Chien Chen
Abstract
The purpose of histopathologic evaluation of neck dissections is to identify metastatic disease to locoregional lymph nodes. The standard protocol for processing neck dissections is effective for the evaluation of macrometastases and works well for clinically positive necks, but clinically negative necks (cN0) present a clini- copathologic dilemma. While many clinically negative necks harbor occult macroscopic and microscopic disease, intensive histopathologic evaluation of neck dissections to identify microscopic disease is neither economically nor technically feasible. Sentinel node biopsy is a sensitive strategy for identifying cN0 patients who require neck dissection and also provides a potential target for more intensive histologic evaluation. Despite relatively low sensitivities, intraoperative evaluation of sentinel lymph nodes is beneficial because it has high specificity and can facilitate neck dissection in roughly half of patients who require one. Although micrometastases and even isolated tumor cells are clinically important, there is no consensus standard protocol for intensive evaluation of sentinel lymph nodes in the head and neck literature. We propose a protocol for the evaluation of sentinel lymph nodes of the head and neck based on the current evidence. Thyroid cancer is discussed separately as it does not require comparable aggressive surgical management of microscopic nodal disease. The pathologic evaluation and workup of nodal metastases of unknown primary is also discussed.
Keywords: neck dissection, pathology, lymph node, sentinel, macrometastasis, micrometastasis, isolated tumor cells, occult primary
27.1 Introduction
The purpose of the neck dissection is to assess for lymph node metastases and excise regional disease. This chapter will discuss the pathologic processing and evaluation of neck dissections, its limitations, and possible solutions for these limitations. Pathologic nodal staging (pN) is based on the American Joint Committee on Cancer (AJCC) staging guidelines.1 For head and neck tumors, the principal criteria that determine nodal staging and therefore prognosis are (1) the presence or absence of metasta- ses, (2) the size of the largest metastasis, (3) the presence or absence of extranodal extension, and (4) the laterality of involvement. Interestingly, the number of involved lymph nodes has relatively little impact on the pathologic nodal stage, although it is a reflection of the burden of disease. Of these criteria, the single most significant driver of prognosis for all head and neck sites is the presence or absence of lymph node metasta- ses.2.3.4 The purpose of histopathologic evaluation is to answer each of these questions regarding nodal involvement.
27.2 Processing and Reporting of Neck Dissections
To understand the histopathologic evaluation of neck dissections, it is important to understand how they are processed.
The processing of neck dissections requires cooperation between the surgeon and the pathologist. Radical neck dissections have anatomic landmarks, which allow the pathologist to orient and separate the neck dissection into individual nodal levels at the grossing bench. However, these landmarks are absent in both modified neck dissections and selective neck dissections; thus, it is important for the surgeon to aid in the orientation of these specimens. This may be readily achieved in two ways. The entire neck dissection can be pinned out on a cork board and oriented by the surgeon using a surgical marker, either in the operating room or subsequently in the surgical pathology gross room. Alternatively, the specimen can be divided into individual levels in the operating room and the individual levels can be sent separately. Orientation by the surgeon ensures that the specimen will be properly separated into the correct nodal levels for pathologic evaluation.
Once separated into individual levels, the pathologic processing and evaluation of neck dissections follows a generic protocol applicable to lymph node dissections in general.5 Specimens should be received and processed fresh rather than in formalin fixative. Fresh tissue is softer than fixed tissue and the difference in the consistency of the lymph nodes and the surrounding soft tissue is more readily appreciable, making them easier to identify. Once identified, each lymph node should be blunt dissected from the surrounding tissue. Fresh processing is particularly relevant for neck dissections as they often contain significant amounts of fibromuscular tissue which can become quite firm after fixation, making blunt dissection difficult. A thorough search for lymph nodes should be made with the intent to identify all lymph nodes in the dissection. Despite the fact that a minimum lymph node count is not part of pathologic nodal staging, it is important to have an accurate lymph node count because it is a surrogate measure of the quality of the neck dissection and is an independent predictor of the risk of recurrence.67 8 If lymph nodes are not identified by palpation, the tissue should be submitted in toto. While this may compromise the integrity of the lymph node count, lymph nodes that are grossly undetectable are unlikely to harbor tumor and in to- to submission increases the sensitivity for detecting lymph nodes, which is paramount. All identified lymph nodes should be submitted for evaluation. For lymph nodes grossly involved by tumor, measurement of largest metastasis and evaluation for extranodal extension for each lymph node with one or more gross sections are generally sufficient.
