.84 96.56 26.88 15.45 1.44 0.18 0.06 24.46 1965.1 2686.6
MEDIAN 99.80 98.70 96.10 27.15 15.50 1.30 0.00 0.00 17.15 1884.0 3084.5
MINIMUM 99.50 98.00 94.70 24.60 13.50 0.00 0.00 0.00 9.50 1634.0 469.0
MAXIMUM 100.00 100.00 99.20 28.50 17.60 3.70 1.00 0.45 58.30 2543.0 4439.0
STANDARD
DEV 0.23 0.76 1.80 1.53 1.27 1.16 0.36 0.16 17.91 274.8 1536.0
Table21. SAVImini data , 15mm<SD<20mm ACUROS (above) Table22. SAVImini data , 20mm<SD<25mm TG43 (below)
TG43 V90 V95 V100 V150 V200 V cav V PTV V
PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 7 7 7 7 7 7 7 7 7 7 7 7
MEAN 99.20 98.27 95.34 27.00 15.2 8.6 50.6 50.4 1629.3 21.8 3234.3 13.8
MEDIAN 100.00 99.90 97.30 28.90 16.8 8.3 50.9 50.9 1703.0 21.5 3980.0 10.5
MINIMUM 94.90 90.00 84.90 19.00 9.5 7.8 48.6 47.8 972.0 20.0 1408.0 7.5
MAXIMUM 100.00 100.00 99.80 30.90 17.7 9.8 52.5 52.5 1887.0 24.4 4745.0 23.1
STANDARD
DEV 1.90 3.68 5.69 4.26 2.94 0.77 1.53 1.73 298.49 1.89 1348.7 6.47
85
ACUROS V90 V95 V100 V150 V200 V
air/seroma V
invag VR Ipsi
Lung d Max
Skin D Max
CW D
% % % cc cc cc cc cc mm cGy cGy
n = 7 7 7 7 7 7 7 7 7 7 7
MEAN 98.93 97.79 95.73 29.23 14.86 1.91 0.00 0.00 18.80 1534.4 3139.6
MEDIAN 100.00 99.70 98.20 27.90 16.40 1.90 0.00 0.00 15.50 1588.0 3885.0
MINIMUM 93.30 88.40 82.70 19.30 9.50 0.00 0.00 0.00 12.50 905.0 1355.0
MAXIMUM 100.00 100.00 99.50 49.50 17.00 3.90 0.00 0.02 28.10 1800.0 4619.0
STANDARD
DEV 2.49 4.21 6.03 9.64 2.74 1.55 0.00 0.01 6.47 288.9 1337.2
Table 23. SAVImini data , 20mm<SD<25mm ACUROS (above) Table 24. SAVImini data , SD>25mm TG43 (below)
TG43 V90 V95 V100 V150 V200 V cav V PTV V
PTVe Max
Skin D Min
Skin Max
CW D Min
CW
% % % cc cc cc cc cc cGy mm cGy mm
n = 4 4 4 4 4 4 4 4 4 4 4 4
MEAN 99.83 99.30 97.35 27.33 15.6 8.4 49.5 49.3 1299.0 35.0 2582.1 15.5
MEDIAN 99.80 99.25 98.20 27.90 16.1 8.4 48.9 48.7 1195.5 30.1 2345.5 16.3
MINIMUM 99.70 99.00 94.50 24.20 13.4 7.8 48.0 47.7 989.0 26.2 1472.2 6.4
MAXIMUM 100.00 99.70 98.50 29.30 16.8 8.8 52.1 52.1 1816.0 53.4 4165.0 23.2
STANDARD
DEV 0.15 0.36 1.91 2.43 1.58 0.42 1.83 2.01 367.25 12.47 1133.3 6.92
86
ACUROS V90 V95 V100 V150 V200 V
air/seroma V
invag VR Ipsi
Lung d Max
Skin D Max
CW D
% % % cc cc cc cc cc mm cGy cGy
n = 4 4 4 4 4 4 4 4 4 4 4
MEAN 99.68 98.93 97.18 26.70 15.33 6.40 0.00 0.00 20.53 1218.0 2508.5
MEDIAN 99.65 98.90 97.30 27.20 15.75 4.15 0.00 0.00 21.25 1117.0 2251.5
MINIMUM 99.50 98.50 96.00 23.80 13.30 1.30 0.00 0.00 11.40 918.0 1409.0
MAXIMUM 99.90 99.40 98.10 28.60 16.50 16.00 0.00 0.02 28.20 1720.0 4122.0
STANDARD
DEV 0.17 0.40 0.87 2.23 1.47 6.89 0.00 0.01 6.92 357.8 1146.8
Table 25. SAVImini data , SD>25mm ACUROS (above) Table 26. Stratified dosimetry for 5mm SD grouping interval (below)
Skin
Distance #
patients Max Skin
Dose (Gy) PTV
reduction
(%) CW Dose (Gy)
mm TG43 Acuros TG43 Acuros
<< 55 13 35.94 33.71 37.0±12.0 32.9±19.5 32.2±20.6
55 << ss..dd..
<<1100 20 33.25 31.77 17.0±8.0 35.2±15.3 34.0±16.1
1100<<ss..dd..<<1155 20 31.1 30.6 4.0±7.0 31.6±19.7 31.0±20.7
1155<<ss..dd..<<2200 8 26.66 25.43 6.0±16.0 28.0±16.0 26.9±15.4
2200<<ss..dd<<2255 7 18.87 18 0.0±1.0 32.3±13.5 31.4±13.4
ss..dd >> 2255 4 18.16 17.2 0.0±1.0 25.1±11.5 24.4±8.8
87
Table 27. SAVImini centralized data (TG43 and ACUROS)
Skin
Distance Max Skin Dose
(Gy) PTV_EVAL Normal Tissue PTV
reduction†
(%) CW Dose (Gy)
mm #
patients TG43 Acuros V90(%) V95(%) V100(%) V150(cc) V200(cc) TG43 Acuros
<< 55 13 35.94 33.71 95.3±2.8 92.9±3.5 90.0±4.1 17.5±3.8 10.0±1.8 37.0±12.0 32.9±19.5 32.2±20.6
55 <<SSBB<<1100 20 33.25 31.77 97.6±1.2 95.9±1.4 93.3±1.9 23.6±2.9 13.7±2.0 17.0±8.0 35.2±15.3 34.0±16.1
1100<<SSBB<<1155 20 31.1 30.6 98.8±1.8 97.6±2.5 95.3±3.2 27.0±3.0 15.9±2.2 4.0±7.0 31.6±19.7 31.0±20.7
1155<<SSBB<<2200 8 26.66 25.43 99.8±0.2 98.8±0.8 96.6±1.8 27.8±1.5 15.5±1.3 6.0±16.0 28.0±16.0 26.9±15.4
2200<<SSBB<<2255 7 18.87 18 98.9±1.9 97.8±3.7 95.7±6.0 27.0±4.3 15.2±2.9 0.0±1.0 32.3±13.5 31.4±13.4
SSBB>>2255 4 18.16 17.2 99.7±0.2 98.9±0.4 97.2±0.9 27.3±2.4 15.6±1.6 0.0±1.0 25.1±11.5 24.4±8.8
88
6.3. Results and Discussions . Original contributions.
Tables such a nd such present the data for both the multi -institutional and single –
insitution studies. The bulk data was stratified based on the distance from the device to the
skin, in increments of 5mm. Our dosimetric coverage criteria for this study was V90>90%,
V150<50 cm3, V200<20 cm3. Additional constraints are placed to try limiting the chest
wall and skin doses to 100%. All dosimetric data was stratified using 5 mm skin -bridge
intervals, therefore differentiating among cases with major or no PTV volume reduction.
6.3.1 Multi -institutional study on all SAVI type devices
The lumpectomy cavity volumes averaged 10.5 ±3.3 cm3 for the 6-1SAVImini
device up to 57.6.5 ±9.4 cm3 for the largest 10 -1 SAVI applicator.
PTV_EVAL volumes averaged 45.1±9.3 cm3 for the smallest applicator to
105.5 ±18.1 cm3 for the largest . V90 values averaged 96.5±13.2% of the PTV_EVAL
volume across all applicator sizes , which is well within the criteria imposed by the
NSABP B -39/RTOG 0413 protocol. V95 averaged 93.7±4.5% and V100 averaged
89.9±5.7%. V150 averaged 23.2±5.7 cm3 for the smallest device, and 41.7±6.9 cm3 for the
largest. V200 average s ranged from 12.5±3.0 cm3 for the smallest device, to 17.5±3.8 cm3
for the largest.
The Mean, Median, Maximum, Minimum and SD values for V90, V95, V100,
V150 and V200 of PTV_EVAL are reported in Table 28. Similar data for Cavity, PTV,
and PTV_Eval Volumes is reported in Table 29.
Skin spacing varied widely with 460 reported values ranging from 0.5 mm to 76.4
mm (12.1 ± 9.5mm). Skin spacing in these patients were: ≤3 mm (9.4%), 3 -5 mm
(13.2%), 5 -7 mm (10.9%), >7 mm (66.6%). The minimum skin bridge for these patients
averaged 12.1 mm, although each model was used in patients with ≤3.0 mm (44 total).
The maximum skin dose (n=687) was 85%±28% of prescription (mean ± SD).
From smallest to largest models, the values were: 82%±26%, 84%±24%, 87%±29%,
84%±28%.
Chestwall bridge varied widely (n=317), ranging from 0.0 mm to 93.5 mm (15.8 ±
13.8mm), and was similar regardless of model. For each applicator size, the maximum
dose to the chest wall was 75% ± 34% of the prescription dose.
89
Table 28. Multi -institutional, all SAVI type devices – values for V90, V95,
V100, V150 and V200 of PTV_EVAL
Parameter V90 V95 V100 V150 V200
(units) % % % cc cc
MEAN 96.5 93.7 89.9 28.6 14.0
MEDIAN 97.5 94.8 90.9 26.7 13.8
MINIMUM 75.2 70.9 65.7 8.2 3.7
MAXIMUM 100.0 100.0 100.0 70.1 38.7
SD 3.2 4.5 5.7 9.1 4.1
Table 2 9. Multi -institutional, all SAVI type devices – values for Cavity, PTV,
and PTV_Eval Volumes
Parameter V(cavity) V(PTV) V(PTV_Eval)
(units) cc cc cc
MEAN 28.7 90.9 66.9
MEDIAN 24.6 78.8 62.0
MINIMUM 2.8 14.6 14.2
MAXIMUM 78.4 202.8 160.1
6.3.2 Single institution study results, on SAVI6 -1mini device – TG43
The lumpectomy cavity volumes for the SAVI 6-1mini device, the device of
interest for our initial study on this SAVI device subtype (Morcovescu et. al , 2009) ,
averaged 8. 3±0.9 cm3.
PTV_EVAL and PTV volumes averaged 43.2±9.0 cm3 and 49.5±3.8 cm3,
respectively. V90 values averaged 98.5±1.9% of the PTV_EVAL volume, which is again
well within the criteria imposed by the NSABP B -39/RTOG 0413 protocol. V95 averaged
97.4±2.8% and V100 averaged 95.3±3.4%. V150 averaged 24.4±4.7 cm3 while V200
averaged 14.1±2.9 cm3.
