Patient outcomes and clinician perspectives following one year of ad hoc implementation of neoadjuvant endocrine therapy in early breast cancer

Introduction

Background

Endocrine therapy is a cornerstone of management of estrogen receptor (ER)-positive, HER2-negative breast cancer, which accounts for over 75% of incident cases of breast cancer (1). Endocrine therapy is now standard of care for patients with ER-positive (luminal) breast cancer in the adjuvant setting, conferring a reduction in mortality from breast cancer at 15 years by up to a third (2). The literature supports a prolonged course of treatment for prophylaxis of recurrence (3-5).

The extent to which endocrine therapy could be harnessed as a neoadjuvant therapy has been studied in more recent years (6-10), particularly in terms of aiding with prognostication and selecting patients who may receive little additional benefit from adjuvant chemotherapy. Ki-67 has been shown to be both a predictive and prognostic biomarker in breast cancer (11-15). Recent reviews have suggested that pre-menopausal patients with a Ki-67 of >20% should receive an extended course (>5 years) of adjuvant endocrine therapy (16). A correlation between Ki-67 and risk of relapse of surgically resected HR+/HER2 early breast cancer has also been identified, with suggested cutoff values for low risk of relapse of <20% and high risk of relapse of >30% (17). Unfortunately, interpretation of Ki-67 values is highly variable due to differing cutoff values between labs, different analysis techniques, delays in time to fixation and different reporting pathologists (18-25).

Neoadjuvant endocrine therapy (NAET) was shown to produce changes in Ki-67 between core biopsy and excision specimen that can help to predict patient response to endocrine therapy in the adjuvant setting in the POETIC trial in 2022 (26). In some cases, response to NAET predicted a response to adjuvant endocrine therapy so profound that adjuvant chemotherapy could be omitted without negatively affecting relapse-free survival. This therapy provides a unique opportunity for an in-vivo test of disease responsiveness or lack of responsiveness to therapies which are widely available, well-tolerated and cheap.

Several other validated tools exist to predict disease response to adjuvant therapies. One of these is the Oncotype DX test, a 21-gene assay which predicts 10-year risk of metastatic recurrence in node-negative, ER-positive breast cancer patients (27). Others include EPClin and Prosigna/PAM50 which do incorporate Ki-67 and Breast Cancer Index (BCI) which do not. A recent review of these tools concluded that while performing gene expression risk scores on core needle biopsies is a valid prognostic tool, performing these tests at initial biopsy with the goal of administering neoadjuvant chemotherapy (NACT) or NAET to reduce tumour size and potentially achieve complete pathological response should be avoided for patients who are candidates for conservative surgery at diagnosis (28). The oncotype DX test gives a recurrence score where a score of less than 15 indicates that adding adjuvant chemotherapy confers no survival benefit over endocrine therapy alone and a score of 26 to 100 indicates that adding chemotherapy will provide a substantial benefit in relapse-free survival (29). Patients with intermediate recurrence scores on the oncotype DX test have been demonstrated to show no statistically significant differences in pathologic outcomes at the time of surgery based on whether they received NAET or NACT (8). The test costs around 4,000 Australian dollars (AUD) and has a 4-week turnaround time. It is not currently funded in the Australian healthcare system (30), but has been adopted in other healthcare systems around the world (31) in order to reduce unnecessary administration of chemotherapy and the associated costs.

In comparison, a 2-week course of anastrazole costs 20 to 25 AUD (32), and Ki-67 response has been observed after just 2 weeks of NAET (10,26). Using change in Ki-67 from core biopsy to excision specimen, for example from greater than 10% to less than 10% as a marker of response may have similar benefits in terms risk stratification and predicting need for chemotherapy (33) as the oncotype DX test, but at a much lower cost. However, confirming this would require a prospective study, of which none are currently in progress. It can also be observed over a shorter period of time while patients are awaiting surgery. The Medicare billing cost of specimen analysis including Ki-67 ranges from $71.50 to $86.35 (AUD) (34).

Rationale and knowledge gap

This treatment is being considered in academic and clinical settings (35-37), but the development of protocols to use NAET as a prognostic tool is of vital urgency, as patients are being recommended for harmful chemotherapy which may be unnecessary (38). This presents an opportunity to provide individualized care that can spare patients from severe and debilitating side effects which may not reduce their recurrence risk. Most protocols found in large scale randomized controlled trials use varying cutoff points for Ki-67 expression as an inclusion criterion (5-7,9,10,26,33,35,36,39-42), but interpretation of Ki-67 is known to be problematic.

Given that NAET is a relatively low risk treatment and one that patients are likely to commence post-operatively anyway, it is commonly commenced by individual clinicians who are aware of the supporting literature without development of a formal protocol and the associated multidisciplinary planning. While this is common practice, it represents a suboptimal implementation strategy. When implementation strategies are used in practice, they often lack a theoretical foundation which can lead to problems in replicability and generalisability of the intervention (43). Implementation science literature suggests that in order to optimise the adoption of new interventions, clinicians should work with behavioural scientists to gain expertise in programme theory, which combines an interventions components with a narrative about structures, behaviours, processes and contextual features that will be needed to achieve the aims and actions of the intervention (44).

Programme theory should ideally be used prospectively, but is of use even whilst a project is underway, making links between theory and interventions explicit. It can also be beneficial retrospectively, contributing towards standardisation of intervention strategies, furthering understanding of intervention effects, and enhancing the opportunities for accurate replicability and generalisability across other settings (43).

One possible model for the implementation of novel treatment has been developed by KidGen. A network of multi-disciplinary kidney genetic clinics to support the delivery of genomic testing was established (45,46) which saw an increase in wait times for patients due to increased clinician awareness of the benefits of genomic testing in kidney disease. To address this, they implemented a Clinical Change Program to support broader access for patients using a “hub and spoke” model where peripheral units acted as the spoke with direct access to the hubs (tertiary units with) to offer support for implementation of genomic testing (47). This model could be applied to other novel treatments such as NAET, with protocols being developed by one multidisciplinary team for specimen handling, Ki-67 testing and prescribing of NAET for use in other units.

