Chemical Exchange Saturation Transfer MRI for Detection of Cell Death
in Breast Cancer Xenografts
by
Jonathan Klein
A thesis submitted in conformity with the requirements for the degree of Master of Science
Department of Medical Biophysics University of Toronto
© Copyright by Jonathan Klein 2018
ii
Chemical Exchange Saturation Transfer MRI for Detection of Cell
Death in Breast Cancer Xenografts
Jonathan Klein
Master of Science
Department of Medical Biophysics University of Toronto
2018
Abstract
Detecting cell death after chemotherapy could help optimize treatment regimens and improve
outcomes. Breast cancer xenografts (MDA-MB-231) were scanned before and after
chemotherapy to investigate parameters of chemical exchange saturation transfer (CEST) MRI
that can differentiate regions of cell death from viable tumour. The CEST effect at 0.5 μT
saturation amplitude was measured using the magnetization transfer ratio (MTR) at 1.8 and -3.3
ppm frequency offsets. An MTR cutoff of 0.12 at 1.8 ppm was able differentiate between viable
tumour and cell death regions (p<0.0001) and the detected patterns of cell death closely matched
those detected with ISEL staining. Using this cutoff, the mean increase in cell death index (±
standard error of the mean) after chemotherapy was 4±4%, 10%±7%, 10±8%, and 4±9% at 4, 8,
12, and 24 h, respectively. These results suggest that CEST MRI can detect cell death from
chemotherapy in breast cancer.
iii
Acknowledgments
I am deeply indebted to so many people for their patience, expertise, and assistance, without
which this project never could have been completed. First, thank you to my supervisors Dr. Greg
Stanisz and Dr. Gregory Czarnota for their guidance through my graduate experience. I would
also like to thank Dr. Wilfred Lam and Dr. Wendy Oakden for training me on the MRI imaging
platform and helping me with my scans and to Dr. Kim Desmond for her previous work in
helping design the MRI sequences used for these experiments. Thank you also to the rest of the
members of the Stanisz lab including Kayla Sutton, Dr. Lech Skorski, and Dr. Hatef Mehrabian.
Thank you to Linda Nghiem and Katerina Molnarova for their expert help with animal care and
to Margaret Koletar for help with implanting, growing and monitoring the xenografts. Thank you
also to Christine Tarapacki, Farah Hussein, Anoja Giles and Dr. Azza Al-Mahrouki for teaching
me how to prepare, store, and analyze histological specimens. All of you were so kind to provide
your time and expertise to help me.
Also, thank you to my advisory committee members, Dr. David Goertz and Dr. Martin Yaffe for
their advice and discussions of how best to design and implement my project goals. And, finally,
I would like to thank my family, especially my wife, Casandra Campbell, for all the support,
patience and love that has allowed me to advance and succeed within my graduate program.
The work described in this thesis was supported by a grant from the Terry Fox Research Institute
and a Canada Graduate Scholarship from the Canadian Institutes of Health Research, as well as
the University of Toronto Department of Medical Biophysics.
iv
Table of Contents
Acknowledgments.......................................................................................................................... iii
Table of Contents ........................................................................................................................... iv
List of Figures and Tables.............................................................................................................. vi
Chapter 1
Background ......................................................................................................................................1
1.1 Introduction ..........................................................................................................................1
1.2 Physiologic Imaging for Treatment Response Monitoring ..................................................3
1.2.1 Non-MRI-based techniques ........................................................................................3
1.2.2 MRI-based techniques ................................................................................................6
1.3 Chemical Exchange Saturation Transfer MRI .....................................................................9
1.4 Structure of the Thesis .......................................................................................................14
Chapter 2
Chemical Exchange Saturation Transfer MRI to Assess Cell Death in Breast Cancer Xenografts at 7T ........................................................................................................................15
2.1 Introduction .........................................................................................................................15
2.2 Methods ...............................................................................................................................17
2.2.1 Animal Model ...........................................................................................................17
2.2.2 MRI Imaging .............................................................................................................20
2.2.3 Region of Interest Definition ....................................................................................20
2.3 Results .................................................................................................................................23
2.3.1 Data Analysis ............................................................................................................23
2.3.2 Defining MTR Characteristics of Viable Tumour and Cell Death ...........................25
2.3.3 Comparison of Viable Tumour to Cell Death ...........................................................29
v
2.4 Discussion ...........................................................................................................................32
2.5 Conclusions .........................................................................................................................35
Chapter 3
Summary and Future Perspectives .................................................................................................36
3.1 Summary .............................................................................................................................36
3.2 Future Work ........................................................................................................................37
3.3 Conclusions .........................................................................................................................39
References ......................................................................................................................................41
vi
List of Figures and Tables
Chapter 1
Figure 1 Graphical representation of the value of the magnetization
transfer ratio (MTR). 11
Chapter 2
Figure 2 CEST MRI pulse sequence and timing of scans. 18
Figure 3 Representative images of different methods of tumour analysis
employed in the study. 22
Figure 4 CEST spectra. 24
Figure 5 Histograms of MTR values. 26
Figure 6 Differences in cell death regions defined at different MTR cutoffs. 27
Table 1 Cell death index measured by ISEL staining and MTR values. 28
Figure 7 CEST spectra comparison between cell death and viable tumour regions. 30
Figure 8 Change in cell death index by time after chemotherapy administration. 31
1
Chapter 1 Background
1.1 Introduction
As with any medical procedure, treatments for cancer come with associated risks of side effects.
These side effects are caused by the cytotoxicity of modern treatments like radiotherapy and
chemotherapy. Because the primary mode of cell death is through DNA damage, rapidly
dividing tissue is more sensitive to these therapies than is more indolent tissue.1 With careful
dose titration, healthy cells can be allowed to recover while cancerous cells are unable to do so
leading, hopefully, to eradication of the cancer and ultimate cure.
Many cancers, including locally advanced breast cancer and rectal cancer, are now routinely
treated with so-called “neoadjuvant” regimens whereby some treatments (such as chemotherapy
and/or radiotherapy) are given prior to surgery, instead of afterwards as in more traditional
treatment methods. These neoadjuvant treatments may be given to attempt to shrink the tumour
to permit a less extensive, and therefore less debilitating, surgery to be performed. For example,
a combination of neoadjuvant chemotherapy and radiation therapy for rectal cancer increases the
chance that a sphincter-sparing surgery can be performed, which significantly improves patients’
quality of life after treatment.2 Neoadjuvant treatment is also associated with less disease
recurrence compared to administering chemotherapy and radiation post-operatively.3 Similarly,
neoadjuvant chemotherapy given for breast cancer can increase the likelihood that a breast-
conserving surgery can be pursued,4 although survival has not yet been shown to improve with
neoadjuvant chemotherapy compared with post-operative administration.3,5 Methods to predict
the ultimate tumour response to these neoadjuvant therapies early in the treatment course (or
even in advance of beginning treatment) would permit ineffective therapies to be switched for
more effective ones, increasing the chance of cure and avoiding potentially debilitating side
effects with no clinical benefit.
Monitoring or predicting the response of cancer during and after the course of treatment could
also provide prognostic information for expected survival after completion of therapy. Large
2
studies have shown that breast cancer patients who achieve a complete pathological response to
neoadjuvant chemotherapy (defined as no presence of cancer either in the breast or regional
lymph nodes at the time of surgery) experience significantly improved survival compared with
patients who have residual cancer after neoadjuvant chemotherapy.5-7 This association is
considered to be robust enough that ongoing clinical trials are now using complete response as a
surrogate endpoint for survival to shorten the time it takes to obtain and present clinical data. As
well, given good survival outcomes after complete response to chemotherapy, randomized trials
are now investigating whether post-operative radiotherapy can be safely omitted in patients who
achieve a complete response to neoadjuvant chemotherapy.
Standard magnetic resonance imaging (MRI) techniques, such as T1 and T2-weighted sequences,
can measure tumour size and demonstrate macroscopic changes in tissue structure, but they lack
the ability to detect physiologic data or microscopic changes in tumour composition.8
Traditionally, monitoring response of cancer to therapy is done by physical exam and/or
radiographically: the tumour is initially measured prior to treatment, such as via physical
examination, ultrasound, or MRI. The treatment is then administered, which can last up to 4-6
months for modern chemotherapy regimens, with additional time added for the patient to recover
from side effects and to allow for immunologic removal of the dead tissue. Afterward, a repeat
test is performed and the size is compared to measurement obtained before treatment; if the
tumour has shrunk then a response (or non-response) is defined using criteria such as the
Response Evaluation Criteria in Solid Tumours (RECIST).9 Monitoring the patient for onset of
new signs and symptoms may also suggest that disease has spread (for example, new skin
changes or pain). Clearly, many months may elapse before a definite determination of response
can be established using tumour size measurements. If the treatment ends up being ineffective,
the patient may have suffered toxicity for little benefit and tumour proliferation in the meantime
may have reduced or even eliminated the chance of cure.10
Some studies have been carried out involving mid-treatment imaging to detect early changes in
tumour size and adjust treatment according, but anatomic imaging retains the downside of
requiring significant time before adequate treatment assessment can take place.11 One such study
performed serial contrast-enhanced MRI imaging on 33 patients undergoing three cycles of
neoadjuvant chemotherapy before initiation of therapy, after the first cycle and just prior to
3
surgery. That study suggested that a measured reduction of 8.8% or lower after the first cycle of
chemotherapy in the sum of largest diameters of the tumour as measured on MRI could be used
as a cutoff to identify patients who will likely not respond to chemotherapy. However, among the
10 patients in this cohort who had an early size reduction of 8.8% or less, 2 patients eventually
became responders. The fact that contrast enhancement can be seen both in viable tumour and in
post-chemotherapy sclerosis could contribute to the unreliability of anatomical monitoring
methods.12 Response prediction methods that are more reliable and become apparent earlier in
the treatment course are needed.
1.2 Physiologic Imaging for Treatment Response Monitoring
1.2.1 Non-MRI-Based Techniques
Physiologic imaging techniques have been studied in an attempt to detect changes earlier and
predict response to therapy earlier in the treatment course than anatomic methods. One such
modality is positron emission tomography (PET), which is based on the introduction of a
positron-emitting radionuclide into the body. When a positron is emitted, it rapidly annihilates
with a nearby electron, leading to emission of a pair of gamma ray photons in anti-parallel
directions. An array then detects the arrival of these photon pairs and, based on the arrival of
many different photon pairs at the array, computes the location of emission. If the radionuclide is
attached to a molecule with biological activity or which tends to accumulate in a given area of
the body, the flow or distribution of the molecule can be tracked by the PET technique.