Grossly uninvolved lymph nodes should be serially sectioned at 2-mm intervals and submitted in toto (Fig. 27.1). Smaller lymph nodes (< 2 cm) should be sectioned perpendicular to the shortest axis (longitudinal), whereas larger lymph nodes should be sectioned perpendicular to the longest axis (transverse). Care should be taken to place the sections in the cassette sequentially such that different surfaces are presented for cutting to ensure that the lymph node is consistently being evaluated at 2-mm intervals. While the Association of Directors of Anatomic and Surgical Pathology (ADASP) recommendations accept a thicker section (3-4 mm),5 this risks missing small macrome- tastases (>2mm) and current guidance from the College of American Pathologists favors the 2-mm gross section thickness.9 If multiple lymph nodes are submitted in a cassette, the surface of the lymph node should be differentially inked prior to sectioning to identify which sections belong to which lymph node so that positive lymph nodes are only counted once.
Fig.27.1 Protocol for histologic processing of nonsentinel lymph nodes.
Only lymph nodes from the same nodal level/sublevel should be submitted together in a cassette. Lymph nodes too small to be grossly sectioned should be submitted whole with similarly sized lymph nodes from the same nodal level. Cassette summaries should indicate how many lymph nodes are present in each cassette, if and how they have been inked and sectioned, and from which nodal level they originate. One hematoxylin and eosin (H&E) slide containing multiple histologic sections cut at 2- pm thickness is generally considered sufficient for histologic evaluation. If a submandibular gland is identified, it should be serially sectioned and grossly evaluated for abnormalities, and a representative section should be taken to document the presence of the salivary gland and any identified abnormality.
Pathologic reports should include the following information: the total number of lymph nodes identified in each nodal level and the number involved by tumor, the presence of extranodal extension, the size of the largest metastasis (not the size of the involved lymph node), the presence of any soft-tissue deposits without evidence of lymph node architecture, and the presence of submandibular glands and associated pathology.5
27.3 Limitations of Histopathologic Evaluation
The greatest limitation of histologic evaluation of neck dissection specimens is that it only evaluates a minute fraction of the tissue received. When tissue is received from the operating room, based on gross evaluation, only representative gross sections are usually taken to evaluate critical information such as margin status, histologic type and grade, lymphovascular invasion, and mitotic activity. Because of this, the taking of representative gross sections requires both judgment and experience to maximize yield. It is a testament to the importance of nodal status that grossly uninvolved lymph nodes are generally submitted in toto. Even then, only a tiny fraction of the submitted tissue is ever histologically evaluated. Given 2-mm gross sections, the fact that a histologic section is 2 pm means that only 0.1% of the tissue is evaluated in each histologic section. The standard protocol essentially takes a representative “snapshot” of the tissue at 2-mm intervals, and can therefore identify all lesions larger than 2 mm in smallest dimension if the gross section thickness is 2 mm and the sections are properly oriented and embedded. Smaller lesions may never enter the plane of sectioning and can therefore be missed altogether.
So why do we accept this limitation? Essentially, complete histologic evaluation is impractical. True complete histologic evaluation of even a single 1cm x 0.6 cm x 0.6 cm lymph node would require 3,000 histologic sections, more than the entire daily workload of the average academic surgical pathology laboratory and would thus be both prohibitively expensive and technically infeasible. Even with representative histologic evaluation, each incremental improvement in sensitivity increases the histopathologic workload geometrically. This is a big deal because the typical neck dissection contains approximately 20 lymph nodes which currently require 20 to 40 H&E slides to evaluate so workloads can quickly spiral out of control. The current processing protocol is a tradeoff between yield and resource utilization and can identify all macrometastases (>2mm) but have a significant risk of missing micrometastasis (2 to 0.2 mm) and largely miss isolated tumor cells (ITCs; < 0.2 mm).
Another limitation of standard processing is the difficulty of identifying small tumor deposits on routine H&E stain. Small groups and single tumor cells may resemble histiocytes or endothelial cells, which are normal components of the lymph node. Similarly, basaloid and small round blue cell tumors can blend into the background lymphoid tissue at scanning power. This limitation can usually be overcome by careful examination and judicious use of immunohistochemical (IHC) stains. However, routine use of IHC stains to evaluate lymph nodes is cost prohibitive and can generate false-positive results.
Despite these limitations, standard processing and reporting works well for neck dissections in which gross disease is evident. Once gross tumor is identified in a lymph node, the main additional pieces of information needed are the size of the largest metastasis and whether there is contralateral involvement. Since the standard grossing protocol can find all macrometastases, high N stage disease can be readily assessed in this manner.
