Tr a n s l a t i o n a l O n c o l o g y Volume 9 Number 1 February 2016 pp.

8–17 8
www.transonc.com

Early Prediction and Evaluation Alina Tudorica*, Karen Y Oh*, Stephen Y-C Chui †, ‡,
Nicole Roy*, Megan L Troxell ‡, §, Arpana Naik ‡, ¶,
of Breast Cancer Response to Kathleen A Kemmer †, ‡, Yiyi Chen ‡, #,
Megan L Holtorf ‡, Aneela Afzal**,
Neoadjuvant Chemotherapy Charles S Springer Jr. ‡,**, Xin Li** and Wei Huang ‡,**
Using Quantitative DCE-MRI1
*Diagnostic Radiology, Oregon Health & Science University,
Portland, OR, USA; † Medical Oncology, Oregon Health &
Science University, Portland, OR, USA; ‡ Knight Cancer
Institute, Oregon Health & Science University, Portland, OR,
USA; § Pathology, Oregon Health & Science University,
Portland, OR, USA; ¶ Surgical Oncology, Oregon Health &
Science University, Portland, OR, USA; # Public Health and
Preventive Medicine, Oregon Health & Science University,
Portland, OR, USA; **Advanced Imaging Research Center,
Oregon Health & Science University, Portland, OR, USA

Abstract
The purpose is to compare quantitative dynamic contrast-enhanced (DCE) magnetic resonance imaging (MRI) metrics
with imaging tumor size for early prediction of breast cancer response to neoadjuvant chemotherapy (NACT) and
evaluation of residual cancer burden (RCB). Twenty-eight patients with 29 primary breast tumors underwent DCE-MRI
exams before, after one cycle of, at midpoint of, and after NACT. MRI tumor size in the longest diameter (LD) was
measured according to the RECIST (Response Evaluation Criteria In Solid Tumors) guidelines. Pharmacokinetic analyses of
DCE-MRI data were performed with the standard Tofts and Shutter-Speed models (TM and SSM). After one NACT cycle
the percent changes of DCE-MRI parameters K trans (contrast agent plasma/interstitium transfer rate constant), ve
(extravascular and extracellular volume fraction), kep (intravasation rate constant), and SSM-unique τi (mean intracellular
water lifetime) are good to excellent early predictors of pathologic complete response (pCR) vs. non-pCR, with univariate
logistic regression C statistics value in the range of 0.804 to 0.967. ve values after one cycle and at NACT midpoint are also
good predictors of response, with C ranging 0.845 to 0.897. However, RECIST LD changes are poor predictors with C =
0.609 and 0.673, respectively. Post-NACT K trans, τi, and RECIST LD show statistically significant (P b .05) correlations with
RCB. The performances of TM and SSM analyses for early prediction of response and RCB evaluation are comparable. In
conclusion, quantitative DCE-MRI parameters are superior to imaging tumor size for early prediction of therapy response.
Both TM and SSM analyses are effective for therapy response evaluation. However, the τi parameter derived only with SSM
analysis allows the unique opportunity to potentially quantify therapy-induced changes in tumor energetic metabolism.

Translational Oncology (2016) 9, 8–17

Address all correspondence to: Wei Huang, PhD, Advanced Imaging Research Center, Oregon
Introduction Health & Science University, 3181 SW Sam Jackson Park Road, Portland, OR 97239.
Neoadjuvant chemotherapy (NACT) is commonly used as standard E-mail: [email protected]
1
of care treatment for locally advanced breast cancer with the primary Financial Support: This study was supported by NIH grants U01-CA154602 and
R44-CA180425.
clinical goals of downstaging the disease, improving operability, and
Received 7 September 2015; Revised 20 November 2015; Accepted 23 November 2015
allowing breast conserving surgery [1,2]. Though NACT does not
improve overall survival in comparison with postoperative adjuvant © 2016 The Authors. Published by Elsevier Inc. on behalf of Neoplasia Press, Inc. This is an open
access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).
therapy [3–5], one advantage of NACT is to provide the opportunity 1936-5233/16
for assessment of pathologic response to the treatment. Pathologic http://dx.doi.org/10.1016/j.tranon.2015.11.016
Translational Oncology Vol. 9, No. 1, 2016 DCE-MRI Evaluation of Breast Cancer Therapy Response Tudorica et al. 9

complete response (pCR) or minimal residual disease following SSM-unique τi parameter (mean intracellular water lifetime), an
NACT has been shown to be prognostic for survival [6–9]. The inverse measure of cellular metabolic activity [40,41], were evaluated
majority of the patients, however, do not achieve pCR with the pCR and compared between the TM and SSM and with imaging tumor
rate reported in the range of 6% to 45% depending on breast cancer size measurement, for early prediction of pathologic response to
subtypes and treatment regimens [6,7,10,11]. Furthermore, the NACT and assessment of residual disease.
pathologic response status is generally determined only from the
surgical specimen after NACT completion. Therefore, there is a Materials and Methods
genuine and unmet need for reliable and minimally invasive imaging
methods to provide early prediction of response to NACT. In the Patient Cohort and Study Schema
emerging era of precision medicine, early prediction of NACT In this institutional review board–approved and HIPAA-compliant
response may allow rapid, personalized treatment regimen alterations study, twenty-eight women who were diagnosed with 29 grade 2 to 3
for non-responding breast cancer patients and spare them from invasive breast tumors (one patient had two independent primary tumors)
potential short and long term toxicities associated with ineffective and to undergo NACT were consented to participate in a longitudinal
therapies. Additionally, accurate evaluation of residual disease after research MRI study that includes DCE-MRI. Twenty one (22 primary
NACT is vital for surgical decision making and could result in surgical tumors) of the 28 patients were treated with standard of care therapy
treatment plans more tailored to individual patients. regimens that include four cycles of doxorubicin-cyclophosphamide every
Imaging tumor size change with guidelines such as RECIST 2 weeks followed by four cycles of a taxane every 2 weeks, or six cycles of
(Response Evaluation Criteria In Solid Tumors) [12] is routinely used the combination of all three drugs every 3 weeks. The targeted agent,
in clinical trial settings to assess tumor response to treatment. trastuzumab, was added to the regimen for tumors with positive HER2
However, size change in response to therapy is often found to (human epidermal growth factor receptor 2) receptor status. The other
manifest later than changes in underlying tumor functions [13–16], seven patients (seven primary tumors) were enrolled in the NACT ISPY-2
such as vascularization and vascular permeability, cellularity, and trial (http://ispy2.org), where patients were randomized to receive
metabolism. Recognized as a minimally invasive imaging method for standard of care regimen or standard of care regimen plus experimental
evaluation of perfusion and permeability, dynamic contrast-enhanced drugs. The ISPY-2 standard of care regimen starts with a taxane, followed
magnetic resonance imaging (DCE-MRI) is increasingly used in by doxorubicin-cyclophosphamide. If used, the experimental drug is
research and early phase clinical trial settings to measure and, usually added to the taxane. Four of the seven patients were placed in the
importantly, predict tumor response to treatment [13,14]. Over the treatment arm with experimental drugs: three (patient 7, 12, and 13,
last decade, substantial evidence has accumulated [17–19] showing Table 1) received neratinib, a tyrosine kinase inhibitor, and the other
the utility of DCE-MRI for assessment and early prediction of breast (patient 24) received ganitumab, a human monoclonal antibody against
cancer response to NACT. Despite large variations in DCE-MRI data type 1 insulin-like growth factor receptor (IGF1R). The clinicopathologic
acquisition protocol details (temporal resolution, spatial resolution characteristics of the studied patient cohort are presented in Table 1.
and coverage, acquisition time length, etc.) and data analysis methods The MRI exams for this research study were performed pre-NACT
(semi-quantitative vs. quantitative pharmacokinetic analysis of signal (Visit 1, V1), after one cycle of NACT (V2), at midpoint of NACT
intensity time-course data), many studies have shown that changes in (V3; usually after 3 or 4 cycles of NACT, or before change of NACT
several semi-quantitative [20–25] or quantitative [26–35] DCE-MRI agents), and after NACT completion but prior to surgery (V4). The
metrics during the course of NACT can provide good early prediction ISPY-2 trial includes a different MRI protocol conducted at the same
of pathologic response after one to two NACT cycles, and valuable four time points. For the 7 patients who participated in the ISPY-2
clinical evaluation of overall response and prognosis. In correlating trial, care was taken to ensure there were at least 24 hours between the
DCE-MRI parameters with pathologic response endpoints, most research and ISPY-2 MRI studies to allow CA, used in both, clearance
studies use binary discrimination of pCR and non-pCR with few [23] from the body. For the V2 – V4 studies, the MRI scan was
reporting relationships between post-NACT imaging metrics and undertaken at least 7 days after the administration of the previous
pathologically measured residual disease burden, which could have NACT cycle to allow time for drug effects.
important implications for surgical decision making. Among studies
that performed quantitative pharmacokinetic analyses of DCE-MRI DCE-MRI Data Acquisition
data, most employed the nuclear medicine, tracer kinetic model based All breast MRI studies were performed using a 3 T Siemens Tim
Tofts model (TM) [36,37] with inherent neglect of the effects of Trio system with the body coil and a four-channel bilateral
intercompartmental water exchange kinetics. The water molecule is phased-array breast coil as the transmitter and receiver, respectively.
not the signal molecule in nuclear medicine imaging, but in In each MRI session, following pilot scans and pre-CA axial
DCE-MRI it is. Considering the two-compartment model of intra- T2-weighted MRI with fat-saturation and axial T1-weighted MRI
and extra-cellular spaces, for example, since contrast agent (CA) without fat-saturation, axial bilateral DCE-MRI images with
molecules generally reside in the extracellular space, the cross-cell fat-saturation (using the approach of water excitation only) and full
membrane water exchange kinetics needs to be accounted for when breast coverage were acquired with a 3D gradient echo-based TWIST
converting MRI signal intensity change to tissue CA concentration (Time-resolved angiography WIth Stochastic Trajectories) sequence,
change in pharmacokinetic analysis of DCE-MRI data. which employs the strategy of k-space undersampling during
In this paper we report our initial results in DCE-MRI assessment acquisition and data sharing during reconstruction [42]. Compared
of breast cancer response to NACT. The DCE-MRI data were to conventional full k-space sampling gradient echo sequences, the
analyzed using both the TM and the Shutter-Speed model (SSM), TWIST sequence enables acceleration of image acquisition and
which takes into account the finite intercompartmental water preserves signal-to-noise ratio, while the image artifacts can be
exchange kinetics [38,39]. DCE-MRI parameters, including the minimized using appropriate k-space undersampling and view sharing
10 DCE-MRI Evaluation of Breast Cancer Therapy Response Tudorica et al. Translational Oncology Vol. 9, No. 1, 2016

