Intrinsic subtype distribution should vary according to institutions.

We read with interest the article by Maranta et al. [1], which described the distribution of Ki-67 values at a single institution in an effort to elucidate the cut-off value discriminating luminal A-like and B-like tumours. Despite of their impressively consistent data, this study has several concerns requiring further consideration.

Firstly, it is unclear how the biopsy samples (30% of the subjects) were chosen. Ki-67 staining tends to be heterogeneous and as such, is usually assessed at a hot spot. For this reason, surgical specimens may enable a more accurate selection of an assessment field [2]. Furthermore, we previously found the Ki-67 labelling index can change dramatically during neo-adjuvant chemotherapy (NAC) [3] and recommended biopsy specimens be examined in these patients. Information pertaining to NAC is therefore crucial, but is not provided in the current study.

Secondly, intrinsic subtype cannot be determined immunohistochemistry, it is only a substitute for gene profiling in practical use. Moreover, the variance in intrinsic subtype distribution between patient cohorts makes the attempt to correspond them pointless, and this approach may cause confusion and misunderstanding among clinicians. Considering the distinctive nature and background of Ki-67, an institution-specific cut-off value can rightfully exist and be established by examining patient outcomes.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of competing interest

The authors have no conflicts of interest to declare.