27.4 The Dilemma of the cN0 Neck
Clinical N0 necks (cN0) present a problem. It is well documented that a significant portion of cN0 necks contain tumor on pathologic evaluation.3,10 The risk of involvement is dependent on tumor site, tumor (T) stage, and tumor thickness. In high T stage cancers (T3 and T4), a substantial fraction of cN0 necks contain tumor and elective neck dissection is generally considered to be prudent regardless of the clinical nodal status.4,10 The real challenge lies with low T stage (T1 and T2) lesions, which are estimated to have an approximately 20 to 50% risk of nodal metastases despite being cN0, depending on primary site.4 That these clinically unapparent nodal metastases affect survival has been amply demonstrated both by comparison of watchful waiting versus elective prophylactic neck dissection11,12 and by comparison of pN0 and pN1 patients given elective neck dissection despite cN0 status.13 In these low T stage patients, routine elective neck dissection risks overtreatment with its associated morbidity and expense but watchful waiting risks undertreatment and worse disease-free and overall survival. This has led to significant research into strategies to predict the risk factors for nodal metastases in low T stage tumors, clinically, radiologically, and pathologically.
27.5 Sentinel Node Biopsy
The premise of sentinel lymph node biopsy is that the lymphatic drainage of a tumor can be identified using tracers (radioactive, colorimetric, fluorescent, etc.), which, if injected into the tumor site, will enter the lymphatics around the tumor and accumulate in the first lymph node(s)—the sentinel lymph node(s)—that drain the tumor site. Tumor cells travelling through the lymphatics either take residence in the first lymph node(s) they encounter or “skip” metastasizes, i.e. bypass the first encountered lymph nodes. Skip metastases are rare. If the “sentinel lymph nodes” are negative, the likelihood of nodal involvement in the entire nodal basin is low. This strategy has been used successfully in other cancers and has in fact become the standard of care in breast cancer and melanoma. While not widely used for squamous cell carcinoma of the head and neck in the United States, many studies, mostly in the European and Asian literature, suggest that this strategy has the potential to solve the dilemma of the cN0 neck in low T stage head and neck tumors.14151617
There are a number of theoretical limitations to sentinel lymph node biopsy in head and neck tumors that may explain its slow adoption in the United States.1017 It has been asserted that the proximity of the lymph node basins to the tumor site, and therefore to the site of tracer injection, can obscure the detection of the sentinel lymph node, particularly in floor-of- mouth tumors. Also, mass effect by a tumor might obstruct draining lymphatics and redirect tracers to nonsentinel lymph nodes. Additionally, some studies suggest that there is a high false-negative rate if only a single sentinel lymph node is evaluated and more extended tracer injections identify multiple lymph nodes, typically two to three, the removal of which may not be superior to an ultraselective nodal dissection in terms of morbidity and expense. Furthermore, the small surgical window used to extract the sentinel lymph nodes might risk injury to adjacent structures, which would not occur with a larger operative field. Despite these limitations, several multicenter randomized clinical trials have shown that sentinel lymph node biopsy is safe and highly reproducible, and has high sensitivities and negative predictive values.14,15,16,17,18,19
27.5.1 Intraoperative Evaluation of Sentinel Lymph Nodes
Ideally, once a sentinel lymph node is identified, intraoperative evaluation could allow for immediate triage to neck dissection if tumor is identified. The two primary mechanisms for intraoperative evaluation are frozen section and touch imprint cytology. For frozen section, the sentinel lymph node is bisected perpendicular to the shortest axis and the tissue is embedded in optimum cutting temperature compound and frozen in a cryostat. Two to three frozen sections are cut at 5- to 10- pm thickness and are H&E stained for evaluation. For touch imprint cytology, the sentinel lymph node is bisected or serially sectioned longitudinally, the cut surfaces are touched to a glass slide, and the slide is H&E stained.
Much of the literature regarding the performance of frozen section versus touch imprint cytology is derived from the breast cancer literature,9 but the findings are likely applicable to the head and neck. Frozen section benefits from greater general familiarity in the general surgical pathology community but it is both more expensive and time consuming (15-20 minutes per frozen section block processed). This significantly limits the number of sections that can be evaluated within the real-time constraints of a surgery. Additionally, frozen section artifact can both compromise permanent H&E evaluation and interfere with ancillary IHC studies.
Touch imprint cytology is significantly faster (a few minutes) to process, allows rapid evaluation of multiple cut surfaces, and could potentially evaluate a much larger volume of the sentinel lymph node if it is serially sectioned. Unfortunately, it suffers from less familiarity in the surgical pathology community, resulting in a slightly decreased overall sensitivity for touch imprint cytology as compared to frozen section.