Objective

This exploratory project aims to identify important issues and generate hypotheses that should be considered when developing protocols for the administration of NAET in real-world settings. This was done by observing the outcomes in our unit during a window of opportunity where the treatment was initiated on an ad hoc basis without the adoption of a structured protocol by the unit as a whole, mimicking real-world translation from research to practice. The study investigates the feasibility of implementing NAET recommendation for post-menopausal patients with hormone receptor-positive disease and Ki-67 >10%, and investigates whether change in Ki-67 plays a role in the recommended adjuvant therapy for these patients.

During implementation, breast surgeons and pathologists were interviewed to investigate what they perceived as barriers to initiating the treatment and possible limitations of the use of Ki-67 as both an inclusion criterion and a marker of response. While there is a large body of research on the efficacy of NAET (2,7,33,35-37,40,41,48), this study will examine how one particular protocol can affect adjuvant therapy recommendations in a real-world setting, while gaining valuable feedback on possible limitations and areas for future study from clinicians. These aims are summarized below:

• Evaluate the feasibility of implementing short-course NAET on an ad hoc basis using Ki-67 as a response marker without a formal treatment pathway;
• Assess oncological outcomes, including Ki-67 reduction and its impact on adjuvant therapy decisions;
• Identify barriers to implementation from the perspective of surgeons;
• Identify issues with Ki-67 interpretation that may exacerbate variability in interpretation, affecting eligibility for treatment;
• Provide insights for the development of Ki-67 and NAET protocols to improve reproducibility and clinical decision-making;

We present this article in accordance with the STROBE reporting checklist (available at https://tbcr.amegroups.com/article/view/10.21037/tbcr-25-19/rc) (49).

Methods

This was a mixed methods study with a quantitative and qualitative arm carried out in parallel.

Quantitative methodology

The quantitative component was an observational, case-control study comparing patients who received NAET, with patients who did not, in terms of changes in Ki-67 from core biopsy to excision specimen and recommended adjuvant therapy. Eligible patients were identified for consideration in the study using the Breast Cancer Multi-Disciplinary Meeting Agendas from the period of the study, May 2023 to May 2024. Inclusion criteria were determined by the breast Unit based on the POETIC study (26) which used the following inclusion criteria:

• Post-menopausal patients with newly diagnosed breast cancer;
• ER-positive, HER2-negative cancer;
• Ki-67 of >10% on core biopsy.

All eligible patients had early breast cancer and were planned for breast conserving surgery. Patients who had any clinical features suggesting that they would benefit from neoadjuvant treatment for any other reason (bulky tumour, high grade or metastatic disease) were excluded.

Eligible patients were divided into a group who received NAET (case) and a control group who did not at the discretion at each surgeon as per routine clinical practice. We then recorded and compared their change in Ki-67 between core biopsy and excision specimen. We also reviewed whether or not they were recommended for adjuvant chemotherapy. Data collection and follow-up took place continuously throughout the period of the study. The recommendation from the POETIC trial was that adjuvant endocrine therapy alone could be considered for post-menopausal patients with, ER-positive disease if the Ki-67 went from above 10% in the core biopsy (before NAET) to below 10% in the excision specimen (following NAET). Patients in whom this was achieved were said to have shown a “significant reduction” (26). Patients in our cohort who achieved this were therefore labelled as having a “reduction across the threshold value of 10%”.

Data sources included the unit MDM agendas during the period of the study, clinical notes and histopathology reports. This was an unmatched cohort study which included 25 patients in the exposed group and 20 patients in the unexposed group.

Outcomes included change in Ki-67, whether change in Ki-67 moved across the 10% threshold and adjuvant therapy recommended. Although the study includes a small sample size, patients who were recommended for adjuvant chemotherapy were compared with patients who were not in terms of whether they received NAET or not, the change in Ki-67 that was seen, and their sentinel node status.

Exposures in this study were NAET (anastrozole) and various treating surgeons. Predictors included baseline Ki-67 (if Ki-67 was very high on core biopsy, patients may be less likely to have a “reduction across the 10% threshold” response even if they do see a reduction as their reduced value is likely to still be greater than 10%), histological grade, patient age, treatment adherence, and treating surgeon preference. Potential confounders included disease stage, overall health status/performance status, patient preference and concurrent medications.

Effect modifiers included age and negative sentinel node status. Older patients would be expected to have a poorer tolerance for chemotherapy (50), potentially increasing the effect of NAET exposure in preventing them from being recommended for adjuvant chemotherapy. If patients had N2/N3 disease, this is classified as clinically high risk (51) and clinicians may be less confident in prescribing adjuvant endocrine therapy alone and may be more likely to recommend chemotherapy for these patients.

The MDM provided a valuable and comprehensive data source for change in Ki-67, adjuvant therapy recommended, histological grade, patient age, menopausal status, treating surgeon, disease stage, overall health/performance status and other tumour biomarkers. Duration of NAET exposure was calculated from date of clinic appointment where NAET was prescribed to date of surgery as given in operative histopathology report. This was equal to the normal time the patient would have been waiting for surgery if the study had not been carried out. Treatment adherence was assessed by reviewing clinic notes but patients were not interviewed so there is a possibility that adherence is overestimated. These methods of data collection were consistent across both groups.

A comparison of NAET vs. non-NAET patient demographics and pathology was carried out to determine whether any obvious differences emerged between the groups.

The main source of bias in this study was in the non-random selection of patients who received NAET. This was at the discretion of the treating surgeon, introducing selection bias. All unit surgeons were invited to participate in the qualitative arm of the study, but not all surgeons chose to participate, introducing possible selection bias in the qualitative data. Given that this is not a randomized controlled trial, efforts were taken to limit bias in the data collection and analysis as much as possible. Standardized protocols for data collection used with all MDM agendas during the period of the study being reviewed and all eligible patients being included in order to limit selection bias. A variety of different clinical laboratories analysed the core biopsies of the patients before they were referred to the unit MDM, and it is possible that samples analysed at one lab may have met the cut-off for inclusion of exclusion based on Ki-67 may not have been included or excluded if they had been analysed at a different lab.

Given that this study involves a small sample size, descriptive statistics were used.

