 |
|
Sentinel Node Localization in Oral Cavity and Oropharynx Squamous Cell Cancer
Rodney J. Taylor, MSPH, MD;
Richard L. Wahl, MD;
Pramod K. Sharma, MD;
Carol R. Bradford, MD;
Jeffrey E. Terrell, MD;
Theodoros N. Teknos, MD;
Earl M. Heard, MD;
Gregory T. Wolf, MD;
Douglas B. Chepeha, MSPH, MD
Arch Otolaryngol Head Neck Surg. 2001;127:970-974.
ABSTRACT
| |
Objective To evaluate the feasibility and predictive ability of the sentinel node
localization technique for patients with squamous cell carcinoma of the oral
cavity or oropharynx and clinically negative necks.
Design Prospective, efficacy study comparing the histopathologic status of
the sentinel node with that of the remaining neck dissection specimen.
Setting Tertiary referral center.
Patients Patients with T1 or T2 disease and clinically negative necks were eligible
for the study. Nine previously untreated patients with oral cavity or oropharyngeal
squamous cell carcinoma were enrolled in the study.
Interventions Ulfiltered technetium Tc 99m sulfur colloid injections of the primary
tumor and lymphoscintigraphy were performed on the day before surgery. Intraoperatively,
the sentinel node(s) was localized with a gamma probe and removed after tumor
resection and before neck dissection.
Main Outcome Measures The primary outcome was the negative predictive value of the histopathologic
status of the sentinel node for predicting cervical metastases.
Results Sentinel nodes were identified in 9 previously untreated patients. In
5 patients, there were no positive nodes. In 4 patients, the sentinel nodes
were the only histopathologically positive nodes. In previously untreated
patients, the sentinel node technique had a negative predictive value of 100%
for cervical metastasis.
Conclusions Our preliminary investigation shows that sentinel node localization
is technically feasible in head and neck surgery and is predictive of cervical
metastasis. The sentinel node technique has the potential to decrease the
number of neck dissections performed in clinically negative necks, thus reducing
the associated morbidity for patients in this group.
INTRODUCTION
FOR PATIENTS with head and neck cancer, the histopathologic status of
cervical lymph nodes is an important prognostic factor.1
Thus, accurate staging of the cervical lymph nodes is important for the treatment
of this population. Even when patients have clinically negative lymph nodes,
there is still a significant chance that they may harbor occult metastases.2-3 For patients with head and neck cancer
who are treated surgically, management of the clinically negative neck usually
involves an elective neck dissection.4-5
Unfortunately, the performance of an elective neck dissection may frequently
leave an aesthetic and functional impairment. Cosmetic deformities due to
neck dissection procedures may result from the lengthy scar and the ensuing
asymmetry of the neck. Furthermore, acute and chronic shoulder dysfunction
can be important sequelae of neck dissections.6-8
If sentinel node localization can predict which patient requires a neck dissection,
then neck dissection may be avoided in patients with negative sentinel nodes,
thus reducing the associated cosmetic deformities and shoulder dysfunction.
Noninvasive evaluations, such as clinical examination, radiographic
studies, and ultrasonagraphy, are very useful but still are not sufficiently
sensitive or specific to serve as stand-alone, staging modalities for predicting
the histopathologic status of cervical lymph nodes for head and neck tumors.5, 9-10 More than 2 decades
ago, investigation with cervical lymphography was proposed as a potentially
less extensive means of determining if more comprehensive lymphadenectomies
were necessary.11 In 1992, Morton12
used blue dye to identify sentinel nodes in patients with cutaneous melanoma,
while other investigators began using radioactive tracers and gamma probes
to identify sentinel nodes in patients with early breast cancer.13-14
The principle underlying sentinel node surgery is rooted in the concept that
the sentinel node is predictive of a primary tumor that has a propensity to
metastasize. Thus, if the sentinel node is positive, the other regional nodes
may be positive as well. Limited success in the use of sentinel node localization
has been reported in patients with head and neck cancer.15-16
To evaluate this potentially valuable diagnostic adjunct, we conducted a prospective
efficacy study to determine whether sentinel node localization is technically
feasible and whether the histopathologic status of the sentinel node is predictive
of the status of the other regional lymph nodes for squamous cell carcinoma
(SCC) of the upper aerodigestive tract.