PTV reduction m ounted up to 37.0±12.0% for the cases where the skin distance
(SD) was < 5 mm, especially where combined with reduced Chest Wall bridges (CWB).
This can result in dramatic drops of the CI (conformity index) values for PTV_EVAL,
90
where air/seroma is present, down to 61.1%. Though, across the entire cohort, CI values
averaged 96.6±5.7 %.
Skin and CW sculpting of PTV is always employed when creating PTV_EVAL
structures. The PTV volume reduction PTV VR averaged 13.0±16.0%, with min and max
values of 0.0% (no reduction) and 59.0% (when both SB and CWB were < 5mm)
respectively.
Thirteen (1 3) patients had a skin bridge (S D) of less than 5 mm. For these patients,
the V90 (n=1 3) was 96.1±2.8% (mean±standard deviation). Chest wall bridge (CWB)
varied widely, ranging from 0.3 mm to 61.0 mm (15.2±11.7 mm).The maximum dose to
the chest wall, over the entire cohort of patients, was 94.5±49.5% of the prescription dose.
Dosimetric data for a later study of ours (Morcovescu et al., 2014) , which i ncluded 121
patients, is shown in Table 30. No major variation from the numbers pr eviously published
is observed.
The maximum skin dose for the patients where the skin distance SD was less than
5 mm was 96.0±4.9%. Chest Wall bridge varied widely, ranging from 0.3 mm to 61.0 mm
(15.7 ± 12.0mm). The maximum dose to the chest wall, over the entire cohort of patients,
was 94.5% ± 49.5% .
The V150 “hotspots” averaged 24.4±4.7 cm3, while V200 averaged 14.1±2.9 cm3.
The average minimum ski n distance was 13.5 mm, but the applicator was used in patients
where the skin bridge was as lo w as 1mm. The average maximum skin dose (MSD) was
72.8% of the prescription dose. The average minimum CWB was 15.2 mm, with the
shortest of less than 0.3mm, and the average maximum dose to the chest wall of 90.5% of
the prescription dose.
Table 30. Stratified dosimetry for 5mm skin -distance grouping interval
Skin Bridge
(mm) # Patients Max Skin
Dose (Gy) V90 (%) V200 (cm3) PTV VR (%)
< 5 16 35.94 96.3±2.3 10.1±1.7 36.0±12.0
5 <SB <10 34 33.25 98.4±1.2 13.4±1.7 16.0±7.0
10 <SB <15 31 31.78 99.3±1.2 16.0±1.9 3.0±6.0
15 <SB <20 14 26.66 99.9±0.2 15.7±2.2 4.0±12.0
20 <SB <25 16 18.87 99.6±2.5 15.6±2.1 1.0±3.0
SB > 25 10 18.16 99.6±0.4 16.4±2.3 0.0±1.0
91
The PTV -VR values are largest for the cases with reduced skin bridges, where
there is a significant reduction of the volume of PTV_EVAL compared to PTV. Dose
coverage in those marginal cases was still excellent, with average V90 of 96.3 %.
The Mean, Med ian, Maximum, Minimum and SD values for V90, V95, V100,
V150 and V200 of PTV_EVAL are reported in Table 31. Similar data for Cavity, PTV,
and PTV_Eval Volumes is reported in Table 32.
Table 31. Single -institutio n, SAVImini type device – values for V90, V95, V100, V150
and V200 of PTV_EVAL
Parameter V90 V95 V100 V150 V200
(units) % % % cc cc
MEAN 98.0 96.5 94.1 24.4 14.1
MEDIAN 98.9 97.1 94.6 25.9 14.7
MINIMUM 89.9 86.3 82.2 10.9 6.5
MAXIMUM 100.0 100.0 99.7 31.3 20.0
SD 2.3 3.1 3.9 4.7 2.9
Table 32. Single -institution, SAVImini type device – values for Cavity, PTV,
and PTV_Eval Volumes
Parameter V(cavity) V(PTV) V(PTV_Eval)
(units) cc cc cc
MEAN 8.3 49.5 43.2
MEDIAN 8.3 49.6 47.1
MINIMUM 6.2 25.7 14.2
MAXIMUM 10.4 57.9 53.9
SD 0.9 3.8 9.0
The PTV to PTV_EVAL volume reduction may be large for SB < 5mm
(37.0±12.0), but plans still met all other dosimetric criteria. Less PTV reduction, 17.0±8.0
(5mm<s.d.<10mm), and 4.0±7.0 (10mm<s.d<15mm) results in better dose
conformity/critical structure sp aring.
92
6.3.3 Single insti tution study results, on SAVI mini device – ACUROS
When using the ACUROS formalism, our comparison study (Morcovescu et. al .,
2012) revealed the following results.
V90 values averaged 98.0±2.3% of the PTV_EVAL volume, V95 averaged 96.5
±3.1% and V100 averaged 94.1±3.9%. V150 averaged 23.9±5.4 cm3 while V200 averaged
13.8±2.8 cm3. The values indicat e a small but quantifiable degradation of the PTV_EVAL
coverage compared with when employing the TG43 protocol for calculations .
Dosimetric data for all 72 patients is also shown in Table 13.
The maximum skin dose for the patients where the skin distance SD was less than
5 mm was 91.7±5.2%.
The maximum dose to the chest wall, over the entire cohort of patients, was 92.0%
±51.3% of the prescription dose.
Stratified dosimetry for all 72 patients is centralized in Table 33.
Table 3 3. Stratified dosimetry for 5mm skin -distance grouping interval
Skin Bridge
(mm) # Patients Max Skin
Dose (Gy) V90 (%) V200 (cm3) Max CW
Dose (Gy)
< 5 13 33.71 95.3±2.8 9.9±2.0 32.2±20.6
5 <SB <10 20 31.77 97.6±1.2 13.2±2.0 34.0±16.1
10 <SB <15 20 30.6 98.8±1.8 15.4±2.0 31.0±20.7
15 <SB <20 8 25.43 99.8±0.2 15.5±1.3 26.9±15.4
20 <SB <25 7 18 98.9±2.5 14.9±2.7 31.4±13.4
SB > 25 4 17.2 99.7±0.2 15.3±1.5 24.4±8.8
The SAVI 6-1mini strut -based device is an excellent APBI solution for patients
with reduced skin and/or rib bridges, and it accommodated volumes unfeasible for other
types of single -entry brachytherapy devices. The Acuros data indicates a minor
degradation of the RTOG 0413 reference V90 coverage of the PTV_EVAL (0.5% on
average) while revealing that TG43 slightly overestimates the average dose values for skin
(by 4.1%) and for chest -wall (by 2.6%).
93
6.3.4 Overall Results and Discussion.
All data reported and pertaining to both the multi and single institution studies
clearly show that all SAVI device s allow for excellent PTV_EVAL coverage, in all
encountered clinical situations. This is achieved while concomitantly being able to keep
the skin and chest wall maximum doses below protocol imposed values (NSABP Protocol
B-39/RTOG 0413 Protocol , 2007 ). Because of its versatility in dose shaping and adaptable
device design, the SAVI6 -1mini was successfully used in the treatment of a stenotic distal
vagin a as well (Morcovescu et al., 2016).
Our study on the smallest SAVI6 -1mini device indicate that it can efficiently be
used in clinical situations where stringent limitations are imposed by close proximity of
the lumpectomy cavity either to the skin or ches t wall. Even for situations where SB < 5
mm all PTV coverage criteria are met, while avoiding skin overexposure.
Another dosimetry study of our s on 108 all -type SAVI devices (Reiff et al., 2013)
used on patients where the distance from the applicator to both skin and ribs were less than
7 mm indicates that the doses on all critical structures were within tolerable limits, as
recommended by standard protocols or society accepted standards (Smith et al. 2009).
The use of the ACUROS formalism, when dose inhomogeneity corrections are
applied in such a complex anatomical environment, where the presence of air pockets,
seroma, breast tissue of different densities, proximity of the skin surface or of the rib cag e
can change the scatter and back scatter conditions dramatically, falls in line with
theoretical predictions, showing that overall coverages and reported maximum doses to
critical organs degrade slightly, more for skin average maximum reported doses ( ~ 4%)
than for c orrespondent chest wall doses ( ~ 2%). These results are confirmed by more
recent studies on the impact of heterogeneity correction implications on brachytherapy
planning, all indicating minimal effects on the average values of PTV coverage and
maximum dose usually in the range of 2% , and 4% resp ectively (Loupot et al ., 2014 ;
Slessinger et al., 2012 ), but sometimes as high as 5% , and 7% or 10% respectively ( Zhang
et al., 2012; Likhacheva et al ., 2016; Thrower et al., 2016), with variations due to
placement of the device or heterogeneity information.
As other biological model based studies indicate (Bovi et al ., 2007), the clinical
significance of these differences we also observed is unclear in terms of variations of
tumor control.
94
CHAPTER 7 COMPREHENSIVE EVALUATION OF A STRUT –
ADJUSTED -VOLUME -IMPLANT SAVI DEVICE QUALITY
ASSURANCE PROGRAM
7.1 Study motivation
Considering the complexity of the entire treatment process involving APBI cases and
other brachytherapy procedures in general , there is a need of designing a comprehensive
quality assurance program , as already done for other specific brachytherapy procedures
(Brown et al. 2016) but also specifically for APBI (Cui et al., 2011 ; Chilukuri, 2016) , that
deals with all stages of this process taking place in a radiotherapy department: 1) the QA
of the cavity and device localization and reconstruction, 2) the QA of the treatment plan,
3) the QA at the time of treatment and 4) the post treatment QA. We discuss all those in
sequence and we highlight the common practices and the extra measure s we included into
our customized QA program, in an attempt to incorporate our clinical experience with the
device into a comprehensive QA program capable of preventing and addressing even the
least frequent clinical situations. A number of recent retrospe ctive studies (Iftimia et al.,
2015; Shah et al., 2016; Pinn -Bingham et al., 2011) clearly outline a variety of clinical
factors and situations that can have a major impact on the outcome of high -dose-rate
brachytherapy treatments, encompassing all stages of a QA program, from pre -delivery to
post-delivery, and emphasizing its importance in the cases of any type of breast
brachytherapy type applicators (Kyriacou, 2016 ; Pinder et al., 2016 ).