The most common PET radiotracer is 18-fluoro-deoxyglucose (18F-FDG), which accumulates in
active tumour through increased metabolism and consequent increased glucose uptake compared
to normal tissue. It is commonly used for staging and restaging solid tumours including
gastrointestinal, gynecological, breast, and head and neck malignancies. PET is highly
reproducible using the standard uptake value (SUV) parameter to accurately assess tumour
glucose uptake.13 Treatment response monitoring using 18F-FDG PET has suggested a decrease
in tumour SUV, which is significantly greater in breast cancer patients who ultimately responded
to chemotherapy than those who did not respond after two courses of chemotherapy14,15 and
4
possibly even as early as the first cycle.13 Response prediction using FDG-PET has been
demonstrated for both local disease (which is potentially curable)16 and metastatic disease.13
Early changes in SUV with FDG-PET correlate with ultimate response as assessed by both
histopathology and conventional size-based measurements using computed tomography (CT)
imaging or MRI.13 Response and survival prediction by changes in FDG-PET uptake have also
been demonstrated for esophageal17 and ovarian cancers.18
While FDG-PET is well established in the diagnosis and staging of cancer, concerns regarding
sensitivity, specificity and image resolution have arisen since treatment-induced inflammatory
responses can cause transient increases in tracer uptake mimicking residual disease. At present,
standard FDG-PET is neither sufficiently sensitive nor does it have sufficient resolution to serve
as a primary assessment modality for malignancies such as breast cancer.19 As a result, alternate
PET methods have been studied to counteract this effect. The cellular proliferation marker 3’-[F-
18]fluoro-3’-deoxythymidine (FLT) accumulates in proliferating cells as it is an analogue of the
DNA nucleoside thymidine. A clinical study of 14 patients with breast cancer (metastatic or
localized) demonstrated changes in FLT uptake that correlated with ultimate response to
treatment measured anatomically via CT.20 However, FLT-PET is still in the developmental
stages and is not widely available.
Ultrasound is a non-invasive imaging technique with a wide range of clinical applications.
Standard B-mode ultrasound imaging is useful for visualizing tumours in the breast, prostate,
liver, and other areas, but image quality is variable between machines and is dependent on
operator skill to obtain high quality images. Tumour measurements and response prediction with
such techniques also suffer from the limitations of other anatomical measurement methods,
namely the delay between tumour response and measureable changes in size. Quantitative
parameters of tumour backscatter properties are independent of the machine operator and
consistent between machines, and so represent a promising treatment prediction modality. In
preclinical models, quantitative ultrasound (QUS) parameters such as average acoustic
concentration (AAC) correlated with cell death after chemotherapy21 while ultrasound
backscatter22 and spectral slope23 have been demonstrated to specifically correlate with
apoptosis. An early clinical study of 24 patients with locally advanced breast cancer using
conventional frequency ultrasound (~7 MHz) used parameters such as mid-band fit and 0-MHz
5
intercept and reported 100% sensitivity and 83% specificity in discriminating responders from
non-responders by the fourth week of neoadjuvant chemotherapy.24 A more recent clinical study
of 100 patients undergoing neoadjuvant chemotherapy for breast cancer demonstrated that QUS
parameters estimating the diameter of acoustic scattering elements within the tumour as well as
their concentration and spacing between them can predict ultimate treatment response to
neoadjuvant chemotherapy with greater than 80% accuracy as early as 1 week into therapy.
Those patients classified as responders on the basis of QUS had significantly improved survival
compared with patients classified as non-responders.25
Although optical imaging is an older concept, having been studied for breast cancer as early as
1929, it has recently been revived as new technological and computing advances have promised
improvements in sensitivity and specificity.26 Near-infrared diffuse optical spectroscopy (IR
DOS) measures near infrared absorption and derives functional images of oxyhemoglobin,
deoxyhemoglobin and water concentration, which correlate in general with levels of
angiogenesis, tumour proliferation, and hypoxia. Optical spectroscopy thus provides potential for
non-invasive functional image without the need for exogenous contrast. While breast cancers
have been shown to contain up to twice the hemoglobin concentration as normal breast tissue,
some benign neoplasms such as fibroadenomas may have similar optical characteristics as
malignant tumours. Small clinical studies have shown significant differences in the concentration
of oxy- and deoxyhemoglobin as well as water between breast cancer patients who respond to
neoadjuvant chemotherapy and those who do not respond. One study of 10 breast cancer patients
demonstrated differences in some optical parameters (scattering power and concentrations of
oxyhemoglobin and deoxyhemoglobin), but not in water content, between responders and non-
responders 4 weeks after chemotherapy.27 Another study of 11 patients with breast cancer
suggested that DOS can predict eventual response to neoadjuvant chemotherapy with 100%
specificity and sensitivity only 1 week after treatment start.28
6
1.2.2 MRI-Based Techniques
Morphological MRI (such as T1 and T2-weighted imaging) can show tumour size and
macroscopic tumour characteristics.8 In a comparison with clinical examination, mammography,
and ultrasound, morphological MRI has shown better agreement with pathological
response.29 However, morphological MRI may tend to overestimate that response, as shown in
one study in which MRI-based response assessment was discordant with all pathologic non-
responders while it showing better agreement with patients who did respond to treatment.30
Magnetic resonance spectroscopy (MRS) can help diagnose breast cancer based on the detection
of choline-containing compounds, which may correlate with active cell replication. Several small
prospective trials have assessed MRS as a predictive marker for tumour response. In one trial of
20 women receiving doxorubicin-based neoadjuvant chemotherapy, patients underwent an MRS
scan before the start of treatment as well as after 1-2 cycles and after 4 cycles of neoadjuvant
chemotherapy. Patients were then categorized as responders or non-responders based on change
in tumour size as assessed by gadolinium-enhanced MRI scanning. At the first follow-up (after
1-2 cycles of chemotherapy), significant reductions in both tumour size and the amount of MRS-
detected choline-containing compounds were seen for patients characterized as responders to
therapy while no significant change was seen for the non-responder group. This study also
concluded that metabolic changes, as measured by MRS, were greater than changes in
morphological size, suggesting that MRS-detected changes may become evident sooner,
allowing for earlier prediction of ultimate response using these techniques.31 Another study of 16
women with breast cancer suggested that MRS may be able to detect significant changes and
predict ultimate response as early as 24 h after chemotherapy administration.32 Pre-clinical work
has suggested that MRS can be used to monitor changes in tumour lactate levels which can act as
an early indicator of response.33 Despite this early promise, the applicability of MRS to clinical
situations is potentially limited, however, by poor spatial resolution and limited availability and
expertise in the technology by clinical radiologists.11,34,35
Dynamic contrast-enhanced (DCE) MRI can produce high-resolution images depicting perfusion
and the permeability of capillaries within a tumour.36 DCE-MRI can differentiate between non-
vascularized fibrosis and viable tumour.37,38 DCE-MRI parameters have also been shown to be
7
superior predictors of breast cancer response to chemotherapy than tumour size.39 One study of
thirty women undergoing neoadjuvant chemotherapy for breast cancer found that DCE-MRI
assessment after two cycles of chemotherapy yielded 93% accuracy in predicting patients who
would achieve pathological complete response.36 Another study found 94% sensitivity and 82%
specificity with DCE-MRI when assessed after two cycles of chemotherapy. Tumour size
measured by MRI did not predict response in this study.40 A larger study of 188 breast cancer
patients investigated whether response prediction using DCE-MRI would vary based on the
subtype of breast cancer as defined by hormone and HER2-neu receptor status. This study found
that DCE-MRI was an effective predictor of ultimate response for hormone receptor negative
breast cancer (HER2-neu positive or negative), but not for hormone receptor positive cancers.41
Notable limitations of DCE-MRI include the need for injected contrast, delays of up to several
weeks following chemotherapy to allow sufficient changes to be seen31 and the relative difficulty
with image analysis which makes it not well-suited for routine clinical use.11 Studies have shown
variable sensitivity and specificity of DCE-MRI for evaluation of cancer, questioning the
reliability of the technique in applications like breast and prostate cancer.42,43
Cost and difficulty with administration of contrast agents has led to discovery of endogenous
contrast mechanisms capable of creating high quality images without the need for injection of
exogenous contrast media. Diffusion-weighted MRI (DW-MRI) measures change in the
Brownian motion of water within tumour tissue. Since tumours cells restrict diffusion of fluid in
tissue, a decrease in cellularity (such as caused by cytotoxic therapy like chemotherapy or
radiation) should lead to an increase in diffusion, which can be represented by the apparent
diffusion coefficient (ADC).10 Thus, ADC values increase after cytotoxic therapy. Study of ADC
in liver metastases after chemotherapy treatment suggests that ADC values in tumours that
ultimately responded to treatment significantly differ from baseline as early as 4 or 11 days after
treatment (depending on the analysis used), while those that do not respond never show a
difference compared to baseline values.10 Other studies have confirmed that ADC changes
measured as early as after the first chemotherapy cycle can predict ultimate tumour response in
breast44,45 and rectal cancer.46
8
However, ADC is sensitive to different types of tissue changes, not only those related to cell
death.47 Edema or inflammation, which may be induced in normal tissues by cancer treatments,
also affect ADC, limiting its utility as a cancer treatment monitoring technique. As well, many
new molecular targeted agents such as tyrosine kinase inhibitors and angiogenesis inhibitors are
primarily cytostatic in action, arrest cell development and proliferation rather than directly
killing cells.48 This lack of cytotoxicity and subsequent decrease in overall cellularity may inhibit
the ability of diffusion-weighted imaging to predict response to such new treatment methods.45
Paradoxical findings of decreased ADC in rectal tumours that respond to chemotherapy have
also been reported,46,49,50 perhaps suggesting a lack of consistency in response between tumour
sites or even individual tumours.
As described above, significant effort has been made to develop imaging-based monitoring
techniques for tumour response. While early trials of several such modalities have shown
promise, most published trials are based on small samples of tumours and questions of
sensitivity, specificity, inter-rater reliability, and spatial resolution remain. Other drawbacks
include the requirement for injectable contrast media before each scan, which is expensive, can
be associated with potentially severe reactions, and requires additional time from the patient and
staff. As well, standard T1 and T2-weighted techniques lack the ability to show microscopic or
functional changes in the tumour microenvironment8 and optimal tumour visualization with T1-
weighted imaging requires injection of gadolinium-based contrast agents, increasing costs and
requiring clinical monitoring for sensitivity reactions. However, early tumour changes to therapy,
including cellular swelling and blood flow restrictions, may cause transient increases in ADC
values on DWI, limiting the sensitivity of DWI to predict ultimate tumour response early in a
course of treatment.51 A modality which does not require contrast injection but is sensitive and
specific to early cellular changes predictive of cell death across a range of different cancer
histologies would offer significant advantages for treatment response monitoring.
9
1.3 Chemical Exchange Saturation Transfer MRI
Chemical exchange saturation transfer (CEST) MRI is a novel metabolic imaging technique that
reflects changes in the tumour microenvironment, such as those brought about by cell death52,53
(apoptosis or necrosis) or changes in cellular metabolism.54-56 It thus presents a promising
method for detecting tumour growth and predicting response to treatment.52 The CEST effect
derives from mobile hydrogen nuclei (commonly referred to as protons) in solute molecules
exchanging rapidly (correlation time tc ~ 10-10 s)57 with those in water molecules within a
solution either through direct exchange or by the nuclear Overhauser effect (NOE).52 For in vivo
systems such as tumours, the main solutes are proteins, which present a variety of chemical
microenvironments for the protons, such as amide, amine, or aliphatic groups among others.54,58
However, because water represents the vast majority of the molecules present in a tumour, most
of the protons that contribute to this signal are present on water molecules. When placed in an
external magnetic field (such as is present in an MRI scanner), a net magnetization is created in
these nuclei.