PATIENTS AND METHODS
PATIENTS
The eligibility criteria for this prospective study included patients
with previously untreated oral cavity or oropharyngeal SCC (American Joint
Committee on Cancer stage I and II disease) and excluded patients who had
previously undergone neck surgery or radiation therapy of the head and neck
or who had a history of any noncutaneous malignancy. For midline tumors, sentinel
node localization and neck dissection were performed bilaterally. When the
primary tumor treatment involved surgery, our approach was to perform a neck
dissection for primary tumor depth of greater than 2 mm. Primary sites other
than the oral cavity and pharynx were excluded because of the inaccessibility
for precision injection of the primary site. Informed consent was obtained
for all participating patients. The institutional review board of the University
of Michigan, Ann Arbor, approved the protocol.
From October 1998 to January 2000, we enrolled 9 previously untreated
patients (4 men and 5 women; mean age, 61.9 years; age range, 22-80 years),
including 7 with unilateral lesions and 2 with bilateral lesions. All but
1 of the primary lesions were oral cavity SCCs (oral tongue and floor of the
mouth); the remaining lesion was oropharyngeal (tonsil). The tumor stage in
all patients was American Joint Committee on Cancer stage II (T2 N0). Two
patients had midline lesions; consequently, the sentinel nodes were identified
and localized bilaterally. Thus, the sentinel node localization technique
was used for 11 necks among 9 patients.
Two patients were excluded from analysis for eligibility or protocol
violations. One patient was excluded when it was discovered that he had a
remote history of radiation therapy for a previous SCC of the oral cavity.
The second patient was excluded because of a previous noncutaneous malignancy:
he had a tonsillar lymphoma, which was treated with chemotherapy.
Approximately two thirds of the patients who were approached for enrollment
in the protocol elected to participate. Those patients who declined enrollment
reported concern regarding spending the additional time required to participate
in a clinical study. Some patients also expressed concern regarding pain associated
with tumor injection.
MATERIALS
On the afternoon before surgery, the patients arrived at the nuclear
medicine suite for tumor injection with 3 mCi of unfiltered technetium Tc
99m sulfur colloid radiotracer. At least 1 member from the Department of Otolaryngology
and 1 from the Department of Nuclear Medicine were present at the time of
each injection. A total volume of 0.20 mL was prepared from a commercially
available sulfur colloid kit (CIS-US Inc, Bedford, Mass). Because the injections
elicited a painful, burning sensation in the patients, 2% tetracaine hydrochloride
(Pontocaine) solution was used as topical anesthetic. Four injections of equal
volume (0.05 mL) were meticulously placed submucosally at 4 equidistant points
within 2 mm of the peripheral tumor margin.
We elected to perform tumor injection on the afternoon before surgery,
with a first case start in the operating room the following morning, so that
a reliable time between tumor injection and surgery could be established.
The time elapsed between tumor injection and surgery was approximately 16
hours.
Immediately after the injection, the patients were taken for scintigraphic
imaging. Anterior and lateral images were obtained with a single-head gamma
camera (E CAM; Siemen's Medical System Inc, Hoffman Estates, Ill). Serial
images were taken up to 1 hour after injection and once again the following
morning before surgery. During imaging, the primary injection site was lead
shielded so that the sentinel nodes could be more easily identified. The lymphoscintigrams
were used to provide preoperative information on the general location and
number of sentinel nodes, rather than specific detailed anatomical data. For
patients with unilateral, nonmidline lesions that were found to have contralateral
radiotracer uptake on the lymphoscintigrams, we did not pursue localization
of the contralateral node and contralateral neck dissection because bilateral
neck dissection for unilateral, nonmidline primary tumors in clinically negative
necks is not the current standard of care. A senior staff nuclear medicine
physician provided lymphoscintigraphic interpretations before surgery.
INTRAOPERATIVE IDENTIFICATION
Preoperatively, lymphoscintigraphic scans were examined to aid in identification
of the sentinel node. The procedure began with an attempt to transcutaneously
identify the sentinel node with a gamma probe (Navigator Gamma Guidance System;
Auto Suture Co, Norwalk, Conn); however, the "shine-through" effect of the
primary tumor frequently made this difficult. The higher radioactivity from
the injection site, termed the shine-through artifact,
obscures localization of the sentinel nodes in the first echelon. Therefore,
extirpation of the primary tumor site was always performed before sentinel
node localization and neck dissection. The sentinel node was then assessed
transcutaneously. A neck dissection incision was made, and the sentinel nodes
were identified with the aid of the gamma probe and individually dissected
and removed. After the removal of the sentinel nodes, the tissue bed was reevaluated
to confirm the removal of all sentinel nodes. Subsequently, a selective neck
dissection (levels I-IV) was performed.
