Tumor-infiltrating lymphocytes and mitotic index in metastatic melanoma as predictors of patient survival

Bogunovic et al. (1) studied gene-expression profiles, mitotic index (MI), tumor-infiltrating leukocytes (TILs), and CD3+ T cell counts in metastatic melanoma and found that the parameters studied were predictive of postrecurrence survival; furthermore, they were better predictors of postrecurrence survival than Tumor, Node, Metastasis (TNM) staging. If validated in larger independent studies, the results may assist the management of melanoma patients. However, there are a number of salient points with regard to the methods and histologic interpretation that are important to consider when evaluating their results.

Studies have shown that the type of recurrence (regional lymph node, in transit, and distant recurrence compared with local recurrence) is a significant predictor of poorer survival (2, 3). However, Bogunovic et al. (1) stated that “many of the tissue samples were from lymph node metastases,” but they did not provide details about the types of recurrences in their cohort, which may have had a bearing on patient outcomes and therefore may have merited inclusion as an additional parameter in the multivariable analysis.

In this study, tumor slides were examined by two pathologists, and “MI was established by counting mitoses in 10 high power fields (HPF) per tumor section and then averaging the number by HPF (1.96 mm²)” (1). The method of calculating MI as described is not entirely clear. Was the area of one HPF 1.96 mm² or did this figure represent the area of 10 HPFs? The American Joint Committee on Cancer Staging Committee recommends that mitotic rate in primary melanoma be expressed as the number of mitoses per millimeter squared of invasive tumor (4), and it is usually measured using the hot-spot method, which has been shown to have excellent interobserver reproducibility (5). The rationale for adopting the methodology of measuring MI performed by Bogunovic et al. (1) was not provided, and the level of interobserver agreement using this method has not been reported, to our knowledge. Also, the authors did not describe if each of the two pathologists independently assessed MI for each case, and if they did so, the authors did not report the level of interpathologist agreement and the method of dealing with discrepant MI scores.

Finally, the authors stated that they assessed TIL counts and CD3+ T cell counts in the portion of tumor at least 2 HPFs away from the tumor’s interface with normal lymph node parenchyma. Although they made a laudable attempt to assess these parameters without contamination of lymphoid cells from adjacent lymph node parenchyma, our experience is that reliable assessment of TILs in lymph node metastases and avoidance of confusion with overlying or “carried-over” nodal lymphocytes can be extremely difficult and sometimes impossible. Furthermore, the edges of metastases may not be regular or well-demarcated from the adjacent parenchyma, and the 2-HPF distance from the presumed edge of the metastasis may not necessarily avoid contamination by nodal lymphoid cells. Another related question is whether or not there was any correlation between TILs count, CD3+ T cell counts, and the site of recurrence, particularly in nodal metastases. These issues require consideration and further validation in future studies on this subject.