Research ethics

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was granted via the non-HREC (Human Research Ethics Committee) Pathway by the research department at the Northern Hospital on 13/12/2024. This involved review of potential negative effects on patients, of which none were identified. Methods to ensure patient privacy were also incorporated into the application. The ethics committee determined that participants in the qualitative section did not need to formally provide consent to participate as implicit consent was given by participating in the interview.

Qualitative methodology

The qualitative component involved gaining feedback from consultant breast surgeons and breast pathologists in semi-structured interviews which were carried out in the second half of the year period of the study. This arm of the study was designed to address aim 3 of our study by interviewing surgeons in order to understand their decisions as to whether to implement NAET or not for eligible patients. To address aim 4, pathologists were interviewed in order to discover their concerns about the limitations of the test itself. A combination of the results of these two studies was intended to address aim 5. Questions focused on their opinions on this intervention, the use of Ki-67 as an inclusion criterion and marker of response and any perceived barriers to its implementation.

These interviews took place both in person and via Microsoft Teams. The following open-ended questions were asked followed by an unstructured discussion:

• What is your opinion on the use of Ki-67 response as an indicator of need for adjuvant chemotherapy?
• What barriers do you see in the implementation of NAET for this group of patients? and
• Do you think that this intervention for this group of patients is worthwhile?

All interviews were conducted by the lead researcher (S.F.) and interviews were transcribed. The transcribed data was reviewed and annotated with codes. These codes were then grouped into categories and each category was analysed and compared with other categories. Common themes emerged when analysing the different categories in relation to each other. A focused literature review was performed based on these themes.

Common themes were identified from the surgeon interview transcripts when a particular coded point emerged in the feedback of more than one surgeon. The two participating pathologists were interviewed together and therefore a common theme was noted when a particular coded point was either mentioned by both pathologists or appeared twice or more within the transcript.

By combining the findings from the quantitative and qualitative arms of the study, several recommendations for future practice were identified.

Results

A total of 44 patients were initially referred to the Breast Unit MDM who were eligible for NAET between May 2023 and May 2024, being post-menopausal, having hormone receptor-positive breast cancer and having a Ki-67 of >10%. Twenty-four of 44 (55%) eligible patients received NAET (Figure 1), all of whom were post-menopausal. All patients received anastrazole as their endocrine therapy. All available Ki-67 values were noted in the MDM discussions, but other disease and patient factors predominantly impacted choice of adjuvant therapy.

Figure 1 Proportion of eligible patients prescribed NAET during the study period. NAET, neoadjuvant endocrine therapy.

NAET group results

Ki-67 testing was performed on 22/24 core biopsies and 24/24 excision specimens in the NAET group. Two patients who did not have Ki-67 values reported on their core biopsy specimens but were post-menopausal and ER-positive were included in this group. Eighteen of 24 (75%) had a reduction in Ki-67 from core biopsy to excision specimen. 2/24 (8%) had an increase and 2/24 (8%) saw no change. Eleven of 24 (46%) of patients had a reduction from greater than 10% to less than 10% (Figure 2). The change in Ki-67 from core biopsy to excision specimen for each patient can be seen in Figure 3. The average change in Ki-67 across the group was a reduction of 18.86% with a sample standard deviation of 22.5%.

Figure 2 NAET group Ki-67 responses from core biopsy to excision specimen. NAET, neoadjuvant endocrine therapy.

Figure 3 Change in Ki-67 from core biopsy to excision specimen for each NAET patient (green arrow indicates reduction, red arrow indicates increase). NAET, neoadjuvant endocrine therapy.

Five of 24 patients (21%) in this group were recommended for adjuvant chemotherapy. Two of those recommended for chemotherapy had had an increase in Ki-67, while one had no change, one had a reduction to a value above 10% and one had a reduction across the 10% threshold. Both patients who had a reduction in Ki-67 had positive sentinel nodes.

Of 9 patients who had positive sentinel nodes, 4 were recommended for adjuvant chemotherapy.

The patients in the NAET group ranged in age from 42 to 84 years. Ki-67 value on core biopsies ranged from 10% to 70%. Ki-67 was recorded on all excision specimens, with values ranging from 0% to 44%. Duration of NAET ranged from 11 to 63 days.

Three of 24 patients in the NAET group were recommended for adjuvant endocrine therapy alone (without radiotherapy). Sixteen of 24 patients were recommended for adjuvant endocrine therapy and radiotherapy. Four of 24 patients were recommended for adjuvant chemoradiotherapy (Figure 4). One patient was recommended for adjuvant endocrine therapy and chemotherapy, but they ultimately left the country for treatment elsewhere. This patient was included in the group who were recommended for adjuvant chemotherapy although she did not actually receive it in this hospital.

Figure 4 Adjuvant therapy recommendations for patients who received NAET. NAET, neoadjuvant endocrine therapy.

All patients who were commenced on NAET were presumed to have continued to take it until their surgery date. For patients who had Ki-67 values for both core biopsy and excision specimen, the duration of treatment showed no correlation to the reduction in Ki-67, with a correlation coefficient of −0.1999 (Figure 5).

Figure 5 Reduction in Ki-67 compared with duration of NAET showing no clear correlation, with a correlation coefficient of −0.1999. NAET, neoadjuvant endocrine therapy.

Non-NAET group results

In the non-NAET group, Ki-67 testing was performed on all core biopsy specimens and 19/20 excision specimens. Seven of 19 (37%) had a reduction in Ki-67 from core biopsy to excision specimen. The average change in Ki-67 in this group was increase of 2.05% with a sample standard deviation of 20.82%.

Eight of 19 (42%) had an increase and 3/19 (16%) had no change. Five of 20 (26%) had a reduction (Figure 6).

Figure 6 Ki-67 response in eligible group who did not receive NAET. NAET, neoadjuvant endocrine therapy.

Eight of 19 (42%) were recommended for adjuvant chemotherapy. Of these 8, 1 had a reduction across the 10% threshold, 2 had a reduction to a value above 10% and 5 had an increase in Ki-67 from core biopsy to excision specimen. Three of the patients recommended for adjuvant chemotherapy had a positive sentinel node and 5 did not. Of the 5 patients who had a negative sentinel node and were recommended for adjuvant chemotherapy, 3 had an increase in Ki-67 and 2 had reductions to values above 10%.

