The Performance of Ki-67 Labeling Index in Different Specimen Categories of Invasive Ductal Carcinoma of the Breast Using 2 Scoring Methods

In breast carcinoma proliferative rate is not only prognostic, but also predictive factor. Ki-67 labeling index (Ki-67 LI) is a superior and more reproducible way to assess proliferation than counting mitoses. There are different Ki-67 LI cut-off values proposed for stratification. The best method of scoring Ki-67 LI is still controversial. Our aim was to test the consensus of Ki-67 LI read in the 2 most common specimens, diagnostic core biopsies (CNB) and surgical excision (SE), first using the hot spot (HS) for counting, and then the average (Av) Ki-67 LI. Cases diagnosed as having invasive duct breast carcinoma on CNB followed by SE over 6 years were recruited (n = 96). Ki-67 LI was counted in both specimen types using HS and Av scoring methods. For both methods, agreement between the 2 specimen categories was tested using different cut-off values. Intraobserver reproducibility was also measured for both scoring methods. Ki-67 LI was significantly lower in CNBs compared with SEs (mean difference: - 13.3 and - 6.3, P < 0.001, using HS and Av methods, respectively). The agreement between Ki-67 LI in CNBs and SEs using 10, 14, and 20 as cut-off value was superior when we used Av method (for kappa = 0.793, 0.814, and 0.745; vs. for HS: kappa = 0.542, 0.525, and 0.672, respectively). Intraobserver reproducibility ranged from very good to perfect for both methods. Our results suggested that specimen-specific cutoff value should be applied for both scoring methods.