This contrast mechanism is intrinsic to the cell, and thus no injectable contrast medium is
required, reducing costs and eliminating the requirement of medical oversight of an injection
process and possible allergic reactions. CEST is also a promising modality for cell death
detection because it is sensitive to small changes in metabolite concentration, as the exchange of
protons between water and dissolved metabolites creates a large pool of protons that amplify the
CEST effect.59
When a pulse of radiofrequency (RF) energy is introduced to the system, protons whose resonant
frequency matches that of the RF pulse will absorb that energy (termed “saturation”), resulting in
attenuation of the net magnetization signal. The solute protons can exchange with those in water
multiple times during saturation, so a large pool is created of protons which retain the magnetic
properties they possessed when bound to protein and which amplifies the contributions from the
solute molecules to the net magnetization. If no protons resonate at the frequency of the RF
pulse, no energy is absorbed and the measured magnetization remains unchanged; this is defined
as the reference signal strength S0.52
10
The magnetization transfer ratio (MTR) is a simple metric which combines the contributions
from all contrast mechanisms, including CEST and the magnetization transfer contrast (MTC)
phenomenon, that can be manipulated by way of radiofrequency (RF) energy absorption and
saturation effect. The MTR is defined as
1
where S is the measured strength of the MRI signal at a specific frequency offset of RF
saturation and S0 is the strength of the MRI signal when no RF saturation is applied.60 The ratio
is called the “normalized signal” value. By varying the RF pulse frequency and recording the
net magnetization as a function of frequency, a spectrum known as the CEST spectrum (also
known as a Z-spectrum) is generated. Changes in relative concentrations of different solutes and
chemical groups, including contribution from large semisolid macromolecules from MTC,60 can
then be characterized and used to differentiate different tissue types.61 Figure 1 shows a graphical
representation of the MTR measurement on a sample Z-spectrum.
The RF absorption and subsequent signal attenuation creates negative peaks in the Z-spectrum
which are characteristic of the different absorption frequencies of chemical groups as well as a
large peak centered around the resonant frequency of water. The central frequency of a
Z-spectrum is the resonant frequency of bulk water, given by the formula
γ2π
where f0 is the resonant frequency, γ is the gyromagnetic ratio, and B0 is the magnetic field
strength. However, the resonant frequency of a given proton species, measured in hertz, is
dependent on the magnetic field strength of the MRI scanner being used.
11
Figure 1: Graphical representation of the value of the magnetization transfer ratio (MTR).
This figure shows the MTR value at the 2 ppm offset of a sample Z-spectrum.
Resonant frequencies are given in units of parts per million (ppm), which is independent of MRI
field strength, and given by the formula
∆ ,ref
,refx10
where ∆is the frequency offset (in ppm), f0 is the resonant frequency of interest, and f0,ref is the
resonant frequency of a reference compound. Contrary to MRS (which uses tetramethylsilane as
the reference), the convention in CEST analysis is to use water as the reference, so the frequency
offset of water is defined as 0 ppm.
In cancerous tissue, the main contributors to the observed peaks within the Z-spectrum are from
proteins. While CEST is sensitive to protein content and tissue pH,56 previous research suggests
negligible change in tumour pH after treatment, so changes in the Z-spectrum should reflect
changing protein concentration and thus be predictive of ultimate treatment response.8
12
Specifically, the main contributors are protons found in amine groups in guanidine groups (e.g.,
in creatine)62 which resonate at a frequency offset around 2 ppm, amide groups from the peptide
backbone (3.5 ppm), and aliphatic groups in both the side chains and backbone (-3.5 ppm).52,63
The CEST imaging technique has many of the characteristics of an ideal treatment response
mechanism. First, because of the signal amplification effect due to the transfer of protons from
water to solute, CEST MRI should be extremely sensitive to small changes in tumour
composition, allowing for very early detection of cell death and response to treatment. Second,
the quantitative nature of CEST MRI analysis should make analysis mostly independent of the
user and operator skill. Third, because it is MRI-based, the lack of ionizing radiation eliminates
the risk of inducing second cancers as seen with CT imaging. CEST does not require injection of
exogenous contrast, obviating the risk of allergic reactions and the need to monitor for them.
And finally, standard assessment and treatment algorithms for many cancers (such as breast,
rectal, head and neck, cervix, and prostate) already involve MRI so adding CEST imaging to
these pathways is relatively simple and would not significantly increase costs or time.
Clinically, CEST MRI can differentiate between regions of viable tumour and radiation necrosis
within the brain.53,63 In a preclinical model, Sagiyama et al.64 used amide proton transfer (APT)
imaging, which is a CEST-based imaging protocol specifically focused on the amide region of
the Z-spectrum (3.5 ppm). After administration of temozolomide (TMZ), a standard alkylating
chemotherapy agent used to treat glioblastoma multiforme (GBM), to mice with tumours grown
using a human GBM cell line, the strength of the APT signal was found to decrease after one
dose of TMZ while it continued to increase in untreated controls. This study suggested that APT
(and other chemical exchange-based imaging techniques) can be sensitive to changes in tumours
induced by chemotherapy. In another pre-clinical study, Zhou et al.63 used amide proton transfer
(APT) imaging, which is a CEST-based imaging protocol specifically focused on the amide
region of the Z-spectrum (3.5 ppm) to study human GBM xenografts. This study demonstrated
large differences in APT signal between areas of viable tumour identified by conventional T1-
weighted contrast-enhanced MRI imaging and both radiation-induced brain necrosis and normal
brain tissue. The difference in APT signal between normal brain tissue and radiation necrosis
was much smaller than for tumour-necrosis or tumour-brain signals, but this difference was still
statistically significant. APT signal changes were observed early after radiation therapy (3 and 6
13
days after treatment) while other imaging techniques like T1, T2, and DW-MRI showed no
change at these time points.
In a clinical study, Mehrabian et al.53 studied human patients with brain metastases from a
variety of primary cancers (6 breast, 5 lung, 3 kidney, 2 melanoma) previously treated with
stereotactic radiosurgery. As defined by the mean MTR values, the largest separation between
CEST signals of regions of necrosis and regions of tumour progression were seen in the amide
and NOE (aliphatic) regions of the CEST spectra, suggesting that CEST can differentiate
between tumour and necrosis in the setting of brain metastases. CEST has also been studied as a
method to predict response of brain metastases undergoing stereotactic radiosurgery treatment.65
In this study, changes in CEST parameters within the tumour in the aliphatic and amine regions
of the Z-spectrum at 1 week after treatment correlated with ultimate change in tumour size at 1
month after treatment. CEST can also differentiate progression of glioma from
pseudoprogression, a benign phenomenon which mimics the MRI characteristics of glioma
progression after chemoradiotherapy. 66
A preliminary clinical study of 3 patients with locally advanced breast cancer has shown an
increase in APT signal in a patient whose cancer did not response to neoadjuvant chemotherapy
(as demonstrated by ultimate growth in tumour size), while the APT signal decreased in a patient
who ultimately developed a complete tumour response.8 A preliminary report of CEST MRI for
breast cancer patients67 compared CEST results at 1.2 – 1.8 ppm with results from DCE-MRI. In
3 of the 6 patients in this cohort, high CEST signal correlated well with tumour identified using
DCE-MRI and CEST signal values were higher in tumour than in surrounding fibroglandular
tissue.
These preclinical and clinical studies suggest that CEST can be used to detect changes induced
by chemotherapy and predict ultimate response to therapy. However, much of this previous work
has focused on the CEST characteristics of exchanging protons from protein amide groups (i.e.,
APT imaging), although differences may be evident in other regions of the Z-spectrum.
Building on this work, differences in Z-spectrum characteristics may be evident between viable
tumour tissue and areas of cell death. Differences between tumour and other tissue types should
14
also be detectable for other cancer types, such as breast cancer. The work described here sought
to study the CEST characteristics of breast cancer xenografts and determine if CEST (either in
the APT region or elsewhere in the Z-spectrum) can differentiate between viable tumour and
necrosis in a murine breast cancer model. Such differences could help form the basis for future
CEST-based cell death detection and treatment response monitoring protocols for breast cancer.
1.4 Structure of the Thesis
The purpose of the work described in this thesis is to characterize the CEST parameters of breast
cancer xenografts. In particular, the goal is to define differences in the CEST spectra of viable
tumour tissue and regions of cell death. Such differences could be studied to develop methods for
distinguishing viable and dead tissue and to detect onset of cell death early in the treatment
course in an attempt to predict ultimate treatment response and guide clinical decision making.
Chapter 2 describes the main experiment dedicated to studying CEST parameters of viable
tumour tissue and necrosis within breast cancer xenografts and to developing a CEST-based
method to differentiate between the two. Groups of xenografts were scanned with a CEST
protocol before chemotherapy and at a variety of times afterwards (ranging from 4 to 24 h). The
CEST data were then compared with ISEL stained histology specimens to determine appropriate
MTR values to differentiate cell death from viable tumour and to study the time course of cell
death response to chemotherapy.
Although previous experiments have studied CEST parameters after radiotherapy treatments,
chemotherapy represents the most common treatment modality given for breast cancer prior to
surgery. Detecting response (or non-response) to neoadjuvant chemotherapy would provide the
most relevant information for prognosis and treatment selection in the locally advanced breast
cancer setting. For this reason, CEST parameters before and after chemotherapy administration
were studied in the experiments described here. Chapter 3 summarizes the relevant findings of
the study and suggests future work that can expand on the results.
15
Chapter 2 Chemical Exchange Saturation Transfer MRI to Assess Cell
Death in Breast Cancer Xenografts at 7T
2.1 Introduction
Locally advanced breast cancer, generally defined as tumours which are i) larger than 5 cm, ii)
invading skin or the chest wall, or iii) have matted or multiple involved axillary lymph nodes, is
an aggressive form of cancer associated with poor survival.68 Modern treatment approaches
increasingly use chemotherapy before surgery (“neoadjuvant chemotherapy”) followed by
surgery and then radiotherapy.69 Although chemotherapy can shrink or even eliminate cancer
within breast and lymph nodes on microscopic examination, recurrence is common even after
maximal treatment. However, the degree of response to chemotherapy has been shown to
correlate with survival outcomes.5
Standard assessment of tumour response to therapy involves anatomical measurements of tumour
size before and after therapy, typically using magnetic resonance imaging (MRI) or ultrasound.
Unfortunately, the cancers of some patients do not respond well to chemotherapy, which may
lead to 4-6 months of ineffective treatment associated with potentially significant side effects
such as alopecia, nausea/vomiting, long-term cardiac toxicity, and weakness or numbness in the
hands and feet.70-72 A tumour may also grow and/or metastasize during this time, decreasing or
even eliminating the chance of cure. A method to detect response to chemotherapy either prior to
treatment or early in a course of treatment could potentially improve outcomes for these patients
while sparing side effects during ineffective therapy. One way to predict such a response would
be to detect cell death, such as through apoptosis or necrosis, quickly after administration of
cytotoxic therapy. Those tumours whose cells are not killed by cytotoxic treatment may require
additional or alternate treatments to achieve adequate cytotoxicity. Similarly, cell death
processes such as apoptosis or necrosis may not be the only physiological responses undergone
by a tumour receiving cytotoxic therapy. The totality of these changes and the subsequent
changes in the chemical microenvironment within the tumour may be detectable by functional
16
imaging techniques, and such changes can be correlated with ultimate tumour response and
patient survival.