Eligible patients who did not receive NAET ranged in age from 45 to 92 years. Ki-67 value was reported on all core biopsies ranging from 10% to 70%. Ki-67 was reported was reported on 19/20 excision specimens with values ranging from 5% to 70%. Three patients were recommended for adjuvant endocrine therapy alone. Nine patients were recommended for adjuvant endocrine therapy and radiotherapy. Eight patients were recommended for adjuvant chemotherapy and radiotherapy (Figure 7).

Figure 7 Adjuvant therapy recommendations in non-NAET group. NAET, neoadjuvant endocrine therapy.

Cohort comparison

All patients included in the study were pre-operatively stage 1–2. NAET group ages ranged from 43 to 85 years with a mean of 67.71 years. Non-NAET group ages 46 to 93 years with a mean of 66.65 years. No obvious differences were observed between the two cohorts. This was compared with results from the qualitative arm which suggested that the main factor considered by surgeons when deciding who to commence the treatment for was perimenopausal status and their own beliefs about the utility of the treatment.

Rate of NAET prescribing by surgeon

Seven consultant breast surgeons had patients included in this study, and they prescribed NAET at varying rates which are summarized in Table 1 and Figure 8. All consultants prescribed NAET to at least one eligible patient during the period of the study.

Figure 8 NAET prescribing rate by consultant. NAET, neoadjuvant endocrine therapy.

Qualitative results

All 7 consultant breast surgeons were asked for their opinions on the feasibility of NAET for these patients and the value of this intervention, and 3 surgeons gave interviews in person. One surgeon and 2 breast pathologists gave interviews over video-call. Three surgeons declined to give interviews. Surgeons and pathologists annotated interviews were coded and these were put into six different categories (Table 2). A summary of the themes under each category can be seen in Figure 9.

Figure 9 Qualitative code categories and themes. AI, artificial intelligence. approx., approximately.

There was no overlap in key themes between surgeons and pathologist key themes, demonstrating that surgeons are not cognisant of the limitations of Ki-67 interpretation. While some concerns were raised about a perceived lack of benefit to this therapy due to short duration, surgeons overall did not have a strong opposition to their patients receiving this treatment while waiting for their surgery, provided it did not delay the surgery.

Pathologist themes

Pathologists raised very interesting points around the various limitations of Ki-67 interpretation. They felt that a “significant response” as described in the POETIC study (52), with a reduction in Ki-67 from >10% in core biopsy to <10% in excision specimen would be very difficult to confirm in a reproducible way by conventional techniques of Ki-67 testing. The theme of tumour heterogeneity was referred to by the pathologist when expressing concerns about Ki-67 “hotspots” in the tumour, but this issue did not emerge as a theme.

Two consultant pathologists raised concerns about the use of Ki-67 as a marker to plan adjuvant therapy. They noted that the length of time that a specimen takes to fix can affect Ki-67 values, meaning that larger excision specimens may give different Ki-67 values to smaller core biopsy specimens despite having similar values at baseline. Furthermore, the time to fixation in formalin can also affect this value, and breast specimens can often be kept longer in theatre or be sent to radiology before being fixed in formalin, which does not happen with core biopsies. Furthermore, long fixation times can also lead to lower Ki-67 values. This effect has been demonstrated in the literature with concerns raised about resulting variability in Ki-67 interpretation (53,54).

They also noted that different laboratories use different techniques which can cause variance in Ki-67 values. For example, some larger labs use artificial intelligence (AI) to determine Ki-67 by counting 500 to 1,000 tumour cells, whereas other labs often do so by counting approximately 100 tumour cells by eye. A recent study suggests that a majority of pathologists find AI reliable, increasing accuracy, consistency and trust in Ki-67 interpretation (55). Both pathologists explained that they may count cells in one hotspot or across two or more hotspots giving an average, which increases inter-user variability and could affect the final value.

They raised some concerns with the inclusion criteria, noting that at values less than 20%, it is very difficult to be precise about Ki-67. They estimated that any value between 5% and 30% is not precise, suggesting that the choice of above 10% to below 10% as a threshold for reduction may not be ideal. They would be confident that a true reduction had taken place in cases where the core biopsy Ki-67 was reported as greater than 40% and the excision specimen was reported as less than 10%, but anything in between these values could be imprecise.

Finally, they noted the importance of inter-user variability in interpreting these results, and suggested that in order to ensure reproducibility, ideally there would be clear protocols on the timing of fixation of specimens in formalin, the technique of reporting used and the pathologist responsible for carrying it out. It would also be helpful for pathologists to document their technique clearly, for example describing whether they counted cells in one hotspot or took an average across two or more.

Pathologists did not suggest a specific alternative Ki-67 cutoff that would be preferable.

Surgeon themes

One theme from the surgeon interviews was concern that the time frame (from consultation to operation) was not long enough to make a significant difference to long term outcome and recurrence risk. This surgeon felt that very few patients who are newly diagnosed with early ER-positive breast cancer will end up being recommended for adjuvant chemotherapy and therefore the effect of the intervention may be quite small. However, this surgeon had no opposition to their patients receiving the treatment and in fact did prescribe NAET to one of their patients pre-operatively. Another surgeon noted that in some cases they would be reluctant to prescribe aromatase inhibitors to perimenopausal patients.

It was noted that surgeons would be more comfortable prescribing the treatment if they had resources to provide to patients on possible side effects. They would also be in favour of involvement of the unit Cancer Care Nurse.

Surgeons did not mention any concerns about specimen handling or need for staff education to facilitate the adoption of NAET.

Discussion

Key findings

Short-course NAET was implemented for post-menopausal women on an ad hoc basis using Ki-67 as a response marker without a formal treatment pathway, with over half of eligible patients receiving the treatment during the period of the study. Although the sample size was small, Ki-67 reductions were observed in both groups, and this reduction was considered in the multidisciplinary meeting, although other disease and patient factors primarily determined adjuvant therapy decisions.

Surgeons noted barriers to prescribing including need to counsel patients about side effects of aromatase inhibitors during short clinic appointment, concern that the time frame from consultation to surgery may be too short to make have an effect and concern that this group of patients are already unlikely to require adjuvant chemotherapy whether NAET is given or not.

