TY - JOUR
AU - Saunders, John R.
AB - ObjectiveManagement of the N0 neck in head and neck squamous cell carcinoma is an important issue for the head and neck surgeon. Experience with radionuclide-labeled colloid injection to identify a sentinel node in malignant melanoma suggests a high level of accuracy for this approach to identify microscopic metastasis when present. We set out to explore the feasibility of using the handheld gamma probe to identify radiolabeled sentinel nodes in oral squamous cell carcinoma.Patient PopulationFive individuals with N0 necks and accessible oral or oropharyngeal primary sites from a major tertiary referral center.MethodsRadiolabel with unfiltered technetium Tc 99m sulfur colloid was injected in quadrants around the primary site followed by immediate dynamic lymphoscintigraphy. Open biopsy of the sentinel node was accomplished within 2 hours of injection after extirpation of the primary site. Regional or complete neck dissection was performed after sentinel node biopsy.ResultsSentinel node biopsy accurately identified one or several nodes in 2 cases, including nodes containing metastatic cancer in 1. In the other 3 cases, the radiolabel failed to identify the sentinel node despite the presence of metastatic disease in the nodes at final pathologic study in 2.ConclusionsDetection and biopsy of the sentinel node are feasible for selected patients with oral head and neck squamous cell carcinoma with N0 necks. There is a potential savings of time, cost, and morbidity with this approach. However, several substantial problems were encountered with the technique in this limited series of patients. Establishing the reliability of lymphoscintigraphy in this setting would require testing in a much larger patient cohort. Our experience suggests that such an investment may not be warranted.OVER THE past decade, the use of radiolabel with technetium Tc 99m sulfur colloid injection to locate the first echelon nodal basin for biopsy has gained increasing acceptance in the management of malignant melanoma. The first echelon nodes were given the name "sentinel nodes"(SNs) indicating their utility for predicting the status of the nodal basin.Cutaneous melanoma greater than 0.76 mm in depth has a 20% to 30% incidence of occult nodal metastasis. In response to the ongoing debate over the efficacy and necessity of nodal dissection in this setting, the concept of sampling the SN to direct the treatment of nodal basins was proposed. Initially, the dispersal of the radiotracer was followed with the use of a gamma camera to indicate the nodal basins serving the region of the primary tumor. Dissection of these basins followed. Particularly in melanoma of the head and neck, a substantial number of cases demonstrated drainage patterns judged to be discordant with those that were considered at risk on the basis of clinical experience. In several cases, the discordant nodal basin contained metastatic disease.Later, inert blue dyes were used to locate SNs intraoperatively. The dye was injected after the lymphoscintigraphy was completed, a few moments before surgery, and staining of lymphatic channels and nodes was sought after incision and flap elevation. This approach was found to be highly accurate. In several series, SNs could be identified in more than 90% of cases.Reintgen et aldemonstrated the sensitivity of this approach showing that no case with a negative SN had metastatic disease in any other node on full-neck dissection. Many SNs (34 of 42) did not contain metastatic disease. However, there were no cases in which a node closer to the primary site than the SN was involved with tumor. Thompson et alhad similar results using the same approach, but reported 2 cases in which a positive node was found in the neck dissection specimen when the SN was negative. The failure was blamed on technical difficulty with dye localization and the positive nodes were judged to be additional SNs missed on initial dissection. In 2 other series, neck dissection was performed if the SN was positive, but cases with negative SN biopsy results were managed expectantly. As a result, clinical metastatic disease developed later in several cases.Again, the authors reported difficulty with finding the stained lymphatics intraoperatively in several cases.The difficulty with intraoperative localization of stained nodes has been addressed by the use of a handheld gamma probe to direct the nodal biopsy. The feasibility of this technique was first demonstrated in experimental