The implementation of a comprehensive quality assurance of a brac hytherapy
program is mandatory by state and federal regulations (Kutcher et al .,1995) , and its
effectiveness and thoroughness vital for a successful delivery of brachytherapy treatments
(Wazer et al., 2006). A number of studies on partial breast irradiatio n applicators clearly
indicate the variety of clinical situations raising concerns and requiring special attention
that need to be accounted for in a n all -inclusive QA program. This is because of the
potential harmful effect of improperly evaluated and add ressed apparently infrequent
clinical situations that can properly be flagged and identified in various stages of the QA
program, before, during or after the time of treatment, such as: impact of breast
augmentation (Akhtari et al., 2015; Bloom et al., 201 1), inter -fractional displacement or
motion of SAVI devices ( Chandrasekara et al., 2015; Morcovescu et al., 2011; Park et al.,
2011 ; Hyvarinen et al., 2014) , presence of air/fluid single or cluster cavities (Harmon et
95
al., 2013; Richardson et al ., 2010 ; Huang et al ., 2010 ), irregularly and asymmetrically
shaped volumes of breast tissue (Lu et al., 2012; Lu et al., 2010) , geometric anatomical
uncertainties when using the SAVI device (Akino et al ., 2014) , proximity of rib cage
(Brown et al., 2012) or the presence of pacemakers (Jacob et al., 2011 ; Kalavagunta et al.,
2015 ). All these studies are clearly indicative of the impact all these possible factors can
have on the final outcome of these type of treatments, and emphasizes the importance of
having a carefully design ed QA program.
7.2 Pre-treatment Q uality Assurance
7.2.1. Imaging and documentation for treatment planning
The imaging sequence is an important part of the planning process. In our clinic, a
patient is usually scheduled for a CT planning simulation and a CT is acquired the day
following the implant surgery , in order to allow time for adequate tissue conformance
around the SAVI device (Liu et al., 2012) . The acquired images become the reference set
of images all subsequent pre -fractional images will be compared against. It is important to
have a thorough verification process of the position and placement of the SAVI device
because the patient should be treated in the same position as planned.
During the initial planning scan done us ing a GE Lightspeed, large -bore, 4 slice CT
scanner (GE Medical Systems, Waukesha, Wisconsin), two CT scouts, one anterior and
one lateral, are acquired. For a high quality image reconstruction, we use 2.5 mm slice
thickness for CT planning scan, extending at least 5 cm superiorly and inferiorly to the
SAVI defined cavity area, typically resulting in a number of transverse images of around
65 slices. We employ a breath -hold during the CT scan, or at least we instruct the patients
having breathing difficulti es to exert a shallow breath pattern during the actual scan, in
order to minimize respiratory motion artifacts.
M2, M4 and M6 markers are identified with the built -in markers of the SAVI
devices, pertaining to their corresponding catheter number, 2, 4 or 6. We evaluate
distances between the three pairs of strut markers in the AP or Lateral CT scouts, Figure
29, and compare, record and review these values prior to each fractional treatment.
96
Figure 29. Pre-fractional comparison and review of implant placement: daily CT scout
(right image) compared against original planning CT scout (left image)
An in -house form was developed in order to document and record relevant position
parameters for the initial planning scan and for all subsequent pre -treatment verification
scans. Several items need to be recorded on this form, which we generically refer to as the
“SAVI daily C T checks and measurements” sheet. The standardized form is printed front
and verso. The information on the front page is:
F1) Patient Name;
F2) Patient unique identifier or ID;
F3) Breast Side: LEFT or RIGHT; needs to be circled.
F4) SAVI applicator type;
F5) The direction of the catheter exit: MEDIAL or LATERAL.
F6) Number of CT (transverse or axial) slices above or below the SAVI device.
F7) the LONG axis length;
F8) The M2 -to-M4 marker distance (mm);
F9) The M2 -to-M6 marker distance (mm);
F10) Maximum Ex pansion (mm)
97
F11) Min. SAVI – chest wall distance (mm) – evaluated in an axial CT image;
F12) Min. SAVI – skin distance (mm) – evaluated in an axial CT image;
F13) CT Tech initials – the initials of the CT scanner operator;
F14) PHYSICIAN: the initials of the Radiation Oncologist care provider, supervising
the treatment ;
F15) the A distance – the reading, on the SAVI device shaft’s scale, from the hub to
the skin entrance;
F16) position of the #2 catheter in reference to the incision site on the skin, viewed
along the shaft of the device;
The information on the verso page of the “SAVI daily CT checks and measurements”
sheet is:
V1) the lengths of the individual catheters of the SAVI device, a minimum of 7 (for
the SAVI6 -1mini or SAVI6 -1) or maximum of 11 (for the SAVI8 -1 or SAVI10 -1)
catheters.
This information allows for the verification of the patient’s identity (F1 -F2),
implant and positioning setup (F3 -F5), and for monitoring the device’s inter -fractional
position variations or deviations from the standard position, assigned to the very first set of
images recorded at the time of the planning simulation scan. All F6 -to-F12 values will
help monitoring b oth rotational and translatio nal variations of the particular device, for a
particular patient.
98
Figure 30. Front page (above) and Verso page (below) of our QA form
99
There is need for a very detailed quality assurance program for SAVI treatments,
patient specific, that is performed in addition to the routine daily and quarterly HDR
machine QA.
One important test to be performed on the day of the planning CT simulation is the
verification of the SAVI device catheter lengths. This is usually performed while
connecting the re -usable treatment catheters to the device and measuring and confirming
the tre atment lengths, usually found to be approximately within ± 2mm of the
manufacturer’s expected values. In parallel with the device positioning reproducibility
test, evaluated using the anterior or lateral CT scouts, we employ a more visual
verification proc ess of the device position, by recording the distance from the hub to the
skin and marking the position of the 2nd strut on the skin.
7.2.2 Treatment Time Nomogram for Strut -Based Accelerated Partial Breast
Applicators. Original Contributions.
7.2.2.1 Stud y motivation
The purpose of this work was to generate a tabular nomogram of treatment times
for strut -based APBI applicators. The nomogram is intended to provide guidance and a
pre-treatment quality assurance check for clinics establishing new treatment techniques or
transitioning from balloon -based applicators.
7.2.2.2 Materials and Methods
A retrospective analysis was conducted of 486 patients receiving APBI using the
SAVI strut-based applicators at three separate institutions. Patient data was organized
based on applicator size (a surrogate of treatment volume) and number of organs at risk.
Three organs at risk categories were determined based on the proximity of the
device to the patient’s skin and/or chest wall (0, 1, or 2 OARs). Organs at ri sk were
defined as being < 5 mm from cavity wall/peripheral struts. A tabular nomogram of
treatment time (based on nominal 10 Ci source strength) was generated from descriptive
statistics of each combination of applicator size and organs at risk category.
100
Figure 3 1: Boxplot of treatment times of various combinations of applicator size and
Number of nearby organs at risk.
7.2.2.3 Results and Discussion
The treatment time was observed to be directly proportional to applicator size and,
to a lesser extent, inversely proportional to the number of nearby organs at risk.
Distributions of treatment times observed for each combination of applicator size and
organs at risk category are depicted in Figure 31. The tabular nomogram featuring average
treatment times with standard deviations is presented in Table 34.
Table 34: Tabular nomogram of average treatment times in seconds. Standard
deviations are presented in parentheticals.
Number of Nearby Organs at Risk
0 1 2
Applicator
Size 6-1 mini 148.0 (9.4) 137.0 (14.4) 126.6 (19.1)
6-1 175.7 (13.9) 165.7 (19.9) 152.1 (24.5)
8-1 226.4 (16.0) 229.0 (25.0) 189.5 (25.0)
10-1 287.2 (24.1) 269.4 (30.0) 252.6 (25.9)
101
Strut -based APBI treatment times were observed to depend in a consistent manner
on applicator size and number of nearby organs at risk. Information provided in the
nomogram presented here represents several hundred treatments performed at multiple
institutions. This data can serve as guidance or quality assurance for institutions with
limited experience using strut -based APBI applicators. This work distills considerable
multi -institutional experience with strut -based APBI applicators into an accurat e and
pragmatic clinical tool that promotes treatment consistency and quality.
7.3 During and post treatment Quality Assurance – Interfractional Variance .
Original Contributions.
7.3.1 Materials and Methods
In a study we presented in 201 1 at the ASTRO (Morcovescu et al ., 2011) , we
evaluate d the effect on target coverage and organ of risk sparing caused by inter fraction
variance of a SAVI6 -1mini APBI applicator in patients with small breast tissue volumes
and limited skin/chest wall spacing. Daily setup combined with tissue conformance
variations are confirmed to potentially have a big impact on dosimetric parameters in
SAVI -based breast brachytherapy, as a number of recent studies indicate (Liu et al., 2012;
Chandrasekara et al ., 2015) , therefore the outmost importance of a thorough QA of the
planning process (Pella, 2016). Mammographic evaluation of an breast implant is also
referred to as post treatment evaluation of an APBI treatment or plan (Ojeda -Fourier et al.,
2011) , but in this study we ref er to the actual on -going dosimetric evaluation of a plan on
subsequent CT data acquired during the treatment course itself.
The study considered three (3) patients treated with a SAVI6 -1mini device in our
clinic. Individual Initial treatment plans were ge nerated based on the original CT -SIM
scans and saved as template plans for fractional re -planning. Pertinent dosimetrical
parameters were evaluated and recorded (i.e., V200, V150, V100, V95 and V90,
maximum skin and chest wall dose). The template plan was then superimposed on the
subsequent pre -treatment fractional CT -Sim and tomography -based treatment plans were
generated for Fractions 1 -10. All pertinent structures were reconstructed as to closely fit
the volumetric profiles of the initial plan structures . Relevant dosimetric comparison was
102
then performed using the maximum doses to organs at risks, %PTV_EVAL and % PTV
coverage, and target dose homogeneity index.
A Reference Line was created in the Initial CT -sim& TemplatePLAN (Figure 32).
This structure was used for further rotational shifts evaluations in subsequent fractional
CT-sim&PLANS.
Figure 3 2. The Reference Line created in the Initial CT -sim&TemplatePLAN
A total of 11 plans were generated for each patient, a total of 33 plans for the entire
current study. Accurate reconstruction of the SAVI planning structure was an initial study
objective, thus only variations of +/ – 0.2 cc from the reference SAVI device volume in the
initial CT -Sim scan were accepted. Minor PTV_EVAL volume changes of +/ – 1.5 cc w ere
then estimated. The SAVI6 -1mini applicator average rotational axial shift, over the course
of the entire treatment, was 2.8o ± 2.1o, hardly assessable or detectable based on
prefractional CT -sim and external skin marker positional checks.
103
7.3.2 Results and Discussion
We found that d ue to interfraction variance in local anatomy (i.e., altered SAVI to
chest wall and skin distances, slightly changed skin surface contours, air/seroma profile
difference), skin and chest wall dose escalations of up to 30% and PTV_EVAL V90 and
V95 coverage degradation for and of up to 5% of the reported values on the initial plan
can be observed.
Figure 33. Absolute value inter -fractional variations of planning parameters
Variations on monitored parameters were recorded and trended, Figure 33. The
highly sensitive parameters to rotational variances are the Maximum Doses to Skin and
Chest Wall, which can mount to escalations of those values of up to 140% of the
prescription dose.
The treatment time reduction, due to imposed to necessary cooling off of the dose on
the skin or chest wall, and preservation of planning obje ctives, was up to 4. 8% of the
initial treatment time, when replanning was employed. A summary of the data analysis for
Patient 1 is shown in Table 35 below.
