Evaluation of Immunohistochemical Fine Sectioning for Sentinel Lymph Node Biopsy in Oral Squamous Cell Carcinoma

Methods

Our study was approved by the University of Miami Human Subjects Research Office Institutional Review Board, and patients included in this study provided appropriate consent. We retrospectively reexamined sentinel lymph node tissue samples of oral squamous cell carcinoma obtained from patients treated between 1996 and 2004 that were previously sectioned at 2- to 3-mm intervals in our pathology department. Patients who had previously received finer sectioning on an internal protocol were naturally excluded, as no tissue was available. Patients who had undergone thicker 2-mm sections were obtained primarily from a group placed on a multi-institutional trial from our institution. These tissue blocks originally generated and processed with H&E and IHC staining were now reexamined following exhaustive serial sectioning and staining at 150-µm intervals. The experimental findings were then compared with the patient status after selective neck dissection, as documented in the medical record and prospectively collected data file, to determine any prognostic significance of the findings obtained through exhaustive sampling of the sentinel node.

Results

Ten patients with oral squamous cell carcinoma who previously underwent sentinel lymph node biopsy (SLNB) with sentinel lymph node (SLN) examination at 2-mm intervals were exhaustively fine sectioned at 150-µm intervals. These SLNs were reexamined using H&E and IHC staining at this finer section interval. A total of 35 SLNs were reexamined using this protocol. Table 1 provides the primary tumor sites, tumor stage, number of sentinel lymph nodes obtained, and sentinel lymph node positivity.

OPEN IN VIEWER

Table 1.

Characteristics of Primary Tumor and Sentinel Lymph Node Examination

Patient	Primary Tumor Site	Tumor T Stage	Number of SLNs	SLN H&E Status	SLN Cytokeratin Status
1	Right tongue	T2	4	0	0
2	Floor of mouth	T1	9	0	0
3	Anterior left tongue	T1	2	0	0
4	Left buccal mucosa	T1	2	0	0
5	Oral tongue	T1	4	0	0
6	Right tongue	T1	1	0	0
7	Left lower lip	T1	4	0	0
8	Right lateral tongue	T1	5	0	0
9	Left lip	T1	0	0	0
10	Lower lip	T1	4	0	0

Abbreviations: H&E, hematoxylin and eosin; SLN, sentinel lymph node.

No additional micrometastatic disease was discovered during the more extensive fine sectioning performed at 150-µm section intervals in these 10 patients’ 35 SLNs. The validity of SLNB in comparison with the gold standard cervical lymphadenectomy as documented in patient charting further demonstrates congruity of results with no missed metastatic disease in the SLNs.

Discussion

To date, there is no standard sentinel lymph node processing for examination of the SLN of clinically negative necks of patients with early T-stage oral mucosal squamous cell carcinoma. The Second International Conference on Sentinel Node Biopsy in Mucosal Head and Neck Cancer recommends step serial sectioning of the entire SLN at 150-µm intervals. ¹ However, other authors use sectioning of the sentinel lymph node at 2- to 3-mm intervals. ^4,5 The significant difference in labor involved led us to transition away from finer sectioning at our institution. Our retrospective reexamination of SLNs obtained to evaluate the clinically negative neck for early T-stage oral squamous cell carcinoma suggests that finer sectioning of the SLN at 150-µm intervals would improve the diagnosis of micrometastatic disease infrequently in comparison to SLN sectioning at 2-mm intervals, and the significant additional cost needs to be carefully considered. These results suggest that exhaustive fine sectioning of the SLN at 150-µm sections may not actually contribute to a change in clinical decision making; that is, with no change in SLN status based on finer sectioning, the patient would still not be offered a therapeutic neck dissection and his or her regional disease control would remain the same as if 150-µm SLN sections were employed.

A limitation of this study is the number of patient specimens that were reprocessed. An extensive amount of time and resources is necessary to exhaustively fine section the SLN, stain the sections with H&E and anticytokeratin as per protocol examination of these SLNs, and then subsequently examine all new slides obtained by this protocol. For example, sectioning at 2-mm intervals would generate 3 longitudinal slices from a typical 6-mm-wide SLN, as demonstrated in Figure 2 . These 3 slices are each placed in a separate block, and 4 representative sections are typically taken from each block for H&E and IHC staining. Thus, 3 blocks and 12 slides are generated for each lymph node. If there are 3 sentinel nodes, which is common in the head and neck region, 9 blocks and 36 slides are generated. In contrast, processing of the same 3 SLNs exhaustively at 150-µm intervals ultimately yields 120 slides for staining and evaluation. ⁶ A protocol of 150-µm intervals can thus become quite time-consuming and labor intensive with an associated increased cost. Significantly increased detection of micrometastasis would indeed justify this increased effort and expenditure.

