What is the utility and prognostic value of Ki-67 as a biomarker to guide treatment decisions regarding adjuvant abemaciclib in combination with endocrine therapy for patients with hormone receptor (HR)-positive / HER2-negative early breast cancer?
The FDA approval of abemaciclib for patients with high-risk HR-positive / HER2-negative early breast cancer represents a significant advancement in the field, and it is notably the first regulatory approval in the adjuvant HR-positive/HER2-negative setting after more than 15 years. Based on the results of the phase III monarchE trial, the FDA approved the use of abemaciclib in combination with endocrine therapy (tamoxifen or an aromatase inhibitor) for the adjuvant treatment of adult patients with node-positive, HR-positive/HER2-negative early breast cancer and a high risk of recurrence based on clinicopathologic features (either ≥ 4 axillary lymph nodes or 1-3 positive axillary lymph nodes with tumor grade 3 and / or tumor size ≥ 50 mm) and a Ki-67 score ≥ 20%, as determined by an FDA-approved test. The details of the approval, specifically the inclusion of a Ki-67 score of ≥ 20% as a requirement, has created some controversy in the oncology community.
The monarchE trial results demonstrated that patients with the highest risk of recurrence—specifically those with both high-risk clinical features and a high Ki-67 score—derived the greatest absolute benefit in terms of invasive disease-free survival with the addition of abemaciclib to endocrine therapy. Not surprisingly, this subgroup was the basis for the FDA approval. However, the study also demonstrated that although a high Ki-67 score is prognostic for recurrence, it is not a predictive marker for benefit from abemaciclib, as patients with a low Ki-67 score also demonstrated improved invasive disease-free survival with the addition of abemaciclib to endocrine therapy. Some in the community feel that a longer follow-up period is needed for the population of patients with a low Ki-67 score to better understand the role of adjuvant abemaciclib in this patient group. Historical data suggest that this group of patients with a low Ki-67 score is more likely to experience later recurrence compared with the group of patients with a high Ki-67 score. In view of this, both the National Comprehensive Cancer Network and American Society of Clinical Oncology guidelines currently recommend the consideration of treatment with adjuvant abemaciclib in otherwise eligible patients regardless of the Ki-67 score. This underscores the notion that the most significant impact of novel agents such as CDK4/6 inhibitors is likely to be in the localized, curable breast cancer setting, where they could reduce the risk of progression to metastatic disease.
The use of the Ki-67 score in the monarchE trial is a natural choice given that it is a marker of cellular proliferation and therefore mechanistically linked to CDK4/6 inhibitors. Also, the nuclear staining of the Ki-67 protein as a prognostic biomarker for HR-positive/HER2-negative breast cancer is well established, widely used, and inexpensive. However, the scoring methodology for Ki-67 stains remains controversial. Data from the International Ki67 in Breast Cancer Working Group (IKWG) have shown dramatic heterogeneity in Ki-67 assessment across laboratories, and the IKWG has expressed serious concerns regarding its routine use. The IKWG has systematically analyzed the causes of Ki-67 variability across laboratories and developed guidelines for its assessment using immunohistochemistry (IHC). The IKWG is of the opinion that consistency in Ki-67 scoring is difficult to achieve in the 5% to 30% range. Of note, the FDA approval of a Ki-67 score using a ≥20% cutoff as a companion diagnostic has not been generally embraced in the breast oncology community.
Ki-67 is a pan-cell cycle marker that is expressed in all tissues, albeit at low levels. This has led to issues regarding antibody titration for IHC analysis. A consequence of this is that antibodies are used at different dilutions, with resultant differences in nuclear staining intensities. The FDA approval of the Ki-67 IHC MIB-1 assay now requires the use of a ready-to-use standard concentration of an antibody. Similarly, it provides standardized criteria and analytical methods for defining Ki-67 positivity. Regarding the monarchE trial and the FDA-approved criteria, a nucleus is considered positive for Ki-67 if: (1) the signal is unequivocally brown; (2) the staining corresponds to a nucleus; (3) the staining covers the whole chromatin distribution within the nucleus; and (4) the nuclear staining is observed in viable, nonapoptotic cells. This definition differs from the one proposed by the IKWG in that the IKWG defines a nucleus as positive if it is not blue; therefore, grey nuclei are considered positive. In addition, focal nuclear positivity is considered sufficient. Both the IKWG- and FDA-approved methods recommend the analysis of the entire slide, although the IKWG method is more involved in requiring counting of larger areas and a higher number of cells. Both methods exclude necrotic tumor areas, foci of carcinoma in situ, edge effects, and fixation- and processing-related artifacts.
Overall, CDK4/6 inhibitors have revolutionized the treatment for patients with HR-positive/HER2-negative advanced breast cancer, and the FDA approval of adjuvant abemaciclib plus endocrine therapy in the early breast cancer setting is a significant advancement in the field.
Rodrigo Arrangoiz MS, MD, FACS, FSSO 