Both frozen section and touch imprint cytology techniques have broad range of reported sensitivities, ranging from 50 to 90%, likely as a result of variable sectioning protocols, patient populations, tumor sites, types and sizes, and overall experience of the evaluating pathologists.9 It is therefore important to recognize the limitations of intraoperative evaluation and to communicate to the patient that the decision to perform a neck dissection may need to be deferred to permanent processing of the sentinel lymph node(s). Despite these limitations, the fact that intraoperative sentinel lymph node evaluation can triage roughly half of patients requiring nodal dissection to immediate dissection represents a substantial cost saving and justifies its use. Other modalities for intraoperative evaluation have been studied including one-step RT-PCR (reverse transcription polymerase chain reaction) and ultrafast immunohistochemistry, but these methods are both time consuming and expensive and are therefore not currently ready for routine clinical use.9
27.5.2 Permanent Evaluation of Sentinel Lymph Nodes
The use of sentinel lymph node biopsy is attractive from a histopathologic standpoint because it provides a smaller target to concentrate more extensive histologic evaluation. As previously discussed, intensive histologic evaluation of a full neck dissection is both technically and financially infeasible. However, if evaluation can be focused on one or a few sentinel lymph nodes, a more aggressive approach may be taken. How extensive this evaluation should be is dependent on the clinical significance of macrometastases, micrometastases, and ITCs. The prognostic significance of macrometastases and the therapeutic value of neck dissection in this setting are well documented and serve as the basis for both the current AJCC pathologic nodal staging system and the standard histopathologic grossing protocol.
A growing body of evidence suggests that micrometastases and even ITCs also have clinical prognostic significance.20,21,22 Two main strategies have been employed to identify microme- tastases and ITCs, step sectioning and IHC. Step sectioning is the process of taking histologic sections at discrete intervals through the paraffin-embedded gross section. Suggested intervals range from 50 to 500 pm. Obviously, smaller intervals increase sensitivity but this must be balanced with practical considerations. To avoid missing any micrometastases, the section interval would have to be 200 pm or less; to find all ITCs is technically infeasible. However, even a small lymph node with a shortest axis of 6 mm would require 30 histologic sections to evaluate at this interval. Given that an average of 3 sentinel lymph nodes are identified, this translates to 90 histologic sections, roughly the current histologic workload equivalent of a full neck dissection, per sentinel lymph node procedure. This would place a significant workload and financial burden on pathology departments given the current reimbursement environment, so its feasibility remains to be seen. IHC can detect smaller lesions than routine H&E but is more time consuming and expensive and thus becomes infeasible at even relatively large sectioning intervals.
27.5.3 Proposed Protocol for Evaluation of Sentinel Lymph Nodes of the Head and Neck
Work from the breast literature provides promise for a feasible histologic protocol for sentinel lymph nodes. In breast cancer, micrometastases and ITCs within lymph nodes tend to be multiple. In a classic study, sentinel lymph nodes initially called negative on routine H&E were completely serially sectioned at 150 pm to identify occult micrometastases and ITCs. In 95% of these sentinel lymph nodes, at least one micrometastasis or ITC was identified within the first three levels by immunohisto- chemistry. Importantly, much of this effect was due to additional sampling rather than the increased sensitivity of IHC because 80% of these micrometastases and ITCs would have been identifiable by H&E.23 If the same holds true for head and neck tumors, it will provide a feasible strategy for histologic evaluation of sentinel lymph nodes.
The feasibility of representative sampling in head and neck tumors is further supported by a recent evaluation of the distribution of micrometastases in the sentinel lymph nodes of patients with head and neck squamous cell carcinoma.24 This study shows metastatic foci are asymmetrically distributed and tend to aggregate near the afferent pole along the central axes of the lymph node. As a result, representative sampling in this region of the lymph node identifies most occult metastases. In this study, at least 90% of micrometastases and 80% of ITCs were identified by evaluation of the first four slices of a sentinel lymph node sectioned at 150 pm along the shortest axis.
Based on these two studies, the following protocol (Fig. 27.2) is recommended to evaluate sentinel lymph nodes in head and neck tumors. The sentinel lymph node should be grossly bisected or sectioned at 2-mm intervals (if larger than 5 mm in shortest diameter) perpendicular to the shortest axis with the plane of bisection as the central sectioning plane, and embedded to insure that the central slices are cut starting at the center at the plane of bisection and stepping outward but all other slices present a different surface for cutting. Note that this is different from the embedding protocol for nonsentinel lymph nodes. Two unstained slides at 2 to 4 pm should be cut at each of four levels at 150- pm intervals. One slide from each level should be routinely stained with H&E and evaluated. If all of these H&E slides are negative, one or more of the unstained slides should be subjected to IHC for evaluation. This strategy maximizes sensitivity within the confines of feasible representative sectioning and minimizes the use of more expensive and time-consuming IHC studies without loss of sensitivity. Studies by Broglie et al show 96% long-term neck control in patients with cN0 necks who have negative sentinel lymph nodes as assessed by a similar protocol.20,21
Once micrometastases and ITCs are identified histopathologically, the question is what to do about them. In the breast literature, there is convincing evidence that while small micrometastases and ITCs in sentinel lymph nodes predict a low but real risk of nonsentinel lymph node axillary involvement (13-26%),25 they do not appear to affect overall survival.26 The presence of micrometastases and ITCs in the sentinel lymph nodes of head and neck tumors also predicts a similar likelihood of involvement (13-20%)27 of nonsentinel lymph nodes. While we know that microscopic disease in cervical lymph nodes affects survival in head and neck tumors, it is unclear if elective neck dissection in these cases would necessarily improve survival. However, the behavior of head and neck micro- metastases and ITCs cannot be assumed to be analogous to those of the breast. Numerous systemic adjuvant chemotherapy options are used routinely for breast cancer, while surgical intervention or radiotherapy remains the mainstay of disease control in the head and neck. Therefore, until a reliable nomogram, analogous to those used in breast cancer, can be formulated to estimate the risk of additional nonsentinel lymph node metastases, elective neck dissection or adjuvant radiation to the neck is prudent for all cases with identified metastases to sentinel lymph nodes regardless of size.