animalsand then in several series of patients with melanoma.The relative power of the gamma probe to identify the SN was shown in a third report comparing it with the blue dye technique. The gamma probe successfully located a positive SN that would have been missed using blue dye alone in several cases.Our experience with melanoma of a variety of cutaneous sites, including the head and neck, concurs with the literature that this tool permits a minimally invasive, highly accurate, stepwise approach to the management of occult nodal disease. The handheld probe is easy to use and leads to a definite SN in most cases. We have been encouraged by cases in which the SN contained microscopic metastases and was the only node with disease. In all cases with a negative SN, no other metastatic site has been identified to date.Lindberg'sclassic treatise on the pattern of nodal metastasis in head and neck squamous cell carcinoma (HNSCC) serves as the basis for predicting the likely location of metastatic disease for a given primary site. For primary tumors of most sites in the upper aerodigestive tract, the jugular-digastric nodes at levels II-III are the most likely to contain metastases. However, clinical experience may not provide fail-safe information to direct therapy for individual cases. Byers et alreport that 16% of 277 patients with squamous cell carcinoma (SCC) of the oral tongue had "skip metastases" that bypassed the expected first echelon nodal basin.A number of tools have been used to gather information about the nodal status of patients with HNSCC with no clinical neck disease and include computed tomography (CT), magnetic resonance imaging, positron emission tomography, and ultrasound-guided fine needle aspiration.We wondered whether the use of radionuclide lymphoscintigraphy and intraoperative localization of SNs with use of the gamma probe could be extended to use in HNSCC.The literature contains some precedents for using lymphoscintigraphy in HNSCC. The initial studies did not involve injection of the primary site directly. However, the nodal drainage system could be visualized after injecting the skin of the neck, scalp, or retromolar trigone.Alex and Kraghave reported successful use of the gamma probe approach in one case of supraglottic SCC. The tumor site was injected at the time of direct laryngoscopy, and radionuclide uptake was found in 2 of 3 positive nodes. To our knowledge, this is the only published reference to direct injection of the mucosa of the upper aerodigestive tract for lymphoscintigraphic analysis of the nodal basin. The feasibility and efficacy of SN biopsy for oral SCC has not been rigorously established. We set out to gain more information about the applicability of the technique.PATIENTS AND METHODSPATIENT SELECTIONFive consecutive patients with a T1 or T2 SCC in a location amenable to injection of the radioisotope under topical anesthesia and a clinically N0 neck as assessed by findings from physical examination and CT were identified at Johns Hopkins Hospital and The Greater Baltimore Medical Center, Baltimore, Md. Each individual provided informed written consent for lymphoscintigraphy and for SN biopsy in conjunction with regional or comprehensive neck dissection and surgical extirpation of the primary oral lesion.INJECTION OF RADIOISOTOPETopical anesthesia was provided using tetracaine (Pontocaine) spray and 20% benzocaine paste. In one case a regional block of the lingual nerve was accomplished using 1% lidocaine (Xylocaine). Unfiltered technetium Tc 99m sulfur colloid was injected into the mucosa in 3 to 4 locations around the perimeter of the oral lesion while the patient was in the nuclear medicine suite. Each injection volume was approximately 0.1 mL and the total dose used was less than 1 mCi (in SI notation, 1 mCi=37 MBq) per patient.DYNAMIC LYMPHOSCINTIGRAPHYBeginning approximately 2 minutes after the first of 4 injections, a large field-of-view gamma camera set at 20% window peak energy and fitted with a low-energy, high-resolution parallel hole collimator was used to follow the disbursement of the radionuclide in the lymphatics. A hot marker or a transmission scan was used to outline the body for one or more images. Counts were accumulated for 1 minute per frame for up to 30 minutes, imaging in both the anterior and lateral positions. A point-of-source marker was used to estimate the location of any visualized lymph node, and the site was marked with a skin pen.INTRAOPERATIVE LOCALIZATION OF SNA handheld gamma probe