Sentinel lymph node identification in colon and rectal cancer using a radioactive and fluorescent tracer

Colorectal cancer (CRC) is one of the leading causes of cancer related deaths
in the Western World. In the Netherlands there are over 10.000 new cases each
year. This number will increase to more than 14.000 caused by more awareness of
the disease, aging, growing population and nationwide screening programmes
which will start this year. Complete resection of the primary tumour and
accurate staging of the lymph nodes are the determining factors in patient
survival. Nowadays, 50% of the patients with colorectal carcinoma are diagnosed
with a stage I or II tumour, which are node negative by definition. The 5-year
survival rates for patients with stage I and II colorectal carcinoma are
respectively 90% and 75%. Up to 20-30% stage I/II patients will nonetheless
develop distant metastases and the majority will eventually die from colorectal
carcinoma. It is possible that in this group of patients, small lymph node
metastases have been missed, resulting in understaging. This may be due to an
inadequate surgical lymphadenectomy or insufficient pathological examination.
Metastatic nodes can be missed when not enough lymph nodes are being examined.
For accurate staging a minimum of 7-14 lymph nodes is recommended. This is also
the mean number of nodes we examine at the VU University Medical Center.
However, some authors suggest that when examining only 14 lymph nodes
metastases will be missed. Therefore about 40 lymph nodes need to be examined
for a 85% probability that the patient is node negative.Better detection and
pathologic staging of the lymph nodes could contribute to a better survival of
colorectal cancer patients. With conventional H&E staining, micrometastases are
not detected. They can be detected by multislicing and immunohistochemical
staining (IHC). However, these ultrastaging techniques are labour and cost
intensive, so it is highly impractical to ultrastage all the nodes in a given
specimen. A solution for better detection, harvesting and optimalization for
histopathological examination is the sentinel lymph node (SLN) procedure. This
procedure aims to identify the first draining lymph node(s) from the primary
tumour, which have the highest risk of harbouring metastases. These SLNs can be
analyzed with the more sensitive histopathologic techniques as mentioned above.
Since the introduction of the SLN principle by Morton, the SLN procedure has
become a diagnostic staging procedure in multiple types of cancer like melanoma
and breast cancer. Identification of the SLN provides information about the
rest of the lymph nodes. A SLN with metastasis is an indication for additional
adjuvant postoperative chemotherapy. However, a negative SLN would justify a
wait and see policy. Recently we have published a meta-analysis of 52 studies
with 3767 SLN procedures; 2961 colon and 806 rectum carcinoma. We noticed that
between the studies there were big differences in used technique, type and
amount of dye, number of performed procedure, patient selection and
pathological examination of the harvested nodes. The mean overall detection
rate was 0.94 (0.95% CI0.92-0.95), at pooled sensitivity 0.76 (0.72-0.80). In a
subgroup of 8 high methodological quality showed a higher detection rate and
sensitivity in colon cancer patients (resp. 0.96, 0.95% CI 0.90-0.95 en 0.90
(0.86-0.93) and rectum cancer (resp. 0.95 (0.75-0.90) and 0.82 0.60-0.93)). We
concluded that for implementation, the SLN procedure needs optimization and
standardization. Therefore we made a view recommendations:
1. Colon and rectal cancer should be considered as a distinct type of cancer,
as it has a different pattern of spread and recurrence. Secondly, there is a
wide variation in the used SLNM techniques.
2. Standardization of the used tracer, injection and harvesting technique.
Identification of the SLNs in colorectal cancer is usually guided by the in
vivo or ex vivo injection of blue dyes, a radioactive tracer or a combination
of both. Most experience with the SLNM technique has been reported with blue
dye. The particle size of blue dye is relatively small so its passages quickly
through the lymphatic vessels and nodes. As a result, all lymph nodes will
stain in a few minutes. This makes it impossible to identify the true SLN.
Another drawback is the limited tissue penetration depth which makes it
impossible to detect SLNs in the fatty mesocolon.
Using only radioactive tracer presents the problem of signal interference of
the injection site so we are not able to distinguish between the injection spot
and the SLNs. SLN mapping with a near-infrared (NIR) dye is a new and promising
technique. In breast cancer and melanoma is has shown excellent results.
NIR-dyes can be used during the surgical procedure and can be seen during the
operation. This provides the opportunity for the surgeon to localize the lymph
node in the abdomen and make sure there is an adequate resection. In the last
year, we included 27 patients with CRC by which a SLNM procedure is performed
using the NIR dye Indocyanin Green (ICG). Out of this study we concluded that a
submucosal injection technique at the base of the tumour was superior compared
to a subserosal peritumoral injection. Although we found fluorescent nodes in
all patients, the techniques needs improvement. Penetration depth in fatty
tissue is better with ICG than blue dye, but still limited. Fluorescent
guidance to the region of interest could benefit from additional guidance
provoked by a radioactive component. Preoperative injection of a tracer which
consist of a fluorescent and radioactive component would allow for preoperative
planning and intraoperative detection of the SLNs.
Preoperative imaging of the SLNs using PET/CT provides useful preoperative
anatomic information regarding the location. It can produce a road map to
detect all SLNs even if there is an unusual pattern of lymph drainage from the
primary tumour. Pharmacokinetics and biodistribution can also be monitored
accurately with preoperative imaging. The multimodal SPECT-tracer
ICG-99mTc-NanoColl is already used in SLNM in prostate cancer. The authors
experienced the value of the real-time fluorescence guidance in areas where
accurate gamma tracing was hindered by background signals. The preoperative
imaging performed with SPECT/CT provided anatomic information which improved
the SLN detection accuracy to 98%. Out of results from SLNM in floor and mouth
tumours, we know that the resolution of gamma-or SPECT camera is not always
sufficient to visualize and localize SLNs. Probably because of the *shine-
through* of the injection site (i.e. primary tumour) which makes detection of
SLNs near the tumour impossible. Intraoperative gamma-probe detection also
fails if it can not differentiate between SLN and injection site. The
department of Otolaryngology and Nuclear Medicine & PET research of VU
University Medical Center Amsterdam (Prof.dr. G.A.M.S. van Dongen and Prof.dr.
R. de Bree) have developed a new PET/CT-tracer 89Zr-Nanocoll.
Preliminary results showed that preoperative detection and localization of SLNs
with PET/CT was superior compared to SPECT/CT. Another advantage of the 89
Zr-nanocoll compared to the common used SPECT-tracer 99m Tc-Nanocoll is the
longer half-life.
For so far, we do not know the kinetics of the tracer in colorectal carcinoma.
Therefore we prefer a tracer which can be detected long after administration.