27.6 Special Considerations for Thyroid Cervical Metastases
Unlike other head and neck sites, thyroid carcinoma (except anaplastic carcinoma) benefits from effective systemic therapy in the form of radioactive iodine (RAI), which can ablate microscopic metastatic disease. As such, the purpose of neck dissection is to de-bulk gross disease to maximize the effectiveness of RAI. Therefore, clinically positive necks are treated much as they would be for any other head and neck site but cN0 necks are not aggressively pursued. The 2015 American Thyroid Association (ATA) guidelines28 recommend ultrasound evaluation of the central and lateral neck compartments prior to surgery for well-differentiated thyroid carcinoma and biopsy of sonographically suspicious lymph nodes, but does not recommend sentinel lymph node biopsy in clinically and radiographically unremarkable necks. Postsurgical use of a scanning dose of RAI may help identify occult disease and a therapeutic dose of RAI may be used in patients with high-risk features, but prophylactic neck dissection for the cN0 neck is generally not warranted. The primary means of pathologic evaluation for clinically or radiologically suspicious lymph nodes is ultrasound-guided fine needle aspiration (Fig. 27.4a, b) with or without thyroglobulin assay.
Fig. 27.2 Proposed protocol for histologic processing of sentinel lymph nodes.
27.7 Cervical Metastases of Unknown Primary
In general, metastases to the cervical lymph nodes are largely from sites in the head and neck, particularly the mucosal surfaces, the thyroid, the skin, and the salivary glands. Less common considerations are metastases from the lung and breast; other sites are quite rare except in advanced disease. Additionally, lymphoma should be considered in patients with diffuse bulky lymphadenopathy.
True occult primaries, in which the initial clinical presentation is a positive cervical lymph node, are most commonly associated with squamous cell carcinomas arising in Waldeyer’s ring (~ 90%) and to a much lesser extent papillary carcinomas of the thyroid and melanomas of the scalp. While they can be occult, cervical metastases from tumors of the salivary gland, breast, and lung tend to be a late finding in advanced disease. Nevertheless, it is sometimes necessary to differentiate among a number of possible known and suspected primary sites to appropriately stage and treat a patient. The standard means of initial evaluation for a neck mass of unknown primary is fine needle aspiration, although excisional biopsy may be warranted for more extensive workups.
27.7.1 Cytomorphology
The cytomorphology of the metastatic tumor is often helpful in narrowing the likely primary site (Table 27.1). This cytomorphology can be broken down into five general categories: keratinizing squamous cell carcinoma, well/moderately differentiated adenocarcinoma, basaloid carcinoma, poorly differentiated carcinoma, and mononuclear malignancy. While spindle cell malignancies can metastasize to the cervical nodes, they are rare and almost always spindle cell variants of squamous cell carcinoma or melanoma. Metastases from true mesenchymal malignancies to the cervical neck are quite rare and almost never occult.
Table 27.1 Cytomorphology and immunophenotype of metastatic tumors to the cervical lymph nodes
Keratinizing squamous cell carcinoma morphology generally suggests a mucosal, skin, or pulmonary primary site (Fig. 27.3). It is usually impossible to determine the site of origin of keratinizing squamous cell carcinomas pathologically because of extensive morphologic and immunophenotypic overlap, so careful clinical and radiologic evaluation is crucial.
A well-to-moderately differentiated adenocarcinoma suggests primaries of the thyroid, salivary gland, lung, or breast (Fig. 27.4). Papillary thyroid carcinoma tends to be a well-differentiated adenocarcinoma with sheetlike cytological architecture and distinctive nuclear features including elongation, clearing, grooves, and pseudoinclusions (Fig. 27.4a), whereas Hurthle cell carcinoma is composed of sheets of oncocytic cells with dense granular cytoplasm, sharp cytoplasmic borders, and prominent nucleoli (Fig. 27.4b). Follicular carcinoma (non-Hurthle cell type) rarely metastasizes to the cervical lymph nodes, and poorly differentiated and undifferentiated thyroid carcinoma metastases in lymph nodes are rare. Salivary gland primaries often have distinctive cytologic appearances (Fig. 27.4c-e) but more poorly differentiated ones may overlap with breast and lung primaries, which are generally moderately to poorly differentiated adenocarcinomas with three-dimensional architecture (Fig. 27.4f, g). These can often be differentiated on the basis of immunophenotype.