As previously described, various imaging techniques have been studied for their ability to
identify changes in tumours early during treatment and to predict tumour response and patient
outcomes. However, the various limitations to these methods suggests that a superior alternative
would be beneficial. CEST MRI possesses many characteristics of such an alternative. Previous
work has demonstrated that it can predict tumour response and differentiate between regions of
active tumour and apoptotic and necrotic tissue in lung cancer xenografts. CEST has also been
shown to successfully evaluate treatment in patients with brain metastases and differentiate
between tumour progression and radiation-induced necrosis.53
Because of both the limitations of other imaging techniques for clinical response monitoring and
the promise of CEST MRI imaging for these purposes, a proof-of-principle study was conducted
to study the CEST properties of MDA breast cancer xenografts both before and after treatment
with chemotherapy. Next, the CEST properties of a larger sample of breast cancer xenografts
were characterized using MTR to determine methods for differentiating viable tumour tissue and
cell death. These methods were then used to study the time dependence of the tumour response
to chemotherapy.
17
2.2 Methods
2.2.1 Animal Model
Tumours were grown by injecting 100 μL of solution containing up to 5 × 106 MDA-MB-231
tumour cells (American Type Culture Collection, Manassas, VA; henceforth referred to as
“MDA”) into the hind legs of CB-17 SCID mice (Charles River Laboratories, Canada, Saint-
Constant, QC).
Animal care protocols were approved by the local Animal Care Committee at Sunnybrook
Research Institute. Mice were anesthetized during scanning by inducing anesthesia with 3-4%
isoflurane. Thereafter, respiratory rate was monitored by a pneumatic pillow. Isoflurane
concentration was titrated to maintain a breathing rate of 60-90 breaths per minute; generally, 1-
2% isoflurane concentration was sufficient to maintain this rate. Temperature was monitored
with a probe placed in constant contact with the skin of the mouse’s stomach. Constant external
temperature was maintained using a warm water circulating bath on the ventral surface of the
mouse’s thorax and abdomen.
Due to the known propensity for MDA xenografts developing necrotic cores,27 tumours were
scanned when they reached approximately 5 mm in diameter as evaluated by measuring the
visible tumour using calipers. Doxorubicin (50 mg/m2) and paclitaxel (100 mg/m2)
chemotherapy was used, as these drugs form the basis of standard, modern, clinical neoadjuvant
chemotherapy regimens. The chemotherapy was administered via tail vein catheter immediately
after completion of each pre-chemotherapy scan. Tumours were rescanned at a pre-determined
time (4, 8, 12, or 24 h) after chemotherapy injection. Scans were timed such that the desired time
point (e.g. 4 hours after chemotherapy injection) occurred at some point during the CEST scan,
as illustrated in Figure 2.
18
Figure 2: CEST MRI pulse sequence and timing of scans.
A) CEST MRI pulse sequence. The line labelled “RF” shows radiofrequency pulse application.
The lines labelled Gslice, Gphase, and Gread show the imaging gradients.
Adapted from: K. L. Desmond, “Endogenous Chemical Exchange Saturation Transfer:
Quantitative Modelling and Application in Cancer.”
B) Timing of scans. Each tumour was scanned before chemotherapy, and then either at 4, 8, 12,
or 24 h after chemotherapy injection. The scan was arranged such that the appropriate time after
injection occurred between the beginning of the first CEST scan and the end of the second CEST
scan.
19
Immediately after completion of the post-chemotherapy scan, animals were sacrificed under
anesthesia by cervical dislocation. Tumours were excised, leaving the skin overlying the tumour
and a layer of muscle underneath the tumour intact. The tumour was cut in half at the point of
largest diameter. The proximal portion of the tumour tissue was fixed in 10% formalin and then
transferred to a solution of 70-80% ethanol for storage until processing; the distal portion was
frozen in liquid nitrogen and stored in a -80°C freezer for future retrieval. The proximal portion
of the tumour was sectioned into 5 μm slices and the largest (closest to the cut surface) slices
were stained with hematoxylin and eosin (H&E) for morphological identification and in situ end
labeling (ISEL) assay for identification of cell death (apoptosis and necrosis). As apoptosis and
necrosis both stain positively (dark purple) using ISEL, this report will refer to regions stained by
ISEL as regions of “cell death.”73-75
High-magnification images of the ISEL-stained histology slides were obtained using a Leica
DC100 microscope with a 40x objective and a Leica DC100 camera connected to a 2-GHx PC
running Leica IM1000 software (Leica GmbH, Wetzlar, Germany). Using the ImageJ program
(National Institutes of Health, Bethesda, MD), the entire tumour area and the region of cell death
staining using ISEL (dark purple staining) were delineated and the cell death index (CDI) was
calculated by
Where Acell death is the area of the cell death region and Atumour is the area of the entire tumour.
20
2.2.2 MRI
Animals were imaged before and after chemotherapy injection on a 7T preclinical MRI system
(BioSpec 70/30 USR, Bruker BioSpin, Billerica, MA). A volume coil was used for transmission
and a 20-mm diameter surface coil was used for reception. The tumours were positioned at the
isocenter of the magnet for optimal shimming. A high-resolution, T2-weighted Rapid
Acquisition with Relaxation Enhancement (RARE) image58 (RARE factor 8, TR/TE = 2500/50
ms) was acquired with 11 slices and the tumour volume identified to perform field map-based
shimming using Bruker’s Map Shim functionality. A correction to account for spatial
inhomogeneity in the B0 field was also performed.31
The MRI sequence used began with a single rectangular off-resonance RF pulse of 490 ms which
was followed by a single slice 2D FLASH sequence with TR/TE = 501/3.1 ms at a resolution of
0.31 mm × 0.31 mm × 1 mm and a matrix size of 64 × 64. Saturation pulse amplitudes of 0.5 µT
was used. Measurements were made at frequency offsets between -1800 Hz (-6 ppm) and
1800 Hz (6 ppm) in increments of 30 Hz between -180 Hz (-0.6 ppm) and 180 Hz (0.6 ppm) and
increments of 90 Hz outside this region. Reference images at 200 kHz offset were interleaved
every 5 offsets throughout the acquisition to correct for signal drift. While previous signal drift
reports showed exponential decay of the reference signal over time,31 our decay showed linear
characteristics, which were used for the correction methods.
2.2.3 Region of Interest Definition
To define the regions of interest for analysis, the structural and CEST images were co-registered.
An area encompassing the tumour, as visualized on the structural image, was manually
delineated on the CEST image. An example is shown in Figure 3, including the corresponding
H&E and ISEL stained histology slides. This area (the “mask”) was intentionally drawn
conservatively to ensure that the mask remained within the tumour over the entirety of the scan,
accounting for small amounts of motion over the length of the scan.
21
The MTR was then calculated for each voxel within the masks for a given frequency offset using
the formula
1
where S is the strength of the MRI signal measured at a given frequency offset of RF saturation
and S0 is the strength of the MRI signal when no RF saturation is applied.60 The voxels were then
assigned as cell death or tumour based on the MTR. Once the mask was defined, a histogram
was created by assigning each voxel into bins by MTR at a given frequency offset. The
histogram was fit to a Gaussian distribution to define cutoffs to segment the masks in viable
tumour and necrotic/apoptotic tissue.
Using these masks, the cell death index (CDI) was calculated for each tumour by
Where Nbelow is the number of voxels with MTR below the cutoff (indicating the presence of cell
death) and Ntotal is the total number of voxels within the mask encompassing the tumour.
The statistical significance between differences in MTR was tested using paired t-test to compare
pre- to post-chemotherapy scans and using unpaired t-test to compare viable tumour to cell
death.
22
Figure 3: Representative images of different methods of tumour analysis employed in our study.
A) T2-weighted “structural” MRI image.
B) CEST MRI image divided into pixels for analysis. Overlaid in orange is the mask defining the
region of interest for CEST analysis.
C) ISEL stained histology slide: blue indicates viable tumour, purple indicates cell death.
D) H&E stained histology slide.
E) Map of MTR for each mask pixel at 1.8 ppm frequency offset.
All scale bars indicate 1 mm.
23
2.3 Results
2.3.1 Data Analysis
Fourteen xenografts were scanned. All tumours were scanned before chemotherapy was
administered. Three were scanned at 4 h after chemotherapy, four at 8 h, four at 12 h, and three
at 24 h. A fourth tumour was scanned in each of the 4 and 24 h groups, but the images were
discarded due to extensive motion during the scan. All fourteen xenografts had histology
preparation and staining after the post-chemotherapy scan.
Two initial analyses were performed. For the first, three tumours with identifiable necrotic cores
were chosen based on visual assessment of the structural T2-weighted images. Masks were then
created to estimate the areas of viable tumour and cell death. The mean Z-spectrum of the three
viable tumour regions was compared to that of the three cell death regions, as shown in
Figure 4A. Although large separations between the spectra were seen at 1.8, 0.6, -0.5,
and -3.3 ppm, only the difference at 1.8 ppm was statistically significant in this analysis (p =
0.03).
The second initial analysis examined the CEST spectra of the entire xenograft region, making no
attempt to differentiate between viable tumour and cell death regions. For this analysis, masks
were created encompassing the entire xenograft (i.e., both areas of viable tumour and regions of
cell death) based on visual analysis of the structural T2-weighted images. The mean CEST
spectra of all pre-chemotherapy scans were then compared to the post-chemotherapy scans. As
seen in Figure 4B, the difference in MTR values between these two groups were much smaller in
magnitude than the differences between the areas of viable tumour and cell death compared in
Figure 4A. The difference at -3.3 ppm did reach statistical significance (p = 0.035), while
differences at other offsets such as 1.8, 0.6 and -0.5 ppm did not (p > 0.05).
24
Figure 4: CEST spectra.
A) CEST spectra (solid lines) averaged over the regions of viable tumour (blue) and cell death
(red) as defined by visual assessment of the T2 structural images with co-registration of the
CEST data. Dashed lines indicate standard deviations.
B) CEST spectra (solid lines) averaged over the entire region of interest mask for all pre-
chemotherapy scans (green) and post-chemotherapy scans (black). Dashed lines indicate
standard deviations.
25
2.3.2 Defining MTR Characteristics of Viable Tumour and Cell Death
Based on the above results, analysis of CEST characteristics was directed toward the 1.8
and -3.3 ppm frequency offsets. Using the initial masks, which encompassed the entire area of
the tumour, including any areas of cell death, the MTR was calculated for each voxel in each
scan. At 1.8 ppm, the MTR for all voxels ranged from 0.076 to 0.24. At -3.3 ppm, the MTR
ranged from 0.077 to 0.23. Histograms of voxel MTR values are presented in the top row of
Figure 5A and 5B.
Cutoffs to label tumour and viable tissue based on the MTR were then determined. The bottom
row of Figure 5 shows scatter plots of the histogram data at 1.8 ppm (Figure 5A) and -3.3 ppm
(Figure 5B) offset with several candidate tumour-cell death cutoffs defined: the mean of the
distribution (labelled in purple), 1 standard deviation below the mean (1 SD; orange) and 0.5
standard deviations below the mean (0.5 SD; green).
Figure 6A-C shows an example of the tumour and cell death mask areas using the three different
cutoffs (with MTR values measured at 1.8 ppm) compared with the structural T2-weighted MR
image and ISEL stained histology slide for the same tumour. Figure 6D shows a T2-weighted
structural image is seen with the corresponding tumour mask and ISEL stained histology slide.