Pathologists noted issues with Ki-67 interpretation that may exacerbate variability in interpretation, affecting eligibility for treatment such as variability in time to fixation of specimens and analysis of specimens in different laboratories by different pathologists using different techniques. A Ki-67 cut off should be decided in conjunction with the pathology department based on the Ki-67 analysis techniques that are used in their laboratory.

The implementation of this treatment in other units would benefit from thorough multidisciplinary planning to allow for standardization of specimen management to pathological analysis improve reproducibility and clinical decision making. Feasibility would also be optimized by development of a pathway to ensure adequate time for counselling and prescribing for patients in the surgical clinic.

Strengths and limitations

While the study investigates feasibility in a breast unit which is quite typical of Australian tertiary hospitals with similar MDM systems and budgets, the small sample size compromises the generalizability of some of the findings, necessitating caution in drawing broad conclusions. Additionally, the absence of feedback from all consultants may limit the comprehensiveness of the study’s insights into the decision-making process. Future survey data on surgeon perspectives on NAET will add useful context to these findings. Comprehensive data regarding patient co-morbidities was not collected and these may have impacted surgeon recommendations for NAET, although all patients in the study were deemed fit for surgery.

The lack of a standardized protocol for Ki-67 interpretation between clinicians and laboratories introduces possible bias in Ki-67 interpretation. Moreover, incomplete data regarding Ki-67 further challenges the robustness of the study’s findings. In this study, patients were assumed to have commenced anastrozole on the day it was prescribed and to have taken it up until the day before their surgery unless clinic notes stated otherwise which may lead to inaccuracy.

These limitations underscore the need for future research with larger and more diverse samples, comprehensive feedback from all involved consultants, development of and adherence to standardized protocols for Ki-67 interpretation, and thorough data collection to further elucidate the impact of NAET on adjuvant therapy recommendations. This study did not require funding support.

Comparison with similar research

Ki-67 as a marker of invasiveness of breast cancer has become more widespread in recent years, but concerns about variability in interpretation have prevented it from being more widely adopted for use as a marker of prognosis and for prediction of therapeutic response (20,21). In 2019, the International Ki-67 in Breast Cancer Working Group (IKWG) updated their previous recommendations from 2011 (19), recommending that “Ki-67 values of 5% or less, or 30% or more, can be used to estimate prognosis, but values within that range should not be used to make clinical decisions” (56). They also recommend measures for handling specimens intended for Ki-67 testing including minimization of prefixation delays, division of surgical specimens to 5- to 10-mm slices for fixation, and fixation in neutral buffered formalin for 6–72 hours as per the American Society of Clinical Oncology and the College of American Pathologists (ASCO and CAP) joint guidelines for handling of specimens intended for HER2 and hormone receptor testing (57,58). The IKWG also recommend that Ki-67 immunohistochemistry (IHC) be performed within 5 years of tumour fixation due to concerns about antigen decay. They have developed a standardized scoring method for Ki-67 which is available on their website (59) which reduces inter-user variability.

The issue of inter-user variability has been raised repeatedly as the use of Ki-67 in clinical practice has become more widespread (22-25). Given that the main concerns around Ki-67 interpretation relate to inter-user variability, some studies have been carried out to investigate the option of using AI and machine learning to standardize reporting of Ki-67, with some models successfully obtaining accurate overexpression rates of HER2, ER, progesterone receptor (PR), and Ki-67, as well as an effective determination of molecular subtypes, with sensitivity up to 97.6% and an average specificity of 96.1% (60).

Another possible option for standardizing Ki-67 testing has been proposed in a recent study from Japan, incorporating a flow cytometric assay that employs tissue dissociation, enzymatic treatment and a gating process to analyse Ki-67 in formalin-fixed paraffin-embedded (FFPE) breast cancer tissue, reducing concerns around inter-laboratory and inter-user variability in results (61).

While this treatment’s safety and efficacy have been demonstrated in the literature, it is still only being used in a minority, with 0.4–1.2% of eligible patients in the real world receiving NAET (42,62). An obstacle to integration of NAET from the academic literature into real-world clinical practice is the lack of studies examining the feasibility of specific protocols, as shown by the fact that this is the primary outcome of a large scale ongoing randomized controlled trial (37), while previous trials have focused on demonstrating efficacy and safety (9,26,35,36,41,48,62). This project addressed this important research gap, demonstrating feasibility of an ad hoc implementation process, mimicking real-world adoption of low-risk treatments from research.

When comparing the different durations of treatment observed in the study, there was no correlation between duration of treatment and reduction in Ki-67. While this is not a statistically significant finding, the lack of positive correlation between this variable and outcome is interesting, and supports the current evidence that NAET may not need to be given for long periods of time to have this effect (62). Many different durations of NAET have been trialled in recent years. The POETIC trial demonstrated response after an average of 2 weeks of NAET with a minimum of 10 days. Taylor et al. demonstrated changes in oncotype DX scores, which incorporate Ki-67 values, after NAET durations of 5.5 and 10 months but does not specifically recommend these as minimum durations for response (8). López-Velazco et al. reported a significant reduction in Ki-67 together with other tumour biomarkers after a similar period of 7 months (7).

There is currently no consensus on an optimal minimal duration of NAET. A meta-analysis of studies investigating the prognostic value of Ki-67 after NAET included twelve studies including 7,897 patients (63), with durations of NAET ranging from 2 weeks (26) to 4–6 months (64). Poor Ki-67 response was correlated with poorer recurrence free and overall survival. A recommendation was not given on optimal duration. Several studies have reported on clinical response following NAET without specifically including Ki-67.

Goldbach et al. found no significant differences in clinical response (based on clinical and pathologic tumor and nodal staging data) following short (<9 weeks), moderate (9–27 weeks) and long (>27 weeks) durations, with objective response rates of 56.7%, 52.1%, and 49%, respectively, but Ki-67 was not used as a marker of response as it was not available in the National Cancer Database (62).

Semiglazov et al. noted a median time to clinical response based on palpation, imaging and qualification for breast conserving surgery and radiotherapy of 57 days in patients receiving endocrine therapy (65).

Data from the Edinburgh Breast Unit demonstrate high response rates with reductions in tumor volume greater than 80% in some cases of postmenopausal women with ER-rich tumors after 3 months of NAET (66). Fontein et al. demonstrated improved clinical response based on palpation, imaging and breast conservation rates following 6 months of NAET when compared to three (67).

