Tumour mitotic rate—a poor predictor of sentinel lymph node status in cutaneous melanoma: should we select using staging criteria?

Abstract

Selection for sentinel lymph node biopsy (SLNB) in cutaneous melanoma is currently based on histopathological variables specified in prognostic staging criteria. In 2009, these criteria were updated replacing Clark level with the tumour mitotic rate (TMR) when defining thin T1 tumours. This study aimed to determine the histopathological variables independently predictive of sentinel lymph node involvement and discusses which variables should influence our selection for SLNB. One hundred and fifty-three patients with primary invasive localized cutaneous melanoma who underwent SLNB between 2003 and 2007 were reviewed from a prospectively collected database. We performed a multivariate binary logistic regression analysis to ascertain which variables independently predict sentinel status. SLNB positivity rate was 17.3%. No patient with a thin T1 tumour was SLNB positive. Breslow thickness was the only independent predictor of sentinel lymph node (SLN) status to reach significance (x 2, 10.555; p, 0.001). TMR and ulceration were not independent predictors of sentinel node status (x 2, 0.988; p, 0.320 and x 2, 2.082; p, 0.149, respectively). Breslow thickness is the only variable that is consistently reported as an independent predictor of SLN status. The use of TMR and ulceration in staging criteria is intended to establish prognosis, not select for SLNB. Given the important implications of a positive SLNB on surgical management, accurate selection is imperative. Selection based solely on Breslow thickness, rather than staging criteria, may be valid. However, there is an urgent need for a well-designed large multi-centre analysis to validate this. Until then, we continue to select for SLNB within a multidisciplinary team accounting for multiple factors. Level of Evidence: Level II, Prognostic/Risk Study.