with automated counter (C-Trak System, Carewise Medical Products Corporation, Morgan Hill, Calif) was used to determine the position of SNs prior to and during the operative procedure. Patients were scheduled in the operating room 90 to 120 minutes after lymphoscintigraphy. Prior to excision of the primary site, the SN could be located by carefully directing the probe away from the oral cavity. The primary site was then resected transorally in 3 of 4 cases. The primary site of the fourth case was not amenable to wide local excision, but no SN was identified in this case. After preparing the neck for biopsy, the probe was again used to locate the SN, which now gave off a greatly reduced signal from the primary bed. A small (3-to 4-cm) incision was made through skin and platysma and the probe was used to direct exposure of the labeled node(s). These were removed and sent for frozen section. If the SN had metastatic disease, a complete modified neck dissection was performed (cases 1 and 3). In our second case the SNs were negative and a selective (bilateral suprahyoid) neck dissection was performed.RESULTSCASE 1A 45-year-old male cigar smoker presented with an SCC of the right lateral tongue, staged T2 N0 M0. Findings from a CT scan and physical examination revealed a small (<1-cm), soft, right jugular-digastric node without ring enhancement or central necrosis (Figure 1, A). Five minutes after injection with technetium Tc 99m sulfur colloid, kinetic images of the region demonstrated the presence of a single region of uptake in the right midjugular node (Figure 1, B). Continued gamma camera monitoring for a total of 15 minutes failed to demonstrate additional areas of tracer uptake.Case 1. A, Axial computed tomographic image of cervical adenopathy at the level of the sentinel node. B, Lymphoscintigraphy image collated over minutes 11 to 15 of the dynamic imaging after injection in the primary tongue site. C, Photomicrograph of microscopic metastatic implant of squamous cell carcinoma in the sentinel node. The diameter of the implant is noted to be similar to the size of a nodal follicular center (hematoxylin-eosin, original magnification ×200).The patient was taken to the operating room 1.5 hours after injection. After induction of general anesthesia, the handheld gamma probe was used to scan the oral cavity and neck. Gamma probe counts were automatically tabulated for the primary site and neck (Table 1). The primary site was then removed by wide local excision. A curvilinear incision was made in the right aspect of the neck, 4 cm in length, in a location amenable to extension for complete neck dissection. The gamma probe provided a signal to direct the initial incision and was useful in guiding the dissection, quickly exposing a small, benign-appearing node in the level III jugular chain. This node was 2 cm below the clinically palpable jugular digastric node.Handheld Gamma Probe Results: Case 1See table graphicOn excision of this SN, the gamma probe confirmed its high concentration of the radiotracer. A neighboring level III node had a lower number of counts, and was also excised. When the SN was examined by frozen section, a focus of metastatic SCC was identified (Figure 1, C).A right modified radical neck dissection ensued. The clinically palpable level II-III node was removed and found to have very little tracer uptake. A single level IV node was identified to have uptake by the gamma probe. Both of these nodes were negative for tumor while the smaller level III node with 1295 counts per 10 seconds also contained metastatic SCC. The final pathology report showed 2 of 26 nodes to be involved with metastatic cancer. The patient received postoperative radiation therapy.CASE 2A 75-year-old woman presented with a SCC involving the mandibular gingiva from the left distal bicuspid to the right canine tooth. Findings on CT and physical examination demonstrated no suspicious nodes (T2 N0). Lymphoscintigraphy scanning for 30 minutes showed no nodal uptake of radionuclide separable from the injection sites. Use of the handheld gamma probe in the operating room was also compromised because of the continuity of the primary site and the submental nodal bed. With surgical exploration of the suprahyoid, a submental node was measured as emitting 216 counts per 10 seconds after excision (background=0 counts per 10 seconds). Two other level I and 1 level III nodes also had low levels of radioactivity. Frozen section analysis showed no metastatic cancer and all lymph nodes