How Do Pathologists in Academic Institutions Across the United States and Canada Evaluate Sentinel Lymph Nodes in Breast Cancer?

A Practice Survey

Jaya Ruth Asirvatham, MBBS; Julie M. Jorns, MD


Am J Clin Pathol. 2021;156(6):980-988. 

In This Article


Our study identified four areas of uniformity and four areas of dichotomy in the evaluation of SLNs among academic breast pathologists in North America. In our study, IOE was performed for all cases in 10 academic institutions. In almost an equal number of institutions, the frequency of FS was limited to select cases (in general, those who did not meet ACOSOG Z0011 criteria), in an attempt to omit ALND.

The key question of the ACOSOG Z0011 clinical trial was whether ALND improved survival or local control compared with observation in patients with clinically node-negative (cN0), early stage (T1-T2, tumor size ≤5 cm) breast cancer with one to two positive SLNs undergoing breast-conserving therapy. The trial included 856 women with median follow-up of 9.3 years and demonstrated that in this group of patients, SLN was not inferior to ALND.[1] Trial exclusion criteria included mastectomy, neoadjuvant chemotherapy (NACT), partial or no radiation, palpable adenopathy or gross extranodal disease, more than three positive SLNs, matted nodes, and metastasis identified by immunohistochemistry (IHC). Recent studies have shown that the impact of the ACOSOG Z0011 clinical trial has been a decrease in the frequency of IOE and ALND.[7–9]

In most academic institutions, FS is the preferred method of IOE. Per current National Comprehensive Cancer Network (NCCN) guidelines, if the patient is clinically node negative at the time of diagnosis, and the SLN demonstrates only micrometastasis or isolated tumor cells, no further axillary surgery is required, even in patients who do not meet the ACOSOG Z0011 criteria. Hence, cytologic techniques, which cannot determine the size of metastasis, may not be an appropriate form of evaluation for this group of patients.

Extranodal extension (ENE) is an emerging predictor of tumor involvement of nonsentinel nodes.[10–12] In a meta-analysis, ENE was associated with a higher risk of mortality (relative risk [RR], 2.51; 95% CI, 1.66–3.79, P < .0001) and recurrence (RR, 2.07; 95% CI, 1.38–3.10, P < .0001).[10] In the ACOSOG Z0011 trial, patients with gross ENE were excluded, and the presence of microscopic ENE was not evaluated. Recent studies have shown that the presence and extent of ENE significantly correlated with nodal tumor burden, suggesting that greater than 2 mm of ENE may be an indication for ALND or radiotherapy when applying ACOSOG Z0011 criteria to patients with one to two positive lymph nodes. Choi et al[11] evaluated the impact of the size of ENE (less than or greater than 2 mm) in patients meeting ACOSOG Z0011 criteria and found in a study of 208 patients that ENE on SLN is associated with N2 disease. Despite the increased nodal burden, ENE 2 mm or smaller was associated with recurrence and survival rates similar to those of patients with no ENE. Hence, evaluation of ENE at FS has implications for subsequent treatment. Most study respondents indicated that fat around the lymph node was not completely trimmed. In one institution, the trimmed tissue was separately submitted for evaluation at permanent sections. However, in most surveyed institutions, the presence of ENE was documented after evaluation of permanent sections.

The most common practice was to submit the entire SLN for IOE after sectioning at 2-mm intervals, similar to processing for permanent sections. This is in keeping with grossing protocols outlined by both the CAP and Royal College of Pathologists (RCPath) designed to identify all macrometastases (>2 mm).[13,14] Careful gross and microscopic evaluation is necessary to determine if patients meet criteria for conservative management. In the NSABP B-32 clinical trial, participating sites were instructed to slice SLN at 2-mm intervals, embed all tissue in paraffin blocks, and examine one H&E section from each block, with goal of identifying (or not missing) macrometastasis.[15] In some institutions, due to potential loss of tissue during FS and/or challenges in obtaining an appropriate section of fatty tissue, only a representative sample is submitted (bisected with both halves submitted for FS and later sectioned at 2-mm intervals for permanents, bisected with one half submitted at 2-mm intervals for FS and the other half reserved for PS, grossly suspicious foci, or one or two larger sections of lymph node with no overt metastatic disease). In all institutions, non-SLNs are sectioned at 2-mm intervals and entirely submitted for PS.