Visual comparison suggests that the heterogeneously-enhancing core of the tumour on the T2
image correlates with the shape and size of the dark-stained cell death region of the ISEL
histology slide. Similarly, the shape and size of the cell death region seen using T2-weighted
imaging and ISEL more closely matches the cell death region determined using the 0.5SD cut-
off (Figure 6B) than the 1SD cut-off (which tended to underestimate the amount of necrosis;
Figure 6A) or the mean value cut-off (which tended to overestimate the amount of necrosis;
Figure 6C) in this example.
26
Figure 5: Histograms of MTR values.
Histograms of MTR for each pixel from all scans combined (pre- and post-chemotherapy). MTR
are counted in bins of 0.005, for 40 bins in total ranging from 0.05 to 0.25.
A) Histograms generated at 1.8 ppm.
Top: Bar graph showing histogram data.
Bottom: Scatter plot of same data as top with Gaussian curve fit to data (red). Vertical lines
indicate the mean (purple; MTR = 0.14), 0.5 standard deviations below the mean (green; MTR =
0.12) and 1 standard deviation below the mean (yellow; MTR = 0.10).
B) As in (A) but generated at -3.3 ppm.
Top: In the bottom subplot, mean (purple) has MTR = 0.10, 0.5 standard deviations below the
mean (green) has MTR = 0.12, and 1 standard deviation below the mean (orange) has MTR =
0.15.
27
Figure 6: Differences in cell death regions defined at different MTR cutoffs.
Example of definitions of viable tumour (orange) and cell death (yellow) regions using different
candidate MTR cutoffs:
A) 1 standard deviation below the mean (MTR = 0.10).
B) 0.5 standard deviations below the mean (MTR = 0.12).
C) Mean (MTR = 0.14).
D) From left to right: the T2-weighted structural image, pixelated CEST image with mask region
overlaid in orange, and ISEL stained histology images for reference.
All scale bars indicated 1 mm.
28
To validate this qualitative observation, the CDI that was measured using the ISEL stained
images was compared with the CDI using different MTR cutoffs. For each time point, CDI was
calculated by
| |
where is the mean CDI calculated using ISEL for all post-chemotherapy xenografts at a
given time point and is the mean CDI calculated using MTR. The sum of all
calculations (CDI Total) were then compared to determined which MTR cutoff most closely
agreed with the CDI measured using ISEL (lower CDI Total suggests better agreement). The CDI
using the 0.5 SD cutoff for both the 1.8 ppm and -3.3 ppm offset most closely agreed with the
CDI measured using ISEL, as shown in Table 1.
Table 1: Cell death index measured by ISEL staining and MTR values.
1.8 ppm cutoff -3.3 ppm cutoff
Chemo
time
ISEL Histogram
mean
0.5 SD 1 SD 0.5 SD 1 SD
4h 0% 32.6% 5.1% 0% 2.1% 0%
8h 12.7% 37.7% 10.2% 1.3% 13.7% 7.4%
12h 20.5% 42.7% 13.6% 5.7% 11.1% 5.2%
24h 29.5% 47.7% 19.1% 2.7% 23.5% 13.5%
CDI Total 98.2 24.8 52.9 18.5 36.5
Percent values represent the mean CDI of all post-chemotherapy xenografts from a given time
point.
29
2.3.3 Comparison of Viable Tumour to Cell Death
The difference in CEST parameters amongst tumours that had identifiable cell death was then
examined. Masks defining regions of viable tumour and cell death were created using the MTR
map at 1.8 ppm (Figure 7A) and -3.3 ppm (Figure 7B); the cutoff between tumour and cell death
used to define these regions was set at the 0.5 SD cutoff for each offset (MTR = 0.12 at 1.8 ppm,
MTR = 0.125 at -3.3 ppm).
The spectra for tumour and cell death regions are shown in Figure 7A using masks generated at
1.8 ppm and in Figure 7B using masks generated at -3.3 ppm. Regardless of which offset was
used to define the masks, the maximum separation between the curves outside of the direct effect
region was observed at 1.8 and -3.3 ppm. The mean MTR of the masks for each individual
xenograft are shown in Figure 7C (using MTR at 1.8 ppm to define the masks) and 7D (using
MTR at -3.3 ppm to define the masks). The differences in MTR were statistically significant for
all shown cases (p ≤ 0.001).
Figure 8 shows the mean change in measured CDI as a function of time after chemotherapy.
Although no differences between experimental times reached statistical significance, a trend is
evident with the maximum cytotoxic effect at 8-12 h after chemotherapy administration.
30
Figure 7: CEST spectra comparison between cell death and viable tumour regions.
A) CEST spectra (solid lines) averaged over the regions of viable tumour (blue) and cell death
regions (red) as defined by the MTR for each voxel at 1.8 ppm using MTR = 0.12 (0.5 standard
deviations below the mean of the calculated histogram) as the cutoff. Dashed lines indicate
standard deviations.
B) Mean MTR of the masks for each individual xenograft used in Section A. The tumour and
cell death masks are differentiated using MTR = 0.12. The MTR difference between the masks at
the 1.8 ppm and -3.3 ppm cutoffs are both statistically significant using this cutoff (p ≤ 0.001).
C) CEST spectra (solid lines) averaged over the regions of viable tumour (blue) and cell death
regions (red) as defined by the MTR for each voxel at -3.3 ppm using MTR = 0.125 (0.5
standard deviations below the mean of the calculated histogram) as the cutoff. Dashed lines
indicate standard deviations.
D) Mean MTR of the masks for each individual xenograft used in Section C. The tumour and
cell death masks are differentiated using MTR = 0.125. The MTR difference between the masks
at the 1.8 ppm and -3.3 ppm cutoffs are both statistically significant using this cutoff (p ≤ 0.001).
T = viable tumour regions; CD = cell death regions
31
Figure 8: Change in cell death index (CDI) by time after chemotherapy administration
Average change in cell death index from pre- to post-chemotherapy scans as defined at 1.8 ppm
frequency offset using the MTR = 0.12 cutoff for viable versus dead tumour.
Error bars denote standard error of the mean. The differences between groups did not reach
statistical significance.
32
2.4 Discussion
This study investigated methods for differentiating viable tumour from tumour regions
containing cell death using CEST MRI. Statistically significant differences in MTR were
identified at 1.8 and -3.3 ppm between regions of viable and dead tissue. An MTR cutoff of 0.12
at 1.8 ppm or 0.125 at -3.3 ppm most closely approximated the cell death pattern shown by
histological assessment. Using this cutoff to determine CDI, a maximum increase in cell death
was observed between 8-12 h after chemotherapy, after which the CDI diminished. We have here
confirmed the previous findings of Desmond et al. (which used a small sample size of MDA
tumours) that MTR analysis can differentiate viable tumour from cell death in this cell line.
Previous pre-clinical research has demonstrated that CEST MRI can be used to differentiate
between tissue types, including differentiating between muscle and tumour, different tumour cell
lines, and between viable tumour and cell death. In a sample of 20 Lewis lung carcinoma (LLC)
xenografts and four MDA breast cancer xenografts, Desmond et al.52 have studied a variety of
MRI parameters, including T1 and T2 relaxation; diffusion (ADC); and CEST parameters such
as MTR and Lorentzian curve peak amplitudes corresponding to amide, amine, and aliphatic
groups within CEST spectra. These MTR analyses were focused on 3.5 ppm to maximize the
contribution of amide protons. Those results indicated that differentiation between viable tumour
and necrotic tissue for both MDA breast cancer and LLC lung cancer xenografts could be
obtained by measuring the amplitude of Lorentzian peaks fitted to the Z-spectrum centered on
the resonance frequencies of amide (3.5 ppm), amine (2 ppm) and aliphatic (-3 ppm) protons.
MTR at 5 ppm was the only other metric that could statistically significantly differentiate
between the two tissue types.
Zhou et al.63 compared APT imaging with anatomical (T1 and T2-weighted) and DW-MRI after
treating human GBM xenografts with radiotherapy. Changes in APT signal were observed at 3
and 6 days after treatment, while the other techniques showed no change at these times points.
As well, APT was able to differentiate between radiation necrosis and both glioma and
gliosarcoma xenografts; neither gadolinium-enhanced T1 nor T2-weighted imaging could
differentiate glioma from radiation necrosis while gliosarcoma could only be differentiated by
33
T2-weighted MRI. This data suggests that CEST-based imaging may be better than other MRI
techniques at differentiating cell death from viable tumour.
A clinical study by Mehrabian et al.53 of tumour progression versus radiation-induced cell death
following stereotactic radiosurgery for brain metastases showed that maximum MTR difference
between cell death and progressive tumour has been found in the amide and aliphatic regions of
the CEST spectra, corresponding to 3.5 and -3.5 ppm, respectively. The -3.5 ppm offset used by
Mehrabian et al. is similar to the -3.3 ppm offset with maximum separation between the pre- and
post-chemotherapy spectra in this study. This finding may reflect the importance of the NOE,
hypothesized to be the contrast mechanism of aliphatic groups in amino acid side chains. In the
work presented here, comparing MTR for viable tumour and cell death at -3.3 ppm also showed
a statistically significant difference.
Schmitt et al. 53 reported on a small cohort of six women with breast cancer imaged with CEST
MRI. Their CEST technique used saturation RF energy between 1.2 – 1.8 ppm. In the 3
analyzable patients in the cohort, high CEST signal correlated well with tumour identified using
DCE-MRI and CEST signal values were higher in tumour than in surrounding fibroglandular
tissue. In the work presented here, similar findings were demonstrated, with significantly higher
MTR values (i.e. higher CEST signal) measured for viable tumour at 1.8 ppm compared with cell
death regions. These findings suggest that CEST around the 1.8 ppm frequency offset is of
particular interest in detecting viable breast cancer.
Imaging methods other than CEST MRI can detect cell death in vivo, albeit at later stages of
advanced necrosis. When these methods have been applied at varying times after treatment, a
trend is evident whereby the cell-death inducing effect of the treatment increases to a point after
which it begins to decrease. Tadayyon et al.21 used high (20 MHz) and low frequency (7 MHz)
QUS to study cell death in MDA-MB-231 xenografts using the same chemotherapy regimen
used in the work here. Histological analysis showed an increase in CDI up to 24 h after
chemotherapy, with the CDI at 48 h lower than at 24 h, although still statistically significantly
increased over baseline. A similar pattern was demonstrated for the change in average acoustic
concentration (ΔAAC), which was highest at 24 h after chemotherapy followed by a reduction at
48 h. A separate study23 which treated HTB-67 melanoma xenografts with photodynamic therapy
34
(PDT) and used high frequency (26 MHz), QUS showed similar patterns in the parameters of
mid-band fit and spectral slope, which have been correlated with cell death.76,77 The peak effect
was observed between 12-20 h after treatment, followed by a decline. In the work described here,
the CDI calculated using the 0.5 SD cutoff to define necrotic tissue demonstrated a similar trend.
The largest average increase in CDI was seen 8-12 h after chemotherapy, with the increase
reduced after 24 h. However, the differences between the time points did not reach statistical
significance.
The time required to set up and conduct each scan was approximately 3 hours. As this
experiment was primarily intended to demonstrate proof-of-concept, preparing and scanning a
large number of tumours (e.g. 5 or more) per time point would take an unnecessarily large
amount of time and resources, such as machine time and animal specimens. Therefore, 3-4
specimens per post-chemotherapy time period was deemed a reasonable compromise between
experimental expediency and sample size.