Table 1. Clinicopathologic Characteristics of the Study Patient Cohort Treated with Neoadjuvant Chemotherapy

Patient Age (yr) Tumor Tumor Pre-treatment Receptor Treatment Regimen Pathologic RCB Class
Number Type Grade Size (cm) Status(ER,PR,HER2) Response

1 27 IDC 2 3.7(Mammo) −, −, + Docetaxel + carboplatin + trastuzumab non-pCR III

2 27 IDC 2 1.3(MRI) +, +, + Docetaxel + carboplatin + trastuzumab pCR 0
3 61 IDC 3 2.4(MRI) −, −, − Adriamycin + cyclophosphamide, then paclitaxel non-pCR I
4 39 IDC 2 2.6(MRI) +, +, + Paclitaxel + trastuzumab, then cyclophosphamide + adriamycin pCR 0
5 63 IDC 2 2.3(US) −, −, + Docetaxel + carboplatin + trastuzumab pCR 0
6 56 IDC 3 4.4(US) −, −, − Paclitaxel, then cyclophosphamide + adriamycin non-pCR I
7 62 IDC 3 2.1(Mammo) −, −, − Carboplatin + neratinib, then cyclophosphamide + adriamycin non-pCR II
8 65 IDC 2 2.1(Mammo) +, +, − Cyclophosphamide + adriamycin, then paclitaxel non-pCR II
9 46 IDC 2 2.5(Mammo) +, +, − Cyclophosphamide + adriamycin + docetaxel non-pCR II
10 33 IDC 2 2.0(US) +, +, + Docetaxel + carboplatin + trastuzumab non-pCR II
11 41 IDC 2 3.0(MRI) +, +, - Cyclophosphamide + adriamycin + docetaxel non-pCR II
12 35 IDC 2 2.8(MRI) −, +, + Paclitaxel + neratinib, then cyclophosphamide + adriamycin pCR 0
13 39 IDC 3 2.7(MRI) +, −, + Paclitaxel + neratinib, then cyclophosphamide + adriamycin non-pCR I
14 42 IDC 2 1.6(US) +, +, + Doxorubicin + cyclophosphamide, then docetaxel + trastuzumab non-pCR II
IDC 2 1.8(US) +, −, − non-pCR I
15 34 IDC 2 5.0(MRI) +, +, − Paclitaxel, then cyclophosphamide + adriamycin non-pCR II
16 45 ILC 2 11.8(MRI) +, +, − Paclitaxel, then cyclophosphamide + adriamycin non-pCR III
17 38 IDC 3 3.6(US) −, −, + Docetaxel + carboplatin + trastuzumab non-pCR III
18 59 IDC 3 2.0(US) +, +, + Docetaxel + carboplatin + trastuzumab non-pCR II
19 46 IDC 3 1.7(US) −, −, + Docetaxel + carboplatin + trastuzumab non-pCR I
20 59 IDC 2 3.0(MRI) −, −, + Paclitaxel + trastuzumab, then cyclophosphamide + adriamycin pCR 0
21 51 IDC 2 3.2(Mammo) +, +, − Cyclophosphamide + Adriamycin + docetaxel non-pCR I
22 75 ILC 2 2.5(Mammo) −, −, − Adriamycin + cyclophosphamide, then paclitaxel non-pCR II
23 34 IDC 3 2.1(MRI) −, −, + Adriamycin + cyclophosphamide, then paclitaxel + trastuzumab non-pCR I
24 32 IDC 3 5.9(MRI) +, +, − Paclitaxel + ganitumab, then cyclophosphamide + adriamycin non-pCR II
25 44 IDC 2 2.8(MRI) +, +, − Adriamycin + cyclophosphamide, then paclitaxel non-pCR I
26 37 IDC 3 9.8(MRI) −, −, − Cyclophosphamide + Adriamycin + docetaxel non-pCR II
27 48 IDC 3 2.9(Mammo) −, +, − Adriamycin + cyclophosphamide, then paclitaxel non-pCR III
28 31 IDC 3 1.6(US) −, +, − Adriamycin + cyclophosphamide, then paclitaxel non-pCR I
IDC: invasive ductal carcinoma; ILC: invasive lobular carcinoma; Pre-treatment size: imaging tumor size in the longest diameter before treatment; Mammo: mammography; US: ultrasound; ER: estrogen
receptor; PR: progesterone receptor; HER2: human epidermal growth factor receptor 2; pCR: pathologic complete response; RCB: residual cancer burden.

