Non-invasive nodal staging in breast cancer with MRI Lymphography using gadofosveset; a pilot- study.

Lymph node status is one of the most important prognostic factors in breast
cancer and is of particular value in choosing adjuvant therapy.
About 30% of breast cancer patients have histopathologically positive axillary
lymph nodes.

Nowadays, all breast cancer patients undergo a mammography, breast ultrasound,
histology, and a breast MRI. After the diagnosis breast cancer has been
determined, the current nodal staging consists of an axillary ultrasound and
possible cytologic or histologic punction followed by sentinel lymph node
biopsy (SLNB) and/or an axillary lymph node dissection (ALND). After
introduction of the MRL, first the accuracy has to be established. If MRL is
equally accurate to the current nodal staging workup (most important is no
false positive cases), skipping partly the SLNB and its histology might make
this workup more efficient. So, accuracy will be researched by comparing the
current, invasive, nodal-staging with non-invasive MRL in relation to the
golden standard, histological examination of the SLNB and/or ALND.

In addition, MRL might be able to detect extra-axillary lymph node metastases
which could induce a change in treatment regimen for an unknown number of
patients. Studies on a positive parasternal SLNB show a change in adjuvant
treatment in 0.9% of the patients. Furthermore, it is known that the SLNB has a
false negative rate of 9.8%. In this pilot study we will only investigate the
axillary region with MRL. We know that a possible reason for a false negative
SNLB is a low number of specimens removed during the SLNB procedure. MRL might
be able to detect the nodes missed by SLNB.

The aim of this pilot-study is to examine the accuracy of MRL compared to
current nodal staging methods. We expect an acceptable accuracy of the MRL
based on earlier studies with gadofosveset enhanced MRI in rectum cancer
patients. A positive MRL could eventually replace the SLNB and the next step in
the nodal staging of these patients should be an ALND. It is unlikely that the
MRL can replace all SLNB*s considering the fact that it might not be able to
detect nodal metastases smaller than 0.2mm. Detection of these metastases is
crucial since they change adjuvant treatment regimen.

This pilot-study is designed as a single-center prospective cohort study. We
plan to include 10 patients in this pilot-study. Patients will be recruited in
the Maastricht Universitair Medisch Centrum (MUMC).
We expect a period needed to recruit of 5 months.

Beside the regular MRI with gadolinium contrast agent (Gadovist) patients will
undergo an extra MRI with gadofosveset contrast agent at least 3 days later.
With the MRI and coils used in this study we only investigate the axilla on the
site of the breast cancer.

The accuracy of MRL will be determined on the basis of a node-to-node matching
of imaged nodes to the definitive histopathology. The pathologic examination of
the SNLB or ALND will be regarded as the golden standard for nodal involvement.

The pathologic examination will take place following the regular procedure.
During microscopic examination each node will be recorded as benign, or
isolated tumor cell (ITC) ( pN0(i+) ) (<= 0.2mm ), or micrometastase (pN1mi)
(0.2 <= 2.0mm) or macrometastase (pN1) (>2.0mm).
Overall nodal status will be reported following the regular procedure. With
this result we can make the patient by patient analysis.

The results of the pathologic examinations will be compared to the results of
the MRL. The MRL will be assessed by two radiologists, who will independently
read all the images and they are blinded for earlier investigations. In the
meantime they will receive feedback about the pathologic results, in order to
make a learning curve possible.
Each lymph node visible on MRL will be scored as benign or malignant using a
confidence level score (0= definitely benign, 1= probably benign, 2= possibly
benign, 3= probably malignant, and 4= definitely malignant). Criteria for
malignancy on gadofosveset-MRI will be low signal intensity and absence of a
*relief* sign.
The diagnostic performance of gadofosveset-MRI will be analyzed on lesion by
lesion basis and on patient by patient basis. In this way we can obtain precise
validation of imaging findings with the underlying histopathology. The lesion
by lesion results will be translated to a patient by patient validation in
order to obtain a clinically more relevant assessment of the diagnostic
performance on a patient basis.

The patient who participates in this study will undergo all regular
investigations to come to a proper staging of the breast cancer. The current
procedure on staging of breast cancer patients is extensive. The procedure
includes in most cases a MRI scan of the breast. For this study we will perform
an extra MRI with gadofosveset contrast agent of the breast and axilla a couple
of days after the initial MRI scan. Further investigation needed for this study
will not extra burdensome the patients, because all further procedures are
already included in the regular treatment.