During the scan time, some tumour movement could have been experienced such as due to slow
drifts in the equipment position or deflation of pads and pillows used to set up the mouse on the
scanner. Image registration was employed in the fitting algorithms to minimize the effects of
such motion. Registration is more accurate when multiple slices are acquired (allowing 3D
registration). In this work, however, only single slices were acquired in Z-spectra, limiting
registration to in-plane.
Resource management and patient comfort considerations make long scans untenable in human
trials. Reducing the number of frequency offsets used in clinical trials, for example by obtaining
data from several offsets around 1.8 ppm while minimizing the data taken in other offset regions,
would permit the use of shorter scans, consequently reducing scan costs and improving patient
satisfaction by not requiring long periods of cooperation lying in an MRI scanner. Measurements
at fewer offsets may also allow for longer RF saturation times given the availability of multiple
RF amplifiers, which generally have limited duty cycles, on a clinical scanner. This data can be
used to guide decisions to optimize scan protocols for future planned clinical trials.
35
Detection of cell death in vivo provides a promising avenue for early response assessment and
prediction for patients undergoing neoadjuvant chemotherapy for locally advanced breast cancer.
This data further supports the ability of CEST to detect cell death in breast cancer. Differences in
MTR measurements at 1.8 ppm should be a point of interest in studies attempting to translate
CEST MRI analysis into clinical practice and may be investigated alone or in combination with
previously studied metrics such as Lorentzian peak amplitude to develop prediction algorithm
based on multiple CEST parameters. Further study, in both animal models and humans, can
combine CEST MRI with other validated imaging modalities to further refine detection methods
to detect cell death and improve predictive models for response and clinical outcomes.
2.5 Conclusions
Analysis of magnetization transfer ratio using CEST MRI can differentiate between viable
tumour and cell death in MDA-231 xenografts. Maximal tumour response to chemotherapy is
seen at 8-12 h after administration
36
Chapter 3 Summary and Future Work
3.1 Summary
This thesis investigated the use of CEST MRI to distinguish between areas of viable tumour and
cell death in vivo in breast cancer xenografts. The first chapter described previous clinical efforts
to develop imaging-based methods to monitor the response of cancer to therapy and to predict
ultimate response early in the treatment course. It also described the downsides presented by
previously studied methods including questions regarding sensitivity and specificity of signal
changes after cytotoxic chemotherapy, the need for injected contrast agents, cost, and patient
inconvenience in integrating new modalities into existing clinical pathways. This chapter also
reviewed the physics of CEST MRI contrast and the characteristics that make it a very promising
modality for treatment response monitoring.
Chapter 2 described experimental efforts to characterize the in vivo CEST parameters. Tumours
were scanned before administration of chemotherapy and then again after a series of different
intervals after chemotherapy administration (4, 8, 12, and 24 h). Histological specimens were
obtained from each tumour and stained to differentiate areas of viable tumour and cell death.
Visual comparison between these stained histological specimens with high resolution MRI
provided a method to distinguish between viable tumour and cell death on the images.
Registration of the CEST images with the high-resolution images allowed the delineation of
areas of cell death and viable tumour on the CEST image.
Preliminary comparison of the Z-spectra of viable tumour areas with cell death regions suggested
that the 1.8 ppm frequency offset showed maximum separation between the two regions
compared to all other frequencies. A second frequency of interest, -3.3 ppm, was found by
comparing the mean Z-spectrum of the entire pre-chemotherapy xenografts to the mean
Z-spectrum of the entire post-chemotherapy xenografts. Histograms of the MTR values for each
voxel at these two frequencies were generated. Segmentation of the tumours was then performed
37
using a variety of threshold values to differentiate cell death and viable tumour based on these
histograms. Using the 1.8 ppm offset, the threshold MTR value of 0.12 was found to provide the
most accurate definition of these regions (using ISEL staining as the gold standard), while a
threshold value of 0.125 provided the most accurate definition using the MTR measurements
from the -3.3 ppm offset; both these cutoffs represented the value 0.5 standard deviations below
the mean of the calculated histogram for that offset. Regions defined using this threshold showed
significant differences in MTR values between the two regions.
Finally, establishing a threshold MTR value to define areas of cell death allowed the mean
change in CDI between the pre-chemotherapy and post-chemotherapy scans to be measured as a
function of interval between chemotherapy administration and post-chemotherapy scan time. An
increase in CDI to 8-12 h followed by a decrease at 24 h was measured, although the changes did
not meet statistical significance, likely due to the relatively small numbers of tumours studied.
This time course is similar to that measured after treatment by other methods.78,79
3.2 Future Work
The results described here represent a basis for detection of cell death both before treatment and
after administration of chemotherapy (e.g., apoptosis and/or necrosis). Future work expanding on
these findings should focus on three avenues of inquiry: 1) validating the results, 2) translating
CEST protocols for response detection and monitoring for human breast cancer, and 3) further
expansion to other cancer types.
The first step is to validate the results in a larger sample of xenograft tumours. The study
described in this thesis was intended as a proof-of-concept, and so a balance was sought between
developing a robust sample size and experimental expediency given the difficulties of growing
tumours and long scan times required. The results obtained were promising, as seen in the large
separation in MTR values at the 1.8 ppm offset between tumour and necrosis regions were
shown and the ability of MTR mapping at this offset to approximate regions of cell death
determined histologically, changes in CDI measured at each time point after chemotherapy
38
administration did not reach statistical significance. Continuing this experimentation with
additional xenografts to increase the sample size would add confidence in the results.
These findings may also inform the development of clinical protocols for response detection and
monitoring in human patients with locally advanced breast cancer undergoing pre-operative
chemotherapy. One challenge to translation of CEST MRI from preclinical to clinical studies is
the long scan time required. In the work presented here, a single CEST scan with RF pulses
across the entire frequency spectrum takes ~35 min with additional time required for structural,
inversion recovery, B0 mapping, and other ancillary scans. A clinical scan requiring a similar
length of time would present a significant burden on patients’ time. It would also tax existing
MRI scanners and staff which are often stretched to accommodate all patients requiring imaging
services or would introduce significant capital and operating costs to obtain and run additional
MRI scanners. A reduction in scan time would greatly aid uptake of CEST into the clinic.
One way to reduce scan time is to refine the understanding of differences in CEST parameters so
as to use only those RF frequencies which provide maximum contrast to answer a given clinical
question. The experiments described in this thesis suggest that the region around 1.8 ppm
provides the maximum MTR difference between cell death and viable tumour. Therefore, future
CEST MRI protocols could focus on RF frequencies around this offset, which would
dramatically reduce scan time.
Finally, the experimental methods and experience developed through this experiment can be
used to expand CEST MRI for use among other cancer types. Prostate cancer, for example, often
presents clinical dilemmas in determining optimal treatment for patients. Most localized prostate
cancer is amenable both to surgical resection of the prostate or to radiotherapy, either via
external beam radiotherapy or brachytherapy insertion of radioactive sources.80 However, no
directly comparative randomized controlled trials have been performed. One way to guide
appropriate treatment would be to determine the sensitivity of a given patient’s tumour to
radiation therapy; patients with more radiosensitive tumours could be offered radiotherapy-based
treatments while patients with radiation resistant tumours could be steered toward surgery.
Because rapidly dividing tissues are both more susceptible to radiotherapy and more
metabolically active than more slowly dividing tissue, using CEST MRI to measure metabolic
39
activity81 could also serve as a marker for radiation sensitivity. Preclinical work on this topic is
currently ongoing within the Stanisz laboratory in the University of Toronto Department of
Medical Biophysics.
Further development of CEST MRI and refinement of the ability to detect cell death early in a
treatment course and predict ultimate treatment response will provide valuable information for
clinicians to optimize treatment protocols (and make appropriate real-time changes if initial
decisions prove suboptimal), maximize cure and control rates and avoid unnecessary toxicity.
The promise of CEST MRI for detecting changes in tissue microenvironment, such as after
cytotoxic therapy, without attendant risks of other modalities makes it a very promising modality
for a range of clinical applications, including treatment response monitoring and prediction.
Although there is much work remaining to refine these protocols and applications, the work
presented here provides another proof-of-principle for the use of CEST in detecting cytotoxicity
which can eventually lead to more involved treatment monitoring algorithms.
3.3 Conclusions
This thesis has demonstrated that CEST MRI can be used to differentiate cell death from viable
tumour in an in vivo breast cancer model. It determined the RF saturation frequency (1.8 ppm)
which provides the maximum contrast between these two regions, which should provide a basis
for future work to refine the frequency regions scanned in CEST analyses (thus shortening scan
time) and to translate this preclinical work into early phase clinical trials. A threshold MTR value
which can distinguish between these two regions was also determined and the measured shows a
characteristic trend in that the maximum increase in cell death was seen at 8-12 h after
chemotherapy while less increase in cell death was measured 24 h after chemotherapy.
The results presented here should serve as an important basis for translation of the CEST MRI
technique to clinical trials. Investigations of CEST parameters and early detection of cell death
and treatment response can use the RF frequencies and threshold values identified as providing
insight into cell death to define their protocols and to reduce scan time by eliminating the need to
40
scan the entire frequency spectrum. Such protocols will hopefully allow for early detection of
cell death and early prediction of ultimate treatment response, allowing for better personalization
of cancer treatment and improving patient outcomes while reducing side effects.
41
References
1. Baskar R, Dai J, Wenlong N, Yeo R, Yeoh KW. Biological response of cancer cells to radiation treatment. Front Mol Biosci. 2014 Nov 17;1:24. 2. Bordeianou L, Maguire LH, Alavi K, Sudan R, Wise PE, Kaiser AM. Sphincter-sparing surgery in patients with low-lying rectal cancer: Techniques, oncologic outcomes, and functional results. J Gastrointest Surg. 2014 Jul;18(7):1358-72. 3. Sauer R, Becker H, Hohenberger W, Rödel C, Wittekind C, Fietkau R, Martus P, Tschmelitsch J, Hager E, Hess CF, Karstens JH, Liersch T, Schmidberger H, Raab R; German Rectal Cancer Study Group. Preoperative versus postoperative chemoradiotherapy for rectal cancer. N Engl J Med. 2004 Oct 21;351(17):1731-40. 4. Killelea BK, Yang VQ, Mougalian S, Horowitz NR, Pusztai L, Chagpar AB Lannin DR. Neoadjuvant chemotherapy for breast cancer increases the rate of breast conservation: Results from the National Cancer Database. J Am Coll Surg. 2015 Jun;220(6):1063-9. 5. Rastogi P, Anderson SJ, Bear HD, Geyer CE, Kahlenberg MS, Robidoux A, Margolese RG, Hoehn JL, Vogel VG, Dakhil SR, Tamkus D, King KM, Pajon ER, Wright MJ, Robert J, Paik S, Mamounas EP, Wolmark N. Preoperative chemotherapy: Updates of National Surgical Adjuvant Breast and Bowel Project Protocols B-18 and B-27. J Clin Oncol. 2008 Feb 10;26(5):778-85. 6. Kuerer HM, Newman LA, Smith TL, Ames FC, Hunt KK, Dhingra K, Theriault RL, Singh G, Binkley SM, Sneige N, Buchholz TA, Ross MI, McNeese MD, Buzdar AU, Hortobagyi GN, Singletary SE. Clinical course of breast cancer patients with complete pathologic primary tumor and axillary lymph node response to doxorubicin-based neoadjuvant chemotherapy. J Clin Oncol. 1999 Feb;17(2):460-9. 7. Guarneri V, Broglio K, Kau SW, Cristofanilli M, Buzdar AU, Valero V, Buchholz T, Meric F, Middleton L, Hortobagyi GN, Gonzalez-Angulo AM. Prognostic value of pathologic complete response after primary chemotherapy in relation to hormone receptor status and other factors. J Clin Oncol. 2006 Mar 1;24(7):1037-44. 8. Dula AN, Arlinghaus LR, Dortch RD, Dewey BE, Whisenant JG, Ayers GD, Yankeelov TE, Smith SA. Amide Proton Transfer Imaging of the Breast at 3 T: Establishing Reproducibility and Possible Feasibility Assessing Chemotherapy Response. Magn Reson Med. 2013 July; 70(1):216-24. 9. Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, Rubinstein L, Verweij J, Van Glabbeke M, van Oosterom AT, Christian MC, Gwyther SG. New guidelines to evaluate the response to treatment in solid tumors. European Organization for Research and Treatment of Cancer, National Cancer Institute of the United States, National Cancer Institute of Canada. J Natl Cancer Inst. 2000 Feb 2;92(3):205-16. 10. Theilmann RJ, Borders R, Trouard TP, Xia G, Outwater E, Ranger-Moore J, Gillies RJ, Stopeck A. Changes in water mobility measured by diffusion MRI predict response of metastatic breast cancer to chemotherapy. Neoplasia. 2004 Nov-Dec;6(6):831-7.