Aim of the present study is to investigate if a combination of a radioactive
and fluorescent tracer can increase the sensitivity and specificity of the SLNM
technique in colon and rectal cancer by utilizing the radioactive component
for preoperative imaging (PET/CT) and guidance to the SLNs by using the near
infrared fluorescence imaging of the nodes during the surgical procedure. The
results of this study (scan protocol, amount of conjugate etc.) will be the
fundamental part to set up a new trial with larger number of patients.

Single center prospective pilot study

The proposed study is a prospective feasibility study to demonstrate the
feasibility of SLN mapping by using a radioactive and fluorescent tracer.

Study setting:
The proposed study will be conducted at the VU University Medical Center in
Amsterdam, The Netherlands. Patients who meet the inclusion criteria will be
asked to participate during their standard appointment at the outdoor patient
clinic.

Administration of 89Zr-Nanocoll and ICG
The 89Zr-Nanocoll tracer will be administered by colonoscopy ± 48 hrs before
start of the operation. The day before tracer administration the sigmoid will
be cleaned with Klean-Prep 4L or Moviprep 2L. For injection we use a
colonoscopic sclerosing needle. An earlier performed *paste-test* showed that
at least 90% of the radioactivity will be administrated. A total of 0.5-4.0 ml,
maximum 20 MBq 89Zr-Nanocoll will be administrated in 1-4 injections at the
base of the tumour, depending on the size of the tumour. Indocyanin Green 25 mg
(Pulsion, Munich, Germany) will be diluted in 9 ml 0.9% saline and 1 ml 200
g/l human albumin. This will be administrated during surgery by colonoscopy
when the patient is under general anesthesia. Therefore patient will receive
two enemas the day before surgery.

Patients will undergo three PET/ CT scans during this study, independent of the
used tracer.
The first will take place directly after tracer administration. The second one
± 24 hrs after tracer administration and the third just before the surgical
procedure, respectively ± 48 hrs after tracer administration.

The PET/CT images will be compared with respect to the total number and
location of foci and, if visible, lymphatic vessels. During surgery, PET/CT
images will be visible on an extern monitor. The images of the SLNs identified
with PET/CT scan will be used as guidance for localization of the SLNs during
surgical exploration, which makes it easier to identify the nodes.
All patients will undergo laparoscopic surgery. Laparoscopic access will be
obtained in the traditional way and abdominal exploration shall be performed to
rule out the intra-abdominal metastasis. The involved colonic segment will be
localized and mobilized. The SLN procedure will then be performed.
First the segment will be inspected for fluorescent nodes with the NIR-
laparoscope. Fluorescent nodes are marked with a suture.
Thereafter the PET/CT images will be used as a roadmap, to detect SLNs which
are not visible with the NIR laparoscope. These nodes will also be marked with
a suture in a different colour then the fluorescent nodes. The involved segment
of the colon and regional lymph nodes will be resected like the conventional
way.
Ex vivo the surgical specimen will be inspected for fluorescent and radioactive
nodes which are not detected during surgery. All the identified nodes will be
taken out ex vivo and stored separately. The number of identified nodes during
the surgical procedure will be compared with the number of nodes found by the
PET/CT scans.
The entire specimen will be submitted for pathologic examination. All
identified fluorescent and/or radioactive nodes will be stained with
hematoxylin-eosin (H&E). If the fluorescent or radioactive SLNs are negative
after routine H&E staining, they will be sliced in multiple parts and examined
with IHC with the specific marker CAM5.2.
The pathologist uses palpation to identify the remaining non-fluorescent and/
or radioactive lymph nodes. Nodes found by palpation will be screened for
fluorescence by the NIR-laparoscope and radioactivity by placing them on the
tip of a gamma-probe.
Lymph nodes with metastasis will be analysed with the the Q-prodit; an
interactive video morphometry system (Leica, Cambridge, UK) to determine the
amount of tumour tissue.

All participating patients will receive conventional resection of the tumour
and follow-up according to normal standards in our hospital. The main goal of
this study is to optimize the SLN mapping technique in colon cancer. If we are
able to identify the true SLN this could lead to better staging and survival of
patients with this type of cancer. Patients undergo one or two colonoscopies
for injection of the tracers and three additional PET/CT scans. Because of the
colonscopy ± 48 hrs before surgery, patients stay in the hospital will be
prolonged with one day. The additional risks and exposure to radiation for
participating patients are calculated and can be considered as negligible.