Identifying the optimal cutoff point of Ki-67 in breast cancer: a single-center experience

Abstract

Objective

Ki-67 is associated with breast cancer subtypes, but the optimal cutoff point of Ki-67 has not been established in our center. We evaluated the cutoff point of Ki-67 in breast cancer and analyzed the associations among Ki-67, clinicopathological features, and prognosis.

Methods

The clinicopathological data and prognostic information of patients with breast cancer treated in our center were retrospectively collected, and the optimal cutoff point of Ki-67 was determined by univariate and multivariate survival risk analyses. The cutoff point was used to group the patients, and the differences in the clinicopathological features and prognosis were analyzed between the two groups.

Results

In total, 609 patients with estrogen receptor-positive and human epidermal growth factor receptor 2-negative primary breast cancer were enrolled. The mean Ki-67 value was 22.3% ± 15.4%, the median was 20%, and a cutoff point of 30% was an independent factor influencing recurrence-free survival. When 30% was used as the cutoff point, patients with a Ki-67 value of ≤30% had a better prognosis and lower tumor malignancy.

Conclusion

The optimal cutoff point of Ki-67 in breast cancer in our center is 30%.

Keywords

Ki-67 cutoff point breast cancer molecular subtype prognosis malignancy

Introduction

Genetic testing has confirmed the presence of heterogeneity in breast cancer. For example, breast cancer subtypes are divided into luminal A, luminal B, normal breast-like, human epidermal growth factor receptor 2 (HER2)-enriched, and basal-like.¹ Some researchers have divided breast cancer into four subtypes according to the Prediction Analysis of Microarray 50 (PAM50).² The correct classification helps formulate reasonable treatment plans. For example, the luminal A subtype is sensitive to endocrine therapy, and the HER2-enriched subtype is sensitive to anti-HER2 targeted therapy. However, because of economic costs and other factors, it is impossible to perform genetic testing on every patient in the clinical setting; therefore, clinicians usually use immunohistochemical methods instead of genetic testing to classify breast cancer.³^,⁴

Immunohistochemical methods are used to determine breast cancer subtypes by detecting estrogen receptor (ER), progesterone receptor (PR), HER2, and Ki-67. According to the recommendations of the St. Gallen International Expert Consensus in 2013, luminal A breast cancer is ER- and PR-positive (PR expression of ≥20%) and HER2-negative and shows low expression of Ki-67 (Ki-67 <14%), whereas luminal B (HER2-negative) breast cancer is ER-positive and HER2-negative, has PR expression of <20%, or has high expression of Ki-67.⁴ Therefore, Ki-67 is important for distinguishing luminal A from luminal B (HER2-negative) breast cancer. Moreover, Ki-67 is an important predictor. For example, in a clinical study involving 2887 patients with breast cancer with a median follow-up duration of up to 10 years, Ki-67 was confirmed to be a predictive factor of docetaxel, suggesting that only patients with ER-positive cancer exhibiting high Ki-67 expression may receive benefits from taxane therapy.⁵

No consensus has been reached regarding the optimal detection method and cutoff point for Ki-67.⁶ Ki-67 is a nuclear protein that was first reported by Gerdes et al.⁷ in 1983. This protein is related to cell proliferation and is expressed in all cell cycle stages except G0.⁷^,⁸ Breast cancer cells express Ki-67, and high expression of Ki-67 is associated with a poor prognosis.⁹ Inconsistencies in Ki-67 detection and interpretation methods among laboratories have led to variances in detection results, adversely affecting the accuracy of breast cancer subtyping.¹⁰ For example, the cutoff point of Ki-67 was set at 14% at St. Gallen in 2013⁴ and was later revised to 20% to 29% at St. Gallen in 2015, as determined by each laboratory.¹¹ Standardizing Ki-67 assays is challenging for two main reasons. First, although a certain degree of variation exists among pathologists’ interpretations of staining results, Ki-67 is a continuous marker that is extremely robust against such variations.¹² Second, pre-analytical factors (e.g., tissue handling, fixation) may lead to differences in Ki-67 measurement results between presurgical biopsies and postsurgical specimens.¹³

Our center has not yet defined the cutoff point of Ki-67, leading to some difficulty in the subtyping of breast cancer. Therefore, in the present study, we used the prognosis as the observation index to define the optimal cutoff point of Ki-67 in our center and analyzed the relationship between Ki-67 and the clinicopathological features of breast cancer.