104
Table 3 5. Summary of the data analysis for Patient 1
Parameter Average SD MAX MIN
Max Skin DOSE (cGy) 100.2 8.34 115 89
Max CW DOSE (cGy) 123.2 8.19 144 115
PTV_EVAL Volume (cc) 31.3 0.77 32.7 30.2
PTV_EVAL V95(%) 89.8 1.35 92.3 88.3
PTV_EVAL V90(%) 92.3 1.31 94.9 91.0
PTV_EVAL V150cc) 15.7 0.24 16.2 15.2
PTV_EVAL V200(cc) 9.4 0.22 9.8 8.9
PTV V95(%) 71.5 3.32 79.4 68.6
PTV V90(%) 75.4 2.51 79.7 72.7
SAVI Volume (cc) 7.5 0.06 7.6 7.4
Individual differential source dwelling times adjustments, building up to as low as
3% of the total treatment time of the Initial plan, were sufficient for recovering the desired
standard APBI planning objectives.
Routine pre -treatment CT scan checks are unlikely to point out hard -to-quantify local
anatomy inter -fraction variances, which are identified as the main impact factor for APBI
treatments in patients with reduced chest wall and skin sparing.
Therefore, c arefully conducted pre -treatment QA procedures are mandatory for th e
correct delivery of a SAVI6 -1mini APBI treatment and for the achievement of a high
degree of conformance between the initial reference TPS plan and the clinically delivered
plan (Morcovescu et al., 2016) . Inter -fractional TPS plan re -evaluations should become a
routine when dealing with c ases where the applicator is very proximal to organs of risk
and where highly asymmetrical PTV_EVAL volumes are employed, in order to avoid
overlooking prohibitive sensitive structure doses and unacceptable treatment volume
coverages.
7.4 Overall Results a nd Discussion
The SAVI6 -1mini strut -based device proves to be a highly adaptable and versatile
APBI solution for patients with reduced breast and lumpectomy cavity volumes, and skin
105
and/or chest wall bridges. Inside the framework of a detailed and clear QA program, when
it is appropriately elected as the APBI device of choice, optimally implanted, and
comprehensively monitored during the course of treatment, all SAVI device types, but the
SAVI6 -1mini in particular, indeed offer a very effective and highly reproducible tool for
the treatment of complex breast cancer cases.
The SAVI (the Strut Adjusted Volume Implant ) device is one of the most novel
devices used in Accelerated Partial Breast Irradiation (APBI). Our analysis demonstrates
the dosimetric versatility and outl ines the clinical implementation process of the SAVI
brachytherapy device, especially for APBI cases that require more flexible dose
optimization, for both coverage of PTV volumes and sparing of dose to adjacent critical
structures.
106
CHAPTE R 8 CLINICAL RESULTS
Accelerated partial breast irradiation (APBI) has been widely embraced as an
acceptable modality for delivering adjuvant radiation therapy for a selected group of
patients undergoing breast -conserving therapy. The recently published GEC -ESTRO trial,
a Phase III randomized prospective trial, and other recent studies (Polgar et al., 2004;
Bitter et al., 2016) demonstrated superior cosmesis and noninferiority of APBI with
brachy therapy when compared to whole breast irradiation (WBI) . Several recently
published APBI studies demonstrat ed that local control that is equivalent to WBI with 5
to 10 years of followup (Vicini et al., 1997). Multicatheter interstitial therapy was
originally the method of choice , and all interstitial data reveal outstanding results (Vicini
et al ., 1999; Wazer et al ., 2006). More recently introduced s ingle -entry intracavitary
devices, like the MammoSite balloon, improved ease of use, but early expe rience proved
to be challenging in cases of inadequate skin and chest wall spacing , where there is a
difficulty in dosimetri cally modulating the dose, which in turn can lead to significantly
increased toxicity (Keisch et al., 2003; Hong et al., 2010; Kuske et al., 2014 ; Kamrava et
al., 2014) . The introduction of multilumen catheter devices allowed for a more
customized sculpt ing of the dose , better conformed to the woman’s anatomy, increasing
the number of women eligible for PBI, which in term resulted in a number of excellent
comparative studies trying to determine the dose modulation capacities, pros and cons of
all available multilumen devices ( Cross et al., 2008; Patel et al., 2010; Raffi et al., 2010;
Yashar, 2011; Rana et al ., 2015; Zaker et al., 2015) In addition to multicatheter and
single -entry devices, external beam APBI using photons or protons is also described in
the literature (Vaidya et al., 2004; Njeh et al., 2010) .In this chapter we indicate the results
of our own 5 year retrospective study on all SAVI device types (Yashar et al ., 2016) ,
reporting them in terms of local control, toxicity, and survival for the first 250 patients
treated across multiple institutions .
Toxicity
Toxicity was graded according to the Common Terminology Criteria for Adverse Events
version 3.0 , commonly used in prospective toxicity analysis for breast cancer studies
107
(Rehman et al ., 2016), except seroma and fat necrosis, which were modified by the
consortium to more appropriately fit APBI as shown in Table 36.
Tabel 36. Toxicity Definitions
Toxicity (CTCAE version 3.0) Grade 2
Hyperpigmentation Slight or localized
Induration Marked increase in density and firmness on palpation with or without minimal
retraction
Telangiectasias Moderate number
Breast pain Moderate, pain, or analgesics interfering with function, but not ADL
Toxicity (modified from
CTCAE version 3.0) Grade 0 Grade 1 Grade 2 Grade 3
Seroma Not mentioned or not
present Radiographic or clinical but
asymptomatic Symptomatic Aspirated or excised for
symptoms Fat necrosis
The collected data included comprehensive information such as disease status,
recurrence and whether true recurrence/marginal miss (TR/MM) or elsewhere and
histology of recurrence, contralateral and regional recurrence, DM, date of last
mammogram/ultrasound/magnetic resonance imaging, cosmesis score, and grading of late
effects including hyperpigmentation, induration, hormonal therapy, erythema, fibrosis,
breast pain, seroma, fat necrosis, telangiectasias, breast asymmetry an d cause (surgical or
radiation), and other toxicities. Cosmesis was graded excellent, good, fair, and poor
according to the Harvard scale (Rose et al., 1989) . Of note, in some centers, ultrasound
was routinely done, and fat necrosis or seroma was routinely indicated on followup
mammograms, but in others, changes expected after lumpectomy with or without radiation
were simply reported as ‘‘postlumpectomy changes.’ ’ This difference prompted the
modification of the grading of seroma and fat necrosis as per the included table to
represent those that were symptomatic.
The median age of the 250 patients in this study was 62 years (range, 40 -85), out of
who 73.5% had invasive disease and the remaining 26.5% had pure ductal carcinoma in
situ.). The majority (57%) of these women were older than 60 years , although 36 (14.4%)
were younger than 50 years (11% were <45 years). Most patients were postmenopausal
(84%), had estrogen receptor positive tumors (90%), received endocrine therapy (65%),
and did not receive chemotherapy (91%).
108
Tumor size was reported in 94% of subjects with 85% of patients having
pathologic tumor sizes in the range of 1 – 20 mm (2% were >30 mm). Median followup in
these patients was 59.5 months. Of this group, 80% had ≥ 3 years of followup and 70%
had ≥ 4 years of followup.
For the 250 patients, the mean V 90 was 96.1 ± 3.7%, V 95 was 93.5 ± 6.1%, V100
was 90.4 ± 5.5 cc, V150 was 30.5 ± 11.0 cc, and V 200 was 14.7 ± 5.1 cc. These
dosim etric variables differed by de vice. Skin dose mean was 269 cGy (79.1% of
prescription dose). For those patients with a skin bridge >3 mm and ≤5 mm or ≤3 mm, the
mean values were 274 cGy and 281 cGy (80.7 and 82.7% of the prescription dose),
respec tively. The mean rib dose was 273 cGy (80.4% of the pre scription dose).
In this cohort, all four sizes of applicator were used: 10 -1 (21%), 8 -1 (32%), 6 -1
(38%), and 9% received the SAVI6 -1mini size. More than half the patients had skin
spacing less than or equal to 10 mm (44% > 10 mm) with 12% and 17% having skin
bridges of 3 -5 mm and ≤3 mm, respectively. Of the entire population, 85.9% had
excellent/good cosmesis at 60 months. Ten of the 11 reporting sites had excellent/good
cosmesis in 96.2% , and in one center, excellent/ good c osmesis was reported in 57.9%.
Grade 2 or higher adverse events at any time for hyperpigmentation, induration,
erythema, telangiectasia, breast pain, symptomatic seroma, and symptomatic fat necrosis
were 0.4%, 3.0%, 3.5%, 3.0%, 3.9%, 4.8%, and 1.3%, respec tively. Time course of
toxicity is shown in Table 37. Infection rate was 3.7%, with some centers giving
prophylactic antibiotics.
Table 37. Toxicity onset at any time
Toxicity
(≥ Grade 2) At any time,
N (%) 0
12 Months >12
18
Months >18
24
Months >24 Months
Subjects at risk 230 230 220 217 214
Hyperpigmentation 1 (0.4%) 0 (0%) 0 (0%) 0 (0%) 1 (0.5%)
Induration 7 (3.0%) 2 (0.9%) 0 (0%) 1 (0.5%) 4 (1.9%)
Erythema 8 (3.5%) 6 (2.6%) 0 (0%) 0 (0%) 2 (0.9%)
Telangiectasia 7 (3.0%) 0 (0%) 1 (0.5%) 0 (0%) 6 (2.8%)
Breast pain 9 (3.9%) 3 (1.3%) 1 (0.5%) 3 (1.4%) 2 (0.9%)
Seroma 11 (4.8%) 6 (2.6%) 0 (0%) 0 (0%) 5 (2.3%)
Fat necrosis 3 (1.3%) 1 (0.4%) 0 (0%) 0 (0%) 2 (0.9%)
109
Dosimetric targets were met in virtually all patients and are similar to other
published series with excel lent target coverage and normal tissue avoidance. With the
struts adjacent to the target tissue, the allowable target for V 200 is more similar to the
interstitial target than the balloon target. But, as demonstrated in this serie s, this has not
led to increased toxicity. A recent single institutional series compared a large cohort of
patients (n 5 594) with APBI using different single -entry devices. This report also
observed outstanding target coverage with excellent skin and rib sparing (Rana et. al ,
2015) . Associations between dose and telangiectasia were made possible based on
univariate analysis . Since dose and skin spacing are related, and other publications have
demonstrated a similar correlation , this is was of no surprise. Other possible correlations
are made between V90 and V 150 and symptomatic seroma, and V 90 and fat necrosis, but
the numbers were overall too small to trigger for a prompt alteration of guidelines .
The strengths of this report include its multi -institutional participation and robust
numbers ( n = 250) with the longest followup for a single -entry multilumen breast
brachytherapy device. The s election criteria was practically unbiased, free of any screening
filters, since it basical ly included the first 250 subjects accrued . It is limited by its
retrospective nature, which may confound data as institutiona l toxicity reporting and
treatment policies may differ.
110
CONCLUSIONS
Based on the results of our studies , we can conclude that t he use of multilumen
applicators clearly simplif es brachytherapy APBI compared to interstitial brachytherapy,
allowing the advantages of brachytherapy over other forms of accelerated partial breast
radiation therapy accessible to more women.