strategies [42]. DCE-MRI acquisition parameters included 10 o flip images and the baseline images from the DCE series [42,43]. The
angle, 2.9/6.2 ms TE/TR, a parallel imaging acceleration factor of 2, ROI-averaged and voxel (within the tumor ROI) DCE-MRI
30 to 34 cm FOV, 320×320 in-plane matrix size, and 1.4 mm slice time-course data were separately fitted with different pharmacokinetic
thickness. The total acquisition time for a DCE-MRI series was models: once with the one-compartment-two-parameter TM [36]
~ 10 minutes for 28 to 38 image volume sets of 96 to 128 slices each and once with the two-compartment-three-parameter fast-exchange-
with 14.6 to 20.2 s temporal resolution. The variations in number of regime (FXR)–allowed SSM [38]. Equations (1) and (2) represent the
image volumes, number of slices per volume, and temporal resolution pharmacokinetic time-course expressions for the TM and FXR-SSM,
were due to differences in breast size. The intravenous injection of the respectively:
CA, Gd(HP-DO3A) [ProHance (Bracco Diagnostic Inc.)] 8 9
< Zt
(0.1 mmol/kg at 2 mL/s), by a programmable power injector was  trans  0=
R 1 ð tÞ ¼ r 1 K
:
trans 0
Cp ðt Þ exp −K 0
=ve ðt–t Þ dt þ R10 ;
;
ð1Þ
timed to commence after acquisition of two baseline image volumes, 0
followed by a 20-mL saline flush.
For quantification of the pre-CA T1 value, T10, proton
"
density-weighted images were acquired immediately before and ( Zt
 
spatially co-registered with the DCE-MRI scan [42,43]. The data R1 ðtÞ ¼ ð1=2Þ 2R1i þ r1 Ktrans =ve Cp ðt0 Þ exp −Ktrans =ve ðt–t0 Þ dt0
acquisition sequence and parameters were the same as for the 0
)
DCE-MRI scan except for 5° flip angle and 50 ms TR.
þðR10 –R1i þ 1=τi Þ=ve
DCE-MRI Data Analysis (" Zt ð2Þ
Breast tumor regions of interest (ROIs) were drawn by two – 2=τi þ ðR1i – R10 −1=τi Þ=ve −r1 Ktrans =ve C p ð t0 Þ
experienced breast radiologists on post-CA (approximately 90 to 0 #
#2 )1=2
120 s after the CA injection) multi-slice DCE images covering the  
 exp −Ktrans =ve ðt–t0 Þ dt0 þ 4ð1–ve Þ=τi 2 ve ;
entire contrast-enhanced tumor. They also measured the longest
diameter (LD) of the tumor from these images using the RECIST
guidelines [12]. To avoid within-subject inter-observer variations, all where R1(t) is the tissue longitudinal relaxation rate constant, Cp(t′) is
images from the longitudinal study of a given patient were interpreted the arterial plasma CA concentration time course, or arterial input
by one radiologist. function (AIF), R1i is the intrinsic intracellular longitudinal relaxation
For the purpose of pharmacokinetic modeling of DCE-MRI data, rate constant, R10 (= 1/T10) is the pre-CA tissue R1, r1 is the tissue
the tumor ROI- and voxel-based T10 value was determined by CA relaxivity, K trans is the rate constant for CA plasma-to-interstitium
comparing signal intensities between the proton density-weighted transfer, ve is the volume fraction of extravascular and extracellular
Translational Oncology Vol. 9, No. 1, 2016 DCE-MRI Evaluation of Breast Cancer Therapy Response Tudorica et al. 11

space, and τi is the mean intracellular water molecule lifetime. Both results. The same RCB index value ranges were used for defining
model fittings return K trans and ve parameter values, and the CA in-breast RCB classes.
intravasation rate constant, kep, can be calculated as kep = K trans/ve.
However, only the SSM fitting returns the τi parameter. Since the Statistical Analysis
TM neglects the finite intercompartmental water exchange kinetics, Descriptive statistical analysis was conducted to summarize the
assuming the water exchange MR system is always in the fast-ex- pharmacokinetic parameter (returned by each model fitting) and
change-limit condition, the linear relationship between R1(t) and RECIST LD values at each visit, as well as the percent changes of
tissue CA concentration, Ct(t), shown in Equation (3) and implicit in these imaging metrics relative to baseline (V1). In assessing the ability
Equation (1), is used: for early prediction of therapy response, the univariate logistic
regression (ULR) analysis was used to correlate V1, V2, V3 MRI
R1 ðtÞ ¼ r1 Ct ðtÞ þ R10 ð3Þ metrics, and the corresponding V21% and V31% changes, with
The SSM accounts for the finite water exchange kinetics during the CA dichotomous pathologic response endpoints, pCR vs. non-pCR. A
passage through the tissue of interest, and consequently R1(t) is not ULR C statistics value, a measure equivalent to the area under the
linearly related to Ct(t). The FXR-SSM takes into account transcyto- receiver operating characteristic curve, in the range of 0.9 to 1.0
lemmal water exchange kinetics in the two-compartment model of intra- indicates an excellent predictor; 0.8 to 0.9, a good predictor; 0.7 to
and extra-cellular compartments (in the extravascular space), but assumes 0.8, a fair predictor; b 0.7, a poor predictor. A two-sample t test was
single exponential longitudinal MR signal decay [38,39,41,44]. used to evaluate the differences in imaging metrics and the
A population-averaged AIF was used for pharmacokinetic analysis corresponding percent changes between the pCR and non-pCR
of DCE-MRI data for each patient and each visit. This AIF was groups. In assessing the ability for evaluation of RCB (and in-breast
obtained by averaging individually measured AIFs from an axillary RCB) following NACT completion, the ULR analysis was used to
artery in a previous sagittal breast DCE-MRI study with higher estimate the capabilities of V4 and V41% MRI metrics for
temporal resolution [44], which employed the same CA injection discriminating RCB (and in-breast RCB) class, while the Spearman
protocol, including dose, injection rate, and injection vein (ante- correlation (SC) analysis was used to correlate V4 and V41% MRI
cubital vein). Following the TM and SSM fittings of the DCE-MRI metrics with RCB (and in-breast RCB) index values.
data, voxel-based parametric maps of the derived pharmacokinetic
parameters were generated. The mean whole tumor pharmacokinetic Results
parameter value estimated with each model at each visit was calculated As shown in Table 1, pathological analyses of the surgical specimens
as the average of single-slice ROI parameter values from the multiple revealed that 5 patients (5 primary tumors) achieved pCR following
image slices covering the whole tumor, weighted by the number of NACT, while the other 23 patients (24 primary tumors) all had pPR, or
voxels in each ROI. For each imaging metric, including pharmaco- non-pCR. The RCB class for each tumor is also presented in Table 1.
kinetic parameters and RECIST LD, the percent changes for later
visits relative to V1, V21% (V2 relative to V1), V31%, and V41%, Early Prediction of Pathologic Response
were calculated. Table 2 lists the mean ± SD whole tumor MRI metric values of the
pCR and non-pCR groups and the corresponding ULR C statistics
Pathological Analysis
values for early prediction of pCR vs. non-pCR. Only the absolute
The status of pathologic response (to NACT) for each breast tumor
pharmacokinetic parameters and the V21% and V31% changes with
was determined by pathological analysis of the post-NACT resection
C ≥ 0.8, representing good to excellent early predictors, are listed. The
specimen. The pathology parameters measured from the resection
C values for V21% and V31% RECIST LD changes are presented for
specimen under light microscopy include: cross sectional tumor size
in two dimensions (d1, d2, measured [in mm] grossly and confirmed
microscopically), estimated invasive tumor cellular density (finv),
number of involved lymph nodes (LN), and the greatest tumor Table 2. Early Prediction of Pathologic Response (pCR vs. non-pCR)