Patients who received NAET avoid a recommendation for adjuvant chemotherapy in greater numbers than those who did not within this group, but numbers are not sufficient to draw a statistically significant conclusion. Further research with larger sample sizes is in progress to elucidate these trends and establish more robust conclusions (8). The observed impact of NAET on the overall likelihood of adjuvant chemotherapy suggests that this treatment approach may influence the subsequent therapeutic decisions in breast cancer management, similarly to the way that the oncotype DX test influences the frequency of recommendations for adjuvant chemotherapy in other healthcare systems (68,69).

The reduction in recommendations for adjuvant chemotherapy aligns with the observed changes in Ki-67, emphasizing the potential role of Ki-67 in guiding adjuvant therapy planning. This implies that the assessment of Ki-67 index changes could be a valuable consideration in tailoring adjuvant treatment strategies for patients who have undergone NAET when considered in conjunction with other disease and patient factors.

The role of the consultant surgeon emerged as a crucial factor in determining whether eligible patients received NAET, as the surgeon is the first clinician that the patient meets on referral from the breast screening programme. The study highlighted significant variability among different breast surgeons in the unit regarding the adoption of NAET for eligible patients. The treating consultant breast surgeon emerged as the most influential factor, with different surgeons exhibiting diverse practices, ranging from widespread utilization of NAET to limited application. This underscores the impact of individual awareness, preferences, and practices among healthcare providers in the implementation of NAET.

The qualitative arm of this study gave valuable insights into some of the barriers to implementing this therapy according to the given inclusion criteria. Consultant surgeons raised important points about their perceived obstacles to prescribing NAET and some ways in which they can be overcome, which were used to generate the recommendations that conclude this section. The pathologists raised very important limitations of the use of Ki-67 as a therapeutic planning tool, given the variability that arises from variation in handling and assessment of these specimens. These will be of great use in designing future studies in this area.

Explanations of findings

This study provides an insight into the feasibility of implementing short term NAET for post-menopausal women with hormone receptor-positive while they await definitive surgery. Ki-67 was used as an eligibility criterion, and more than half of eligible patients received the treatment in the first year of implementation. The inclusion of the qualitative component gave valuable insights into the possible limitations of Ki-67 cutoff as a method of selecting patients for this treatment and as a marker of treatment response. Observing the ad hoc implementation of this treatment and gaining feedback from clinicians and pathologists within the first year gave valuable insights into the possible pitfalls of implementing this treatment without thorough multidisciplinary planning to mitigate the issues which increase variability in Ki-67 interpretation.

Comparison of our experience with implementation literature suggests that other units might benefit from more thorough planning of implementation strategy which incorporates programme theory, combining NAET implementation based on Ki-67 with a narrative about unit structures, clinician and patient behaviours, clinic, theatre and pathology laboratory processes and broader contextual features that will be needed to implement NAET for eligible patients.

Implications and actions needed

Recommendations for other units who plan to adopt this treatment

• Comprehensive planning and development of a standardized protocol before initiating the treatment with patients with representatives from pathologists, medical oncologists, surgeons, interventional radiologists, breast care nurses, theatre nurses and administrative staff;
• Standardized handling and time to fixation of core and excision specimens according to the IKWG guidelines;
• Education of theatre staff on optimal specimen handling and the need for appropriate scheduling of cases to allow for early delivery of specimens to anatomical pathology;
• Specimens to be clearly labelled as NAET patients at excision and handled according to recommendations;
• Standardized technique for Ki-67 analysis in laboratory with the maximal possible number of cells counted in each specimen to increase accuracy and a guidance given on whether it is appropriate to count cells at two or more hotspots and give an average as a result;
• Analysis of all specimens at the same laboratory where possible;
• Develop guidelines for the provision of NAET in the perimenopausal period.
• Frequent audit of the above processes to allow for further optimisation and to keep up with future advances.

Conclusions

This study was carried out to investigate the feasibility and outcomes of implementing surgeon initiated, pre-operative NAET for postmenopausal patients with early breast cancer in a Victorian breast unit on an ad hoc basis. The goals were to determine whether NAET was a feasible intervention, observe whether change in Ki-67 after NAET was considered when recommending adjuvant therapy and to assess clinician and pathologists feedback on the implementation process to generate recommendations for other units seeking to implement this treatment. NAET was shown to be a feasible intervention, with barriers including consultant awareness/preference and belief in the utility of the intervention, peri-menopausal status of the patient, side effects of anastrozole and the surgeon’s own confidence in counselling patients regarding these and surgeon awareness of the benefits of this intervention. Concerns about the interpretation of Ki-67 were raised by the pathologists which could be addressed by the development of standardised protocols to limit inter-user variability. While the small sample size prevented results from being statistically significant, descriptive statistics suggest that patients who received NAET may be less likely to be recommended for adjuvant chemotherapy.

This information is of great value to the breast unit at this hospital and provides generalizable insight that can be of use to other similar breast units in Australia and other countries who want to limit the prescribing of chemotherapy which may not improve relapse-free survival. It is also of use to hospitals in health systems with fewer resources who cannot access tools like the oncotype DX test for prognostication but might be able to afford a relatively cheap intervention like anastrozole. Dissemination of these findings has the potential to have a positive and lasting impact on patients with early breast cancer by preventing the administration of unnecessary toxic therapies without reducing their chances of long periods of relapse-free survival. This will be an important resource for our own breast unit and will provide an example to other breast units about ways that they can implement this tool.

This project has successfully established feasibility of NAET implementation in this unit on an ad hoc basis using the patient eligibility parameters suggested by the POETIC study (26). Despite its small sample size, it has also provided valuable data about the effects that implementing NAET can have on the likelihood of patients being recommended for adjuvant chemotherapy following their surgery. Carrying out this project in tandem with a qualitative study demonstrates the possible barriers to implementation and the importance of thorough multidisciplinary planning in the broader implementation of this treatment, both in this breast unit and elsewhere.

While giving insight into the processes that guide patients through the breast cancer treatment process in our unit, it provides information that is generalizable to breast units both in our own healthcare system and abroad, given the relatively wide availability and low cost of endocrine therapy.

