Comparison between Ki-67 index and mitotic index for predicting outcome in canine mast cell tumours.

OBJECTIVES: To assess correlation between Ki67 index and mitotic index and determine which more accurately predicts survival in canine mast cell tumours. METHODS: Retrospective analysis of cases from three UK referral hospitals. Correlation between Ki67 index and mitotic index was assessed and survival analysis performed. RESULTS: A total of 162 dogs were included: 57 dogs died with 37 due to mast cell tumour. Correlation between Ki67 index and mitotic index was moderate, while the agreement was poor. A high Ki67 index was considered sensitive (86 · 5%) at predicting mast cell tumour-related death, but poorly specific (57 · 9%). Mitotic index(>5) was poorly sensitive (32 · 4%), but highly specific (96%). A mitotic index of ê2 had a 75 · 7% sensitivity and an 80 · 0% specificity. Ki67 index showed a statistically significant survival difference within the mitotic index <2 (P = 0 · 009) group. Ki67 index did not predict survival rate in tumours with mitotic index of ê2. CLINICAL RELEVANCE: Correlation between Ki67 and mitotic index is moderate. High mitotic index accurately predicted death, but many dogs with low mitotic index also died. Low Ki67 accurately predicted survival, but high Ki67 should not be considered a poor prognostic indicator. A three-tier mitotic index assessment may more accurately predict death due to mast cell tumour.

Mastocytosis in children: clinicopathological study based on 35 cases.

Immunohistochemical staining is useful in the diagnosis of bone marrow infiltration in systemic mastocytosis. However, it is not clear if antibody staining may be helpful in the diagnosis of cutaneous mastocytosis (CM). We studied the histological appearance of CM in 35 pediatric patients. Cases were assigned to three basic clinical groups: I--Urticaria pigmentosa (UP, n=29); II--Mastocytomas (n=4); and III--Diffuse Cutaneous Mastocytosis (DCM, n=2). The analysis of clinical information revealed an association between the presence of diarrhea and a higher number of cells/field. Nine doubtful cases, all of them macules, were selected based on the scarcity of mast cells (MC) and the absence or rarity of other inflammatory cells. We compared the number of cells identified in Giemsa and immunohistochemical stains in definite and doubtful cases. The intraclass correlation statistic tested the concordance between each staining method. All 9 dubious cases according to the Giemsa stain had their CM diagnosis confirmed by the immunohistochemistry analysis. The intraclass correlation between Giemsa and c-kit was good (0.7) when the number of MC was high. However, there was no correlation between the mast cells counts in the two different stains in the dubious cases. The immunohistochemistry with c-kit might make CM diagnosis easier, especially in the macular cases, when there is a lower number of MC.