Patients and methods

Patients

The clinicopathological data of consecutive patients with breast cancer treated in our center from 1 January 2012 to 31 December 2019 were collected. The inclusion criteria were (1) primary invasive breast cancer, (2) pTNM stage 1 to 3 (AJCC 8th edition), (3) ER positivity and HER2 negativity, (4) age of >18 years, (5) female, (6) receiving radical surgical treatment (including but not limited to modified radical mastectomy or breast-conserving surgery), and (7) availability of complete clinical and pathological information. The exclusion criteria were (1) ductal carcinoma in situ (DCIS), (2) pT1mi, (3) simultaneous double primary breast cancer, (4) ipsilateral supraclavicular lymph node metastasis, (5) distant metastasis, (6) treatment with neoadjuvant chemotherapy or neoadjuvant endocrine therapy, and (7) other simultaneous or previous malignant tumors. All patients received postoperative adjuvant chemotherapy, endocrine therapy, and radiotherapy. The indication criteria for adjuvant chemotherapy were lymph node metastases (except for pN1mi), invasive tumors of >2 cm, and age of <35 years. The indication criterion for adjuvant endocrine therapy was ER or PR positivity. The indication criteria for adjuvant radiotherapy were breast-conserving surgery, lymph node positivity, ≥5-cm maximum diameter of the primary tumor, and tumor invasion of the breast skin or chest wall. The prognosis of patients with ER-positive, HER2-negative breast cancer is influenced by several factors, including the patient’s age, pT and pN status, PR expression, and Ki-67 level. To ensure statistical validity, a minimum of 20 patients per factor was required, resulting in 100 patients. Accounting for potential missed visits, a sample size of 120 patients was recommended to meet the statistical requirements.

Methods

The reporting of this retrospective observational study conforms to the REMARK guidelines.¹⁴ The clinicopathological data of the patients were collected, and follow-up was conducted. All patient details have been de-identified. The results of ER, PR, HER2, and Ki-67 detection were fully recorded in the pathological data. Ki-67 was detected by immunohistochemistry, and the proportion of positive cells among all tumor cells was estimated visually.¹⁵ Specifically, surgical specimens were routinely dehydrated, embedded in paraffin, and sectioned within 30 minutes after resection. The thickness of the wax slices was 4 μm. Cancer tissue sections containing normal breast tissue were selected for EnVision two-step immunohistochemical detection. The clone number of the Ki-67 antibody was MIB1 (purchased from Beijing Zhongshan Jinqiao Biotechnology Co., Ltd., Beijing, China), and Real EnVision K5007 was selected as the secondary antibody display system (purchased from Hangzhou Xiebo Pharmaceutical Technology Co., Ltd., Hangzhou, China). Twenty fields of view were observed under a high-power microscope, and the proportion of positive cells among all tumor cells was visually estimated without evaluation of the hot-spot. ER and PR were detected by immunohistochemistry, and a ≥1% proportion of positive cells was considered positive.¹⁶^,¹⁷ HER2 was detected by immunohistochemistry and fluorescence in situ hybridization. HER2 negativity was defined as follows: immunohistochemical detection was negative or 1+, or patients with 2+ immunohistochemical detection were further subjected to detection of the HER2 gene by fluorescence in situ hybridization and the result was negative (HER2/CEP17 ratio of <2.0, average HER2 copy number/cell of <4.0).¹⁸

The prognostic indicators were relapse-free survival (RFS), disease-free survival (DFS), breast cancer-specific survival (BCSS), and overall survival (OS). RFS was defined as the time from radical surgery to any of the following events: invasive ipsilateral breast tumor recurrence/progression, local invasive recurrence/progression, regional invasive recurrence/progression, appearance/occurrence of metastases/distant recurrence, ipsilateral DCIS, and death of any cause. Patients who did not experience the above events were counted until the last follow-up. DFS was defined as the time from radical surgery to termination events in addition to all termination events described by RFS; it also included invasive contralateral breast cancer, second primary invasive cancer (non-breast cancer), and contralateral DCIS. BCSS was defined as the time from radical surgery to death of breast cancer recurrence or metastasis. OS was defined as the time from radical surgery to death of any cause.¹⁹^,²⁰ The end of follow-up was 31 December 2022. The median follow-up time was 6 years.