PROCESSING OF SPECIMENS
After removal of the sentinel nodes and completion of the neck dissection,
the entire surgical specimen was stored in the Department of Nuclear Medicine
for 48 hours to allow decay of the radioisotope to background levels. At our
institution, the Department of Pathology does not receive any tissue specimen
that has been exposed to radioactivity for at least 48 hours, regardless of
the amount of radioactivity. Therefore, primary tumor margins could not be
evaluated routinely at the time of surgery. There is ongoing dialogue between
the Departments of Otolaryngology, Pathology, Nuclear Medicine, and Radiation
Safety at our institution regarding this issue. Both the Department of Nuclear
Medicine and the Department of Radiation Safety have stated clearly that the
level of radioactivity used in this protocol is safe for intraoperative use
for both the patient and the surgeon, requiring no special precautions. We
are optimistic that with the growing popularity of this technique and its
established safety, this problem will be resolved at our institution in the
near future. This potentially problematic scenario is addressed in our informed
consent process before patient enrollment in the study.
After the 48-hour period, the neck dissection specimens were grossly
evaluated and processed according to standard protocol at our institution.
Each of the 4 levels of the neck dissection specimen was submitted separately
for permanent analysis. Lymph nodes were identified by palpation and visual
inspection and then bisected. A single representative cross section of each
lymph node was histologically examined by an attending pathologist to evaluate
for cervical metastasis. The sentinel nodes were processed and examined in
an identical fashion to the nodes from the neck dissection specimens. Afterward,
the histopathologic status of the sentinel nodes was compared with that of
the remainder of the neck dissection specimens.
RESULTS
For each of the 9 eligible patients, the sentinel node was successfully
localized and removed. For all previously untreated patients, the sentinel
node technique had a negative predictive value of 100% for the absence of
cervical metastasis. Four of our 9 patients demonstrated cervical metastasis
on permanent pathological analysis; in each instance, the sentinel node(s)
was the only node(s) identified that harbored metastasis. Table 1 provides data on the location of the primary tumors and
sentinel nodes as well as the histopathologic status of both the sentinel
nodes and the remainder of the neck dissection specimens. When assessing the
size of the sentinel nodes, we found no statistical difference between histologically
positive and negative sentinel nodes. The mean size of the positive and negative
sentinel nodes was 1.86 and 1.52 cm (P = .66), respectively.
|
|
|
|
Data on Primary Tumor and Sentinel Node Location as Well as Histopathologic
Status of Both the Sentinel Nodes and the Remaining Neck Dissection Specimen(s)*
|
|
|
The 2 patients who were enrolled but excluded from the analysis owing
to treatment for previous head and neck malignancies did not demonstrate sentinel
nodes. No radiotracer uptake was elicited in the cervical lymphatics of the
patient who had previously undergone radiation therapy to the oral cavity
and cervical region for an oral cavity SCC. He was not found to have cervical
metastasis after neck dissection. The second patient, who had been previously
treated with chemotherapy for tonsillar lymphoma, had radiotracer uptake,
but the sentinel node was not identified intraoperatively. In the latter patient,
cervical metastases were subsequently discovered after neck dissection.
One patient was found to have a positive microscopic primary tumor margin,
which was reported on permanent pathological analysis. This patient subsequently
underwent postoperative radiation therapy because of cervical metastasis.
No complications occurred in association with the sentinel node localization
technique. This patient is presently free of disease.
The lymphoscintigrams demonstrated cervical lymph node radiotracer uptake
and served as a guide for the number and location of sentinel nodes. In 6
of the 7 patients with unilateral lesions, the number of sentinel nodes determined
by lymphoscintigraphy correlated with the number of sentinel nodes identified
intraoperatively in the ipsilateral neck area. Two of the 7 patients with
unilateral primary lesions also demonstrated lymphoscintigraphic uptake on
the opposite side of the neck. According to protocol, no attempts were made
intraoperatively to locate the nodes on the opposite side of the neck. Neither
of the 2 patients with unilateral primary lesions and bilateral uptake has
subsequently developed contralateral cervical metastasis. The 2 patients with
midline primary lesions who underwent bilateral neck dissections demonstrated
multiple bilateral nodes on the lymphoscintigrams. For both of these patients,
1 sentinel node was identified intraoperatively in each side of the neck.
Additional operative time was used for patients undergoing sentinel
node localization. Time spent for data recording and locating sentinel nodes
added approximately 45 minutes to the operative time. Also, some delays were
encountered in operating room start time because of difficulty in coordinating
the second lymphoscintigraphic scan performed in the nuclear medicine suite
on the morning of surgery.