42
11. Thoeny HC, Ross BD. Predicting and monitoring cancer treatment response with diffusion-weighted MRI. J Magn Reson Imaging. 2010 Jul;32(1):2-16. 12. Cheung YC, Chen SC, Su MY, See LC, Hsueh S, Chang HK, Lin YC, Tsai CS. Monitoring the size and response of locally advanced breast cancers to neoadjuvant chemotherapy (weekly paclitaxel and epirubicin) with serial enhanced MRI. Breast Cancer Research and Treatment. 2003;78:51-8. 13. Schwarz J, Bader M, Jenicke L, Hemminger G, Jänicke F, Avril N. Early prediction of response to chemotherapy in metastatic breast cancer using sequential 18F-FDG PET. J Nucl Med. 2005 Jul;46(7):1144-50. 14. Rousseau C, Devillers A, Sagan C, Ferrer L, Bridji B, Campion L, Ricaud M, Bourbouloux E, Doutriaux I, Clouet M, Berton-Rigaud D, Bouriel C, Delecroix V, Garin E, Rouquette S, Resche I, Kerbrat P, Chatal JF, Campone M. Monitoring of early response to neoadjuvant chemotherapy in stage II and III breast cancer by [18F]fluorodeoxyglucose positron emission tomography. J Clin Oncol. 2006 Dec 1;24(34):5366-72. 15. Duch J, Fuster D, Muñoz M, Fernández PL, Paredes P, Fontanillas M, Guzmán F, Rubí S, Lomeña FJ, Pons F. 18F-FDG PET/CT for early prediction of response to neoadjuvant chemotherapy in breast cancer. Eur J Nucl Med Mol Imaging. 2009 Oct;36(10):1551-7. 16. Smith IC, Welch AE, Hutcheon AW, Miller ID, Payne S, Chilcott F, Waikar S, Whitaker T, Ah-See AK, Eremin O, Heys SD, Gilbert FJ, Sharp PF. Positron emission tomography using [(18)F]-fluorodeoxy-D-glucose to predict the pathologic response of breast cancer to primary chemotherapy. J Clin Oncol. 2000 Apr;18(8):1676-88. 17. Weber WA, Ott K, Becker K, Dittler HJ, Helmberger H, Avril NE, Meisetschläger G, Busch R, Siewert JR, Schwaiger M, Fink U. Prediction of response to preoperative chemotherapy in adenocarcinomas of the esophagogastric junction by metabolic imaging. J Clin Oncol. 2001 Jun 15;19(12):3058-65. 18. Avril N, Sassen S, Schmalfeldt B, Naehrig J, Rutke S, Weber WA, Werner M, Graeff H, Schwaiger M, Kuhn W. Prediction of response to neoadjuvant chemotherapy by sequential F-18-fluorodeoxyglucose positron emission tomography in patients with advanced-stage ovarian cancer. J Clin Oncol. 2005 Oct 20;23(30):7445-53. 19. Avril N, Rosé CA, Schelling M, Dose J, Kuhn W, Bense S, Weber W, Ziegler S, Graeff H, Schwaiger M. Breast imaging with positron emission tomography and fluorine-18 fluorodeoxyglucose: Use and limitations. J Clin Oncol. 2000 Oct 15;18(20):3495-502. 20. Pio BS, Park CK, Pietras R, Hsueh WA, Satyamurthy N, Pegram MD, Czernin J, Phelps ME, Silverman DH. Usefulness of 3'-[F-18]fluoro-3'-deoxythymidine with positron emission tomography in predicting breast cancer response to therapy. Mol Imaging Biol. 2006 Jan-Feb;8(1):36-42.
43
21. Tadayyon H, Sannachi L, Sadeghi-Naini A, Al-Mahrouki A, Tran WT, Kolios MC, Czarnota GJ. Quantification of Ultrasonic Scattering Properties of In Vivo Tumor Cell Death in Mouse Models of Breast Cancer. Transl Oncol. 2015 Dec;8(6):463-73. 22. Czarnota GJ, Kolios MC, Vaziri H, Benchimol S, Ottensmeyer FP, Sherar MD, Hunt JW. Ultrasonic biomicroscopy of viable, dead and apoptotic cells. Ultrasound Med Biol. 1997;23(6):961-5. 23. Banihashemi B, Vlad R, Debeljevic B, Giles A, Kolios MC, Czarnota GJ. Ultrasound imaging of apoptosis in tumor response: Novel preclinical monitoring of photodynamic therapy effects. Cancer Res. 2008 Oct 15;68(20):8590-6. 24. Sadeghi-Naini A, Papanicolau N, Falou O, Zubovits J, Dent R, Verma S, Trudeau M, Boileau JF, Spayne J, Iradji S, Sofroni E, Lee J, Lemon-Wong S, Yaffe M, Kolios MC, Czarnota GJ. Quantitative ultrasound evaluation of tumor cell death response in locally advanced breast cancer patients receiving chemotherapy. Clin Cancer Res. 2013 Apr 15;19(8):2163-74. 25. Sadeghi-Naini A, Sannachi L, Tadayyon H, Tran WT, Slodkowska E, Trudeau M, Gandhi S, Pritchard K, Kolios MC, Czarnota GJ. Chemotherapy-Response Monitoring of Breast Cancer Patients Using Quantitative Ultrasound-Based Intra-Tumour Heterogeneities. Sci Rep. 2017 Sep 4;7(1):10352. 26. Ntziachristos V, Chance B. Probing physiology and molecular function using optical imaging: Applications to breast cancer. Breast Cancer Res. 2001;3(1):41-6. 27. Soliman H, Gunasekara A, Rycroft M, Zubovits J, Dent R, Spayne J, Yaffe MJ, Czarnota GJ. Functional imaging using diffuse optical spectroscopy of neoadjuvant chemotherapy response in women with locally advanced breast cancer. Clin Cancer Res. 2010 May 1;16(9):2605-14. 28. Cerussi A, Hsiang D, Shah N, Mehta R, Durkin A, Butler J, Tromberg BJ. Predicting response to breast cancer neoadjuvant chemotherapy using diffuse optical spectroscopy. Proc Natl Acad Sci U S A. 2007 Mar 6;104(10):4014-9. 29.Yeh E, Slanetz P, Kopans DB, Rafferty E, Georgian-Smith D, Moy L, Halpern E, Moore R, Kuter I, Taghian A. Prospective comparison of mammography, sonography, and MRI in patients undergoing neoadjuvant chemotherapy for palpable breast cancer. AJR Am J Roentgenol. 2005 Mar;184(3):868-77. 30. Chen X, Moore MO, Lehman CD, Mankoff DA, Lawton TJ, Peacock S, Schubert EK, Livingston RB. Combined use of MRI and PET to monitor response and assess residual disease for locally advanced breast cancer treated with neoadjuvant chemotherapy. Acad Radiol. 2004 Oct;11(10):1115-24. 31. Baek HM, Chen JH, Nalcioglu O, Su MY. Proton MR spectroscopy for monitoring early treatment response of breast cancer to neo-adjuvant chemotherapy. Ann Oncol. 2008 May;19(5):1022-1024.
44
32. Meisamy S, Bolan PJ, Baker EH, Bliss RL, Gulbahce E, Everson LI, Nelson MT, Emory TH, Tuttle TM, Yee D, Garwood M. Neoadjuvant chemotherapy of locally advanced breast cancer: Predicting response with in vivo (1)H MR spectroscopy--a pilot study at 4 T. Radiology. 2004 Nov;233(2):424-31. 33. Aboagye EO, Bhujwalla ZM, He Q, Glickson JD. Evaluation of lactate as a 1H nuclear magnetic resonance spectroscopy index for noninvasive prediction and early detection of tumor response to radiation therapy in EMT6 tumors. Radiat Res. 1998 Jul;150(1):38-42. 34. Horská A, Barker PB. Imaging of brain tumors: MR spectroscopy and metabolic imaging. Neuroimaging Clin N Am. 2010 Aug;20(3):293-310. 35. Li Y, Park I, Nelson SJ. Imaging tumor metabolism using in vivo magnetic resonance spectroscopy. Cancer J. 2015 Mar-Apr;21(2):123-8. 36. Martincich L, Montemurro F, De Rosa G, Marra V, Ponzone R, Cirillo S, Gatti M, Biglia N, Sarotto I, Sismondi P, Regge D, Aglietta M. Monitoring response to primary chemotherapy in breast cancer using dynamic contrast-enhanced magnetic resonance imaging. Breast Cancer Res Treat. 2004 Jan;83(1):67-76. 37. Loo CE, Teertstra HJ, Rodenhuis S, van de Vijver MJ, Hannemann J, Muller SH, Peeters MJ, Gilhuijs KG. Dynamic contrast-enhanced MRI for prediction of breast cancer response to neoadjuvant chemotherapy: Initial results. AJR Am J Roentgenol. 2008 Nov;191(5):1331-8. 38. Pickles MD, Lowry M, Manton DJ, Gibbs P, Turnbull LW. Role of dynamic contrast enhanced MRI in monitoring early response of locally advanced breast cancer to neoadjuvant chemotherapy. Breast Cancer Res Treat. 2005 May;91(1):1-10. 39. Tudorica A, Oh KY, Chui SY, Roy N, Troxell ML, Naik A, Kemmer KA, Chen Y, Holtorf ML, Afzal A, Springer CS Jr, Li X, Huang W. Early Prediction and Evaluation of Breast Cancer Response to Neoadjuvant Chemotherapy Using Quantitative DCE-MRI. Transl Oncol. 2016 Feb;9(1):8-17. 40. Ah-See ML, Makris A, Taylor NJ, Harrison M, Richman PI, Burcombe RJ, Stirling JJ, d'Arcy JA, Collins DJ, Pittam MR, Ravichandran D, Padhani AR. Early changes in functional dynamic magnetic resonance imaging predict for pathologic response to neoadjuvant chemotherapy in primary breast cancer. Clin Cancer Res. 2008 Oct 15;14(20):6580-9. 41. Loo CE, Straver ME, Rodenhuis S, Muller SH, Wesseling J, Vrancken Peeters MJ, Gilhuijs KG. Magnetic resonance imaging response monitoring of breast cancer during neoadjuvant chemotherapy: Relevance of breast cancer subtype. J Clin Oncol. 2011 Feb 20;29(6):660-6. 42. Gubern-Mérida A, Martí R, Melendez J, Hauth JL, Mann RM, Karssemeijer N, Platel B. Automated localization of breast cancer in DCE-MRI. Med Image Anal. 2015 Feb;20(1):265-74. 43. de Rooij M, Hamoen EH, Fütterer JJ, Barentsz JO, Rovers MM. Accuracy of multiparametric MRI for prostate cancer detection: A meta-analysis. AJR Am J Roentgenol. 2014 Feb;202(2):343-51.