were read as negative for tumor. A bilateral suprahyoid neck dissection was performed, and all nodes were negative for tumor.CASE 3A 43-year-old woman presented with a history of leukoplakia on the left lateral tongue from which a biopsy specimen had been obtained 2 years earlier and had been found to be benign. Subsequently, she had an enlargement of the leukoplakic region to 13 mm in diameter. Her neck had no clinical sign of nodal metastasis, although results of CT showed a 1.3-cm ipsilateral subdigastric node. On the day of surgery, 0.2 mL of technetium Tc 99m sulfur colloid was injected into the submucosa at several sites around the lesion. A lymphoscintigram showed uptake in a single level II lymph node. In the operating room, the level II node was found to emit 27390 counts per 10 seconds, while 2 level I nodes were measured at 7334 and 6216 counts. All 3 nodes were histologically benign on frozen section. A supraomohyoid neck dissection was performed and the remaining 21 lymph nodes were also histologically benign.CASE 4A muscular 60-year-old man presented with a large left tonsil mass. He had no palpable cervical adenopathy, although a CT scan showed a 1.2-cm node in the left midjugular chain. This node had no ring enhancement or central necrosis. Dynamic lymphoscintigraphy showed a node low in the left jugular chain within 15 minutes of injection. After resection of the tonsil, 2 distinct areas of activity in the neck were noted prior to incision. The greater activity was detected in level IV, consistent with the lymphoscintigram (1685 counts per 10 seconds) (background count=6 per 10 seconds). A lower level of activity was noted in level II at the approximate location of the 1.2-cm node noted on CT scan (995 counts per 10 seconds).Two separate incisions were planned so that they could be extended into MacFee incisions if a full neck dissection was needed. The level IV node had 2305 counts per 10 seconds. When excised, frozen section analysis indicated that it was negative for tumor. The higher node was firm and clinically suggestive of cancer once exposed by reflecting the platysma and sternocleidomastoid muscles. It had 932 counts per 10 seconds and was positive for SCC, which had completely replaced the nodal tissue. A modified radical neck dissection ensued. Two other level II-III nodes were also extensively infiltrated by metastatic disease.CASE 5A 53-year-old man presented after receiving full-course radiation therapy for a T2 N0 SCC of the right retromolar trigone. Four months later biopsy confirmed that there was persistent tumor in the primary site. There was a firm consistency to the ipsilateral submandibular gland, consistent with radiation-induced fibrosis, but findings from physical examination and CT revealed no obvious lymphadenopathy. No SN was visible by lymphoscintigraphy. In the operating room, no uptake of radioactivity was noted in the neck, prior to incision nor in any of the exposed nodes found on dissection. An 8-mm submandibular node, and several tiny (3- to 4-mm) level III nodes were found to have been replaced by metastatic disease. A modified radical nodal dissection was performed.COMMENTGamma probe–directed biopsy of the SN offers potential benefits for the management of HNSCC with N0 necks. In most cases of melanoma reported in the literature, when a SN contained metastasis, it was the only node involved. This suggests the possibility of limiting neck dissections to a selected surrounding region even after positive SN biopsy findings. Indeed, selective neck dissection is currently advocated by some for the treatment of patients with HNSCC with a single clinically involved node.If gamma probe detection of the SN were proven to have sufficient sensitivity to permit the cessation of further surgery or postoperative irradiation on the basis of a negative biopsy, the saving of time, cost, and morbidity would be substantial. At our institution, the cost of the nuclear medicine study is $400 and an outpatient deep node biopsy including operating room charges and anesthesia, pathology, and surgical department fees is approximately $3500. In contrast, the cost of a regional nodal dissection and 3 days of hospitalization is about $13500. The cost of resection of the primary malignancy is highly variable and thus should not be not factored into this price comparison.Our preliminary experience with handheld gamma probe–directed SN biopsy for N0 oral SCC provides a number of intriguing observations. The first case demonstrates