Quantitative assessment Ki-67 score for prediction of response to neoadjuvant chemotherapy in breast cancer

Measurement of Ki-67, a marker of cell proliferation, has been associated with response to therapy, but methods of measurement are controversial. Here we use a quantitative objective measurement for Ki-67 to determine the best method for assessment of Ki-67 for prediction of response to neoadjuvant chemotherapy. Analysis was conducted on a cohort of 105 consecutive invasive breast cancer patients that received neoadjuvant therapy between 2002 and 2010, and on whom pre-surgical biopsies were obtainable. Ki-67 expression was measured using quantitative immunofluorescence automated quantitative analysis (AQUA) technology. Images for each specimen were collected for 5 to 115 fields of view (FOVs) and summary scores were obtained, corresponding to the average and maximum of all the FOVs. AQUA scoring (using both intensity and area) was comparable to automated calculation of percentage of positive nuclei for prediction of response to chemotherapy (OR: 2.832 vs 2.712). Both the average and maximum AQUA score showed Ki-67 expression was directly correlated to pathological complete response (pCR; average P=0.0002; maximum P=0.0011). Although examining the maximum FOV was more predictive of response to therapy (OR: 3.546 vs 2.832), averaging all fields provided more sensitivity and specificity (AUC 0.769 vs 0.732). Ki-67 average (P=0.0025) and maximum (P=0.0239) AQUA score were also significant predictors of pCR in a multivariable analysis, including tumor size, nuclear grade, nodal status, ER status, and HER2 status. Measurement of Ki-67 expression by objective quantitative methods shows increased Ki-67 levels are an independent predictor of response to neoadjuvant chemotherapy.