OPEN IN VIEWER

Figure 2.

Method of exhaustive fine step sectioning of a sentinel lymph node.

Another limitation of our study is that none of the previously reported false-negative (FN) SLNs were reexamined here. The SLNs previously diagnosed as FN in comparison to the gold standard cervical lymphadenectomy specimen are of particular interest. Should exhaustive fine sectioning and reexamination of those FN SLNs reveal the initially missed micrometastatic disease, the FN rate of SLNB would decrease. This would suggest that the 150-µm sectioning interval would prove useful in the diagnosis of micrometastatic disease. Of note, no statistically significant data would be gleaned, however, because the FN rate is generally very low. At our institution, there were no available FNs processed at 2-mm intervals, as none had occurred since this protocol was instituted. In fact, a significant number of FNs would need to be examined to achieve statistically significant results. A study designed to demonstrate that fine sectioning of an SLN reduces the FN rate would require approximately 100 cases with gold standard neck dissection proven metastasis. Specifically, if there were a 30% rate of occult nodal disease among patients in our study with a 10% SLNB FN rate shown by standard lymphadenectomy, 302 patients are required, of whom 91 are SLNB and standard neck dissection for validation positive. This study design is based on 90% power and 5% significance with a hypothesis that fine sectioning at 150-µm intervals improves the diagnostic capability of micrometastasis in an SLNB (McNemer’s test of proportions for paired data).

It should be noted that in our original pathology reports as well as in the current work, both isolated tumor cells (ITCs) and micrometastases were interpreted as positives; floating intraluminal tumor cells that were not within tissue were not. The incidence of micrometastases and ITCs in oral SLNB has only recently been defined, and pathology reports did not routinely distinguish these 2 pathologic entities in the past. Some investigators based their reports and recommendations on micrometastasis defined using an upper size limit of 3 mm but no specification of the lower size limit. This lack of a formal definition makes evaluation of the incidence and clinical implications difficult to determine from these prior articles. ⁷ Currently, micrometastasis is defined by the American Joint Committee on Cancer as a metastatic deposit greater than 0.2 mm in diameter but not greater than 2 mm in greatest diameter. Isolated tumor cells are defined as metastatic deposits that are 0.2 mm in diameter or smaller. The incidence of ITCs is quite low, which contributes to the difficulty in determining their true prognostic significance. A recent investigation was published reexamining 86 SLNs and comparing the histopathologic examination to previously performed neck dissection specimens to assess how adoption of the definition of 2 mm as the upper size limit for micrometastasis would affect tumor staging and compare to the formal neck dissection histopathologic diagnosis of micrometastasis and ITCs. The authors identified 5 patients with ITCs within their SLN(s), 1 of whom had metastatic disease in the neck dissection specimen corresponding to an incidence of 20% regional metastatic disease in the presence of a SLN with an ITC. However, given this low number, it is not possible to make a definitive recommendation for performing neck dissection for all patients with the finding of an ITC in their SLNB specimen. ⁸ Additional investigation is warranted to truly define the prognosis of the ITC. Most surgeons at this point would err on the side of caution, interpret ITC as positive, and proceed with neck dissection.

Our data suggest, but do not prove, that a protocol for evaluation of the sentinel lymph node using 2-mm section intervals maximizes the efficiency, accuracy, and expenditure in the detection of cervical micrometastatic disease. Use of the finer 150-µm section interval did not improve the diagnosis of micrometastatic disease of squamous cell carcinoma in the clinically N0 neck in 35 lymph nodes from 10 patients. In the future, we would like to identify and recut a group of 6 FNs generated at other institutions from the American College of Surgeons Oncology Group multi-institutional trial, in which we participated, to determine if these FNs would have been avoided by finer sectioning, thereby improving the FN rate.

Author Contributions

Francisco J. Civantos, research mentor/principal investigator, editing; Gina D. Jefferson, data gathering, writing; David Sollaccio, data gathering; Carmen R. Gomez-Fernandez, pathologic reprocessing and reevaulation of the sentinel lymph nodes.

Disclosures

Competing Interests: None.

Sponsorships: None.

Funding source: None.