Value of Multilevel Sectioning for Improved Detection of Micrometastases in Sentinel Lymph Nodes in Invasive Squamous Cell Carcinoma of the Vulva

Abstract

Clinical usefulness of sentinel lymph node (SLN) biopsy has been demonstrated in the management of early vulvar cancer. However, what constitutes a negative SLN has not been well defined. Furthermore, to what extent the SLNs should be sectioned for the greatest likelihood of detection of micrometastases and whether multilevel sectioning will further increase this detection rate in this setting have not been well studied. We analyzed 280 groin lymph nodes (SLNs=45, non-sentinel [NSLNs]=235) in 14 patients with invasive squamous cell carcinoma (ISCC) of the vulva treated with vulvectomy and inguinal SLN and NSLN dissection at the H. Lee Moffitt Cancer Center (HLMCC) between 1996 and 2001. Each SNL was evaluated for micrometastases by H&E and pancytokeratin AE1/3 (CK AE1/3) immunohistochemical staining. All negative SNLs (N=40) were sectioned times 3 (x3) at 50-micron intervals and independently reviewed by two pathologists in order to assess the utility of this inexpensive and logical approach to identifying additional micrometastases. Also, the Wilcoxon Rank Sum Test was used to determine if there was an association between tumor size, depth of invasion and SNL status. The patient age ranged from 35 to 81 years (mean 59 yrs); size of invasive tumor from 1.0 to 7.0 cm (mean 3.4 cm); depth of invasion from 3 to 25 mm (mean 10.8 mm). Of 45 SLNs examined from 14 patients, 11% (5/45) SNLs were positive for micrometastases on initial H&E and/or CK AE1/3 stains. Eighty-nine per cent (40/45) SNLs were negative in the remaining 9 patients. None of the latter 40 SNLs showed micrometastases on additional multilevel sectioning. Instead 3 of 135 NSLNs examined in these 9 patients revealed micrometastases on H&E (skip-micrometastases). Mean tumor size (cm) and depth of invasion (cm) were 4.06 (s.d. 1.89) and 1.20 (s.d. 0.35) for SLN (+) and 3.02 (s.d. 2.12) and 1.01 (s.d. 0.86) for SLN (-) tumor subsets (p values 0.385 and 0.348, respectively). Conclusion: Following routine H&E and CK AE1/3 stains, multilevel sectioning does not appear to detect additional micrometastases in sentinel lymph nodes in squamous cell carcinoma of the vulva. Even though mean tumor size and depth of invasion were greater in SNL (+) as compared to SLN (-) tumor subsets in our series, this difference did not reach statistical significance.