COMMENT
Our preliminary results provide promising evidence that sentinel node
localization is technically feasible for oral cavity and oropharyngeal SCC
and is predictive of cervical metastasis. Although previous efforts using
sentinel node localization in patients with head and neck cancer have enjoyed
limited success, we believe that there are several technical issues that contributed
to the successful implementation of this technique in this study. Close collaboration
between the otolaryngologist and the nuclear medicine physician was critical.
Because of the difficulty associated with oral cavity injection, it was essential
that an otolaryngologist be involved in the tumor injection process. Similarly,
the presence of the nuclear medicine physician was key to issues relating
to the handling of the radioisotopes and lymphoscintigraphic image acquisition.
Another important distinction in our study was the use of 3 mCi of unfiltered
technetium Tc 99m sulfur colloid, as opposed to 1 mCi used in previous studies.15 We think that the higher activity that we used facilitated
detection of the deep lymphatics in the head and neck many hours after tumor
injection. Furthermore, we believe that it is essential to place the injections
meticulously within the submucosal region at the margin of the tumor and to
use small, precise volumes of radiotracer so that there would be greater likelihood
that the lymphatics that drain the primary tumor would be accessed instead
of having radiotracer diffuse more widely to involve adjacent lymphatics that
were not directly draining the primary site. Finally, removing the primary
tumor first is essential to intraoperative localization of the sentinel node
so that the shine through of the primary tumor does not impede identification
of the sentinel node.
In our study, we elected to include patients with early disease (American
Joint Committee on Cancer stages I and II) and clinically negative necks,
because it is this targeted group of patients in whom the potential of avoiding
an elective neck dissection and its associated morbidity is most practical.
This approach is consistent with the paradigm that has been established with
breast and cutaneous melanoma sentinel node surgery.12, 17
Our initial experience also underscores the importance of careful patient
selection and adherence to eligibility criteria. The sentinel node localization
technique may not be reliable for patients who have been previously treated.
Therefore, the results of this study may only be applicable to previously
untreated patients.
We have discovered with our initial efforts that, while there is reason
for enthusiasm with the sentinel node localization technique, there is a learning
curve with this technique. There are many technical fine points regarding
tumor injection and imaging as well as intraoperative localization and specimen
processing. The inability to assess frozen-section margins was an obstacle
during this study. Because of the current policy at our institution regarding
the handling of specimens that have been exposed to radioisotopes, we were
unable to send intraoperative specimens to our pathology department for frozen-section
analysis. Consequently, 1 patient was found to have a positive microscopic
margin on permanent analysis. This patient proceeded to undergo postoperative
radiation therapy for cervical metastasis.
Our early results show the promise that sentinel node localization may
have in head and neck oncological surgery for previously untreated patients
with clinically negative regional disease. If successful, patients with negative
sentinel nodes would be spared a neck dissection and the associated morbidity.
Thus far, we have shown the technique to be logistically feasible, with a
negative predictive value of 100% in the untreated population. We are continuing
to accrue patients to provide the sample size necessary to determine if this
technique provides sufficient specificity and predictive value to serve as
a staging technique in this population.
AUTHOR INFORMATION
Accepted for publication February 7, 2001.
The University of Michigan General Clinical Research Committee, Ann
Arbor, provided support for Protocol 1577 (Sentinel Node Localization for
Oral Cavity and Oropharyngeal Squamous Cell Carcinoma) via grant MD1-RR00042
from the National Institutes of Health, Bethesda, Md.
Corresponding author and reprints: Douglas B. Chepeha, MSPH, MD,
Department of Otolaryngology–Head and Neck Surgery, University of Michigan,
1500 E Medical Center Dr, TC 1904, Ann Arbor, MI 48109 (e-mail:
dchepeha{at}umich.edu).
From the Department of Otolaryngology–Head and Neck Surgery (Drs
Taylor, Sharma, Bradford, Terrell, Teknos, Wolf, and Chepeha) and the Department
of Internal Medicine, Division of Nuclear Medicine (Drs Wahl and Heard), University
of Michigan, Ann Arbor.
REFERENCES
| |
1. Shah J. Patterns of cervical lymph node metastasis from squamous carcinomas
of the upper aerodigestive tract. Am J Surg. 1990;160:405-409.
ISI
| PUBMED
2. Jones A. Occult node metastases in head and neck squamous carcinoma. Eur Arch Otorhinolaryngol. 1993;250:446-449.
PUBMED
3. Van den Brekel M. Micrometastases from squamous cell carcinoma in neck dissection specimens. Eur Arch Otorhinolaryngol. 1993;249:349-353.
4. Steiner W. Diagnosis and treatment of the N0 neck of carcinomas of the upper aerodigestive
tract. Eur Arch Otorhinolaryngol. 1993;250:450-456.