The clinical implementation of the SAVI device poses various challenges to the
potential users, but within the frame of a robustly designed and implemented quality
assurance program, all standard dosimetrical goals are achieved, including conformance to
the tumor bed coverage and dose minimization to surrounding normal tissues, as indicated
by other similar studies (Scanderbeg et al., 2009 ; Yashar et al., 2008) .
We argue that the PTV_EVAL and PTV coverage goals achievable with the SAVI
device, with mean V95 over 95% and mean V90% over 97% , way above the protocol
required criteria of 90%/90%, on almost all possible clinical situations , is an assurance that
the use of this APBI device can adequately compensate for random and systematic errors
inherent in these type of treatments (Stojadinovic et al., 2008).
Even though the NSABP B -39/RTOG 0413 p rotocol imposes VHI criteria of 0.750
across all brachytherapy type applicators, therefore promoting a rather restric tive criteria
widely accepted and used for multi -catheter interstitial implant (Wu et al., 1988), our
study indicates that the use of this same criteria is unfit for SAVI type devices. Our
recommendation is that this parameter should not be used for the evaluat ion of the
adequacy of plans in breast brachytherapy, when SAVI trype devices are used, and if used,
to relax the threshold value from 0.750 to a more realistic value of 0.500.
The strut op en architecture design and mul tiple catheter options allow dose
sculpting to each patient’s unique anatomy and cavity lo cation. This flexibility helps to
overcome prior concerns with skin spacing and tumor beds positioned between the
overlying skin and chestwall that limited patient eligibilit y. The dosimetric parameters
considered relevant for our studies, and data reporting of those parameters, namely
especially Max Skin Dose, is accurately reflected in our clinical data outcome reports,
which indicates that the definitions in use for these parameters needed no reconside ration,
as some studies suggest (Park et. al , 2016). The clinical report we contributed to with
111
patient data confirms excellent tumor control comparable to other published APBI rates
and survival with low toxicity , based on me dian 59.5 month outcomes for p atients treated
with the strut -based applicator . Compared to external beam techniques for APBI, SAVI
brachytherapy seems to be as effective, with less toxicity.
112
BIBLIOGRAPHY
***, 2007 , NSABP Protocol B -39/RTOG 0413 Protocol, A randomized phase III study of
conventional whole breast irradiation (BWI) versus partial breast irradiation (PBI) for
women with stage 0, 1, or 2 breast cancer. http://atc.wustl.edu/proto cols/nsabp/b –
39/0413.pdf [accesat la data de 13 octombrie 2016]
Akhtari M., Pino R., Scarboro S.B., Bass B.L., Miltenburg D.M., Butler E.B., The B.S.,
2015 , Dosimetric considerations and early clinical experience of accelerated partial
breast irradiation using multi -lumen applicators in the setting of breast augmentation ,
Journal of Contemp orary Brachytherapy, 6, 423 -429.
Akino Y., Slessinger E., Srivastava S., Das I., 2014 , Impact of geometric unvcertainties in
accelerated partial breast irradiation using the strut -adjusted volume implant (SAVI) ,
Medical Physics, 41, 338.
Arthur D.W., Vicini F.A., Todor D.A., Julian T.B., Lyden M.R., 2010 , Improvements in
critical dosimetric endpoints using the Contura multilumen balloon breast brachytherapy
catheter to de liver accelerated parital breast irradiation: preliminary dosimetric findings
of a Phase IV trial , Int. Journal of Radiation Oncology Biology Physics ,79, 26 -33.
Arthur D.W., Vicini F., Kuske R.R., Wazer D.E., Nag S., 2003 , Accelerated partial breast
irrad iation: an updated report from the American Brachytherapy Society , Brachytherapy,
2(2), S361 -S362.
Arthur D.W., Wazer D.E., Koo D., Berle L., Cuttino L., Yunes M., Rivard M., Todor D.,
Tong S., Teneholz T., DiPetrillo T., 2003 , The importance of dose volum e histogram
evaluation in partial breast brachytherapy: a study of dosimetric parameters , Int. Journal
of Radiation Oncology Biology Physics, 57(2), S361 -S362.
Attix, H.F., 1986 , Introduction to radiological physics and radiation dosimetry , John
Wiley&Sons Inc., NJ, USA.
Baglan K.L., Martinez A.A., Frazion R.C., Kini V.R.,. Kestin L.L, Chen P.Y., Edmundson
G., Mele E., Jaffray D., Vicini F.A., 2001 , The use of high -dose-rate brachytherapy alone
after lumpectomy in patients with early -stage breast cancer tre ated with breast –
113
conserrving therapy , Int. Journal of Radiation Oncology Biology Physics, 50(4), 1003 –
1011.
Beaulieu L., Tedgren A.C., Carrier J.F., Davis S.D., Mourtada F., Rivard M.J., Thomson
R.M., Verhaegen F., Wareing T., Williamson J.F., 2012 , Report of the Task Group 186 on
model -based dose calculation methods in brachytherapy beyond the TG -43 formalism:
Current status and recommendations for clinical implementation , Med. Phys., 39(10),
6208 -6236.
Benitez P.R., Chen P.Y., Vicini F.A., et al, 2004 , Partial breast irradiation in breast
conserving therapy by way of interstitial brachytherapy , American Journal of Surgery,
188, 355 -364.
Betel, G.C., 1996 , Radiation Therapy Planning , 2nd ed., McGraw -Hill, USA.
Bitter S.M., Heffron -Cartwright P., Wennerst rom C., Weatherford J., Einstein D., Keiler
L.C., 2016 , WBR T vs. APBI: an interim report of patient satisfaction and outcomes ,
Journal of Contemporary Brachytherapy, 8(1), 17 -22.
Bloom E.S., Kirsner S., Mason B.E., Nelson C.L., Hunt K.K., Baumann D.P., Gifford
K.A., 2011 , Case Report: Accelerated partial breast irradiation using the strut -adjusted
volume implant single -entry catheter in brachytherapy for breast cancer in the settin g of
breast augmentation , Brachytherapy, 10(3), 178 -183.
Bovi J., Qi X.S., White J., Li X.A., 2007 , Comparison of three accelerated partial breast
irradiation techniques: treatment effectiveness based upon biological models ,
Radiotherapy and Oncology, 84, 226-232.
Brashears J.H., Dragun A.E., Jenrette J.M., 2008 , Late chest wall tocixity after
MammosSite breast brachytherapy , Brachytherapy, 8(1), 19 -25.
Brown D.B., Damato A.L., Sutlief S., Morcovescu S ., Park S -J., Reiff J., Shih A.,
Scanderbeg D.J., 2016 , A consensus -based, process commissioning template for high –
dose-rate gynecologic treatments , Brachytherapy, 15(5), 570 -577.
Brown S., McLaughlin M., Haile K., Israel P.Z., 2009 , Initial radiation experience
evaluating early tolerance and toxicities in pati ents undergoing accelerated partial breast
114
irradiation using the Contura Multi -Lumen Balloon breast brachytherapy catheter ,
Brachytherapy, 8(2), 227 -233.
Brown S., Vicini F., Vanapalli J., Whitaker T.J., Pope D.K., Lyden M., Bruggeman L.,
Haile K.L., 2012 , Factors associated with chest wall toxicity after accelerated partial
breast irradiation using high -dose-rate brachytherapy , Int. Journal of Radiation Oncology
Biology Physics, 83(3), 801 -805.
Cember, H., 1983 , Introduction of health physics , 3rd ed. McGr aw-Hill, USA.
Chandrasekara S., Dumitru N., Hyvarinen M., Pella S, 2015 , Dosimetric consequences
from minimal displacements in APBI with SAVI applicators , Medical Physics, 42, 3303
(doi: http://dx.doi.org/10.1118/1.4955995).
Chilukuri M., 2016 , Quality as surance and QMP for brachytherapy treatment delivery
and treatment planning system , Medical Physics, 43, 3776 (doi:
http://dx.doi.org/10.1118/1.4957658).
Cross C, Brown A, Escobar P, Kokal W, Mantz C., 2008 , Partial breast brachytherapy
utilizing the singleentry, multicatheter SAVI device in patients with less than 7 mm
skintocavity distance: Favourable acute skin toxicity outcomes from a phase II trial , Int J
Radiat Oncol Biol Phys., 72, S182.
Cui J., Mayadev J., Stern R., 2011 , Comprehensive quality assurance program for
accelerated partial breast irradiation using the SAVI HDR applicator , Medical Physics,
38, 3545 (doi: http://dx.doi.org/10.1118/1.3612205).
Cuttino L.W., Todor D., Rosu M. et al, 2009 , Skin and chest wall dose with multi -catheter
and MammoSite breast brachytherapy: Implications for late toxicity , Brachytherapy, 8,
223-226.
Dahl R., 2014 , Creation of standardized APBI SAVI cavities using CT registration ,
Medical Physics, 41, 243 .
Devlin P.M., Cormack R.A., Holloway C.L., Stewart A.J. , 2016 , Brachytherapy:
applications and techniques , 2nd ed., Demos Medical Publishing, NY, USA.
115
Dickler A, Kirk M, Choo J, Hsi WC, Chu J, Dowlatshahi K, Francescatti D., Nguyen C.,
2004 , Treatment volume and dose optimization of Mammosite breast brachytherapy
applicator , Int. Journal of Radiation Oncology Biology Physics , 59(2), 469 – 474.
Dou K., Jacobs M., Ottinger M., Seidel M., Reynolds S., 2011 , Dosimetric study of an
HDR applicator of SAVI for partial breast irradiation , Medical Physics, 38, 3571 .
Dou K., Jacobs M., Seidal M., Ottinger M., 2010 , Clinical implementation and quality
assurance of a new HDR applicator for partial breast irradiation , Medical Physics, 37,
3197 .
Dou K., Lerma F., Jacobs M., Ottinger M., Seidel M., Herchko M., Reynol ds S., 2012 ,
Dosimetric study and clinical implementation of an HDR applicator of SAVI for partial
breast irradiation , Medical Physics, 39, 3774 .
Edmundson G.K., Vicini F.A., Chen P.Y., Mitchell C., Martinez A.A., 2002 , Dosimetric
characteristics of the Ma mmosite RTS, a new breast brachytherapy applicator , Int.
Journal of Radiation Oncology Biology Physics, 52(4), 1132 -1139.
Foo M., Rogers K., Raulerson S., Mantz C., 2008 , Comparative dosimetry of 3 single –
entry, afterloading brachytherapy applicators for partial breast irradiation: SAVI,
Contura and MammoSite , Int. Journal of Radiation Oncology Biology Physics, 72(1),
S512.
Fowler J.F., Mount M., 1992 , Pulsed brachytherapy: the conditions for no significant loss
of therapeutic ratio compared with tradition al low dose rate brachytherapy , Int. Journal of
Radiation Oncology Biology Physics , 23, 661 -669.
