PET-CT and DWI-MRI in evaluating and predicting response to neoadjuvant chemoradiotherapy in esophageal cancer patients: PET/MRI-OES-1 study

PET-CT is currently used as standard method in both staging and determining of
treatment response in esophageal cancer patients. For staging and determining
response a standard PET-CT will be performed with additional endoscopic
ultrasonography (EUS) of suspected nodes seen on PET-CT or during EUS. In
20-36% of the patients with esophageal cancer (EC) treated with neo-adjuvant
chemoradiation (CRT), no residual tumor (ypT0N0) can be detected after curative
intended surgery. On the other hand nearly 40% of the patients after CRT had
locoregional failure in the CROSS trial and our study group found residual
tumor in approximately 15% of the patients outside the radiation target volumes
(CTV) after CRT.
Although, 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) and
computed tomography (CT), in particular PET/CT, is able to distinct responders
from non-responders, it still misses significant clinical evidence. Recently,
whole-body diffusion-weighted imaging (DWI) magnetic resonance imaging (MRI)
has shown a potential benefit, which might be enhance by combining both imaging
methods. In assessing treatment outcome in cancer patients, FDG-PET and
whole-body DWI-MRI should be regarded as complementary techniques, as they
address complete different biophysical tissue properties (glucose metabolism
vs. cellular density and vascularity of the tumor).
PET has limitations similar to CT in distinguishing tumor from inflammatory
tissue. The value of adding DWI-MRI (Siemens Healthcare) is an increase in
overall sensitivity up to 94% with fused image data sets of PET-CT with a rapid
total acquisition time of around 30-40 min for PET/MRI depending on the area of
interest. The potential benefit of using both PET/CT and DWI-MRI is a more
precise anatomic and metabolic correlation with FDG-avid lesions and better
assessment of post-treatment changes with reasonable accuracy, especially
regarding nodal staging.
For both techniques it is claimed that they can play an important role in
selection of patients with clinical complete response (cCR) and indirectly with
pathologic complete response (pCR) after CRT, and also in patients with no
clinically effective response, early during CRT. However, the exact role and
complementary effects of both techniques are still unknown.
For an appropriate judgment of nodal response to CRT, secure standard
endoscopic ultrasonography (EUS) with fine needle aspiration (FNA) of suspected
lesions seen on PET/CT or DWI/MRI or during EUS, will be performed at primary
staging and after CRT.
DWI-MRI is based on movement (Brownian motion) of water molecules, which
enables identification of tissue with high cellular density, such as in
malignant tumors. The apparent diffusion coefficient (ADC), as a measure of
water diffusion, is calculated by differences in the rate of changes in signal
intensity at various b values. The degree of diffusion-weighting, the so-called
b-value, adjusts the strength and duration of the diffusion gradients. The term
"apparent" is used because the measured value does not only indicate diffusion,
but also reflects capillary perfusion.
The b value parameter requires optimization based on body part, pathology, and
radiologist preference. So, the ADC value is influenced by the b value.
Perfusion, which is more present at a low b-value, has a larger effect on the
ADC value than diffusion. If the b value is zero, there is no diffusion
weighting and the images are similar to T2-weighted images. Conversely, if the
b value is high, there is heavy diffusion-weighting. At a high b-value the ADC
value mainly depends on diffusion, leading to a low ADC.
DWI-MRI can be effective in the pre-treatment prediction of treatment outcome.
Tumor hypoxia is known to mediate chemoradiation resistance, leading to
selection of aggressive tumor cell clones with their capacity to evade tumor
microenvironment by increasing the anaerobic glycolysis and angiogenesis. ADCs
can be used to determine the likelihood of tumor response to treatment, in
which a high ADC before treatment has shown to predict an unfavorable treatment
response, probably because a high ADC reflects the presence of tumor necrosis
and low oxygen tensions. On the other hand as ADC indirect measures tissue
density, ADC values are low in viable tumor and increase with treatment-induced
apoptosis and necrosis.
Tumors presenting with a lower ADC value on presentation generally respond
better to treatment. A recent study has shown that an increased ADC in patients
during and after CRT could be used to predict early pathological response i.e
discrimination of *responders and non-responders* to CRT. So, a low ADC value
is predictive of good treatment response, while increased ADC values post CRT
suggest reduced cell burden. Response in ADC seems to be more sensitive than
the change in tumor volume size.