Acknowledgments

The authors would like to thank the breast unit and the multidisciplinary team at the Northern Hospital for their support during this study. We are also grateful to the breast unit pathologists for sharing their valuable knowledge and expertise on Ki-67 interpretation, which significantly contributed to the quality of this research.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://tbcr.amegroups.com/article/view/10.21037/tbcr-25-19/rc

Data Sharing Statement: Available at https://tbcr.amegroups.com/article/view/10.21037/tbcr-25-19/dss

Peer Review File: Available at https://tbcr.amegroups.com/article/view/10.21037/tbcr-25-19/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://tbcr.amegroups.com/article/view/10.21037/tbcr-25-19/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. This study was granted via the non-HREC (Human Research Ethics Committee) Pathway by the research department at the Northern Hospital on 13/12/2024. This involved review of potential negative effects on patients, of which none were identified. Methods to ensure patient privacy were also incorporated into the application. The ethics committee determined that participants in the qualitative section did not need to formally provide consent to participate as implicit consent was given by participating in the interview.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Breast Cancer Trials. How Common is Breast Cancer? [Internet]. Research Blog. 2023 [cited 2024 Jul 16]. Available online: https://www.breastcancertrials.org.au/how-common-is-breast-cancer/
2. Early Breast Cancer Trialists' Collaborative Group (EBCTCG). Relevance of breast cancer hormone receptors and other factors to the efficacy of adjuvant tamoxifen: patient-level meta-analysis of randomised trials. Lancet 2011;378:771-84. [Crossref] [PubMed]
3. Burstein HJ, Lacchetti C, Griggs JJ. Adjuvant Endocrine Therapy for Women With Hormone Receptor-Positive Breast Cancer: ASCO Clinical Practice Guideline Focused Update. J Oncol Pract 2019;15:106-7. [Crossref] [PubMed]
4. Rugo HS, Rumble RB, Macrae E, et al. Endocrine Therapy for Hormone Receptor-Positive Metastatic Breast Cancer: American Society of Clinical Oncology Guideline. J Clin Oncol 2016;34:3069-103. [Crossref] [PubMed]
5. Aromatase inhibitors versus tamoxifen in premenopausal women with oestrogen receptor-positive early-stage breast cancer treated with ovarian suppression: a patient-level meta-analysis of 7030 women from four randomised trials. Lancet Oncol 2022;23:382-92. [Crossref] [PubMed]
6. Khorfan R, Vora HP, Namm JP, et al. Expanded Indications for Neoadjuvant Endocrine Therapy in Early-Stage Breast Cancer During the COVID-19 Pandemic. Ann Surg Oncol 2024;31:7562-8. [Crossref] [PubMed]
7. López-Velazco JI, Manzano S, Otaño M, et al. A prospective study on tumour response assessment methods after neoadjuvant endocrine therapy in early oestrogen receptor-positive breast cancer. Breast Cancer Res 2024;26:3. [Crossref] [PubMed]
8. Taylor C, Meisel J, Foreman AJ, et al. Using Oncotype DX breast recurrence score® assay to define the role of neoadjuvant endocrine therapy in early-stage hormone receptor-positive breast cancer. Breast Cancer Res Treat 2023;199:91-8. [Crossref] [PubMed]
9. Iwata H, Yamamoto Y, Sakai T, et al. Phase III study of long-term prognosis of estrogen receptor-positive early breast cancer treated with neoadjuvant endocrine therapy with/without adjuvant chemotherapy. Breast Cancer Res Treat 2023;199:231-41. [Crossref] [PubMed]
10. Martins-Branco D, Nader-Marta G, Molinelli C, et al. Ki-67 index after neoadjuvant endocrine therapy as a prognostic biomarker in patients with ER-positive/HER2-negative early breast cancer: a systematic review and meta-analysis. Eur J Cancer 2023;194:113358. [Crossref] [PubMed]
11. Domagala W, Markiewski M, Harezga B, et al. Prognostic significance of tumor cell proliferation rate as determined by the MIB-1 antibody in breast carcinoma: its relationship with vimentin and p53 protein. Clin Cancer Res 1996;2:147-54.
12. Sahin AA, Ro J, Ro JY, et al. Ki-67 immunostaining in node-negative stage I/II breast carcinoma. Significant correlation with prognosis. Cancer 1991;68:549-57. [Crossref] [PubMed]
13. Viale G, Giobbie-Hurder A, Regan MM, et al. Prognostic and predictive value of centrally reviewed Ki-67 labeling index in postmenopausal women with endocrine-responsive breast cancer: results from Breast International Group Trial 1-98 comparing adjuvant tamoxifen with letrozole. J Clin Oncol 2008;26:5569-75. [Crossref] [PubMed]
14. Clahsen PC, van de Velde CJ, Duval C, et al. The utility of mitotic index, oestrogen receptor and Ki-67 measurements in the creation of novel prognostic indices for node-negative breast cancer. Eur J Surg Oncol 1999;25:356-63. [Crossref] [PubMed]
15. Mandard AM, Denoux Y, Herlin P, et al. Prognostic value of DNA cytometry in 281 premenopausal patients with lymph node negative breast carcinoma randomized in a control trial: multivariate analysis with Ki-67 index, mitotic count, and microvessel density. Cancer 2000;89:1748-57. [Crossref] [PubMed]
16. Cucciniello L, Gerratana L, Del Mastro L, et al. Tailoring adjuvant endocrine therapy in early breast cancer: When, how, and how long? Cancer Treat Rev 2022;110:102445. [Crossref] [PubMed]
17. Garutti M, Griguolo G, Botticelli A, et al. Definition of High-Risk Early Hormone-Positive HER2-Negative Breast Cancer: A Consensus Review. Cancers (Basel) 2022;14:1898. [Crossref] [PubMed]