dimension (dmet) in the largest involved node. The Residual Cancer

MRI Metric pCR non-pCR ULR C value
Burden (RCB) index value was calculated using Equation (4)
published by Symmans et al [9]: Mean ± SD Mean ± SD P value

h i0:17 h   i0:17 V21% Ktrans(TM) −64% ± 9% −14% ± 41% .098 0.967

−77% ± −20% ±
RCB ¼ 1:4 f inv  ðd 1 d 2 Þ1=2 þ 4 1−0:75LN d met ð4Þ V21% kep(TM)
V21% Ktrans(SSM) −71% ±
9%
9% −16% ±
44%
50%
.050
.052
0.957
0.957
V21% τi 41% ± 26% −11% ± 25% .018 0.946
A complete pathologic response (pCR) is defined as the absence of V2 ve(SSM) 0.78 ± 0.10 0.60 ± 0.14 .073 0.897
V21% ve(TM) 80% ± 60% 35% ± 42% .026 0.880
residual invasive tumor (RCB = 0). A pathologic non-response (pNR)
V21% ve(SSM) 72% ± 41% 19% ± 28% .033 0.880
is defined as tumor cell density in resection specimen equal to or V2 ve(TM) 0.70 ± 0.37 0.29 ± 0.11 .018 0.864
greater than that in core biopsy specimen. Pathologic partial response V3 ve(TM) 0.63 ± 0.31 0.32 ± 0.15 .035 0.845
V3 ve(SSM) 0.81 ± 0.11 0.63 ± 0.15 .088 0.845
(pPR) is defined as findings intermediate between pCR and pNR.
V1 τi (s) 0.53 ± 0.16 0.81 ± 0.26 .047 0.826
Non-pCR includes both pPR and pNR and can be further stratified V31% ve(SSM) 80% ± 54% 27% ± 30% .041 0.810
into RCB classes based on RCB index values [9]: RCB-I: 0 b RCB ≤ V21% kep(SSM) −77% ± 12% −11% ± 94% .092 0.804
1.36; RCB-II: 1.36 b RCB ≤ 3.28; RCB-III: RCB N 3.28. Since the V31% ve(TM) 141% ± 115% 65% ± 85% .070 0.804
V31% RECIST LD −35% ± 21% −26% ± 20% .438 0.673
MRI metrics were measured from the primary breast tumor only, the V21% RECIST LD −15% ± 16% −10% ± 11% .320 0.609
in-breast component of RCB (the first term on the right hand side of ULR: univariate logistic regression; SD: standard deviation; P value: two-sample t test; TM: Tofts
Equation (4)) [23] was also computed for correlation with the MRI model; SSM: Shutter-Speed model.
12 DCE-MRI Evaluation of Breast Cancer Therapy Response Tudorica et al. Translational Oncology Vol. 9, No. 1, 2016

comparison. The V21% changes in tumor K trans (TM), K trans (SSM), There are no noticeable changes in the three parametric maps from V1 to
kep(TM), and τi provide excellent (C N 0.9) early discrimination of V2 for the non-pCR, while the considerable decrease in K trans(SSM) and
pCR and non-pCR, while V2 and V3 ve estimated from either increases in ve(SSM) and τi are clearly visible for the pCR.
pharmacokinetic model, and their corresponding V21% and V31%
changes, are good (0.8 b C b 0.9) markers for early prediction of MRI evaluation of RCB after NACT
response. The V21% and V31% changes in RECIST LD, however, Table 3A presents the ULR C statistics values of several
are poor (C b 0.7) early predictors of response. Except for the post-NACT MRI metrics for differentiating RCB and in-breast
RECIST LD changes, the differences in all other listed metrics RCB classes of 0, I, II, and III. Only the metrics with C ≥ 0.7 (fair or
between the two response groups are statistically significant (P b .05) better markers of RCB class) for in-breast RCB classes are listed,
or approaching significance. Other than τi at V1, which is a good which include absolute values of V4 K trans (SSM), K trans (TM),
early predictor of response with a C value of 0.826, no other kep(SSM), kep(TM), τi, and RECIST LD. The discriminative abilities
pre-NACT MRI metrics have a C value greater than 0.7. Figure 1A of these metrics are improved slightly when in-breast RCB class is
shows the mean ± SD column graph of V21% changes in K trans(TM), used in place of RCB class. There is not a single V41% change metric
K trans(SSM), kep(TM), ve(TM), ve(SSM), τi, and RECIST LD for that is at least a fair marker of RCB or in-breast RCB class.
the pCR and non-pCR groups. Note the substantially larger The SC coefficient, R, and the P value for statistical significance are
differences in the pharmacokinetic parameters between the two groups summarized in Table 3B for correlation between RCB (and in-breast
compared to RECIST LD. Figure 1B shows a similar graph for absolute RCB) index value and post-NACT MRI metric value. Only those
TM and SSM ve values of the two patient groups at V2 and V3. metrics with P b 0.1 are listed. As is the case of correlation with RCB
To demonstrate differences in early changes of tumor pharmaco- class, the strength of SC is generally improved slightly when the
kinetic parameters following NACT initiation, Figure 2 shows in-breast RCB index value is used, and no V41% metric has a SC P
examples of K trans(SSM), ve(SSM), and τi maps at V1 and V2 for a value smaller than 0.1. The positive correlations between RCB and
pCR (Figure 2A, patient 12) and a non-pCR (Figure 2B, patient 3) V4 RECIST LD, K trans(SSM), and K trans (TM) are statistically
patient. The color tumor parametric maps are in image slices significant (P b .05) for both RCB and in-breast RCB values. The
approximately through the center of the tumor in every case, and the inverse correlation between RCB and V4 τi is significant (P = .041)
color scales are kept the same through the two visits for each tumor. for the in-breast RCB, while approaching significance (P = .074) for the
RCB value. The positive correlation between RCB and V4 kep(SSM) is
near statistical significance for both the RCB and in-breast RCB values.
Figure 3 shows examples of linear regressions between the RCB (and
in-breast RCB) index values and the V4 MRI metrics of RECIST LD
(Figure 3A), K trans(SSM) (Figure 3B), and τi (Figure 3C).

Discussion
Consistent with several previous studies [26–35] using DCE-MRI to
assess breast cancer response to NACT, our initial findings from this
study of a 28-patient cohort show that changes in tumor functions as
measured by quantitative DCE-MRI are considerably more reliable
early predictors of pCR compared to changes in imaging tumor size
after only one of six or eight cycles of NACT. This suggests that
therapy-induced tumor functional changes precede changes in tumor
size. The percent changes of the K trans, ve, and kep parameters, as well
as the SSM-unique τi parameter, are good to excellent early predictors
of pathologic response. Additionally, the absolute values of ve after
one NACT cycle (V2) or at NACT midpoint (V3) are also good early
predictors of response. Imaging tumor size measurement under the
RECIST guidelines is the current standard of care for evaluation of
tumor response to treatment. However, our results reveal that changes
in RECIST LD after one NACT cycle, or even at midpoint of NACT
(Table 2), are poor early predictors of response. For example, under
the condition of 100% sensitivity for prediction of pCR (i.e.,
correctly classify all five pCRs in the study cohort), the specificities are
92% and 17% for V21% K trans(TM) and V21% RECIST LD,
respectively, meaning misclassification of only two out of 24
Figure 1. Column graphs of the (A) mean V21% change values of non-pCR tumors as pCRs when using V21% K trans(TM) as the
RECIST LD and several DCE-MRI metrics (K trans, ve, kep, and τi, imaging metric versus 20 out of 24 non-pCRs as pCRs when using
estimated from the TM and SSM pharmacokinetic analyses) and V21% RECIST LD as the imaging metric. The ability of minimally
(B) mean V2 and V3 ve values (TM and SSM) for the pCR (black
column) and non-pCR (gray column) patient groups. The error bar invasive imaging parameters, such as the DCE-MRI parameters, to
represents the standard deviation (SD). V21%: percent change of accurately provide early prediction of therapy response may have
MRI metric at visit 2 (V2, after one NACT cycle) relative to visit 1 (V1, profound importance in the emerging era of precision and
pre-NACT); V3: visit 3, midpoint of NACT. personalized medicine. Early identification of non-responders to a
Translational Oncology Vol. 9, No. 1, 2016 DCE-MRI Evaluation of Breast Cancer Therapy Response Tudorica et al. 13

