Published in:
01-06-2011 | Preclinical study
MIB1/Ki-67 labelling index can classify grade 2 breast cancer into two clinically distinct subgroups
Authors:
Mohammed A. Aleskandarany, Emad A. Rakha, R. Douglas Macmillan, Desmond G. Powe, Ian O. Ellis, Andrew R. Green
Published in:
Breast Cancer Research and Treatment
|
Issue 3/2011
Login to get access
Abstract
Histological grade is recognized as one of the strongest prognostic factors in operable breast cancer (BC). Although grade 1 and grade 3 tumours are biologically and clinically distinct, grade 2 tumours bear considerable difficulty in outcome prediction and planning therapies. Several attempts such as genomic grade index have been performed to subclassify grade 2 into two subgroups with clinical relevance. Here, we present evidence that the routinely evaluable immunohistochemical MIB1/Ki67 labelling index (MIB-LI) can classify grade 2 tumours into two clinically distinct subgroups. In this study, growth fractions of 1,550 primary operable invasive breast carcinomas were immunohistochemically assayed on full-face tissue sections using the MIB1 clone of Ki-67. Growth fractions were assessed as number of MIB1 positive nuclei in 1,000 tumour nuclei at high-power magnification and expressed as MIB1-LI. Using a 10% cut-point of MIB1-LI, grade 2 BCs were classified into low (49.8%) and high (50.2%) proliferative subgroups. Univariate and multivariate survival analysis revealed statistically significant differences between these subgroups regarding patients’ BC specific survival (P < 0.001), and metastasis free survival (P < 0.001) which was independent of the well-established prognostic factors (HR = 2.944, 95% CI = 1.634–5.303, P < 0.001). In conclusion, our results further demonstrate that grade 2 BCs may represent at least two biological or behaviourally different entities. Assay of growth fraction in BC using MIB1/Ki67 immunohistochemistry is a robust cost-effective diagnostic tool that subdivides grade 2 tumours into low and high risk populations providing additional prognostic information in planning therapies and outcome prediction.