the potential value of the technique. Use of lymphoscintigraphy and the gamma probe permitted definite SN localization quickly and easily. The node was lower than a small, palpable node judged more likely to be the first echelon node by the surgeon. The radioisotope correctly identified the SN that contained metastatic carcinoma while the more proximal node had no tumor. The ratio of radiolabel in the SN to background radiolabel in the wound was very high (37:1). A second node with a lower but definite uptake of radiolabel was also involved with tumor. Both nodes appeared to be benign by gross inspection even at the time of removal. The gamma probe successfully identified SNs in case 3, although no metastatic disease was present.The other cases demonstrate a number of problems with the application of lymphoscintigraphy to HNSCC. Among these problems are are the following:The primary tumor site in HNSCC is usually very near to the first echelon nodal basin obscuring the dynamic lymphoscintigram and the independent detection of the radioactive node with the gamma probe. This problem was most clearly demonstrated in case 2. The background counts of the primary site remained high even after wide local excision. Angulation of the gamma probe away from the primary site helps to isolate the radioactive signal from the nodal bed, but this is ineffective when examining the submental region for nodal spread from a primary tumor site of the anterior oral cavity.Intramucosal injection of the radiolabeled technetium Tc 99m sulfur colloid is technically more difficult to achieve than is intradermal injection in cases of melanoma. Dynamic lymphoscintigraphy often demonstrated isotope in the esophagus having extruded from the injection site into the saliva.Replacement of a lymph node with bulky metastatic cancer results in the diversion of the lymphatic flow and radiolabeling of downstream nodes. Grossly positive nodes may be missed by clinical evaluation in the setting of the thick neck and soft tissues as seen in case 4. This finding is consistent with the experience with melanoma reported by Pijpers et al.Patients with HNSCC frequently receive primary radiation therapy that can alter the local lymphatic channels rendering lymphoscintigraphy ineffective. Case 5 exemplifies this problem. Prior irradiation, like previous surgery in the primary site, may be a contraindication to SN biopsy that uses lymphoscintigraphy.The inaccessibility of many HNSCC primary sites to local injection without general anesthesia limits the applicability of lymphoscintigraphy for this population.Many issues pertinent to the necessity and efficacy of SN biopsy for HNSCC would need to be addressed to establish the utility of this approach. First, is it necessary to use any technique to identify the SN in HNSCC? If the pattern of spread is accurately predicted by clinical observation, there is no need for radionuclide detection. The discordance seen with melanoma between the lymphoscintigraphy results and the clinical prediction of the site of the first echelon node was unexpected and substantial (present in up to 80% of the cases). Next, what is the degree of accuracy with which the radionuclide injection identifies the SN? Furthermore, the depth of invasion of oral SCC may be greater than that seen in most cutaneous melanomas, making it difficult to accurately map the lymphatic drainage patterns with percutaneous injection of technetium Tc 99m sulfur colloid. Our first case demonstrated the potential benefit of radiolabeled colloid identification of the SN through the accurate localization of small nodes containing subclinical metastases. The other cases raise a number of questions that cast doubt on the general applicability and utility of the technique for HNSCC.Molecular detection of submicroscopic amounts of cancer metastasis may improve the yield of SN biopsy. In an intriguing study, reverse-transcriptase polymerase chain reaction was used to identify tyrosinase messenger RNA produced by rare metastatic melanoma cells in a background of normal lymph cells.Histologically negative SNs that contain submicroscopic amounts of tyrosinase-producing melanoma predict 3 of 4 locoregional recurrences in a series of 74 patients.In a similar way, we have shown that an oligomer probe created to match a tumor-specific p53 mutation in HNSCC can identify submicroscopic levels of SCC in lymph nodes.It would be of great interest to combine molecular detection of rare cancer cells with selected nodal biopsy