Statistical analysis

The data were processed using IBM SPSS Statistics for Windows, Version 22.0 (IBM Corp., Armonk, NY, USA). Measurement data with a non-normal distribution were evaluated using the nonparametric Mann–Whitney U test. Count data were analyzed by Pearson’s chi-square test. The Kaplan–Meier method was used for survival analysis, and the log-rank test was used to test the difference in survival curves between groups. Univariate and multivariate survival analyses were performed using a Cox regression model. All statistical tests were bilateral tests, and the differences were statistically significant at P < 0.05.

Results

Patients

This study was approved by the Ethics Committee of the First Hospital of Jiaxing (approval no. 2022-LY-385). All patients provided written informed consent to participate in the study at the time of first admission. In total, 609 patients with a median age of 54 years were enrolled. The distributions of patients according to menstruation (premenopausal vs. postmenopausal), pathological type, histological grade, PR status, tumor diameter, and axillary lymph node status are shown in Table 1.

Table 1.

Clinicopathological features of the patients.

Clinicopathological feature	Group	n (%)
Menstruation status	Premenopausal	297 (48.8%)
Menstruation status	Postmenopausal	312 (51.2%)
Pathological type	Invasive breast cancer, unspecified	505 (82.9%)
Pathological type	Others	104 (17.1%)
Histological grade	1/2	487 (80.0%)
Histological grade	3	122 (20.0%)
PR status	Positive	560 (92.0%)
PR status	Negative	49 (8.0%)
Tumor diameter	≤2 cm	394 (64.7%)
Tumor diameter	>2 cm	215 (35.3%)
Axillary lymph node status	Positive	222 (36.5%)
Axillary lymph node status	Negative	387 (63.5%)
Chemotherapy	Anthracyclines	164 (26.9%)
	Anthracyclines and taxanes	155 (25.5%)
	Others	33 (5.4%)
	No chemotherapy	257 (42.2%)
Endocrine therapy	AI	293 (48.1%)
	AI and OFS	22 (3.6%)
	TAM	277 (45.5%)
	TAM and OFS	16 (2.6%)
	OFS	1 (0.2%)
Radiotherapy	Yes	284 (46.6%)
Radiotherapy	No	325 (53.4%)

PR, progesterone receptor; AI, aromatase inhibitor; OFS, ovarian function suppression; TAM, tamoxifen.

Distribution of Ki-67

The mean Ki-67 was 22.3% ± 15.4%, and the median was 20% (Figure 1).

Figure 1.

Distribution of Ki-67 (the abscissa is Ki-67, and every 5% is an interval; the ordinate is the number of patients).

Influence of Ki-67 on RFS

Ki-67, tumor diameter, axillary lymph node status, histological grade, and PR status were included in the multivariate survival analysis. Ki-67 was a continuous variable, and the others were categorical variables. The tumor diameter was divided into ≤2-cm and >2-cm groups; the axillary lymph node status was divided into negative and positive groups; the histological grade was divided into grades 1, 2, and 3; and the PR status was divided into negative and positive groups. RFS was chosen as the prognostic indicator in the multivariate survival analysis. The results showed that Ki-67 and the axillary lymph node status were independent influencing factors of RFS (Table 2).

Table 2.

Results of multivariate survival risk analysis for RFS, DFS, BCSS, and OS.

