Trial Information

A Study to Assess Near Infrared Laparoscopy With Indocyanine Green (ICG) for Intraoperative Lymphatic Imaging and Sentinel Lymph Node Identification During Standard Surgical Resection for Colonic Cancer

1. BACKGROUND AND RATIONALE

1.1. Colorectal cancer Colorectal cancer is the third commonest cancer and second highest
cause of cancer death in the UK. Introduction of the National Bowel Cancer Screening Program
and increasing awareness of symptoms of colorectal cancer will lead to an increase in
numbers of cancers detected at an early stage of the disease. Standard surgical management,
however, remains unchanged and that is segmental colectomy with en bloc mesenteric resection
of the regional lymph nodes as the metastatic status of these nodes remains the most
important factor for determining adjuvant treatment.

This of course was appropriate when the majority of the patients diagnosed with colorectal
cancer presented with symptomatic, locally advanced disease. Our practice, however, needs to
be reconsidered in the view of the recently published data from screening centres. Pilot
results demonstrate a significant shift towards an earlier stage of the disease in the
screened population and suggests that, in future, up to 50% of screen detected-cancers will
be T1 or T2 (i.e. early stage disease) 1. Less than 15% of patients with T1 and T2 disease
have nodal involvement 2 but current radiological techniques are not sensitive enough to
identify these patients pre-operatively. Instead, major surgery is performed in all patients
with T1 and T2 stage colonic cancer, and in up to 85%, en bloc mesenteric resection provides
no therapeutic benefit, as they have node negative disease. Major resection also exposes
patients to a significant risk of morbidity and mortality, along with late side effects.

The concept of sentinel lymph node mapping (SLNM) in colorectal cancer is not new and
although it has not had the same impact as in the clinical care of breast cancer, melanoma,
and more recently gastric cancer 3-6, some expert centres have suggested that it may be
clinically relevant 7, 8. The technique is based on the principle that the first possible
sites of metastasis along the lymphatic drainage route from the primary lesion are known as
sentinel lymph nodes (SLNs) and these can be detected using injection of dyes, radioisotopes
or a combination of both. Therefore, evaluation of the SLNs can result in an accurate
assessment of the nodal status in the entire lymphatic basin allowing a tailored surgical
treatment. Although this technique has transformed surgical care for melanoma and breast
cancer patients, its application in patients with colorectal cancer has been questioned due
to discrepancies in the reported results caused by the heterogeneous study designs employed,
variations in surgical experience and differing patient inclusion criteria 9-11.

Detailed analysis of studies performed to date shows that tumour and patient
characterisation as well as surgeons experience are key factors when assessing the utility
of the technique 2, 12 and at least three nodes need to be found in order to maximise this
13. The main research focus so far has been on optimising the technique itself. Initial
research was performed using blue dye, radioactive colloid or both in combination but this
failed to achieve the 95-97% identification rates found in breast cancer and was also
accompanied by unacceptable false negative rates 14-17. Although many authors noted that
their false negatives occurred predominantly in patients with advanced disease (T3 and T4
cancers) 18-20, utilisation of those identification agents has proven frustrating to date.
Radioactive colloids require involvement of a nuclear medicine physician, can be difficult
to localise with a handheld camera and cause signal interference if the SLN(s) is too close
to the injection site. Blue dyes however are difficult to localise in patients with a high
body mass index (BMI), who unfortunately comprise the majority of our patient population,
thus producing only limited (if any) visualisation of afferent lymphatic vessels and SLN(s).

In order to re-examine extensive research that has been reported on SLNM, a secondary
intention analysis of the two largest multicentre SLN databases was undertaken. This
reported that in 225 patients with T1 or T2 disease, accurate detection of SLNs was possible
in 95%, confirming that this staging technique should be most accurate in patients with
early colonic cancer 2. This improved with operator experience and when tumour size was
smaller than 35 mm in diameter. In 196 patients treated by experienced surgeons, 144
(73.4%), would have correctly undergone localised full thickness resection if surgery had
been based only on their SLNs status and 49 (25.0%), would have required conventional
resection either because the SLNs were not identified, or due to detection of metastases.
Only 3 patients (1.5%) would have been inappropriately predicted to be node negative.

1.2. Investigational Agent To overcome the aforementioned issues, a number of researchers
have utilized optical imaging using the near-infrared (NIR) fluorescence lymphatic tracer
indocyanine green (ICG) which enables real-time intraoperative visualisation of lymphatic
channels and SLNs. The technique performs as well as the combination of a radioactive tracer
and blue dye in patients with breast cancer 21, and has also been validated in patients with
early gastric cancer 4-6.

ICG was initially recommended as a tattooing agent for colonic and pancreatic lesions to aid
perioperative localization 22-24, as there were no reports of complications such as focal
peritonitis, abscess formation, postoperative adhesions and ileus which had been seen with
the use of India ink 25-27. However, its use as a tattooing agent is limited to patients
undergoing surgery within 8 days after endoscopic injection, as the dye dissipates through
lymphatic channels due to its small molecular size 22. More recently, its fluorescence
properties have been utilised for intraoperative localisation of colorectal neoplasms and
associated SLNs 28-32 in patients undergoing surgery using either charge-coupled device
(CCD) or a NIR camera. Nagata et al. demonstrated that observation of ICG dye using a NIR
camera was far superior to that by conventional laparoscopy with identification of SLNs
being five times better 29. These results suggest that a SLNM technique using NIR with a
fluorescent tracer may be useful in patients with a high BMI who have large amounts of
mesenteric adipose tissue. All authors reporting the NIR technique have noted its success in
colorectal practice, without adverse effects with a comparable role to its use in early
breast or gastric cancer.