Figure 2. V1 (pre-NACT) and V2 (after one NACT cycle) color parametric maps of K trans(SSM), ve(SSM), and τi from a pCR (A, left breast,
patient 12) and a non-pCR (B, right breast, patient 3) breast tumor. The maps were generated for tumor ROIs defined on multiple image
slices, and the ones on the image slice through the central portion of the tumor are displayed here. For each tumor, the color scale of each
DCE-MRI metric is kept the same between the two visits for easy visualization of NACT-induced changes.

therapy regimen may allow rapid decision making in adjusting the both tumor size and DCE-MRI functional parameters that can
treatment plan, e.g., changing drugs and/or undergoing surgery early, provide even more accurate measure of RCB.
to spare these patients from the morbidity caused by ineffective and Since cytotoxic drugs such as doxorubicin, cyclophosphamide, and
toxic chemotherapy agents. This will have significant positive impact paclitaxel were used throughout the NACT regimen for each patient
on healthcare cost savings and patient wellbeing. Using this cohort as studied, it is to be expected that increases in ve after one NACT cycle or
an example, if V21% K trans(TM) had been used as the imaging at NACT midpoint are found to be good early predictors of response.
marker for early prediction of therapy response in clinical care, 22 out The therapy-induced cancer cell death presumably leads to decrease in
of 24 non-pCRs would have been correctly classified after only one the extravascular, intracellular volume fraction, vi (≡ 1 – ve), which is
NACT cycle and could have been treated with different therapy dominated by decrease in cell density [41], and hence increase in ve.
regimens or stratified for novel therapy trials. However, at the early stage of NACT a decrease in vi is not necessarily
Accurate imaging assessment of RCB after NACT can lead to associated with a decrease in perfused tumor area, or contrast-enhanced
better staging for surgery and more informed decision making in MRI tumor size, the basis for RECIST LD measurement. This
breast conservation surgery versus mastectomy. Though RECIST LD hypothesis on why the ve percent change is a good early predictor of
is not a good early predictor of pathologic response, its value response while the RECIST LD percent change is a poor predictor
measured after NACT completion (V4) is indicative of the RCB class needs to be further tested in future studies. On the other hand, the
and index value (Tables 3A and 3B; Figure 3A). The post-NACT post-NACT (V4) RECIST LD is a better measure of RCB than ve. This
DCE-MRI parameters that are fair to good markers of RCB class and could be due to the fact that the RCB calculation (Equation (4))
correlate significantly with RCB index value include K trans (estimated includes dominant contribution from the product of cross sectional
from either of the two models) and τi. Unsurprisingly, since the MRI tumor sizes d1 and d2 measured from the resection specimen, while the
metrics were measured from the primary tumor only, the correlations association of vi (or 1 – ve) with finv is not as distinct - the former
with RCB are slightly improved when the in-breast RCB class and measures overall cell density [41] while the latter characterizes invasive
value are used (Tables 3A and 3B; Figure 3). With a larger patient cancer cell density.
cohort, meaningful multivariate analysis may be performed to The percent deceases in K trans and kep after one NACT cycle are
potentially identify a combination of imaging metrics including good to excellent early predictors of response, suggesting these

Table 3A. Discrimination of Post-NACT RCB Class Table 3B. Spearman Correlation of Post-NACT RCB Index Value with MRI Metric Value

MRI Metric ULR C value MRI Metric RCB In-breast RCB

RCB class In-breast RCB class R P R P

trans
V4 K (SSM) 0.801 (0.680, 0.922) 0.837 (0.734, 0.940) V4 RECIST LD 0.532 .009 0.485 .019
V4 Ktrans(TM) 0.797 (0.669, 0.925) 0.833 (0.708, 0.958) V4 Ktrans(SSM) 0.463 .022 0.643 .001
V4 τi 0.783 (0.647, 0.919) 0.792 (0.634, 0.950) V4 Ktrans(SM) 0.463 .022 0.618 .002
V4 RECIST LD 0.727 (0.586, 0.868) 0.732 (0.538, 0.870) V4 τi −0.380 .074 − 0.429 .041
V4 kep(SSM) 0.697 (0.540, 0.854) 0.719 (0.520, 0.890) V4 kep(SSM) 0.366 .078 0.376 .070
V4 kep(TM) 0.694 (0.553, 0.835) 0.705 (0.527, 0.857)
R: Spearman correlation coefficient; P b 0.05 indicates statistically significant correlation.
Imaging results are missing from a pCR patient (patient 20), who declined the V4 MRI study due to Imaging results are missing from a pCR patient (patient 20), who declined the V4 MRI study due to
personal reasons. The values in the parentheses are 95% confidence intervals (CIs). personal reasons.
14 DCE-MRI Evaluation of Breast Cancer Therapy Response Tudorica et al. Translational Oncology Vol. 9, No. 1, 2016

Figure 3. Scatter plots of pathologically measured RCB and in-breast RCB index values (from post-NACT resection specimens) against
post-NACT (V4) MRI metrics: (A) RECIST LD, (B) K trans(SSM), and (C) τi. The straight line in each panel represents a linear regression. The
Spearman correlation coefficient R and P values for the three imaging metrics are listed in Table 3B and shown in each panel. Note the
inverse relationship between RCB (and in-breast RCB) and τi. Imaging results are missing from a pCR patient (patient 20), who declined
the V4 MRI study due to personal reasons.

imaging metrics of perfusion and permeability may be sensitive to non-pCRs that, by V3 (NACT midpoint), the percent changes
parameters for prediction of NACT response. The therapy-induced (relative to baseline, V31%) in these metrics of the non-pCRs, though still
changes in these microvascular properties are probably due to NACT smaller, draw near to those of the pCRs. Therefore, it is important to
secondary effects [14], since none of the standard of care or detect early microvascular changes with DCE-MRI, as they are better
experimental drugs used in this cohort is known to be antiangiogenic. indicators of complete response to NACT than later changes.
It has been suggested [45] that cytotoxic chemotherapy agents may The τi parameter is used to account for the effects of
affect tumor vasculature by interfering with endothelial cell function transcytolemmal water exchange kinetics and unique to the SSM
without causing endothelial cell death or interfering with a specific method. A recent NMR spectroscopy study of yeast cell suspension
portion of the angiogenic cascade. The results from this study and [40] shows that the reciprocal of τi, kio (≡ 1/τi), the first-order rate
others [26–34] indicate that it is the early changes in K trans and kep, constant for equilibrium cellular water efflux, is positively associated
not the perfused tumor area as measured by RECIST LD, that are with cellular ATP levels. The in vivo association of cellular ATP
good predictors of therapy response. This observation supports the decrease with kio decrease was demonstrated by a DCE-MRI and 31P
hypothesis that the chemotherapeutics effects tumor vasculature MR spectroscopy study of a murine melanoma model treated with
without causing endothelial cell death [45]. One interesting finding is lonidamine [46]. A series of enzymatic and genetic manipulations on
that though the V31% change in ve is still a good early predictor of cell suspensions, perfused tissue, in vivo animal models, and human
response (Table 2), the V31% changes in K trans and kep are degraded data have shown that kio measures the homeostatic turnover of the cell
to only fair predictors of response (not shown). For example, V31% membrane Na +,K +-ATPase [NKA] [41]. Previously, it has not been
K trans(TM) and V31% K trans(SSM) have ULR C values of 0.756 and possible to measure homeostatic NKA activity in vivo. In this study
0.750, respectively. The decrease in the predictive ability is because the V21% change in breast tumor τi is an excellent early predictor of
the pCRs have such larger decreases in K trans and kep at V2 compared pathologic response (Table 2). The pCRs show significant increase in
Translational Oncology Vol. 9, No. 1, 2016 DCE-MRI Evaluation of Breast Cancer Therapy Response Tudorica et al. 15