Variable	Group	RFS		DFS		BCSS		OS
Variable	Group	P	HR (95% CI)	P	HR (95% CI)	P	HR (95% CI)	P	HR (95% CI)
Ki-67	Continuous	0.012	1.029 (1.006–1.051)	<0.001	1.030 (1.014–1.047)	0.004	1.069 (1.021–1.120)	0.023	1.041 (1.006–1.078)
Tumor diameter	≤2 cm vs. >2 cm	0.115	0.584 (0.299–1.140)	0.851	0.954 (0.586–1.555)	0.279	0.403 (0.078–2.088)	0.834	0.889 (0.295–2.677)
Axillary lymph node status	Negative vs. positive	0.001	0.297 (0.149–0.591)	0.009	0.525 (0.324–0.849)	0.101	0.250 (0.048–1.308)	0.196	0.486 (0.163–1.451)
Histological grade	1/2 vs. 3	0.513	1.335 (0.561–3.174)	0.519	1.240 (0.645–2.387)	0.853	1.172 (0.218–6.301)	0.955	0.961 (0.241–3.828)
PR status	Negative vs. positive	0.064	0.147 (0.019–1.116)	0.011	0.152 (0.036–0.646)	0.989	0.000 (0.000)	0.978	0.000 (0.000)

RFS, relapse-free survival; DFS, disease-free survival; BCSS, breast cancer-specific survival; OS, overall survival; HR, hazard ratio; CI, confidence interval; PR, progesterone receptor.

Optimal cutoff point for Ki-67

Given that Ki-67 was an independent influencing factor of RFS, the next step was to identify the optimal cutoff point of Ki-67. Figure 1 shows that the Ki-67 distribution of most patients was <40%; therefore, a univariate survival analysis was performed using the Ki-67 cutoff point within this range at an interval of 5%. Because the number of patients within the interval of 30% to 35% was 0, the cutoff point of 35% was discarded. The results showed that when the Ki-67 cutoff point was 15%, 20%, 25%, 30%, and 40%, the survival differences between the two groups were statistically significant (Table 3). A multivariate survival analysis was performed using the above cutoff points, and the result showed that the cutoff point of 30% was an independent influencing factor of RFS (Table 4). Using 30% as the cutoff point, the patients were divided into those with Ki-67 of ≤30% and those with Ki67 of >30%, and the RFS, DFS, BCSS, and OS in the two groups were calculated. The differences in these indices were statistically significant, indicating that the prognosis of patients with Ki-67 of ≤30% was better than that of patients with Ki-67 of >30% (Figure 2).

Table 3.

Univariate survival analysis.

Cutoff point of Ki-67	P	HR (95% CI)
5%	0.545	0.747 (0.291–1.919)
10%	0.071	0.525 (0.261–1.057)
15%	0.041	0.498 (0.255–0.971)
20%	0.001	0.330 (0.174–0.623)
25%	<0.001	0.325 (0.173–0.612)
30%	<0.001	0.250 (0.134–0.466)
40%	0.007	0.346 (0.159–0.752)

HR, hazard ratio; CI, confidence interval.

Table 4.

Multivariate survival analysis.

Cutoff point of Ki-67	P	HR (95% CI)
15%	0.488	1.561 (0.443–5.496)
20%	0.465	0.430 (0.045–4.138)
25%	0.944	1.080 (0.126–9.258)
30%	0.041	0.327 (0.112–0.957)
40%	0.819	1.116 (0.436–2.855)

HR, hazard ratio; CI, confidence interval.

Figure 2.

Survival curves when the Ki-67 cutoff point was 30%. (a) Relapse-free survival. (b) Disease-free survival. (c) Breast cancer-specific survival and (d) Overall survival.

Relationship between Ki-67 and clinicopathological features of breast cancer

Using 30% as the cutoff point, the patients were divided into those with Ki-67 of ≤30% and those with Ki-67 of >30%, and the differences in age, menstruation status, pathological type, histological grade, PR status, tumor diameter, and axillary lymph node status between the two groups were analyzed. The results showed that except for the pathological type, all differences were statistically significant. Patients with Ki-67 of ≤30% had a higher median age, a lower proportion of a premenopausal status, a lower proportion of histological grade 3 tumors, a higher proportion of PR-positive tumors, smaller tumor diameters, and a lower proportion of positive axillary lymph nodes (Table 5).

Table 5.

Relationship between Ki-67 and clinicopathological features of breast cancer.