of patients with N0 HNSCC. The clinical utility of such an approach would depend on the timely availability of the molecular analysis and on the demonstration of the impact of submicroscopic disease on the clinical outcome.CONCLUSIONSWe have found that the intraoperative handheld gamma probe–directed SN biopsy is feasible for selected cases of N0 HNSCC in which the primary site is accessible to injection. Two cases suggest that the technique facilitates the identification of SN permitting a minimally invasive, stepwise approach to assess the neck. The other cases demonstrate significant limitations to the applicability of the technique. These limitations may preclude further analysis and prevent general acceptance of this approach to minimally invasive evaluation of the nodal drainage basin in HNSCC.KEWellsCWCruseSDanielsCBermanJNormanDSReintgenThe use of lymphoscintigraphy in melanoma of the head and neck.Plast Reconstr Surg.1994;93:757-761.CJO'BrienRFUrenJFThompsonPrediction of potential metastatic sites in cutaneous head and neck melanoma using lymphoscintigraphy.Am J Surg.1995;170:461-466.DLMortonDRWenLJFoshagACochranIntraoperative lymphatic mapping and selective cervical lymphadenectomy for early-stage melanomas of the head and neck.J Clin Oncol.1993;11:1751-1756.DReintgenCWCruseKWellsThe orderly progression of melanoma nodal metastases.Ann Surg.1994;220:759-767.JFThompsonWHMcCarthyCMBoschSentinel lymph node status as an indicator of the presence of metastatic melanoma in regional lymph nodes.Melanoma Res.1995;5:255-260.JCAlexDNKragGamma-probe guided localization of lymph nodes.Surg Oncol.1993;2:137-143.JCAlexDLWeaverJTFairbanksBSRankinDNKragGamma-probe-guided lymph node localization in malignant melanoma.Surg Oncol.1993;2:303-308.AMudunDRMurraySCHerdaEarly stage melanoma: lymphoscintigraphy, reproducibility of sentinel node detection, and effectiveness of the intraoperative gamma probe.Radiology.1996;199:171-175.JJAlbertiniCWCruseDRapaportIntraoperative radiolymphoscintigraphy improves sentinel lymph node identification for patients with melanoma.Ann Surg.1996;223:217-224.RLindbergDistribution of cervical lymph node metastases from squamous cell carcinoma of the upper respiratory and digestive tracts.Cancer.1972;29:1446-1450.RMByersRSWeberTAndrewsDMcGillRKarePWolfFrequency and therapeutic implications of "skip metastases" in the neck from squamous cell carcinoma of the oral tongue.Head Neck.1997;19:14-19.MWvan den BrekelJACastelijnsHVStelRPGoldingCJMeyerGBSnowModern imaging techniques and ultrasound-guided aspiration cytology for the assessment of neck node metastases: a prospective comparative study.Eur Arch Otorhinolaryngol.1993;250:11-17.STeruiTTerauchiSEbiharaLymphoscintigraphy of head and neck cancer.Angiology.1992;43:925-932.SSri-PathmanathanRRailtonLymphoscintigraphy in the detection of cervical metastases from oral carcinoma: a pilot study.Ann R Coll Surg Engl.1989;71:281-284.JCAlexDNKragThe gamma-probe-guided resection of radiolabeled primary lymph nodes.Surg Oncol Clin N Am.1996;5:33-41.SJTraynorJICohenJGrayPEAndersenECEvertsSelective neck dissection and the management of the node-positive neck.Am J Surg.1996;172:654-657.RPijpersGJColletSMeijerOSHoekstraThe impact of dynamic lymphoscintigraphy and gamma probe guidance on sentinel node biopsy in melanoma.Eur J Nucl Med.1995;22:1238-1241.DReintgenMore rational and conservative surgical strategies for malignant melanoma using lymphatic mapping and sentinel node biopsy techniques.Curr Opin Oncol.1996;8:152-158.XWangRHellerNVanVoorhisDetection of submicroscopic lymph node metastases with polymerase chain reaction in patients with malignant melanoma.Ann Surg.1994;220:768-774.JABrennanLMaoRHHrubanMolecular assessment of histopathologic staging in squamous cell carcinoma of the head and neck.N Engl J Med.1995;332:429-435.Accepted for publication January 13, 1998.Presented at the Annual Meeting of the American Society for Head and Neck Surgery, Scottsdale, Ariz, May 13, 1997.Corresponding author: Wayne M. Koch, MD, The Johns Hopkins Medical Institutions, PO Box 41402, Baltimore, MD 21203-6402 (e-mail: wkoch@gwgate1.jhmi.jhu.edu).
TI - Gamma Probe–Directed Biopsy of the Sentinel Node in Oral Squamous Cell Carcinoma
JF - JAMA Otolaryngology - Head & Neck Surgery
DO - 10.1001/archotol.124.4.455
DA - 1998-04-01
UR - https://www.deepdyve.com/lp/american-medical-association/gamma-probe-directed-biopsy-of-the-sentinel-node-in-oral-squamous-cell-cIHlqLLrnU
SP - 455
EP - 459
VL - 124
IS - 4
DP - DeepDyve
ER -