τi (or decrease in kio) compared to non-pCRs after one NACT cycle, with a larger patient population, especially for new findings specific to
consistent with a substantial decrease in tumor metabolic activities this study such as changes in ve and τi as early predictors of response.
early after NACT initiation being generally a good indicator of Second, as a result of small sample size, the imaging results are not
complete response. It is interesting to note that the pCR group has stratified by receptor status-based breast cancer molecular subtypes in
smaller pre-NACT (V1) τi values and thus greater metabolic turnover correlation with pathology endpoints. The number of patients is
than the non-pCR group (Table 2), and V1 τi is the best predictor of simply not adequate to draw meaningful conclusions in early
response (a good metric with ULR C = 0.826) of all the V1 MRI discrimination of pCR and non-pCR or evaluation of RCB for
metrics. The post-NACT (V4) τi is well correlated with the RCB each breast tumor subtype, and make comparisons between the
index value in an inverse relationship (Table 3B and Figure 3C). subtypes. As such, the initial findings reported here reflect the average
These findings suggest that, over the course of the entire NACT results from a general breast cancer population undergoing NACT.
regimen, the good responders have greater increases in τi, or decreases With continuing subject accrual, we will be able to examine and
in metabolic activity, than the poor responders. In fact, from V1 to compare the utility of DCE-MRI for evaluation of therapy response
V4 the mean τi value of the pCR group is increased from 0.53 to in each breast cancer subtype in the future. Additionally, the small
1.45 s, while that of the non-pCR group remains stable (0.81 to sample size also precludes meaningful multivariate analysis of the
0.82 s). The importance of τi as an imaging parameter for evaluation MRI metrics for further improvement in assessing therapy response.
of breast cancer response to NACT is further confirmed by a Third, mean tumor MRI metric values were used in this study for
pre-clinical DCE-MRI study of a genetically engineered mouse model correlation with the pathology endpoints. It is well known that
of human breast tumor [47], which shows τi increase with almost no malignant tumors are heterogeneous in nature [41] and responses to
vi change following treatment with an experimental targeted, treatment are likely heterogeneous as well. However, the heteroge-
non-cytotoxic drug. neity in breast tumor functional changes in response to NACT was
The DCE-MRI data acquired in this study were analyzed with not captured in computing mean DCE-MRI parameter values.
both the TM and SSM. For the three parameters that can be Recent studies show that texture analysis of tumor heterogeneity
estimated with both models, K trans , ve, and kep, the percent changes manifest in either raw image data [49–51] or parametric maps of
of these parameters in the early stage of NACT and the absolute kinetic features [52,53] can be a valuable tool for evaluation of breast
values after NACT have similar capabilities for early prediction of cancer therapy response. With voxel-based parametric maps of
response and evaluation of RCB, respectively, when comparing the DCE-MRI metrics already generated in this study, texture analysis of
two models. This is likely due to the fact that the underestimation of these maps could potentially be performed. Measurement and
the K trans and ve parameters in malignant tumors by the TM relative integration of changes in both mean values and texture features of
to the SSM is generally systematic [27,44]. The systematic parameter DCE-MRI metrics may further improve the robustness of quanti-
variations between the two models were largely cancelled in percent tative DCE-MRI for assessment of therapy response.
change calculation, or caused parameter values of each tumor to shift In conclusion, in this DCE-MRI study of 29 primary breast
in the same direction when estimated from one pharmacokinetic tumors undergoing NACT, we have shown that changes in
model or the other. As a result, TM and SSM analyses performed quantitative functional MRI metrics estimated by either TM or
equally well in early prediction and evaluation of therapy response. Li SSM analysis of DCE-MRI data provide substantially better early
et al. [28] reported similar findings for early prediction of breast prediction of pathologic response to NACT than changes in imaging
cancer response to NACT. Nonetheless, water exchange across tissue tumor size as measured by RECIST LD. Along with post-NACT
compartments is a real physiological phenomenon. Therefore, the RECIST LD, several post-NACT DCE-MRI parameters are good
SSM analysis approach should be used when T1-weighted DCE-MRI markers of RCB. The TM and SSM analyses perform equally well for
data are acquired with a protocol that is sensitive to water exchange, early prediction of response and evaluation of RCB. However, the
which usually also results in better signal-to-noise ratio [48]. SSM method provides additional assessment of tumor metabolic
Furthermore, the SSM analysis allows estimation of the metabolic activity changes in response to NACT, and the SSM-unique τi
activity imaging metric kio, adding a metabolic dimension to parameter is the only pre-therapy MRI metric that provides good early
DCE-MRI, which is conventionally considered only as a functional prediction of response. Future translation of quantitative DCE-MRI into
imaging method for assessment of tissue microvasculature. As clinical practice may help facilitate personalized treatment regimens for
discussed above, the τi (or kio− 1) parameter is a very good marker for individual breast cancer patients and more informed decision making in
early prediction of response and accurate assessment of residual breast conservation surgery vs. mastectomy.
disease. The ability to characterize synergistic microvascular proper-
ties and cellular energetic metabolism simultaneously may harbor Acknowledgements
great promise for SSM DCE-MRI as an imaging tool to study the The authors thank the breast cancer patients who voluntarily
tumor microenvironment, and its response to treatment. participated in this research study and Mr. William Woodward for
This study has several major limitations. First, with only 29 assistance in breast DCE-MRI data acquisition.
primary breast tumors, the sample size of the study cohort is small.
The discriminative ability of a metric tends to be overestimated when References
the sample size is small. This is possibly the reason why the ULR C [1] Hayes DF and Schott AF (2015). Neoadjuvant chemotherapy: what are the benefits
values of several DCE-MRI metrics obtained in this study for early for the patient and for the investigator? J Natl Cancer Inst Monogr 2015, 36–39.
prediction of breast cancer response to NACT are greater than the [2] Schott AF and Hayes DF (2012). Defining the benefits of neoadjuvant
chemotherapy for breast cancer. J Clin Oncol 30, 1747–1749.
equivalent area under the receiver operating characteristic curve values [3] Fisher B, Bryant J, Wolmark N, Mamounas E, Brown A, and Fisher ER, et al
reported by similar studies with larger cohort sizes [22,23,26,28,29]. (1998). Effect of preoperative chemotherapy on the outcome of women with
Thus, it is important to validate the initial findings from this cohort operable breast cancer. J Clin Oncol 16, 2672–2685.
16 DCE-MRI Evaluation of Breast Cancer Therapy Response Tudorica et al. Translational Oncology Vol. 9, No. 1, 2016