Gifford K.A., Nelson C.L., Kirsner S.M., Kisling K.D., Ballo M.T., Bloom E.S., 2012 , On
the feasibility of treating to a 1.5 cm PTV with a commercial single -entry hybrid
applicator in APBI breast brachytherapy , Journal of Contemporary Brachytherapy, 4(1),
29-33.
Gifford K.A. , Horton J., Wareing T., Failla G., Mourtada F., 2006 , “Comparison of a
finite -element multigroup discrete -ordinates code with Monte Carlo for radiother apy
calcula tions”, Phys. Med. Biol., 51(9):2253 –2265.
116
Goetsch S.J., Attix F.H., Pearson D.W., Thomadsen B.R., 1991 , Calibration of 192Ir high –
dose rate afterloading systems , Medical Physics, 18(3), 462 -467.
Graf M., Scanderbeg D., Cervino L., Yashar C., Jiang S., 2011 , Dose calculation of a
breast cancer brachytherapy treatment using Monte -Carlo simulation , Medical Physics,
38, 3650 (doi: http://dx.doi.org/10.1118/1.3612657 ).
Gurdalli S, Kuske R, Quiet C., 2008 , Dosimetric comparison of three brachytherapy
applicators for partial breast irradiation , Proceedings of World Congress of
Brachytherapy, Brachytherapy, 7(2), 143.
Gurdalli S., Kuske R., Quiet C., 2007 , A new -sinlge -entry multicatheter breast
brachyt herapy applicator – initial clinical experience and dosimetric performance , Int.
Journal of Radiation Oncology Biology Physics, 69(3), S233 -S234.
Gurdalli S., Kuske R.R., Quiet C. A., Ozer M., 2011 , Dosimetric performance of strut –
adjusted volume implant: a new single -entry multicatheter breast brachytherapy
applicator , Brachytherapy, 10(2), 128 -135.
Hall E.J., 2000 , Radiology for the radiologist , Lippincott Williams&Wilkins, Philadelphia,
PA, USA
Harmon J.F., Rice B.K., 2013 , Comparison of planning techniques when air/fluid is
present using the strut -adjusted volume implant (SAVI) for HDR -based accelerated partial
breast irradiation , Journal of Applied Clinical Medical Physics, 14(6), 264 -273.
Hong R.L., Akbari S., Weintritt D., Sebastian M., Martin M., Ji H., Lorio V., McRae D.,
2010 , Accelerated partial breast irradiation at Virginia Hospital Center: post hc analysis
of dosimetry comparing MammoSite, Contura, and SAVI devices , Int. Journal of Radiation
Oncology Biology Physics, 78(3), S249.
Hong R. L., Patel R., Han B., Kuske R., Strasser J., Kormanicky L., Mahalingam S.,
Mohideen N., 2012 , Results of the SAVI research collaborative group registry: correlating
clinical toxicity with dosimetric parameters in patients treated with APBI using strut –
adjusted brachytherapy , Int. Journal of Radiation Oncology Biology Physics, 84(3), S186.
117
Huang Y.J., Blough M., 2010 , Dosimetric effects of air pocket sizes in MammoSite
treatment as accelerated partial breast irradiation for early breast cancer , Journal of
Applied Clinical Medical Physics, 11(1), 46 -56.
Hyvarinen M., Pella S., Leventouri T., Casey C., Dumitru N., Herrera R., Long S., 2014 ,
Investigation of dose variation in high dose radiation brachytherapy , Medical Physics, 41,
374.
Iftimia I., McKee A.B., Nixon A.J., Hunter K.U., Girshovich L., Halvorsen P., 2015 ,
Retrospective review of HDR SAVI breast cases , Int. Journal of Radiation Oncology
Biology Physics, 93(3S), E551.
Jacob D., Chen H., Simpson L., 2011 , High dose brachytherapy planning of a left bre ast
cancer patient with in situ pacemaker , Medical Physics, 38, 3626 (doi:
http://dx.doi.org/10.1118/1.3612558).
Ji H., Lorio V., Cernica G., Han J., Nurhussien M., Nasr N., Hong R., 2016 , Quality
management of Accelerated Partial Breast Irradiation (APBI) plans , Medical Physics, 43,
3755 (doi: http://dx.doi.org/10.1118/1.4957533 ).
Kalavagunta C., Lasio G., Yi B., Zhou J., Lin M., 2015 , Characterization of
pacemaker/ICD dose in SAVI HDR brachytherapy , Medical Physics, 42, 3370 (doi:
http://dx.doi.org/10.1118/1.4924531 ).
Kamrava M., Park S.J., Noor B., Demanes D., 2014 , Dosimetric comparison and early
clinical outcomes of interstitial versus strut assisted volume implant (SAVI) for
accelerated partial breast irradiation , Brachytherapy, 13, S82.
Keisch M, Vicini F, Kuske RR, Hebert M, White J, Quiet C, et al., 2003 , Initial clinical
experience with the Mammosite breast brachytherapy applicator in women with
earlystage breast cancer treated with breastconserving therapy , Int. Journal of Radiation
Oncology Biology Physics , 55, 289 –93.
Khan, F., 2003 , The physics of radiation therapy , 3rd ed. Lippincott W&W, Philadelphia,
USA.
118
Kim Y., Trombetta M., 2011 , Dosimetric evaluation of MammoSite multi -lumen balloon
applicator compared with contura applicator for high dose rate (HDR) breast
brachythearpy , Int. Journal of Radiation Oncology Biology Physics , 81(2), 902 -903.
Knoll, F.K., 2000 , Radiation detection and measurement , 3rd ed. John Wiley&Sons Inc.,
NJ, USA.
Kubo H.D., Glasgow G.P., Pethel T.D., Thomadsen B.R., Williamson J.F., 1998 , High
dose-rate brachytherapy treatment delivery: Report of the AAPM Radiation Therapy
Committee Task Group no. 59 , Medical Physics, 25(4), 375 -403.
Kuske R.R., Young S.S., 2014 , Breast brachytherapy versus whole -breast irradiation:
reported differences may be statistically significant but clinically trivial , Int. Journal of
Radiation Oncology Biology Physics , 88, 266-268.
Kutcher G.J., Coia L., Gillin M., Hanson W.F., Leibel S., Morton R.J., Palta J.R.,
Purdy J.A., Reinstein L.E., Svensson G.K., Weller M., Wingfield L., 1995 ,
Comprehensive QA for Radiation Oncology: Report of the AAPM Radiation Therapy
Committee Task Group No. 40, AAPM Report No. 46 , American Association of
Physicists in Medicine; American Institute of Physics New York, NY.
Kyriacou A., 2016 , Acceptance procedures for SAVI, Contura and ML Mammosite ,
Medical Physics, 43, 3776 (doi: http://dx.doi.org/10.1118/1.4957657 ).
Lewis E.E., Miller W.F., 1984 , Computational methods of neutron transport , Wiley, New
York , USA.
Likhacheva A., Grade E., Sadeghi A., Sokolowski T, 2016 , Clinical implications when
using TG -186 (ACE) heterogenity softw are, Medical Physics, 43, 3466
(http://dx.doi.org/10.1118/1.4956157 ).
Liu Z., Sha S., Hargrove I., Barrett J., 2012 , Evaluation of the dosimetric effect of patient
daily setup variations during strut -adjusted volume implant (SAVI) breast brachytherapy ,
Int. Journal of Radiation Oncology Biology Physics, 84(3), S722 -S723.
Loupot S., Han T., Salehpour M., Gifford K., 2014 , APBI brachytherapy treatment
planning: the impact of heterogeneous dose calculations , Medical Physics, 41, 388.
119
Lu S.M., Scanderbeg D.J., Barna P., Yashar W., Yashar C., 2012 , Evaluation of two
intracavitary high -dose-rate brachytherapy devices for irradiating additional and
irregurarly shaped volumes of breast tissue , Medical Dosimetry, 37(1), 9 -14.
Lu S.M., Scanderbeg D.J., Barna P., Yasha r W., Yashar C.M., 2010 , Use of the SAVI
breast brachytherapy device to treat larger and asymmetric tissue volumes , Int. Journal of
Radiation Oncology Biology Physics, 78(3), S821.
Meredith W.J., Massey J.B., 1972 , Fundamental physics of radiology , 2nd ed. W&W
Company, Baltimore, USA.
Mooij R., Ding X., Nagda S., 2014 , Comparison of Cyberknife versus HDR (SAVI) for
partial breast irradiation , Medical Physics, 41, 370 .
Mooney K., Altman M., Garcia -Ramirez J., Thomas M., Zoberi I., Mullen D., DeWees T.,
Esthappan J., 2015 , Optimization of dosimetric guidelines for accelerated partial breast
irradiation (APBI) using the Strut -Adjusted Volume Implant , Medical Physics, 42, 3 536.
Morcovescu S ., Cosma C., Morton J.D., 2016 , Dosimetrical Evaluation and clin ical
implementation of a strut -adjusted -volume -implant SAVI device used for accelerated
partial breast irradiation , Romanian Journal of Physics, 61(7 -8), 1312 -1320.
Morcovesc u S., Morton J.D., Perry K., 2009 , The SAVI 6 -1 Mini APBI applicator: A
unique solution for small lumpectomy cavities and skin distances , Int. Journal of Radiation
Oncology Biology Physics , 75, S718.
Morcovescu S. , Morton J.D., Perry K., 2011 , Investigation of Interfraction Variance of a
SAVI6 -1 Mini APBI Applicator in Patients with Reduced Chest Wall and Skin Sparing , Int.
Journal of Radiation Oncology Biology Physics , 81(2), S263.
Morcovescu S. , Morton J.D., 2016 , The use of a SAVI strut-based device in the boosting
of a stenotic distal vagina , Brachytherapy, 15, S132.
Morcovescu S. , Morton J.D., 2009 , A comparative clincal study and dosimetry planning
experience with the new Contura multi -lumen versus MammoSite single -lumen high -dose-
120
rate balloon breast applicators under the RTOG 0413 protocol , Brachytherapy, 8(2),
S136 -137.
Morcovescu S. , Morton J.D., 2009 , The miniSAVI multicatheter accelerated partial breast
irradiation applicator and its use in patients with small lumpectomy ca vities and/or
inadequate skin distance: early clinical experience , Brachytherapy, 8(2), S138.
Morcovescu S. , Morton J.D., Perry K., 2012, Comprehensive dosimetric evaluation of a
small strut -based device: a retrospective single -institution study , Radiotherapy and
Oncology, 103(2), S49
Morcovescu S. , Morton J.D., Boleware Y.E., Kerri P., 2014 , Clinical experience with a
miniature accelerated partial breast irradiation device: a 5 -year single institution
comprehensive study , Brachytherapy, 13, S82.
Nath R, Anderson LL, Luxton G, Weaver KA, Williamson JF, Meigooni AS., 1995 ,
Dosimetry of Interstitial brachytherapy sources: Recommendations of the AAPM
Radiation Therapy Committee Task Group No 43 American Association of Physicists in
Medicine , Medical Physics, 22, 209 –234.
Njeh C.F., Saunders M.W., Langton C.M., 2010 , Accelerated partial breast irradiation
(APBI): a review of available techniques , Radiation Oncology, 5, 90.
