Improving radical resection rates in patients with breast cancer by intraoperative imaging using bevacizumab-IRDye800CW - the MARGIN-2 study

Breast conserving therapy combined with radiotherapy has become the gold
standard in breast cancer patients with limited tumor size. BCT consists of
local tumor resection with a small rim of healthy tissue. A major limitation in
breast conserving therapy is the challenging differentiation between benign and
malignant breast tissue intraoperatively, thereby increasing the risk of
tumor-positive margins and incomplete resection of the tumor on the one hand
and removing too much healthy tissue on the other. A tumor-positive margin
results in higher rates of local recurrence and decrease in survival compared
to a radical resection. Current postoperative workflows to assess surgical
margins by microscopy take up to several days, which excludes the possibility
of altering the operative plan (performing a wider local excision) immediately
during surgery. Direct tumor-margin assessment and immediate intervention could
possibly reduce the number of TPMs and thereby the need for additional therapy.
Previously, our multidisciplinary research group Optical Molecular Imaging
Groningen (OMIG) performed a phase-1 fluorescence-guided surgery study
investigating breast cancer surgery (MARGIN-1). This study showed that using
the fluorescent tracer bevacizumab-IRDye800CW tumor-positive margins could be
adequately detected. However, this was a proof-of-concept study in which the
fluorescent images were analyzed and correlated to pathology months after the
initial operation. Strikingly, the study found tumor-positive margins in 30% of
the cases, which potentially could have been detected using fluorescence guided
surgery. This current study aims to obtain real-time tumour visualization and
tumour margin assessment using fluorescence guided surgery with
bevacizumab-IRDye800CW intraoperatively.

Primary objectives
-To determine whether real-time tumor visualization using targeted fluorescent
imaging during breast conserving therapy in breast cancer patients can be
achieved intraoperatively and results in adequate assessment of the tumor
margin.

Secondary objectives
-To determine whether ex-vivo fluorescence imaging can adequately show
tumor-positive margins in resected tissue samples.
-To determine whether in-vivo fluorescence imaging can adequately show residual
tumor in the cavity.
-To determine whether one of the above-mentioned imaging techniques is superior
to the other in assessing tumor margin or that both techniques should be used
together.
-To determine whether the fluorescence images do correlate with pathology and
fluorescence shows the presence of tumor-positive margins as hypothesized.
-To investigate the potential feasibility of novel 3D (microscopic) imaging
techniques on 3D depth assessment of the obtained biopsies out of resected
tissue, if available
-To analyze whether patients that in standard-of-care would have tumor-positive
margins and needed additional treatment postoperatively, had a radical
resection after biopsy using fluorescence guided surgery (number and
percentage).

This study is a non-randomized, prospective, multi-center, interventional and
diagnostic accuracy study. The interventional character of this study is
explained by the fact that an extra biopsy can be performed based on
fluorescence imaging and possible tumor-positive margins. Patients with limited
size breast cancer scheduled for breast conserving therapy will be included.
The total number of included patients will be 60, if at least 9 of these
patients will have tumor-positive margins. Otherwise, the total number of
included patients will be expanded to 70. After written informed consent is
obtained, the tracer bevacizumab-IRDye800CW (10 mg) is intravenously
administered two to three days prior to surgery. On the day of surgery,
standard-of-care procedure is performed, and the tumor is resected. After
resection of the tumor, first the resected specimen will be analyzed ex-vivo
and imaged by an ex-vivo fluorescence imaging system, screening for possible
tumor-positive margins. Then the surgical cavity will be imaged in-vivo by the
other, in-vivo fluorescence imaging system, screening for residual tumor tissue
in the patient. This imaging should not take more than 5-10 minutes. The
fluorescence images will be analyzed directly and if no tumor-positive margin
is suspected, the procedure is ended. However, if a tumor-positive margin is
suspected, the in-vivo fluorescence imaging system will be used to image the
resected tissue specimen as well and biopsies will be taken from the region of
interest of both the cavity and resected tissue specimen (if applicable). After
obtaining the biopsies, the procedure is ended. If feasible determined by the
pathologist team member, biopsies of fluorescent areas will be taken from the
already standard of care resected tissue specimen to further study fluorescent
tracer distribution on novel (3D) imaging techniques. The fluorescence images
will be postoperatively correlated to the histopathology (i.e. hematoxylin and
eosin, H&E staining).

According to current standard-of-care workflow, patients with limited size
breast cancer eligible for breast conserving therapy according to the
multidisciplinary tumor board will be informed about the intended surgery.
During this outpatient clinic visit the patient will be informed about this
study as well. After written informed consent patients will receive intravenous
administration of the fluorescent tracer bevacizumab-IRDye800CW two or three
days prior to surgery. At day of surgery, surgery will be performed according
to standard-of-care. After resection of the tumor, this whole tumor sample will
be imaged ex-vivo screening for possible tumor-positive margins. Using another
imaging device in-vivo cavity imaging screening for possible tumor-positive
margins will be performed. When a tumor-positive margin is suspected, a biopsy
will be obtained from the region of interest in the patient before the
procedure is ended. This way the fluorescence images can be correlated to
pathology results afterwards. The study will extend the whole procedure
approximately by 10-15 minutes. We hypothesize that tumor-positive margins can
be detected by real-time fluorescence imaging, paving the way for future
implementation of fluorescence imaging in standard-of-care and improving
radical resection rates.

Burden: Time investment: Generally, patients visit the hospital (30 minutes)
two or three days day prior to surgery for tracer administration. Operation
duration is extended by max. 15 minutes. Additional procedures: First
additional procedure is the intravenous administration of
bevacizumab-IRDye800CW two or three days prior to surgery. The second procedure
is in-vivo imaging of the cavity after resection of the tumor, screening for a
possible tumor-positive margin. The third procedure is only performed if a
tumor-positive margin is suspected based on ex-vivo and/or in-vivo imaging and
consists of tissue biopsy of the region of interest for correlation of the
fluorescence images to pathology results afterwards. Considering the size of a
possible biopsy, no harm to the patients is expected.
Risk: No allergic reactions or other adverse events related to administration
of bevacizumab-IRDye800CW were seen in previous studies containing more than
200 patients in the Martini Hospital and UMCG, even in higher dose-cohorts. The
fluorescence imaging system used for cavity imaging is CE-approved for clinical
use. Therefore, this study is considered low risk.
Benefit: For now, no clear benefit is expected for patients included in this
study. The study is mostly focused on proof-of-concept of real-time
fluorescence imaging for implementation and reducing tumor-positive margins
(thereby lowering additional treatments) in future patients. Yet, current
patients might benefit by acquiring radical resection after biopsy (and thereby
complete removal) of tumor-positive lesions. However, this is not the aim of
this study, but will be the subject of subsequent studies in the future.