[4] Mauri D, Pavlidis N, and Ioannidis JP (2005). Neoadjuvant versus adjuvant systemic response to neoadjuvant chemotherapy by semi-quantitative analysis of high-temporal
treatment in breast cancer: a meta-analysis. J Natl Cancer Inst 97, 188–194. resolution DCE-MRI: preliminary results. Magn Reson Imaging 31, 1457–1464.
[5] Redden MH and Fuhrman GM (2013). Neoadjuvant chemotherapy in the [23] Hylton NM, Blume JD, Bernreuter WK, Pisano ED, Rosen MA, Morris EA,
treatment of breast cancer. Surg Clin North Am 93, 493–499. Weatherall PT, Lehman CD, Newstead GM, and Polin S, et al (2012). Locally
[6] Bonnefoi H, Litiere S, Piccart M, MacGrogan G, Fumoleau P, Brain E, Petit T, advanced breast cancer: MR imaging for prediction of response to neoadjuvant
Rouanet P, Jassem J, and Moldovan C, et al (2014). Pathological complete chemotherapy – results from ACRIN 6657/I-SPY trial. Radiology 263, 663–672.
response after neoadjuvant chemotherapy is an independent predictive factor [24] Johansen R, Jensen LR, Rydland J, Goa PE, Kvistad KA, Bathen TF, Axelson
irrespective of simplified breast cancer intrinsic subtypes: a landmark and DE, Lundgren S, and Gribbestad IS (2009). Predicting survival and early clinical
two-step approach analyses from the EORTC 10994/BIG 1-00 phase III trial. response to primary chemotherapy for patients with locally advanced breast
Ann Oncol 25, 1128–1136. cancer using DCE-MRI. J Magn Reson Imaging 29, 1300–1307.
[7] von Minckwitz G, Untch M, Blohmer JU, Costa SD, Eidtmann H, Fasching PA, [25] Martincich L, Montemurro F, De Rosa G, Marra V, Ponzone R, Cirillo S, Gatti M,
Gerber B, Eiermann W, Hilfrich J, and Huober J, et al (2012). Definition and Biglia N, Sarotto I, and Sismondi P, et al (2004). Monitoring response to primary
impact of pathologic complete response on prognosis after neoadjuvant chemotherapy in breast cancer using dynamic contrast-enhanced magnetic resonance
chemotherapy in various intrinsic breast cancer subtypes. J Clin Oncol 30, imaging. Breast Cancer Res Treat 83, 67–76.
1796–1804. [26] Li X, Kang H, Arlinghaus LR, Abramson RG, Chakravarthy AB, Abramson VG,
[8] Kong X, Moran MS, Zhang N, Haffty B, and Yang Q (2011). Meta-analysis Farley J, Sanders M, and Yankeelov TE (2014). Analyzing spatial heterogeneity
confirms achieving pathological complete response after neoadjuvant chemo- in DCE- and DW-MRI parametric maps to optimize prediction of pathologic
therapy predicts favorable prognosis for breast cancer patients. Eur J Cancer 47, response to neoadjuvant chemotherapy in breast cancer. Transl Oncol 7, 14–22.
2084–2090. [27] Huang W, Li X, Chen Y, Li X, Chang MC, Oborski MJ, Malyarenko DI, Muzi
[9] Symmans WF, Peintinger F, Hatzis C, Rajan R, Kuerer H, Valero V, Assad L, M, Jajamovich GH, and Fedorov A, et al (2014). Variations of dynamic
Poniecka A, Hennessy B, and Green M, et al (2007). Measurement of residual contrast-enhanced magnetic resonance imaging in evaluation of breast cancer
breast cancer burden to predict survival after neoadjuvant chemotherapy. J Clin therapy response: a multicenter data analysis challenge. Transl Oncol 7, 153–166.
Oncol 25, 4414–4422. [28] Li X, Arlinghaus LR, Ayers GD, Chakravarthy AB, Abramson RG, Abramson
[10] Gonzalez-Angulo AM, Morales-Vasquez F, and Hortobagyi GN (2007). VG, Atuegwu N, Farley J, Mayer IA, and Kelley MC, et al (2014). DCE-MRI
Overview of resistance to systemic therapy in patients with breast cancer. Adv analysis methods for predicting the response of breast cancer to neoadjuvant
Exp Med Biol 608, 1–22. chemotherapy: pilot study findings. Magn Reson Med 71, 1592–1602.
[11] Zambetti M, Mansutti M, Gomez P, Lluch A, Dittrich C, Zamagni C, Ciruelos E, [29] Tateishi U, Miyake M, Nagaoka T, Terauchi T, Kubota K, Kinoshita T, Daisaki
Pavesi L, Semiglazov V, and De Benedictis E, et al (2012). Pathological complete H, and Macapinlac HA (2012). Neoadjuvant chemotherapy in breast cancer:
response rates following different neoadjuvant chemotherapy regimens for operable prediction of pathologic response with PET/CT and dynamic contrast-enhanced
breast cancer according to ER status, in two parallel, randomized phase II trials with an MR imaging—prospective assessment. Radiology 263, 53–63.
adaptive study design (ECTO II). Breast Cancer Res Treat 132, 843–851. [30] Li SP, Makris A, Beresford MJ, Taylor NJ, Ah-See MLW, Stirling JJ, d’Arcy JA,
[12] Therasse P, Arbuck SG, Eisenhauer EA, Wanders J, Kaplan RS, Rubinstein L, Collins DJ, Kozarski R, and Padhani AR (2011). Use of dynamic
Verweij J, Van Glabbeke M, van Oosterom AT, and Christian MC, et al (2000). contrast-enhanced MR imaging to predict survival in patients with primary
New guidelines to evaluate the response to treatment in solid tumors. European breast cancer undergoing neoadjuvant chemotherapy. Radiology 260, 68–78.
Organization for Research and Treatment of Cancer, National Cancer Institute [31] Jensen LR, Garzon B, Heldahl MG, Bathen TF, Lundgren S, and Gribbestad IS
of the United States, Nation Cancer Institute of Canada. J Natl Cancer Inst 92, (2011). Diffusion-weighted and dynamic contrast-enhanced MRI in evaluation
205–216. of early treatment effects during neoadjuvant chemotherapy in breast cancer
[13] O’Connor JPB, Jackson A, Parker GJM, Roberts C, and Jayson GC (2012). patients. J Magn Reson Imaging 34, 1099–1109.
Dynamic contrast-enhanced MRI in clinical trials of antivascular therapies. Nat [32] Ah-See MLW, Makris A, Taylor NJ, Harrison M, Richman PI, Burcombe RJ,
Rev Clin Oncol 9, 167–177. Stirling JJ, d’Arcy JA, Collins DJ, and Pittam MR, et al (2008). Early changes in
[14] Leach MO, Morgan B, Tofts PS, Buckley DL, Huang W, Horsfield MA, functional dynamic magnetic resonance imaging predict for pathologic response
Chenevert TL, Collins DJ, Jackson A, and Lomas D, et al (2012). Imaging to neoadjuvant chemotherapy in primary breast cancer. Clin Cancer Res 14,
vascular function for early stage clinical trials using dynamic contrast-enhanced 6580–6589.
magnetic resonance imaging. Eur Radiol 22, 1451–1464. [33] Yankeelov TE, Lepage M, Chakravarthy A, Broome EE, Niermann KJ, Kelley
[15] Padhani AR and Miles KA (2010). Multiparametric imaging of tumor response MC, Meszoely I, Mayer IA, Herman CR, and McManus K, et al (2007).
to therapy. Radiology 256, 348–364. Integration of quantitative DCE-MRI and ADC mapping to monitor treatment
[16] Harry VN, Semple SI, Parkin DE, and Gilbert FJ (2010). Use of new imaging response in human breast cancer: initial results. Magn Reson Imaging 25, 1–13.
techniques to predict tumor response to therapy. Lancet Oncol 11, 92–102. [34] Padhani AR, Hayes C, Assersohn L, Powles T, Makris A, Suckling J, Leach MO,
[17] Lobbes MBI, Prevos R, Smidt M, Tjan-Heijnen VCG, van Goethem M, and Husband JE (2006). Prediction of clinicopathologic response of breast cancer
Schipper R, Beets-Tan RG, and Wildberger JE (2013). The role of magnetic to primary chemotherapy at contrast-enhanced MR imaging: initial clinical
resonance imaging in assessing residual disease and pathologic complete response results. Radiology 239, 361–374.
in breast cancer patients receiving neoadjuvant chemotherapy: a systematic [35] Pickles MD, Lowry M, Menton DJ, Gibbs P, and Turnbull LW (2005). Role of
review. Insights Imaging 4, 163–175. dynamic contrast enhanced MRI in monitoring early response of locally advanced
[18] Marinovich ML, Sardanelli F, Ciatto S, Mamounas E, Brennan M, Macaskill P, breast cancer to neoadjuvant chemotherapy. Breast Cancer Res Treat 91, 1–10.
Irwig L, von Minckwitz G, and Houssami N (2012). Early prediction of [36] Tofts PS and Kermode AG (1991). Measurement of the blood-brain barrier
pathologic response to neoadjuvant chemotherapy in breast cancer: systematic permeability and leakage space using dynamic MR imaging. Magn Reson Med 17,
review of the accuracy of MRI. Breast 21, 669–677. 357–367.
[19] Wu LM, Hu JN, Gu HY, Hua J, Chen J, and Xu JR (2012). Can [37] Tofts PS, Brix G, Buckley DL, Evelhoch JL, Henderson E, Knopp MV, Larsson
diffusion-weighted MR imaging and contrast-enhanced MR imaging precisely HB, Lee TY, Mayr NA, and Parker GJ, et al (1999). Estimating kinetic
evaluate and predict pathological response to neoadjuvant chemotherapy in parameters from dynamic contrast-enhanced T1-weighted MRI of a diffusable
patients with breast cancer? Breast Cancer Res Treat 135, 17–28. tracer: standardized quantities and symbols. J Magn Reson Imaging 10, 223–232.
[20] Pickles MD, Lowry M, Manton DJ, and Turnbull LW (2015). Prognostic value [38] Yankeelov TE, Rooney WD, Li X, and Springer CS (2003). Variation of the
of DCE-MRI in breast cancer patients undergoing neoadjuvant chemotherapy: a relaxographic “Shutter-Speed” for transcytolemmal water exchange affects the
comparison with traditional survival indicators. Eur Radiol 25, 1097–1106. CR bolus-tracking curve shape. Magn Reson Med 50, 1151–1169.
[21] Woolf DK, Padhani AR, Taylor NJ, Gogbashian A, Li SP, Beresford MJ, Ah-See ML, [39] Li X, Rooney WD, and Springer CS (2005). A unified pharmacokinetic theory
Stirling J, Collins DJ, and Makris A (2014). Assessing response in breast cancer with for intravascular and extracellular contrast agents. Magn Reson Med 54,
dynamic contrast-enhanced magnetic resonance imaging: are signal intensity-time 1351–1359 [Erratum. Magn Reson Med 2006;55:1217.].
curves adequate? Breast Cancer Res Treat 147, 335–343. [40] Zhang Y, Poirier-Quinot M, Springer CS, and Balschi JA (2011). Active
[22] Abramson RG, Li X, Hoyt TL, Su PF, Arlinghaus LR, Wilson KJ, Abramson VG, trans-plasma membrane water cycling in yeast is revealed by NMR. Biophys J
Chakravarthy AB, and Yankeelov TE (2013). Early assessment of breast cancer 101, 2833–2842.
Translational Oncology Vol. 9, No. 1, 2016 DCE-MRI Evaluation of Breast Cancer Therapy Response Tudorica et al. 17