Clinicopathological feature	Ki-67		Statistical value	P
Clinicopathological feature	≤30%(n = 497)	>30%(n = 112)	Statistical value	P
Age in years, median [IQR]	55 [49, 63]	50 [46, 58]	20163	<0.001
Menstruation status
Premenopausal	224	73	14.792	<0.001
Postmenopausal	273	39	14.792	<0.001
Pathological type
Invasive breast cancer, unspecified	406	99	2.900	0.089
Other	91	13	2.900	0.089
Histological grade
1/2	450	37	188.692	<0.001
3	47	75	188.692	<0.001
PR status
Positive	466	94	11.947	0.001
Negative	31	18	11.947	0.001
Tumor diameter
≤2 cm	347	47	31.998	<0.001
>2 cm	150	65	31.998	<0.001
Axillary lymph node status
Positive	159	63	23.219	<0.001
Negative	338	49	23.219	<0.001

IQR, interquartile range; PR, progesterone receptor.

Discussion

The lack of reproducibility of Ki-67 detection by immunohistochemistry among different laboratories due to differences in staining methods (including staining platforms, antigen retrieval, primary antibodies, detection systems, and counterstaining) has led to doubts about the application value of Ki-67 in clinical practice.²¹ The cutoff point of Ki-67 can be used to distinguish between luminal A and luminal B (HER2-negative) breast cancer, but it has been inconsistently reported by various laboratories; it widely ranges from 3.5% to 34.0%, and no consensus has been reached.²² The inconsistency of Ki-67 among different laboratories may be related to the lack of uniform standards for the detection and interpretation of Ki-67.²³ For example, methods for assessing the percentage of Ki-67-positive cells include “eyeballing” (visual scanning and estimating of the staining percentage), formally counting the cells, and using image analysis methods.²⁴ At present, most experts tend to situationally determine the optimal cutoff point for each laboratory.²⁵ These issues are expected to be addressed through the use of new standardized tools.²⁶

To determine the optimal cutoff point of Ki-67, we first delineated the range of 5% to 40% according to the distribution characteristics of Ki-67 in our center and then performed a survival analysis with 5% as the interval. The 5% interval was chosen because previous studies have shown that reported Ki-67 values tended to cluster around those ending in 5 or 0.²⁷ We first analyzed the significance of each candidate cutoff point with a univariate survival analysis; we then performed a multivariate survival analysis using the candidate cutoff points with statistically significant differences. This process revealed a Ki-67 of 30% as an independent influencing factor. This method of determining the optimal cutoff point was used with reference to a study by Tashima et al.²⁸

When selecting prognostic indicators, we believe that Ki-67, as an indicator related to tumor proliferation, is most closely associated with tumor recurrence and metastasis. This led us to choose RFS as a prognostic indicator because it is similar to the recurrence-free interval adopted by Ohara et al.²⁹ Thangarajah et al.³⁰ found that lower Ki-67 was associated with better DFS, which is consistent with our findings; however, we did not use DFS to discern the optimal cutoff point because contralateral primary breast cancer and other primary malignancies were not strongly associated with primary Ki-67, and increasing these events may have confounded the findings. In a study by Tashima et al.,²⁸ OS was used as the prognostic indicator, and an optimal cutoff point of 20% was obtained. Cho et al.³¹ used BCSS as the prognostic indicator. However, this study did not choose OS and BCSS because great progress has been made in endocrine therapy for breast cancer in recent years (such as the application of CDK4/6 inhibitors).³²^,³³ Patients with ER-positive breast cancer still have a chance of achieving long-term survival after recurrence and metastasis; thus, OS and BCSS may be affected by the impact of treatment after recurrence and metastasis.³⁴

After determining 30% as the optimal cutoff point of Ki-67 in our center, we analyzed the associations of Ki-67 with RFS, DFS, BCSS, and OS. The results showed that this cutoff point accurately distinguished the survival risks in terms of these four indices; i.e., Ki-67 of ≤30% was associated with a better prognosis. In the analysis of clinicopathological features, this cutoff point also demonstrated strong discriminative ability in that patients with Ki-67 of ≤30% tended to have less malignancy. Therefore, we believe that 30% should be used as the optimal cutoff point of Ki-67 in our center.

Because this was an observational single-center study, and considering the variability of Ki-67 assays and other factors, we acknowledge that the findings of this study should only serve as a reference for patients who are using the same Ki-67 assay. These results cannot be extrapolated to patients who are using different Ki-67 assays.