PUBMED
5. Freidman M. Rationale for elective neck dissection in 1990. Laryngoscope. 1990;100:54-59.
ISI
| PUBMED
6. Sobol S, Jensen C, Sawyer W. Objective comparison of physical dysfunction after neck dissection. Am J Surg. 1985;150:503-509.
FULL TEXT
|
ISI
| PUBMED
7. Remmler D, Scheetz J, White G, Goepfert H. A prospective study of shoulder disability resulting from radial and
modified neck dissections. Head Neck Surg. 1986;8:280-286.
ISI
| PUBMED
8. Leipzig B. Functional evaluation of the spinal accessory nerve after neck dissection. Am J Surg. 1983;146:526-530.
FULL TEXT
|
ISI
| PUBMED
9. Lenz M. Diagnosis and treatment of the N0 neck in carcinomas of the upper aerodigestive
tract: current status of diagnostic procedures. Eur Arch Otorhinolaryngol. 1993;250:432-438.
PUBMED
10. Ali S. False-positive and false-negative neck nodes. Head Neck Surg. 1985;8:78-82.
ISI
| PUBMED
11. Fisch U, Sigel M. Cervical lymphatic system visualized by lymphography. Ann Otol Rhinol Laryngol. 1964;73:869-892.
ISI
12. Morton D. Technical details of intraoperative lymphatic mapping for early stage
melanoma. Arch Surg. 1992;127:392-399.
FREE FULL TEXT
13. Krag D. Surgical resection and radiolocalization of the sentinel lymph node
in breast cancer using a gamma probe. Surg Oncol. 1993;2:335-340.
FULL TEXT
|
ISI
| PUBMED
14. Giuliano A. Lymphatic mapping and sentinel lymphadenectomy for breast cancer. Ann Surg. 1994;220:391-401.
ISI
| PUBMED
15. Koch WM, Choti MA, Civelek AC, Eisele DW, Saunders JR. Gamma probe–directed biopsy of the sentinel node in oral squamous
cell carcinoma. Arch Otolaryngol Head Neck Surg. 1998;124:455-459.
FREE FULL TEXT
16. Pitman KT, Johnson JT, Edington H, et al. Lymphatic mapping with isosulfan blue dye in squamous cell carcinoma
of the head and neck. Arch Otolaryngol Head Neck Surg. 1998;124:790-793.
FREE FULL TEXT
17. Van Der Veen H. Gamma probe–guided sentinel node biopsy to select patients with
melanoma for lymphadenectomy. Br J Surg. 1994;81:1769-1770.
ISI
| PUBMED
 CiteULike  Connotea  Del.icio.us  Digg  Reddit  Technorati  Twitter
What's this?
RELATED ARTICLE
Archives of Otolaryngology–Head & Neck Surgery Reader's Choice: Continuing Medical Education
Arch Otolaryngol Head Neck Surg. 2001;127(8):1011-1012.
FULL TEXT
THIS ARTICLE HAS BEEN CITED BY OTHER ARTICLES
|
Evaluation of the Efficacy of a Small CdTe {gamma}-Camera for Sentinel Lymph Node Biopsy
Tsuchimochi et al.
JNM 2008;49:956-962.
ABSTRACT
| FULL TEXT
Sentinel Node Biopsy in N0 Squamous Cell Carcinoma of the Oral Cavity and Oropharynx in Patients Previously Treated With Surgery or Radiation Therapy: A Pilot Study
Hart et al.
Arch Otolaryngol Head Neck Surg 2007;133:806-809.
ABSTRACT
| FULL TEXT
Sentinel Lymph Node Biopsy in N0 Squamous Cell Carcinoma of the Oral Cavity and Oropharynx
Hart et al.
Arch Otolaryngol Head Neck Surg 2005;131:34-38.
ABSTRACT
| FULL TEXT
Feasibility and Cost-effectiveness of Sentinel Lymph Node Radiolocalization in Stage N0 Head and Neck Cancer
Kosuda et al.
Arch Otolaryngol Head Neck Surg 2003;129:1105-1109.
ABSTRACT
| FULL TEXT
Assessment of Metastatic Cervical Adenopathy Using Dynamic Contrast-Enhanced MR Imaging
Fischbein et al.
Am. J. Neuroradiol. 2003;24:301-311.
ABSTRACT
| FULL TEXT
The ability of lymphoscintigraphy to direct sentinel node biopsy in the clinically N0 neck for patients with head and neck squamous cell carcinoma
Ross et al.
Br. J. Radiol. 2002;75:950-958.
ABSTRACT
| FULL TEXT
|