Fig. 27.3 Keratinizing squamous cell carcinoma. (a) Diff-Quik stain, 400 x . (b) Papanicolaou stain, 400 x .
Basaloid malignancies include basaloid squamous cell carcinoma of the head and neck, basal cell carcinoma of the skin, the solid variant of adenoid cystic carcinoma, and basal cell adenocarcinoma of the salivary gland, and sometimes small cell carcinoma of the lung (Fig. 27.5). The presence of a magenta metachromatic matrix may suggest basal cell adenocarcinoma (Fig. 27.5d) or a solid variant of adenoid cystic but the first four entities are otherwise quite difficult to differentiate cytologically. In contrast, small cell carcinoma is more dyscohesive and shows nuclear molding, single-cell necrosis, and crush artifact (Fig. 27.5c). IHC studies can be quite helpful in differentiating these entities.
Poorly differentiated carcinomas include poorly differentiated squamous cell carcinoma of the head and neck or lung, and poorly differentiated adenocarcinomas of the salivary gland, lung, and breast (Fig. 27.6). Statistically, a poorly differentiated squamous cell carcinoma of the head and neck is the most likely source but this must be proven. Immuno- profiling may be helpful, but poorly differentiated malignancies often lose tissue-specific markers and, as a result, immunophenotypes are often equivocal.
Mononuclear dyscohesive tumors include melanoma, medullary carcinoma of the thyroid, lobular carcinoma of the breast, small cell carcinoma of the lung, and lymphoma (Fig. 27.7). Melanoma and medullary carcinoma are often described as “great mimickers” due to their variable appearance but, classically, melanoma cells tend to be pleomorphic, with moderate cytoplasm and prominent nucleoli, whereas medullary carcinoma tends to be relatively bland with a plasmacytoid appearance, neuroendocrine chromatin, and indistinct nucleoli. Lobular carcinoma of the breast tends to be bland and can have cytoplasmic vacuoles. Small cell carcinoma has a very high nuclear-to-cytoplasmic (N/C) ratio, shows nuclear molding, and significant crush artifact. Lymphoma is the most dyscohesive of the mononuclear malignancies, but otherwise is similar to small cell carcinoma in its N/C ratio, propensity to crush, and overall uniformity. As cytologic evaluation is relatively insensitive for detecting low-grade lymphomas (Fig. 27.7e, f), patients older than 45 years with enlarged cervical lymph nodes but without overt evidence of other malignancy on fine needle aspiration should be evaluated by flow cytometry to exclude lymphoma.
27.7.2 Immunophenotyping
IHC stains can often further narrow down the primary site in each cytomorphologic category (Table 27.1). For squamous cell carcinomas, positivity for HPV or p16 coupled with basaloid morphology is strongly suggestive of a primary arising from Wal- deyer’s ring29 (Fig. 27.5a), whereas EBER positivity and a poorly differentiated morphology would suggest a nasopharyngeal primary (Fig. 27.6a). For adenocarcinomas, TTF-1 positivity suggests either a thyroid or lung primary and PAX-8 positivity suggests a thyroid primary. Strong diffuse CK7 positivity is seen in thyroid, salivary, lung, and breast primaries but not in squamous cell carcinomas of the head and neck. GATA-3 is seen primarily in breast primaries and some salivary gland primaries (mostly salivary duct carcinomas and mammary analogue carcinomas). CD56 is useful for identifying tumors of the thyroid and small cell carcinoma of the lung. For basaloid malignancies, CK7 reactivity can distinguish salivary gland primaries and small cell carcinoma from squamous cell carcinoma and basal cell carcinoma. Strong diffuse CD117 positivity can be quite helpful for identifying the solid variant of adenoid cystic carcinoma and CD56 positivity can identify small cell carcinoma. For poorly differentiated carcinomas, p63 and p40 are very sensitive for squamous differentiation but also stain many salivary gland tumors; this can often be solved using CK7, which usually strongly stains salivary gland tumors but is either negative or focally positive in squamous cell carcinomas of the head and neck. In contrast, breast and lung primaries would be p63 negative and CK7 positive. Among mononuclear malignancies, S-100 is a sensitive marker for melanoma but HMB-45 and Melan-A are more specific. Calcitonin can identify the vast majority of medullary carcinomas and CD20 will identify most B-cell lymphomas. As such, judicious use of immunohistochemistry can be very helpful in narrowing the search for the primary site within each cytomorphologic grouping. However, since neither the cytomorphology nor the immunoprofile are entirely specific, they cannot substitute for a careful clinical and radiologic examination.