Conclusion

This retrospective study showed that 30% is the optimal cutoff point of Ki-67 for breast cancer in our center. This cutoff point may predict the survival risk and aid in accurate classification of subtypes. Notably, two main limitations of this study are the relatively small sample size (609 cases) and short follow-up period. We will collect more cases and extend the follow-up period to verify our conclusion.

Footnotes

Acknowledgement

The authors acknowledge Mr. Wanxin Wu for his help in interpreting the significance of the results of this study.

Author contributions

Wang Li: Formal analysis, Writing - Original Draft. Ning Lu: Methodology. Caiping Chen: Validation, Supervision, Writing - Review & Editing. Xiang Lu: Conceptualization, Data Curation, Visualization.

Data availability statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Declaration of conflicting interest

The authors declare that there is no conflict of interest.

Funding

This work was supported by the Jiaxing Key Discipline of Medicine (Mastropathy, Innovation Subject, 2023-FC-001) and the Breast Cancer Precision Diagnosis and Treatment Center of the First Hospital of Jiaxing (2021-ZZZX-06).

ORCID iD

Xiang Lu

References

Sørlie

Perou

Tibshirani

, et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci U S A 2001; 98: 10869–10874.

Parker

Mullins

Cheang

, et al. Supervised risk predictor of breast cancer based on intrinsic subtypes. J Clin Oncol 2009 27: 1160–1167.

Russnes

Lingjærde

Børresen-Dale

, et al. Breast cancer molecular stratification: From intrinsic subtypes to integrative clusters. Am J Pathol 2017; 187: 2152–2162.

Goldhirsch

Winer

Coates

, et al. Personalizing the treatment of women with early breast cancer: Highlights of the St Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2013. Ann Oncol 2013; 24: 2206–2223.

Sonnenblick

Francis

Azim

Jr , et al. Final 10-year results of the Breast International Group 2-98 phase III trial and the role of Ki67 in predicting benefit of adjuvant docetaxel in patients with oestrogen receptor positive breast cancer. Eur J Cancer 2015; 51: 1481–1489.

Thomssen

Balic

Harbeck

, et al. Gallen/Vienna 2021: A brief summary of the consensus discussion on customizing therapies for women with early breast cancer. Breast Care (Basel) 2021; 16: 135–143.

Gerdes

Schwab

Lemke

, et al. Production of a mouse monoclonal antibody reactive with a human nuclear antigen associated with cell proliferation. Int J Cancer 1983; 31: 13–20.

Gerdes

Schlueter

, et al. Immunobiochemical and molecular biologic characterization of the cell proliferation-associated nuclear antigen that is defined by monoclonal antibody Ki-67. Am J Pathol 1991; 138: 867–873.

Boyaci

Sun

Robertson

, et al. Independent clinical validation of the automated Ki67 scoring guideline from the International Ki67 in Breast Cancer Working Group. Biomolecules 2021; 11: 1612.

10.

Davey

Hynes

Kerin

, et al. Ki-67 as a prognostic biomarker in invasive breast cancer. Cancers (Basel) 2021; 13: 4455.

11.

Coates

Winer

Goldhirsch

, et al. Tailoring therapies–improving the management of early breast cancer: St Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2015. Ann Oncol 2015; 26: 1533–1546.

12.

Denkert

Budczies

Regan

, et al. Clinical and analytical validation of Ki-67 in 9069 patients from IBCSG VIII + IX, BIG1-98 and GeparTrio trial: Systematic modulation of interobserver variance in a comprehensive in silico ring trial. Breast Cancer Res Treat 2019; 176: 557–568.

13.

Acs

Leung

SCY

Kidwell

, et al. Systematically higher Ki67 scores on core biopsy samples compared to corresponding resection specimen in breast cancer: A multi-operator and multi-institutional study. Mod Pathol 2022; 35: 1362–1369.

14.

Altman

McShane

Sauerbrei

, et al. Reporting recommendations for tumor marker prognostic studies (REMARK): Explanation and elaboration. BMC Med 2012; 10: 51.

15.