Ojeda -Fournier H., Olson L.K., Rochelle M., Hodgens B.D., Tong E., Yashar C.M., 2011 ,
Accelerated partial breast irradiation and posttreatment imaging evaluation ,
Radiographics. 31(6), 1701 -1716.
Park S., DeMarco J., Kamrava M., Demanes D., Low D., 2011 , Evaluation of interfraction
motion of the Strut -Adjusted Volume Implant (SAVI) using 3D reconstruction from CT
scout images , Medical Physics, 38, 3575 (doi: http://dx.doi.org/10.1118/1.3612334).
Park S., Ragab O., Patel S., Kim Y., Demanes J., Kamrava M., 2016 , Is Maximum Skin
Dose a Reliable Metric for Accelerated Partial Breast Irradiation with Brachytherapy? ,
Medical Physics, 43, 3626 (doi: http://dx.doi.org/10.1118/1.4956887 ).
121
Patel R., Wong G., Tran L., 2010 , Multi -lumen hybrid breast brachytherapy: St rut versus
balloon – Dosimetric results in first 100 patients , Brachytherapy, 9, 532 -553.
Patel RR, Becker SJ, Das RK, Mackie TR., 2007 , A dosimetric comparison of accelerated
partial breast irradiation techniques: Multicatheter Interstitital brachytherap y,
Threedimensional conformal radiotherapy and supine versus prone helical tomotherapy ,
Int. Journal of Radiation Oncology Biology Physics , 68(3), 935 –942.
Pella S., 2016 , Planning and Treatment Delivery for APBI Using Savi, Contura and ML
Mammosite , Medical Physics, 43, 3776 (doi: http://dx.doi.org/10.1118/1.4957659 ).
Perez, C.A., 1992 , Principles and practice of radiation oncology , 2nd ed. J.B. Lippincott
Company, Philadelphia, USA.
Pinder J., Chandrasekara S., Pella S., 2016 , The consistency dosim etric analysis of the
accelerated breast brachytherapy , Medical Physics, 43, 3465 (doi:
http://dx.doi.org/10.1118/1.4956154 ).
Pinn-Bingham M.E., Zhang J., Dietrich S., Braggins W., Sehgal V., Al -Ghazi M. , Wong
J., Kuo J., 2011 , Dosimetric parameters and clinical characteristics of early stage breast
cancer patients treated with the SAVI breast brachytherapy device , Int. Journal of
Radiation Oncology Biology Physics, 81(2), S269.
Polgar C., Major T., Fodor J., Nemeth G., Orosz Z., Sulyok Z., Udvarhelyi N., Somogyi
A., 2004 , High -dose rate brachytherapy alone versus whole breast radiotherapy with or
without tumor bed boost after breast -conserving surgery: seven -year results of a
comparative study , Int. Journ al of Radiation Oncology Biology Physics, 60(4), 1173 -1181.
Raffi J.A., Davis S.D., Hammer C.G., Micka J.A., Kunugi K.A., Musgrove J.E., Winston
J.W., Ricci -Ott T.J., DeWerd L.A., 2010 , Determination of exit skin dose for 192Ir
intracavitary accelerated pa rtial breast irradiation with thermoluminiscent dosimeters ,
Medical Physics , 37(6), 2693 -2702.
Rana Z., Nasr M.N., Ji H., Lorio V., Akbari S., Sebastian M., Martin M., Hong R., 2015 ,
Comparative dosimetric findings using accelerated partial breast irradiation across five
catheter subtypes , Radiation Oncology, 10, 160 .
122
Raviner N., Anderson L.L., Luxton G., Weaver K.A., Williamson J.F., Meigooni A.S.,
1994 , Dosimetry of interstitial brachytherapy sources: recommendations of the AAPM
radiation Therapy Committee Task Group No. 43 , Medical Physics, 22 (2), 209 -234.
Rehman S., Agarwal R., Ochoa L., Yoberi J., Margenthaler J., Thomas M., Yoberi I.,
2016 , Prospective analysis of toxicity in patients treated with strut adjusted volume
implant for early stage breast cancer , Brachytherapy, ( doi:
10.1016/j.brachy.2016.04.008) .
Reiff J., Scanderbeg D., Morcovescu S ., Butler E., Imhoff K., 2013 , Dosimetry of 108
strut-based Accelerated Partial Breast Irradiation (APBI) treatments with applicator
distance less then 7 mm from both skin and rib s, Medical Physics, 40, 104 (doi:
http://dx.doi.org/10.1118/1.4814016 ).
Richardson S.L., Pino R., 2010 , Dosimetric effects of an air cavity for the SAVITM partial
breast irradiation applicator , Medical Physics, 37(8), 3919 -3926.
Richardson S. ., Chen L.K., Pino R., Bloch C., Parikh P., 2013, Dosimetric comparison of
TG-43 formalism with Brachyvision Acuros and Monte Carlo Method for patients treated
with the SAVI partial breast applicator , Brachytherapy , 12, S22.
Rivard M.J., Cours ey B.M., DeWerd L.A., Hanson W.F., Huq M.S., Ibbott G. S., Mitch
M.G., Nath R., Williamson J.F., 2004 , Update of AAPM Task Group No. 43 Report: A
revised AAPM protocol for brachytherapy dose calculations , Medical Physics, 31 (3),
633-674.
Robinson R.C., Nelson C.L., Bloom E.S., Kisling K.D., Mason B.E., Fisher G.D., Kirsner
S.M., 2015 , Contralateral breast dose from partial breast brachytherapy , Journal of
Applied Clinical Medical Physics, 16(6), 17 -22.
Rose M.A., Olivotto I., Cady B., 1989 , Conservative surgery and radiation therapy for
early stage breast cancer. Long term cosmetic results , Arch. Surg, 124 , 153 -157.
Scanderbeg DJ, Yashar C, Rice R, Pawlicki T., 2009 , Clinical implementation of a new
HDR brachytherapy device for partial breast irradiation . Radiother Oncol., 90(1), 36 –42.
123
Scanderbeg DJ, Yashar C, White G., Rice R, Pawlicki T., 2009 , Evaluation of three APBI
techniques under NSABP B -39 guidelines , Journal of Applied Clinical Medical Physics,
11(1), 274 -280.
Scanderbeg, D. , Barna P., Jiang S., Pawlicki T., Yashar C., 2009, Dosimetric comparison
of SAVI, Mammosite, Contura and Clearpath for accelerated partial breast irradiation ,
Medical Physics, 36, 2536 (doi: http://dx.doi.org/10.1118/1.3181545 ).
Sehgal V., Zhang J., Dietr ich S., Al -Ghazi M., Wong J., Kuo J., Ramsinghani N., 2011 ,
Dosimetric analysis of patients treated with accelerated partial breast irradiation using
the MammoSite and SAVI applicators , Med ical Phys ics, 38(6), 3575.
Shah C., Wobb J., Manyam B., Khan A., V icini F., 2016 , Accelerated partial breast
irradiation utilizing brachytherapy: patient selection and workflow , Journal of
Contemporary Brachytherapy, 8(1), 90 -94.
Sherman J., Pearson D., Parsai E., 2011 , The comparison of dose modification factors for
two multi -lumen brachytherapy applicators used in partial breast irradiation , Medical
Physics, 38, 3661. ( doi: http://dx.doi.org/10.1118/1.3612707 ).
Slessinger E., Pepin E., Zhao Q., Zhao L., Das I., 2012 , Dose correction in lung for HDR
breast brachytherapy, Journal of Contemp orary Brachytherapy, 2, 106 -110.
Smith B.D., Arthur D.W., Buchholz T.A., Hafty B.G., Hahn C.A., Handenbergh P.H.,
Julian T.B., Marks L.B., Todor D.A., Vicini F.A., Whelan T.J., White J., Wo J.Y., Harris
J.R.., 2009 , Accelerated partial breast irradiation consensus statement from the American
Society for Radiation Oncology (ASTRO) , Int. Journal of Radiation Oncology Biology
Physics , 74, 987 -1001.
Stojadinovic S., Cuttino L.W., Barani I. J., Todor D., Arthur D.W., 2008 , Errors and
uncertainties in accelerated partial breast irradiation (APBI) with balloon brachytherapy:
are current target coverage goals adequate? , Int. Journal of Radiation Oncology Biology
Physics, 2(1), S518.
Thomadsen B. R., 2000 , Achieving quality in brachytherapy , Institute of Physics
Publishing, Bristol, UK
124
Thrower S., Shaitelman F.S., Bloom E., Salehpour M., Gifford K., 2016 , Comparison of
dose distributions with TG+43 and collapsed cone convolution algorithms applied to
accelerated partial breast irradiation patient plans , Int. Journal of Radiation Oncology
Biology Physics , 95(5), 1520 -1526.
Vaidya J.S.,Tobias J.S., Baum M., Keshtgar M., Joseph D., Wenz F., Houghton J.,
Saunders C., Corica T., D’Souza D., Sainsbury R., Massarut S., Taylor I., Hilaris B., 2004 ,
Intraoperative radiotherapy for breast cancer , Lancet Oncol ogy, 5, 165 -173 (doi:
http://dx.doi.org/10.1016/S1470 -2045(04)01412 -3).
Vicini F.,.Kini V.R, Chen P.Y., Horwitz E., Gustafson G., Benitez P., Edmundson G.,.
Goldstein N, McCarthy K., Martinez A., 1999 , Irradiation of the tumor bed alone after
lumpectomy in selected patients with early -stage breast cancer treated with breast
conserving therapy , J. Surg Oncol. , 70(1), 33-40.
Vicini F.A., Chen P.Y., Fraile M., Gustafson G.S., Edmundson G.K., Jaffray D.A, Benitez
P, Pettinga J., Madrazo B., Ingold J.A., Goldstein N.S., Matter R.C., Martinez A.A, 1997 ,
Low-dose-rate brachytherapy as the sole radiation modality in the management of patients
with early -stage breast cancer treated with breast -conserving therapy: Preliminary results
of a pilot trial , Int. Journal of Radiation Oncology Biology Physics , 38(2) , 301 -310.
Wazer D.E., Kaufman S., Cuttino L., DiPetrillo T., Arthur D.W., 2006 , Accelerated
partial breast irradiation: an analysis of variables associated with late toxicity and long –
term cosmetic outcome after high -dose-rate interstitial brachytherapy , Int. Journa l of
Radiation Oncology Biology Physics, 64(2), 489 -495.
Wazer D.E., Arthur D.W., Vicini F.A., 2006 , Accelerated partial breast irradiation:
techniques and clinical implemen tation , Springer -Verlag, Berlin.
White G., Reiff J., Scanderbeg D., 2010 , Treatment feasibility of a new APBI device: SAVI
double helix , Int. Journal of Radiation Oncology Biology Physics, 78(3), S847.
Williamson J.F., Thomadsen B.R., Nath R., 1996, Brachytherapy Physics: AAPM Summer
School 1994 , Medical Physics, 23(5).
Wu A., Ulin K., Sternick E.S., 1987 , A dose homogeneity index for evaluating 192Ir
interstitial breast implants , Medical Physics, 15(1), 104 -107.