Fig. 27.4 Cervical lymph node metastases with well/moderately differentiated adenocarcinoma cytomorphology. (a) Papillary thyroid carcinoma (Papanicolaou stain, 400 x ). (b) Hurthle cell carcinoma (Papanicolaou stain, 400 x).(c) Acinic cell carcinoma (Diff-Quik stain, 200 x ).(d) Adenoid cystic carcinoma (Diff-Quik stain, 200 x ). (e) Mucoepidermoid carcinoma, well differentiated (Diff-Quik stain, 200 x ). (f) Pulmonary non-small cell adenocarcinoma well differentiated (Diff-Quik stain, 200 x ). (g) Breast ductal carcinoma, well differentiated (Diff-Quik stain, 200 x ).
27.8 Conclusion
In short, histopathologic evaluation of neck dissections serves the purpose of assessing for regional nodal involvement by metastatic lesions of the head and neck. This evaluation is limited by the necessity of representative histologic sampling even though all lymph nodes are entirely grossly submitted. This problem is particularly significant for cN0 necks but sentinel lymph node biopsy may provide a viable target for more intensive sampling. Based on the current evidence, we propose a protocol for evaluation of sentinel lymph nodes. Thyroid primaries represent an exception to the need to intensive evaluation of the neck because RAI is an effective adjunct to treat microscopic disease with minimal and acceptable morbidity. Finally, cytomorphologic evaluation and immunohistochemistry can help narrow down possible primary sites for occult meta- stases to the cervical lymph nodes but cannot substitute for a careful and thorough clinical and radiologic evaluation.
Fig. 27.5 Cervical lymph node metastases with basaloid cytomorphology (Diff-Quik stain, 200 x). (a) Basaloid squamous cell carcinoma. (b) Basal cell carcinoma (skin). (c) Solid variant of adenoid cystic carcinoma (salivary gland). (d) Basal cell adenocarcinoma (salivary gland).
Fig. 27.6 Cervical lymph node metastases with poorly differentiated carcinoma cytomorphology (Diff-Quik stain, 400 x ). (a) Poorly differentiated squamous cell carcinoma, (nasopharyngeal carcinoma). (b) Poorly differentiated squamous cell carcinoma, other site. (c) Poorly differentiated salivary gland adenocarcinoma (poorly differentiated mucoepidermoid carcinoma). (d) Poorly differentiated pulmonary non-small cell adenocarcinoma. (e) Poorly differentiated breast ductal adenocarcinoma.
Fig. 27.7 Cytomorphology of metastatic mononuclear malignancies (Diff-Quik stain, 200 x ). (a) Melanoma. (b) Thyroid medullary carcinoma. (c) Lung small cell carcinoma. (d) High-grade lymphoma (diffuse large B-cell lymphoma). (e) Low-grade lymphoma (chronic lymphocytic lymphoma). (f) Reactive lymph node.
References
[1] Amin MB, Edge S, Greene F, et al. AJCC Cancer Staging Manual. 8th ed. New York, NY: Springer; 2016
[2] Coskun HH, Medina JE, Robbins KT, et al. Current philosophy in the surgical management of neck metastases for head and neck squamous cell carcinoma. Head Neck. 2015; 37(6):915-926
[3] Sharma D, Koshy G, Grover S, Sharma B. Sentinel lymph node biopsy: a new approach in the management of head and neck cancers. Sultan Qaboos Univ Med J. 2017; 17(1):e3-e10
[4] Kapoor C, Vaidya S, Wadhwan V, Malik S. Lymph node metastasis: a bearing on prognosis in squamous cell carcinoma. Indian J Cancer. 2015; 52 (3):417-424
[5] ADASP Committee. The Association of Directors of Anatomic and Surgical Pathology. ADASP recommendations for processing and reporting of lymph node specimens submitted for evaluation of metastatic disease. Mod Pathol. 2001; 14(6):629-632
[6] Divi V, Chen MM, Nussenbaum B, et al. Lymph node count from neck dissection predicts mortality in head and neck cancer. J Clin Oncol. 2016; 34 (32):3892-3897
[7] Pou JD, Barton BM, Lawlor CM, Frederick CH, Moore BA, Hasney CP. Minimum lymph node yield in elective level I-III neck dissection. Laryngoscope. 2017; 127(9):2070-2073
[8] Kuo P, Mehra S, Sosa JA, et al. Proposing prognostic thresholds for lymph node yield in clinically lymph node-negative and lymph node-positive cancers of the oral cavity. Cancer. 2016; 122(23):3624-3631
[9] Maguire A, Brogi E. Sentinel lymph nodes for breast carcinoma: a paradigm shift. Arch Pathol Lab Med. 2016; 140(8):791-798