Leung

SCY

Nielsen

Zabaglo

, et al. Analytical validation of a standardised scoring protocol for Ki67 immunohistochemistry on breast cancer excision whole sections: An international multicentre collaboration. Histopathology 2019; 75: 225–235.

16.

Allison

Hammond

MEH

Dowsett

, et al. Estrogen and progesterone receptor testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists Guideline Update. Arch Pathol Lab Med 2020; 144: 545–563.

17.

Wolff

Hammond

MEH

Allison

, et al. Human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists Clinical Practice Guideline Focused Update. Arch Pathol Lab Med 2018; 142: 1364–1382.

18.

Recommended by Breast Cancer Expert Panel. Guideline for HER2 detection in breast cancer, the 2019 version. Zhonghua Bing Li Xue Za Zhi 2019; 48: 169–175.

19.

Gourgou-Bourgade

Cameron

Poortmans

, et al. Guidelines for time-to-event end point definitions in breast cancer trials: Results of the DATECAN initiative (Definition for the Assessment of Time-to-event Endpoints in CANcer trials)†. Ann Oncol 2015; 26: 873–879.

20.

Gourgou-Bourgade

Cameron

Poortmans

21.

Nielsen

Leung

SCY

Rimm

, et al. Assessment of Ki67 in breast cancer: Updated recommendations from the International Ki67 in Breast Cancer Working Group. J Natl Cancer Inst 2021; 113: 808–819.

22.

De Azambuja

Cardoso

De Castro

Jr , et al. Ki-67 as prognostic marker in early breast cancer: A meta-analysis of published studies involving 12,155 patients. Br J Cancer 2007; 96: 1504–1513.

23.

Pathmanathan

Balleine

RL.

Ki67 and proliferation in breast cancer.

J Clin Pathol 2013; 66: 512–516.

24.

Van den Berg

Duarte

Dickens

, et al. Ki67 Immunohistochemistry quantification in breast carcinoma: A comparison of visual estimation, counting, and ImmunoRatio. Appl Immunohistochem Mol Morphol 2021; 29: 105–111.

25.

Sparano

Gray

Makower

, et al. Adjuvant Chemotherapy Guided by a 21-Gene Expression Assay in Breast Cancer. N Engl J Med 2018; 379: 111–121.

26.

Aung

Acs

Warrell

, et al. A new tool for technical standardization of the Ki67 immunohistochemical assay. Mod Pathol 2021; 34: 1261–1270.

27.

Cserni

Vörös

Liepniece-Karele

, et al. Distribution pattern of the Ki67 labelling index in breast cancer and its implications for choosing cut-off values. Breast 2014; 23: 259–263.

28.

Tashima

Nishimura

Osako

, et al. Evaluation of an optimal cut-off point for the Ki-67 index as a prognostic factor in primary breast cancer: A retrospective study. PLoS One 2015; 10: e0119565.

29.

Ohara

Matsuura

Akimoto

, et al. Prognostic value of Ki67 and p53 in patients with estrogen receptor-positive and human epidermal growth factor receptor 2-negative breast cancer: Validation of the cut-off value of the Ki67 labeling index as a predictive factor. Mol Clin Oncol 2016; 4: 648–654.

30.

Thangarajah

Enninga

Malter

, et al. A retrospective analysis of Ki-67 index and its prognostic significance in over 800 primary breast cancer cases. Anticancer Res 2017; 37: 1957–1964.

31.

Cho

Kim

, et al. The long-term prognostic performance of Ki-67 in primary operable breast cancer and evaluation of its optimal cutoff value. Appl Immunohistochem Mol Morphol 2016; 24: 159–166.

32.

Rugo

Finn

Diéras

, et al. Palbociclib plus letrozole as first-line therapy in estrogen receptor-positive/human epidermal growth factor receptor 2-negative advanced breast cancer with extended follow-up. Breast Cancer Res Treat 2019; 174: 719–729.

33.

Johnston

Martin

Di Leo

, et al. MONARCH 3 final PFS: A randomized study of abemaciclib as initial therapy for advanced breast cancer. NPJ Breast Cancer 2019; 5: 5.

34.

Rakha

Chmielik

Schmitt

, et al. Assessment of predictive biomarkers in breast cancer: Challenges and updates. Pathobiology 2022; 89: 263–277.