posters International Association for Breast Cancer Research 2014

Clearing the fog: A novel Ki67-mitosis classifier enables deeper segmentation of the “grey zone” Nottingham Grade II patients (#85)

Vaishali Pannu 1 , Padmashree C.G. Rida 1 , Sergey Klimov 1 , Guilherme Cantuaria 2 , Michelle Reid 3 , Ritu Aneja 1
  1. Department of Biology, Georgia State University, Atlanta, GA, United States
  2. Univeristy of Gynecologic Oncology, Northside Hospital Cancer Institute, Atlanta, GA, United States
  3. Department of Pathology, Emory University Hospital, Atlanta, GA, United States

The Nottingham Grading System (NGS) provides clinically important prognostic information about breast tumor samples by combining glandular differentiation, nuclear pleomorphism and mitotic activity scores. Using NGS, 30-60% of breast tumors end up being classified as Grade II which is a “mixed bag” and is notoriously uninformative for clinical-decision making. Thus, there is a need to enhance the prognostic accuracy of NGS by identifying discriminators that more sharply distinguishes the risk posed by breast tumors. We retrospectively analyzed pathology reports of 1600 breast carcinoma cases at Northside Hospital, Atlanta, GA to determine if a modified metric called Ki67-adjusted mitotic score (KAMS), when incorporated into the tumor grading system could effectively re-stratify Grade II patients and strengthen the prognostic power of tumor grade. Our strategy used KAMS as a criterion to influence shifts in the Mitotic scores and the Nottingham Grade. Mean KAMS value of Grade II patients was used to dichotomize patients into high-KAMS and low-KAMS groups. Essentially, Grade II patients that possessed an above-average KAMS and a high Mitotic score (of 3 or 2) are likely to show good prognosis; hence their mitotic score was reduced by 1. Similarly, Grade II patients with a below-average KAMS and a low Mitotic score (of 1 or 2) are likely to show poorer prognosis; their mitotic scores were therefore increased by 1. Adjusted Nottingham Grades were calculated by adding scores for the 3 histological components- Mitotic score (adjusted), tubule formation, and nuclear pleomorphism. Using this grade adjustment model, 39% of the original Nottingham Grade II patients were re-adjusted into either Grade I or III. There was no significant difference in OS between original and adjusted Grade I (p=0.3047) and Grade III (p=0.3577) patients. Thus, KAMS can provide deeper segmentation of Grade II patient cohort to improve the prognostic accuracy of NGS.