[41] Springer CS, Li X, Tudorica LA, Oh KY, Roy N, Chui SYC, Naik AM, [47] Springer CS, Li X, Jayatilake ML, Pike MM, Rooney WD, Sears RC, and Huang
Holtorf ML, Afzal A, and Rooney WD, et al (2014). Intratumor mapping of W (2015). Metabolic imaging of early tumor therapy. Proc Intl Soc Magn Reson
intracellular water lifetime: metabolic images of breast cancer? NMR Biomed Med 23, 3860.
27, 760–773. [48] Li X, Huang W, and Rooney WD (2012). Signal-to-noise ratio, contrast-to-noise
[42] Tudorica LA, Oh KY, Roy N, Kettler MD, Chen Y, Hemmingson SL, Afzal A, ratio, and pharmacokinetic modeling considerations in dynamic-contrast-
Grinstead JW, Laub G, and Li X, et al (2012). A feasible high spatiotemporal enhanced magnetic resonance imaging. Magn Reson Imaging 30, 1313–1322.
resolution breast DCE-MRI protocol for clinical settings. Magn Reson Imaging [49] Ahmed A, Gibbs P, Pickles M, and Turbull L (2013). Texture analysis in
30, 1257–1267. assessment and prediction of chemotherapy response in breast cancer. J Magn
[43] Huang W, Wang Y, Panicek DM, Schwartz LH, and Koutcher JA (2009). Reson Imaging 38, 89–101.
Feasibility of using limited-population-based average R10 for pharmacokinetic [50] Teruel JR, Heldahl MG, Goa PE, Pickles M, Lundgren S, Bathen TF, and Gibbs P
modeling of osteosarcoma dynamic contrast-enhanced MRI data. Magn Reson (2014). Dynamic contrast-enhanced MRI texture analysis for pretreatment prediction
Imaging 27, 852–858. of clinical and pathological response to neoadjuvant chemotherapy in patients with
[44] Huang W, Tudorica LA, Li X, Thakur SB, Chen Y, Morris EA, Tagge IJ, locally advanced breast cancer. NMR Biomed 27, 887–896.
Korenblit M, Rooney WD, and Koutcher JA, et al (2011). Discrimination of [51] Parikh J, Selmi M, Charles-Edwards G, Glendenning J, Ganeshan B, Verma H,
benign and malignant breast lesions by using shutter-speed dynamic Mansi J, Harries M, Tutt A, and Goh V (2014). Changes in primary breast
contrast-enhanced MR imaging. Radiology 261, 394–403. cancer heterogeneity may augment midtreatment MR imaging assessment of
[45] Miller KD, Sweeney CJ, and Sledge GW (2001). Redefining the target: response to neoadjuvant chemotherapy. Radiology 272, 100–112.
chemotherapeutics as antiangiogenics. J Clin Oncol 19, 1195–1206. [52] Golden DI, Lipson JA, Telli ML, Ford JM, and Rubin DL (2013). Dynamic
[46] Nath K, Paudyal R, Nelson DS, Pickup S, Zhou R, Leeper DB, Heitjan DF, contrast-enhanced MRI-based biomarkers of therapeutic response in triple-
Springer CS, Poptani H, and Glickson JD (2014). Acute changes in negative breast cancer. J Am Med Inform Assoc 20, 1059–1066.
cellular-interstitial water exchange rate in DB-1 melanoma xenografts after [53] Ashraf A, Gaonkar B, Mies C, DeMichele A, Rosen M, Davatzikos C, and Kontos D
lonidamine administration as a marker of tumor energetics and ion transport. (2015). Breast DCE-MRI kinetic heterogeneity tumor markers: preliminary
Proc Intl Soc Magn Reson Med 22, 2757. associations with neoadjuvant chemotherapy response. Transl Oncol 8, 154–162.