1.3. Rationale The ability to utilise this technique and confidently identify the SLNs has
the potential to transform the surgical care of patients with colonic neoplasia. The focus,
however, must be on patients with early disease, thus allowing those who have node negative
disease to be considered for localised full thickness resection of their primary cancer as
definitive surgical therapy. Unfortunately, studies published to date have included patients
with T3 disease in their series 29, 31.

This pilot trial aims to apply a well-described technique using a NIR camera and ICG as the
tracer in order to clarify whether SLNs can be confidently identified in patients undergoing
conventional resection (segmental colectomy and en bloc mesenteric resection) for T1 and T2
colonic cancers. This protocol is based on extensive literature search of studies describing
the technique in patients with early gastric and colorectal cancers 5, 6, 29-31, 33-36.
Provided that the pilot trial is successful, and in order to obtain adequate power, a
further large multicentre trial will be required to determine whether the nodes visualised
by NIR mapping are representative of the patient's lymphatic basin metastatic status. Proven
concordance between the pathology of the SLNs and that of the standard nodal resection would
then justify a more limited surgical intervention, taking just the SLN(s), and allowing a
more precisely tailored operative intervention.

All patients in the pilot trial will undergo standard surgical treatment in addition to the
NIR SLNM technique.

1.3.1. Risk - benefit assessment

Potential benefits This study is designed to assess the effectiveness of NIR mapping with
ICG to identify SLNs in patients who are undergoing the current 'gold standard' treatment of
segmental colectomy and en bloc mesenteric resection; all patients are therefore still
receiving the current established standard of care for their condition. There will be no
direct benefit to the participants in terms of any reduction in the extent of surgery which
they receive, although this could potentially be a benefit of the imaging technique for
future patients in the long term. However, the study design does provide some benefit to
participants in that they will receive additional pathological scrutiny of their disease
above that which is usually performed. In addition to routine standard pathological analysis
of lymph nodes, the patients in this study will have ultra-analysis (including IHC) of the
nodes identified in the mapping. While the significance of micrometastases detected by IHC
is not entirely clear at present, this disease burden has been included in the TNM
classification as an indication for considering additive postoperative therapy. Also, the
addition of lymphatic mapping during the case can be expected to identify first order
draining lymph nodes lying outside the field of standard operative resection and this is
reported to occur in up to 14% of patients37. Excision and analysis of these nodes may
result in upstaging of disease extent and thus allow the recommendation of adjuvant therapy
leading to a potential improvement in prognosis.

Potential risks and their management

There are some small risks associated with the introduction of the additional imaging
procedure. There is a documented risk of allergic (anaphylactic or anaphylactoid) reaction
of less than 1 in 10, 000 to the tracer agent - all participants will be asked about any
prior exposure to the agent or reaction to iodine dyes and shellfish. Case notes will also
be thoroughly reviewed by a member of the research team. However, if allergic reaction
occurs during or after administration of the ICG, standard hospital procedures will be
followed and further active study procedures abandoned. The patient will however be followed
up according to the protocol in order to collect the adverse event data until the adverse
event is resolved. A pregnancy test will be performed in all women of childbearing potential
and those who test positive will be excluded from the trial.

Colonic perforation related to the endoscopy procedure occurs in 1/2000 cases. This
procedure, however, is essential in order to inject the tracer agent. This risk will be
minimized by ensuring that this examination is performed by an experienced endoscopist and
it is significantly diminished with the use of bowel preparation that optimises the
endoscopic view.

Inclusion into the study will add 30 minutes to the total operative time which is thought to
be acceptable by the senior members of the investigative team with minimal risk to the
patient. Experienced clinicians will perform the surgical procedure and we expect that this
added time will decrease significantly as experience is gained with the technique in
question.

Overall benefit-risk assessment

In order to minimise the risks for potential participants, we will only use materials (ICG),
equipment (NIR laparoscopy camera, Olympus) and methods (endoscopic injection) for which the
risks are already known. Previously, the technique of sentinel node biopsy in colorectal
cancer was used to enable a more intensive examination of the lymph nodes after the
operation has taken place. Here, the proposal is that identification of the lymph nodes
before or during the operation could help the surgeon decide exactly which operation to do.
In this initial study we will use a single mapping agent and its fluorescence properties
will allow enhanced sensitivity of detection at the time of surgery. Standard surgery will
then be carried out. In choosing this agent, we have systematically studied the literature
and previous empiric experience with non-fluorescent, non-radioactive agents (primarily
methylene blue) and found this unsatisfactory in patients of average and above-average BMI.
If the results of this 'optimised' technical study are satisfactory, further work will be
proposed with a large multicentre study to assess sensitivity and specificity of the
technique.

ICG is routinely used in radiological and cardiac investigations and is well within safety
recommendations for both patients and the staff entrusted with their care. The Olympus
Infrared Laparoscopic System being used to detect the presence of the mapping agent in lymph
nodes has been CE marked (European Community directive conformity). Therefore, we consider
the overall risk to participants in this study is minimal.

In addition, there are also potential benefits from participating in the trial. Rather than
undergoing standard pathological processing of their resected lymph nodes, patients in this
study will have added intensive study of their sentinel nodes (all other nodes will be
processed in the routine fashion). Aberrant nodes identified with the technique will also be
removed and analysed for evidence of metastases. This will allow highly sensitive detection
of tumour deposits (i.e. small micrometastases) that may be prognostically important in the
nodes most likely to harbour such deposits. Using this technique we also hope to identify a
proportion of patients with aberrant lymphatic drainage which will allow more accurate
staging of the disease and reduce the risk of understaging and therefore avoid depriving
patients of the benefits from adjuvant chemotherapy.