One area of dichotomy in practice was parallel vs perpendicular sectioning of the lymph node. Per current CAP guidelines, "the node is to be sectioned at 2mm intervals along the long axis."[13] This is a new recommendation in the most recent guidelines and not present in prior iterations, in which a recommendation for sectioning at 2-mm intervals without further instruction is provided.[16] In contrast, RCPath recommendations outline sectioning at 2-mm intervals perpendicular to the long axis.[14] Relatively recent CAP updates and conflicting recommendations, coupled with individual and institutional preference, likely resulted in the split in survey responses, with 12 institutions opting for perpendicular, 8 parallel, and 3 variable sectioning in relationship to the long axis of the SLN. Some experts recommend sectioning the node parallel to the longest axis, as it produces fewer slices to examine, and old anatomic data suggest that afferent lymphatics are more likely to enter the lymph node in this plane.[15] They also note that the plane may be difficult to determine, and what is important is ensuring that no slice is thicker than 2 mm, at least one microscopic section is examined every 2 mm, and there is the likelihood of detecting all metastases larger than 2 mm. However, others have suggested that perpendicular sectioning more accurately identifies small SLN metastasis.[17]

Another area of dichotomy in practice was the routine use of multiple levels and cytokeratin stains. In the NSABP B-32 trial, while treatment decisions were based on findings in the initial H&E SLN section, SLN blocks from patients who were negative on the initial H&E were submitted for H&E and cytokeratin stains at a 0.5-mm and 1-mm depth into the block, which resulted in identifying occult metastasis (ITCs and micrometastatic disease not immediately apparent on initial H&E sections) in 15.9% of patients (11% ITC, 4% micrometastasis).[18] Five- and 10-year follow-up of patients with and without occult metastases were statistically significant, but the percent increase in overall and disease-free survival (DFS) and distant disease-free interval was minimal (DFS, 86.4 vs 89.2%; overall survival, 94.6% vs 95.8%). In addition, the ACOSOG Z0010 trial demonstrated that in women undergoing breast-conserving therapy and SLN biopsy, IHC evidence of SLN involvement was not associated with decreased overall survival. In this study, which included 5,119 patients, there was no axillary-specific treatment for H&E-negative SLNs, and clinicians were blinded to IHC results.[19] As additional evaluation does not appear to translate into clinical benefit, the American Society of Clinical Oncology and NCCN do not recommend routine immunostains.[20,21] Other international organizations, such as the National Health Service Breast Screening Program and RCPath, also do not advocate routine use of enhanced methods.[14]

We also identified variability in the interpretation of tumor in pericapsular lymphatic spaces. While some pathologists identified these as lymphovascular emboli, others interpreted these as ITCs. While ITCs are not included in the total number of positive nodes for N classification, in SLN after NACT, there are currently no data to enable axillary dissection to be safely omitted, irrespective of the size of the SLN metastasis.[22,23] An evaluation of 181 patients with post-NACT who had a residual positive lymph node irrespective of the method of detection (17% [1/6] with isolated tumor cells, 64% with micrometastasis, and 62% with macrometastasis) also had additional non-SLN involvement at ALND.[24] The currently accruing Alliance A011202 trial, which randomizes patients with persistent positive axillary disease following NACT to axillary radiation with or without ALND, aims to determine if ALND can be safely omitted in this population.[25]

In conclusion, recent clinical trials have changed axillary management in breast carcinoma toward a more conservative approach. These changes have resulted in decreased utilization of IOE at most institutions. The variation in pathology practice may be driven by variations in surgical practice or expectations of the surgical team. Among pathologists in academic institutions, there are areas of uniformity in practice (preference for FS during IOE, sectioning lymph nodes at 2 mm, submitting the entire lymph node for FS, and documenting ENE after evaluation of permanent section) and areas of dichotomy in practice (complete vs incomplete removal of perinodal fat at FS, perpendicular vs parallel sectioning, routine vs case-by-case multilevel H&E/immunostain examination, and interpretation of tumor in pericapsular lymphatics as LVI or ITC). Data-driven efforts to increase standardization of SLN pathologic evaluation should be undertaken. Additional data on reporting of ENE (for both IOE and permanent section evaluation), plane of gross sectioning, and cost vs benefit for multilevel sectioning best target remaining questions that will help guide development of best practices.