[10] Teymoortash A, Werner JA. Current advances in diagnosis and surgical treatment of lymph node metastasis in head and neck cancer. GMS Curr Top Oto- rhinolaryngol Head Neck Surg. 2012; 11:Doc04
[11] D'Cruz AK, Vaish R, Kapre N, et al. Head and Neck Disease Management Group. Elective versus therapeutic neck dissection in node-negative oral cancer. N Engl J Med. 2015; 373(6):521-529
[12] Tsushima N, Sakashita T, Homma A, et al. The role of prophylactic neck dissection and tumor thickness evaluation for patients with cN0 tongue squamous cell carcinoma. Eur Arch Otorhinolaryngol. 2016; 273(11):3987-3992
[13] Barroso Ribeiro R, Ribeiro Breda E, Fernandes Monteiro E. Prognostic significance of nodal metastasis in advanced tumors of the larynx and hypophar- ynx. Acta Otorrinolaringol Esp. 2012; 63(4):292-298
[14] Liu M, Wang SJ, Yang X, Peng H. Diagnostic efficacy of sentinel lymph node biopsy in early oral squamous cell carcinoma: a meta-analysis of 66 studies. PLoS One. 2017; 12(1):e0170322
[15] Green B, Blythe J, Brennan PA. Sentinel lymph node biopsy for head and neck mucosal cancers - an update on the current evidence. Oral Dis. 2016; 22 (6):498-502
[16] Monroe MM, Lai SY. Sentinel lymph node biopsy for oral cancer: supporting evidence and recent novel developments. Curr Oncol Rep. 2014; 16(5):385
[17] Seim NB, Wright CL, Agrawal A. Contemporary use of sentinel lymph node biopsy in the head and neck. World J Otorhinolaryngol Head Neck Surg. 2016; 2(2):117-125
[18] Schilling C, Stoeckli SJ, Haerle SK, et al. Sentinel European Node Trial (SENT): 3-year results of sentinel node biopsy in oral cancer. Eur J Cancer. 2015; 51 (18):2777-2784
[19] Farmer RW, McCall L, Civantos FJ, et al. Lymphatic drainage patterns in oral squamous cell carcinoma: findings of the ACOSOG Z0360 (Alliance) study. Otolaryngol Head Neck Surg. 2015; 152(4):673-677
[20] Broglie MA, Haerle SK, Huber GF, Haile SR, Stoeckli SJ. Occult metastases detected by sentinel node biopsy in patients with early oral and oropharyngeal squamous cell carcinomas: impact on survival. Head Neck. 2013; 35 (5):660-666
[21] Broglie MA, Haile SR, Stoeckli SJ. Long-term experience in sentinel node biopsy for early oral and oropharyngeal squamous cell carcinoma. Ann Surg Oncol. 2011; 18(10):2732-2738
[22] Cho JH, Lee YS, Sun DI, et al. Prognostic impact of lymph node micrometastasis in oral and oropharyngeal squamous cell carcinomas. Head Neck. 2016; 38 Suppl 1:E1777-E1782
[23] Fréneaux P, Nos C, Vincent-Salomon A, et al. Histological detection of minimal metastatic involvement in axillary sentinel nodes: a rational basis for a sensitive methodology usable in daily practice. Mod Pathol. 2002; 15(6):641-646
[24] Denoth S, Broglie MA, Haerle SK, et al. Histopathological mapping of metastatic tumor cells in sentinel lymph nodes of oral and oropharyngeal squamous cell carcinomas. Head Neck. 2015; 37(10):1477-1482
[25] Bargehr J, Edlinger M, Hubalek M, Marth C, Reitsamer R. Axillary lymph node status in early-stage breast cancer patients with sentinel node micrometasta- ses (0.2-2 mm). Breast Care (Basel). 2013; 8(3):187-191
[26] Giuliano AE, Hawes D, Ballman KV, et al. Association of occult metastases in sentinel lymph nodes and bone marrow with survival among women with early-stage invasive breast cancer. JAMA. 2011; 306(4):385-393
[27] Den Toom IJ, Bloemena E, van Weert S, Karagozoglu KH, Hoekstra OS, de Bree R. Additional non-sentinel lymph node metastases in early oral cancer patients with positive sentinel lymph nodes. Eur Arch Otorhinolaryngol. 2017; 274(2):961-968
[28] Haugen BR, Alexander EK, Bible KC, et al. 2015 American Thyroid Association Management Guidelines for Adult Patients with Thyroid Nodules and Differentiated Thyroid Cancer: The American Thyroid Association Guidelines Task Force on Thyroid Nodules and Differentiated Thyroid Cancer. Thyroid. 2016; 26(1):1-133
[29] El-Mofty SK, Zhang MQ, Davila RM. Histologic identification of human papillomavirus (HPV)-related squamous cell carcinoma in cervical lymph nodes: a reliable predictor of the site of an occult head and neck primary carcinoma. Head Neck Pathol. 2008; 2(3):163-168