Predicting sentinel and residual lymph node basin disease after sentinel lymph node biopsy for melanoma.

Abstract

We read with interest the article by Wagner et al.1 regarding the identification of predictors of metastases to sentinel and residual nonsentinel lymph nodes in patients with melanoma. Wagner et al. demonstrated that one such predictor was a tumor mitotic index (TMI) of ≥ 6 mitoses/high-power field (HPF). We were intrigued by the extraordinarily high values for TMI presented in their study. An exact comparison of data from the Sydney Melanoma Unit (SMU) concerning TMI with those data provided in the study by Wagner et al. is not possible. Nevertheless, we wish to present, from the SMU dataset collected within the same time period, comparable values regarding TMI for patients with melanoma localized to the skin with nonpalpable regional lymph node basins and no evidence of metastatic disease (Table 1). Only those patients who presented with at least five HPFs of assessable tumor were included in Table 1. There was a higher proportion of thin tumors reported in the SMU dataset, which would increase the proportion of tumors with low TMI because a direct correlation has been reported between TMI and tumor thickness.2 Nonetheless, we believe this finding could not explain away entirely the vast differences in the proportion of TMI evident in these two sets of data. It is interesting to note that in Australia, TMI originally was graded in accordance with the “Sydney” classification, provided there were at least 5 HPFs of tumor to assess: low TMI: 0 mitoses/HPF; intermediate TMI: < 1 mitosis/HPF; and high TMI: ≥ 1 mitosis/HPF.3 This classification is in sharp contrast to that given by Wagner et al., in which low TMI indicated ≤ 2 mitoses/HPF, intermediate TMI indicated 3–5 mitoses/HPF, and high TMI indicated ≥ 6 mitoses/HPF. In addition, the currently accepted practice worldwide is to grade TMI based on the number of mitoses per mm2, not the number of mitoses/HPF.4 To our knowledge, this is the first attempt to compare TMI values reported in various series. Helen Shaw Ph.D.*, John Thompson M.D.*, * Sydney Melanoma Unit, Royal Prince Alfred Hospital, Camperdown, Australia