Prognostic significance for metastasis of Ki-67 and mitotic rate in cutaneous melanoma

Abstract

7516 Background: As malignancy is a disorder of cellular growth control, the assessment of cell proliferation rates in tumors has intuitive appeal as a prognostic marker. One such biomarker is Ki-67, a cell cycle dependent protein. Recent reports related to its role as a prognostic factor have been contradictory. Methods: This study includes 610 patients with primary melanomas who had no evidence of metastasis at diagnosis, who were followed for at least 10 years, and for whom paraffin blocks were available. Using immunohistochemistry, Ki-67 expression was quantified as the percentage of melanoma cells binding the monoclonal antibody mib-1 in a dermal “hot spot” of the primary lesion. Mitotic rate was quantified by routine histologic methods. Unadjusted and adjusted odds ratios (OR) were obtained from the logistic regression analyses. A tree-based analysis (including 9 prognostic factors in addition to mitotic rate and Ki-67 expression) was used to identify risk groups. Results: The 10-year metastasis rate was 20.3% where 66% of the patients had thin (< 1.00 mm) invasive melanomas. Both mitotic rate (adjusted OR=1.07, 95% Confidence Interval (CI) =1.02 to 1.13) and Ki-67 expression (OR=1.33, CI=1.08 to 1.63) were significantly associated with ten-year metastasis rates, either independently or controlling for nine other validated prognostic factors. The final classification tree included mitotic rate, Ki-67 expression, thickness and anatomic site. Five risk groups were defined (see table below). Conclusions: Ki-67 is an independent prognostic factor for 10-year metastasis. It can be used to identify a subset of patients with lesions that have low mitotic rates who are at elevated risk for metastasis. No significant financial relationships to disclose.