1. The primary aim of the proposed research is to compare the value of current
best available imaging method (PET-CT) with recently evolved functional imaging
with DWI-MRI alone or in combination with PET/CT in determining response to
neo-adjuvant chemo-radiotherapy (CRT) in potentially curable esophageal cancer
(EC) patients.
Therefore, we will quantified the baseline FDG uptake (max. standardized uptake
value; SUVmax), by FDG-PET and SUV changes (*SUV) during and after treatment
with CRT. As ADC is an indirect measure of tissue density, DWI-MRI may provide
an early marker of response to CRT by detecting both changes in perfusion
(anti-angiogenic effect) and changes in tissue diffusivity (necrosis).
We will determine whether DWI-MRI alone or combined with PET/CT performed at
staging (pretreatment i.e < 6 weeks before start CRT), early during the course
of CRT (i.e 2 weeks after CRT initiation) and late before surgery ( i.e <= 2
weeks before surgery ) can be used to predict early and late
persistent/progressive disease or clinical complete response, respectively.

2. The secondary aim is to determine disease free survival (DFS) after
different response types (complete or partial and no response) at T1 (early
volumetric and functional response at 2 wks after start of CRT) and at T2 (post
CRT=histological response / both volumetric and functional) comparing
pathologic status with blinded reading from two independent
radiologists/nuclear medicine, in order to select patients better for
*definitive* surgery or wait and see policy with eventually *salvage* surgery
(in case of residual or recurrent disease) in the near future. To answer this
question we will need the information obtained by combined PET/CT and DWI-MRI,
including clinical responses (complete = cCR and partial = cPR) at the
different time periods (T1 and T2). Furthermore secure EUS of suspected lesions
on both, PET and DWI or during EUS will be biopsied or punctured for
cytological examination.
Moreover, by comparing PET-CT with DWI-MRI, alone or combined, in correlation
with the pathological examination of the resected esophageal specimen,
according to the RESPECT protocol, we will be able to evaluate the accuracy of
the given radiotherapy by examination of the involved target volumes (Gross =
GTV and clinical = CTV). Furthermore, incorporation of (DWI)-MRI into the
radiotherapy target volume delineation process might improve the accuracy of
the delineation. With this respect both imaging modalities will be performed in
a radiotherapy treatment position.
We will also assess whether TPV=Total proliferative volume (TPV=SUV x TV) will
be a reliable estimation of tumor volume based on the texture features (TF)
estimation in TF index. By using texture features,, including repeating pattern
of local variations in image intensity and the partition of images into regions
of interest in classifying the regions. In this way we can achieve a spatial
arrangement of colors and intensities in those images. This can be performed by
using static and dynamic scans.

Prospective observational non-randomized single institution, comparison pilot
study, that can divide in two main parts.
Part I: consists of a limited exploring study to determine and validate the
most appropriate DWI-MRI b-value, both low (0- 100 sec/mm2) and high (800-1000
sec/mm2) in detecting optimal imaging of EC. The b-value identifies
measurement's sensitivity to diffusion and determines the strength and duration
of the diffusion gradients. This study will be performed in three to five
patients.
Part II: In part two, after we have obtained the optimal b-values, we will use
DWI-MRI to evaluate the additional effect on staging and response evaluation in
15 to 20 patients (pilot study). Furthermore, we will evaluated the effect of
the additional use of DWI-MRI for target volume delineation and radiotherapy
planning.
Depending on the results of this pilot study, the study will be continued (Part
III) after determining a sample size analysis for the genuine study size (new
application or amendment will be submitted at that time ). In this latest study
we will determine the adequacy of PET-CT and DWI-MRI in assessing clinical
tumor response to pathological response in order to select patients better in a
group of early non-responders (early response evaluation; 2 weeks after start
of CRT) and a group of clinical incomplete/partial (cPR) and complete
responders (cCR) vs. pathological complete responders (pCR). In that way
surgery can be planned better in estimating the correlation between DWI-MRI and
pathologic outcome more accurately (monitoring tumor response). Moreover, we
will evaluate whether DWI-MRI will be of additional clinical value in
determining a more accurate radiation (Gross and Clinical) target volumes (GTV
and CTV).

In this study DWI-MRI is added on top of routine work-up and treatment. Adverse
events of MRI are not to be expected. Therefore the burden for the patient is
an extra diagnostic procedure in a machine with a small bore, which may trigger
feelings of claustrophobia.

Relevance
This study will provide insight into the value of response evaluation, both in
early
(<= 2 weeks after the start) and late (<= 2 weeks after the end) of CRT. The
study will give good insight into the usefulness of DWI/MRI combined with
PET-CT as a tool in clinical practice in determining pCR. The responses on
these imaging methods can be used as surrogates or translate to objective cCR
with pCR and subsequently in selecting patients for an *early resection* or a
*wait-see policy*
The additional information of DWI-MRI and PET-CT might also improve target
volume delineation (GTV/CTV) and consequently the radiotherapy treatment
planning and irradiation. Once the program has been developed, it would be
introduced regionally (Managed Clinical Network; MCN esophagus) and nationwide
(DUCA).