125
Yashar C, Quiet C, Scanderber D, Zannis V, Gurdalli S, Rice R, et al. , 2008 , Use of the
breast brachytherapy d evice SAVI to obtain flexible dose modulation for normal structures
in close proximity to device , Breast cancer Symposium.Washington DC: 2008, Sep 5 -7.
Yashar C., 2011 , Counterpoint: the device with the most dose modulation capacity ,
Brachytherapy, 10(6), 436-439.
Yashar C., Attai D., Butler E., Einck J., Finkelstein S., Han B., Hong R., Komarnicky L.,
Lyden M., Mantz C., Morcovescu S ., Nigh S., Perry K., Pollock J., Reiff J., Scanderbeg
D., Snyder M., Kuske R., 2016 , Strut -based accelerated partial breas t irradiation: Report
of treatment results for 250 consecutive patients at 5 years from multicenter retrospective
study , Brachytherapy, 15(6), 780 -787.
Yashar C., Mahmood U., 2010 , Strut -adjusted volume implant: a targeted radiation
treatment in breast cancer , Future Oncology, 6(12), 1813 -1816.
Yashar C., Scanderberg D, Wallace A, Blair S, Barna P, Mantz C., 2009, Accelerated
partial breast irradiation via SAVI: one year followup , American Society of Breast
Surgeons, CA, San Diego, Apr 23 -26.
Yashar C.M., Blair S., Wallace A., Scanderbeg D., 2009 , Initial clinical experience with
the Strut -Adjusted Volume Implant brachytherapy applicator for accelerated partial
breast irradiation , Brachytherapy, 8, 367 -372.
Yashar C.M., Scanderbeg D., Kuske R. , Wallace A., Zannis V., Blair S., Grade E.,
Swenson V.H., Quiet C., 2011 , Initial clinical experience with the Strut -Adjusted Volume
Implant (SAVI) breast brachytherapy device for accelerated partial -breast irradiation
(APBI): first 100 patients with more than 1 year of follow -up, Int. Journal of Radiation
Oncology Biology Physics , 80(3), 765 -770.
Zaker R., Nasr N.M., Ji H., Lorio V., Akbari S., Sebastian M., Martin M., Hong R.L.,
2015 , Comparative dosimetric findings using accelerated partial breast irradiation across
five catheter subtypes , Radiation Oncology, 10, 160 -169.
126
Zhang Z., Dou K., 2012 , Monte Carlo dosimetric study of SAVI applicator for partial
breast irradiation , Medical Physics, 39, 3756 .
127
THE LIST OF PAPERS PUBLISHED DURING THE DOCTORAL PROGRAM
Scientific Papers published in ISI impact journals
1. Yashar C., Attai D., Butler E., Einck J., Finkelstein S., Han B., Hong R., Komarnicky
L., Lyden M., Mantz C., Morcovescu S ., Nigh S., Perry K., Pollock J., Reiff J.,
Scanderbeg D., Snyder M., Kuske R., 2016 , Strut -based accelerated partial breast
irradiation: Report of treatment results for 250 consecutive patients at 5 years from
multicenter retrospective study , Brachytherap y, 15(6), 780 -787 (ISI Thompson impact
factor: 2.088)
2. Brown D.B., Damato A.L., Sutlief S., Morcovescu S ., Park S -J., Reiff J., Shih A.,
Scanderbeg D.J., 2016 , A consensus -based, process commissioning template for high –
dose-rate gynecologic treatments , Brachytherapy, 15(5), 570 -577 (ISI Thompson impact
factor: 2.088)
3. Morcovescu S ., Cosma C., Morton J.D., 2016 , Dosimetrical Evaluation and clincail
implementation of a strut -adjusted -volume -implant SAVI device used for accelerated
partial breast irradiati on, Romanian Journal of Physics, 61(7 -8), 1312 -1320 (ISI
Thompson impact factor: 1.398)
128
List of scientific papers presented at National and International
Congresses and Scientific Meetings
1. Morcovescu S. , Cosma C., Ferenczi J., 1999 , Radon measurement from touristic
underground waters in Maramureș county, Radon in the Living Environment, April
19-23, Athens, Greece, Book of Proceedings, 269 -279.
2. Morcovescu S. , Morton J.D., Perry K., 2009 , The SAVI 6 -1 Mini APBI applicator:
A uniqu e solution for small lumpectomy cavities and skin distances , ASTRO 51st
Annual Meeting, November 1 -5, 2009, Chicago, IL, Int. Journal of Radiation Oncology
Biology Physics , 75, S718.
3. Morcovescu S. , Morton J.D., 2009 , A comparative clincal study and dosi metry
planning experience with the new Contura multi -lumen versus MammoSite single –
lumen high -dose-rate balloon breast applicators under the RTOG 0413 protocol ,
ABS Annual Meeting, May 31 -June 2, 2009, Toronto, Canada, Brachytherapy,
8(2), S136 -137.
4. Morcovescu S. , Morton J.D., 2009 , The miniSAVI multicatheter accelerated
partial breast irradiation applicator and its use in patients with small lumpectomy
cavities and/or inadequate skin distance: early clinical experience , ABS Annual
Meeting, May 31 -June 2, 2009, Toronto, Canada, Brachytherapy, 8(2), S138.
5. Morcovescu S. , Morton J.D., Perry K., 2011 , Investigation of Interfraction
Variance of a SAVI6 -1 Mini APBI Applicator in Patients with Reduced Chest Wall
and Skin Sparing , ASTRO 53rd Annual Meeting, October 2 -6, 2011 , Miami Beach,
FL, USA, (poster)
6. Morcovescu S. , Scanderbeg D., Reiff J., Butler E., Sadeghi A., Graves Y.,
Huaying J., Imhoff K. Webster J., Patra P., Mohideen N., Jacob D., Mantz C.,
2012 , Dosimetric Characteristics of Strut -Based APBI Devices , American College
Radiation Oncology (ACRO) Annual Meeting, February 25, 2012, Fort. Myers,
FL, (poster)
7. Hong R, Lorio V, Huaying J, Han B, Shieh E, Imhoff K, Kuske R, Quiet C,
Sadeghi A, Perry K, Morcovescu S , Yashar C, Scanderbeg D, Graves Y, Strasser
J, Dayee Jacob D, Reiff J, Komarnicky L, Mahalingam S, Webster J, Nigh S,
Mohideen N, Lobo P, Farmer M, Berry M, Patra P, Mantz C, Finkelstein S,
Pollock J, Butler E, Attai D, Patel R, 2012 , Excellent/Good Cosme tic Outcomes in
129
Patients Treated with a Strut -Based Brachytherapy Applicator (SAVI) for
Accelerated Partial Breast Irradiation , National Consortium of Breast Ce nters,
Inc. annual meeting, March 10-14, National Consortium of Breast Centers, Inc.
annual meet ing, Mar ch 10-14, 2012 , Las Vegas, NV (poster)
8. Morcovescu S , Morton J, Perry K, 2012 , Comprehensive dosimetric evaluation of
a small strut -based APBI Device: a retrospective single -institution study , World
Congress of Brachytherapy, May 10 -12, 2012, Bar celona, Spain ( poster)
9. Strasser J., Jacob D., Koprowski C., Attai D., Butler E., Finkelstein S., Han B.,
Hong R., Komarnicky L., Kuske R., Lyden M., Mahalingam S., Mantz C.,
Morcovescu S. , Nigh S., Perry K., Pollock J., Reiff J., Scanderbeg D., Yashar C.,
2012 , Accelerated partial breast irradiation using a strut -based brachytherapy
device: A multi -institutional initial report on acute and late toxicity with greater
than 24 -month follow -up, Breast Cancer Symposium, September 13 -15, 2012, San
Francisco, CA (poster)
10. Perry K, Attai D, Finkelstein SE, Han B, Hong R, Kuske R, Lyden M,
Mahalingam S, Mantz C, Morcovescu S , Nigh S, Pollock J, Reiff J, Scanderbeg D,
Strasser J, Yashar C., 2013 , Outcomes for Accelerated Partial Breast irradiation
with a strut -based brachytherapy applicator: 320 patients with 3 -year median
follow up , American Society of Breast Surgeons 14th annual meeting, May 1 -5,
2013, Chicago, IL (poster)
11. Reiff J. E., Scande rbeg D., Morcovescu S. , Butler E., Imhoff K., 2013 , Dosimetry
of 108 Strut -Based Accelerated Partial Breast Irradiation (APBI) Treatments With
Applicator Distance Less Than 7 mm from Both the Skin and Ribs , American
Association of Physicists in Medicine (A APM) 55th annual meeting, August 4 -8,
2013, Indianapolis, IN
12. Einck J.P., Scanderbeg D., Kuske R., Hong R., Han B., Perry K., Reiff J.,
Mahalingam S., Nigh S., Strasser J., Mantz C., Pollock J., Morcovescu S,
Komarnicky L, Finkelstein S., Yashar C., 2013 , Accelerated partial breast
irradiation using strut -based brachytherapy in ductal carcinoma in situ patients: A
report on 321 patients with median 25 -month follow up. Data updated for 284
patients with median 27 -month follow up , Breast Cancer Symposiu m, September
7-9, 2013, San Francisco, CA (poster)
130
13. Reiff J. E., Scanderbeg D., Morcovescu S. , Butler E., Imhoff K. , 2013 , Dosimetric
analysis of 1007 Strut -based APBI treatments , American Society of Radiation
Oncology (AS TRO) 55th annual meeting, September 22 -25, 2013 , Atlanta, GA
(poster)
14. Morcovescu S. , Morton J.D., Boleware Y.E., Kerri P., 2014 , Clinical experience
with a miniature accelerated partial breast irradiation device: a 5 -year single
institution comprehensive study , American Brachytherapy Society (ABS) Annual
meeting, April 3 -5, 2014 , San Diego, CA , Brachytherapy, 13, S82.
15. A. D. Yock, J. E. Reiff, S. Morcovescu , D. Scanderbeg, 2015 , Treatment time
nomogram for strut -based accelerated partial breast applicators , American
Association of Physicists in Medicine (AAPM) 57th annual meeting, July 11 -16,
2015, Anaheim, CA (poster)
16. Morcovescu S. , Morton J.D., 2016 , The use of a SAVI strut -based device in the
boosting of a stenotic distal vagina , World Congress of Brachytherapy, June 27 –
29, 2016, San Francisco, CA, Brachytherapy 15, S132.
Copyright Notice
© Licențiada.org respectă drepturile de proprietate intelectuală și așteaptă ca toți utilizatorii să facă același lucru. Dacă consideri că un conținut de pe site încalcă drepturile tale de autor, te rugăm să trimiți o notificare DMCA.
Acest articol: BABE Ș-BOLYAI UNIVERSITY CLUJ -NAPOCA FACULTY OF ENVIRONMENTAL SCIENCE AND ENGINEERING Dosimetric and quality assurance procedure evaluation of the… [627283] (ID: 627283)
Dacă considerați că acest conținut vă încalcă drepturile de autor, vă rugăm să depuneți o cerere pe pagina noastră Copyright Takedown.