[48] Mitotic Count Is Sufficient To Predict Prognosis for Extremity Spindle Cell Soft Tissue Sarcomas

Hatim A Khoja, Brendan C Dickson, Jay S Wunder, Peter C Ferguson, Anthony Griffin, David Howarth, Rita A Kandel. Mount Sinai Hospital, Toronto, ON, Canada

Background: The grade of spindle cell sarcomas is considered one of the most important predictors of prognosis. Tumor type (differentiation), mitotic count and tumor necrosis are the parameters used in the French Federation of Cancer Centers Sarcoma Group (FNCLCC) grading system. The objective of this study was to determine which of these histological features have prognostic significance with the goal to simplify the grading system.
Design: 107 incisional biopsies of non-metastatic primary spindle cell sarcomas of the extremities were retrieved from our archives. All cases had a minimum of two year follow-up. Patient data and outcome, and tumor characteristics were recorded. Tumors were reviewed and evaluated according to the FNCLCC grading system. The prognostic significance of mitotic count, tumor necrosis, size and depth were evaluated. Kaplan-Meier survival curves were generated to correlate the grade and each of these parameters with metastases and disease-free survivals for univariate analysis. Independent effect of predictors was assessed by multivariate Cox proportional hazards regression.
Results: There were 52 male and 55 female patients with a mean age of 56 years and an average tumor size of 8 cm. 57% of tumors were deep and 43% were superficial. There was a wide spectrum of tumor types. The percentages of grades 1, 2 and 3 tumors were 10%, 44% and 48%, respectively. Both tumor size (<5 or ≥5cm) and depth (superficial vs deep) were predictive of outcome (eg. metastasis-free survival: p=0.0001 and p=0.006 respectively) confirming that the tumor population was behaving as expected. FNCLCC grading was predictive of outcome (disease-free and metastasis-free) as were the individual parameters of the FNCLCC for necrosis and mitoses. When these parameters were simplified to a mitotic count of either <10 mitoses or ≥10 mitoses /1.7mm2 and necrosis as either present or absent, they were predictive of both metastasis and disease-free survivals by univariate analysis (Mitotic count: p=0.0001; p=0.001 and Necrosis: p=0.013, and p=0.007, respectively). Multivariate analysis suggested that mitotic count was the most significant histological factor in predicting prognosis (p=0.001). Necrosis was not predictive in this analysis.
Conclusions: This data suggests that assessing the mitotic count (<10 or ≥ 10 mitoses /1.7mm2) only may be sufficient to predict prognosis in incisional biopsies of spindle cell extremity sarcomas. A larger prospective study is required to confirm this observation and assess observer reproducibility.
Category: Bone & Soft Tissue

Tuesday, March 5, 2013 1:00 PM

Poster Session IV # 18, Tuesday Afternoon


Close Window