The usefulness of CanAssist Breast over Ki67 in breast cancer recurrence risk assessment

Abstract

AbstractBackgroundAssessment of Ki67 by immunohistochemistry (IHC) has limited utility in clinical practice owing to analytical validity issues. According to International Ki67 Working Group (IKWG) guidelines, treatment should be guided by a prognostic test in patients expressing intermediate Ki67 range, >5%–<30%. The objective of the study is to compare the prognostic performance of CanAssist Breast (CAB) with that of Ki67 across various Ki67 prognostic groups.MethodsThe cohort had 1701 patients. Various risk groups were compared for the distant relapse‐free interval (DRFi) derived from Kaplan–Meier survival analysis. As per IKWG, patients are categorized into three risk groups: low‐risk (<5%), intermediate risk (>5%–<30%), and high‐risk (>30%). CAB generates two risk groups, low and high risk based on a predefined cutoff.ResultsIn the total cohort, 76% of the patients were low risk (LR) by CAB as against 46% by Ki67 with a similar DRFi of 94%. In the node‐negative sub‐cohort, 87% were LR by CAB with a DRFi of 97% against 49% by Ki67 with a DRFi of 96%. In subgroups of patients with T1 or N1 or G2 tumors, Ki67‐based risk stratification was not significant while it was significant by CAB. In the intermediate Ki67 (>5%–<30%) category up to 89% (N0 sub‐cohort) were LR by CAB and the percentage of LR patients was 25% (p < 0.0001) higher compared to NPI or mAOL. In the low Ki67 (≤5%) group, up to 19% were segregated as high‐risk by CAB with 86% DRFi suggesting the requirement of chemotherapy in these low Ki67 patients.ConclusionCAB provided superior prognostic information in various Ki67 subgroups, especially in the intermediate Ki67 group.