45
44. Iwasa H, Kubota K, Hamada N, Nogami M, Nishioka A. Early prediction of response to neoadjuvant chemotherapy in patients with breast cancer using diffusion-weighted imaging and gray-scale ultrasonography. Oncol Rep. 2014 Apr;31(4):1555-60. 45. Pickles MD, Gibbs P, Lowry M, Turnbull LW. Diffusion changes precede size reduction in neoadjuvant treatment of breast cancer. Magn Reson Imaging. 2006 Sep;24(7):843-7. 46. Dzik-Jurasz A, Domenig C, George M, Wolber J, Padhani A, Brown G, Doran S. Diffusion MRI for prediction of response of rectal cancer to chemoradiation. Lancet 2002;360(9329):307-8. 47. Stanisz GJ. Diffusion MR in biological systems: Tissue compartments and exchange. Israel J Chem 2003;43:33–44. 48. Kummar S, Gutierrez M, Doroshow JH, Murgo AJ. Drug development in oncology: Classical cytotoxics and molecularly targeted agents. Br J Clin Pharmacol. 2006 Jul;62(1):15-26. 49. Hein PA, Kremser C, Judmaier W, Griebel J, Pfeiffer KP, Kreczy A, Hug EB, Lukas P, DeVries AF. Diffusion-weighted magnetic resonance imaging for monitoring diffusion changes in rectal carcinoma during combined, preoperative chemoradiation: Preliminary results of a prospective study. Eur J Radiol 2003;45(3):214-22. 50. Kremser C, Judmaier W, Hein P, Griebel J, Lukas P, de Vries A. Preliminary results on the influence of chemoradiation on apparent diffusion coefficients of primary rectal carcinoma measured by magnetic resonance imaging. Strahlenther Onkol 2003;179(9):641. 51. Padhani AR, Liu G, Koh DM, Chenevert TL, Thoeny HC, Takahara T, Dzik-Jurasz A, Ross BD, Van Cauteren M, Collins D, Hammoud DA, Rustin GJ, Taouli B, Choyke PL. Diffusion-weighted magnetic resonance imaging as a cancer biomarker: Consensus and recommendations. Neoplasia. 2009 Feb;11(2):102-25. 52. Desmond KL, Moosvi F, Stanisz GJ. Mapping of amide, amine, and aliphatic peaks in the CEST spectra of murine xenografts at 7 T. Magn Reson Med. 2014 May;71(5):1841-53. 53. Mehrabian H, Desmond KL, Soliman H, Sahgal A, Stanisz GJ. Differentiation between Radiation Necrosis and Tumor Progression Using Chemical Exchange Saturation Transfer. Clin Cancer Res. 2017 Jul 15;23(14):3667-75. 54. Bell LK, Ainsworth NL, Lee SH, Griffiths JR. MRI & MRS assessment of the role of the tumor microenvironment in response to therapy. NMR Biomed 2011;24:612-35. 55. Gillies RJ, Raghunand N, Karczmar GS, Bhujwalla ZM. MRI of the tumor microenvironment. J Magn Reson Imaging 2002;16:430-50. 56. Zhang X, Lin Y, Gillies RJ. Tumor pH and its measurement. J Nucl Med 2010;51:1167-70. 57. Wolff SD, Balaban RS. NMR imaging of labile proton exchange. J Magn Reson. 1990;86:164-9.
46
58. Wen Z, Hu S, Huang F, Wang X, Guo L, Quan X, Wang S, Zhou J. MR imaging of high-grade brain tumors using endogenous protein and peptide-based contrast. NeuroImage. 2010 Jun;51(2):616-22. 59. van Zijl PC, Yadav NN. Chemical exchange saturation transfer (CEST): What is in a name and what isn't? Magn Reson Med. 2011 Apr;65(4):927-48. 60. Stanisz GJ, Odrobina EE, Pun J, Escaravage M, Graham SJ, Bronskill MJ, Henkelman RM. T1, T2 relaxation and magnetization transfer in tissue at 3T. Magn Reson Med. 2005 Sep;54(3):507-12. 61. van Zijl PC, Jones CK, Ren J, Malloy CR, Sherry AD. MRI detection of glycogen in vivo by using chemical exchange saturation transfer imaging (glycoCEST). Proc Natl Acad Sci U S A. 2007 Mar 13;104(11):4359-64 62. Cai K, Singh A, Poptani H, Li W, Yang S, Lu Y, Hariharan H, Zhou XJ, Reddy R. CEST signal at 2ppm (CEST@2ppm) from Z-spectral fitting correlates with creatine distribution in brain tumor. NMR Biomed. 2015 Jan;28(1):1-8. 63. Zhou J, Tryggestad E, Wen Z, Lal B, Zhou T, Grossman R, Wang S, Yan K, Fu DX, Ford E, Tyler B, Blakeley J, Laterra J, van Zijl PC. Differentiation between glioma and radiation necrosis using molecular magnetic resonance imaging of endogenous proteins and peptides. Nat Med. 2011 Jan;17(1):130-4. 64. Sagiyama K, Mashimo T, Togao O, Vemireddy V, Hatanpaa KJ, Maher EA, Mickey BE, Pan E, Sherry AD, Bachoo RM, Takahashi M. In vivo chemical exchange saturation transfer imaging allows early detection of a therapeutic response in glioblastoma. Proc Natl Acad Sci U S A. 2014 Mar 25;111(12):4542-7. 65. Desmond KL, Mehrabian H, Chavez S, Sahgal A, Soliman H, Rola R, Stanisz GJ. Chemical exchange saturation transfer for predicting response to stereotactic radiosurgery in human brain metastasis. Magn Reson Med. 2017 Sep;78(3):1110-11. 66. Ma B, Blakeley JO, Hong X, Zhang H, Jiang S, Blair L, Zhang Y, Heo HY, Zhang M, van Zijl PC, Zhou J. Applying amide proton transfer-weighted MRI to distinguish pseudoprogression from true progression in malignant gliomas. J Magn Reson Imaging. 2016 Aug; 44(2): 456-62. 67. Schmitt B, Zamecnik P, Zaiss M, Rerich E, Schuster L, Bachert P, Schlemmer HP. A new contrast in MR mammography by means of chemical exchange saturation transfer (CEST) imaging at 3 Tesla: preliminary results. Rofo. 2011 Nov; 183(11): 1030-6. 68. Wang M, Hou L, Chen M, Zhou Y, Liang Y, Wang S, Jiang J, Zhang Y. Neoadjuvant Chemotherapy Creates Surgery Opportunities For Inoperable Locally Advanced Breast Cancer. Sci Rep. 2017 Mar 22;7:44673. 69. Giordano SH. Update on locally advanced breast cancer. Oncologist. 2003;8(6):521-30.
47
70. Tachi T, Teramachi H, Tanaka K, Asano S, Osawa T, Kawashima A, Hori A, Yasuda M, Mizui T, Nakada T, Noguchi Y, Tsuchiya T, Goto C. The impact of side effects from outpatient chemotherapy on presenteeism in breast cancer patients: A prospective analysis. SpringerPlus. 2016 Mar 15;5:327. 71. Alarid-Escudero F, Blaes AH, Kuntz KM. Trade-offs Between Efficacy and Cardiac Toxicity of Adjuvant Chemotherapy in Early-Stage Breast Cancer Patients: Do Competing Risks Matter? Breast J. 2017 Jul;23(4):401-9 72. Rugo HS, Klein P, Melin SA, Hurvitz SA, Melisko ME, Moore A, Park G, Mitchel J, Bågeman E, D'Agostino RB Jr, Ver Hoeve ES, Esserman L, Cigler T. Association Between Use of a Scalp Cooling Device and Alopecia After Chemotherapy for Breast Cancer. JAMA. 2017 Feb 14;317(6):606-14. 73. Ohno T, Nakano T, Niibe Y, Tsuji H, Oka K. Bax Protein Expression Correlates with Radiation-Induced Apoptosis in Radiation Therapy for Cervical Carcinoma. Cancer. 1998 Jul 1;83(1):103-10. 74. Ansari B, Coates PJ, Greenstein BD, Hall PA. In situ end-labelling detects DNA strand breaks in apoptoisis and other physiological and pathological states. J Pathol. 1993 May;170(1):1-8. 75. El Kaffas A, Al-Mahrouki A, Tran WT, Giles A, Czarnota GJ. Sunitinib effects on the radiation response of endothelial and breast tumor cells. Microvasc Res. 2014 Mar;92:1-9. 76. Pasternak MM, Wirtzfeld LA, Kolios MC, Czarnota GJ. High-frequency ultrasound analysis of post-mitotic arrest cell death. Oncoscience. 2016 Apr 15;3(3-4):109-21. 77. Tadayyon H, Sannachi L, Gangeh M, Sadeghi-Naini A, Tran W, Trudeau ME, Pritchard K, Ghandi S, Verma S, Czarnota GJ. Quantitative ultrasound assessment of breast tumor response to chemotherapy using a multi-parameter approach. Oncotarget. 2016 Jul 19;7(29):45094-111. 78. Sadeghi-Naini A, Falou O, Tadayyon H, Al-Mahrouki A, Tran W, Papanicolau N, Kolios MC, Czarnota GJ. Conventional frequency ultrasonic biomarkers of cancer treatment response in vivo. Transl Oncol. 2013 Jun 1;6(3):234-43. 79. Gangeh MJ, Sadeghi-Naini A, Diu M, Tadayyon H, Kamel MS, Czarnota GJ. Categorizing extent of tumor cell death response to cancer therapy using quantitative ultrasound spectroscopy and maximum mean discrepancy. IEEE Trans Med Imaging. 2014 Jen;33(6):1390-400. 80. Litwin MS, Tan HJ. The Diagnosis and Treatment of Prostate Cancer: A Review. 2017 Jun 27;317(24):2532-2542. 81. Walker-Samuel S, Ramasawmy R, Torrealdea F, Rega M, Rajkumar V, Johnson SP, Richardson S, Gonçalves M, Parkes HG, Arstad E, Thomas DL, Pedley RB, Lythgoe MF, Golay X. In vivo imaging of glucose uptake and metabolism in tumors. Nat Med. 2013 Aug;19(8):1067-72.