18. Goldhirsch A, Wood WC, Coates AS, et al. Strategies for subtypes--dealing with the diversity of breast cancer: highlights of the St. Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2011. Ann Oncol 2011;22:1736-47. [Crossref] [PubMed]
19. Aung TN, Acs B, Warrell J, et al. A new tool for technical standardization of the Ki67 immunohistochemical assay. Mod Pathol 2021;34:1261-70. [Crossref] [PubMed]
20. Dowsett M, Nielsen TO, A'Hern R, et al. Assessment of Ki67 in breast cancer: recommendations from the International Ki67 in Breast Cancer working group. J Natl Cancer Inst 2011;103:1656-64. [Crossref] [PubMed]
21. Healey MA, Hirko KA, Beck AH, et al. Assessment of Ki67 expression for breast cancer subtype classification and prognosis in the Nurses' Health Study. Breast Cancer Res Treat 2017;166:613-22. [Crossref] [PubMed]
22. Focke CM, Bürger H, van Diest PJ, et al. Interlaboratory variability of Ki67 staining in breast cancer. Eur J Cancer 2017;84:219-27. [Crossref] [PubMed]
23. Shi SR, Liu C, Taylor CR. Standardization of immunohistochemistry for formalin-fixed, paraffin-embedded tissue sections based on the antigen-retrieval technique: from experiments to hypothesis. J Histochem Cytochem 2007;55:105-9. [Crossref] [PubMed]
24. Gong P, Wang Y, Liu G, et al. New insight into Ki67 expression at the invasive front in breast cancer. PLoS One 2013;8:e54912. [Crossref] [PubMed]
25. Lashen A, Toss MS, Green AR, et al. Ki67 assessment in invasive luminal breast cancer: a comparative study between different scoring methods. Histopathology 2022;81:786-98. [Crossref] [PubMed]
26. Smith I, Robertson J, Kilburn L, et al. Long-term outcome and prognostic value of Ki67 after perioperative endocrine therapy in postmenopausal women with hormone-sensitive early breast cancer (POETIC): an open-label, multicentre, parallel-group, randomised, phase 3 trial. Lancet Oncol 2020;21:1443-54. [Crossref] [PubMed]
27. Geradts J, Goodson WH III, Tang CG et al. Molecular prediction of recurrence of breast cancer...Paik S, Shak S, Tang G et al. A multigene assay to predict recurrence of tamoxifen-treated, node-negative breast cancer. N Engl J Med 2004;351:2817-26. N Engl J Med 2005;352:1605-7. Available online: https://research.ebsco.com/linkprocessor/plink?id=9345ccdb-1c77-32fe-991f-5f61c280cfef
28. Garufi G, Carbognin L, Arcanà C, et al. Tailoring neoadjuvant treatment of HR-positive/HER2-negative breast cancers: Which role for gene expression assays? Cancer Treat Rev 2022;110:102454. [Crossref] [PubMed]
29. Sparano J, Gray RJ, Makower DF, et al. Trial Assigning Individualized Options for Treatment (TAILORx): An Update Including 12-Year Event Rates. Cancer Res Cancer Res 2023;83:GS1-05.
30. Choi JDW, Hughes TMD, Marx G, et al. The Utility of the Oncotype DX Test for Breast Cancer Patients in an Australian Multidisciplinary Setting. Breast J 2022;2022:1199245. [Crossref] [PubMed]
31. Kelly CM and Members of the Irish Society of Medical Oncology. Oncotype-DX® gene expression profile and chemotherapy decision-making in patients with early stage breast cancer [Internet]. Irish Society of Medical Oncology; 2011 Apr [cited 2023 Dec 12]. (Evidence-based Series). Report No.: Evidence-based Series #1. Available online: https://www.hse.ie/eng/services/list/5/cancer/profinfo/chemoprotocols/breast/oncdx.pdf
32. The Pharmaceutical Benefits Scheme. Anastrazole [Internet]. Available online: https://www.pbs.gov.au/medicine/item/8179L
33. Ellis MJ, Suman VJ, Hoog J, et al. Ki67 Proliferation Index as a Tool for Chemotherapy Decisions During and After Neoadjuvant Aromatase Inhibitor Treatment of Breast Cancer: Results From the American College of Surgeons Oncology Group Z1031 Trial (Alliance). J Clin Oncol 2017;35:1061-9. [Crossref] [PubMed]
34. Medicare. MBS Online Medicare Benefits Schedule [Internet]. Australian Government Department of Health and Aged Care; [cited 2025 May 4]. Available online: https://www9.health.gov.au/mbs/search.cfm?rpp=10&q=&qt=&cat1=147&cat2=154&cat3=&adv=&&start=1
35. Cutress R. Randomised Controlled Trial Evaluating Effectiveness of Neoadjuvant Endocrine treatment in Post-Menopausal Women [Internet]. United Kingdom; 2021 Jun [cited 2023 Aug 20]. Report No.: EudraCT/CTIS number 2022-000582-40. Available online: https://www.isrctn.com/ISRCTN11896599?q=endonet&filters=&sort=&offset=1&totalResults=1&page=1&pageSize=10
36. Ma CX, Suman VJ, Leitch AM, et al. ALTERNATE: neoadjuvant endocrine treatment (NET) approaches for clinical stage II or III estrogen receptor-positive HER2-negative breast cancer (ER+ HER2-BC) in postmenopausal (PM) women: alliance A011106. J Clin Oncol 2020;38:504.
37. Chien J, Esserman L, Elias A, et al. (Trial in Progress) A phase 2, open-label, randomized multicenter trial to evaluate neoadjuvant lasofoxifene in molecularly-selected HR+/HER2− Clinical Stage 2/3 breast cancer. Breast 2023;68:S66.
38. Weiss A, Mittendorf EA, King TA. Strategies to Optimize Axillary Surgery in Patients With Breast Cancer Receiving Neoadjuvant Endocrine Therapy. Oncology (Williston Park) 2020;34:397-404. [Crossref] [PubMed]
39. Bliss JM, Tovey H, Evans A, et al. Clinico-pathologic relationships with Ki67 and its change with short-term aromatase inhibitor treatment in primary ER + breast cancer: further results from the POETIC trial (CRUK/07/015). Breast Cancer Res 2023;25:39. [Crossref] [PubMed]
40. Sirico M, Virga A, Conte B, et al. Neoadjuvant endocrine therapy for luminal breast tumors: State of the art, challenges and future perspectives. Crit Rev Oncol Hematol 2023;181:103900. [Crossref] [PubMed]
41. Wang X, Fan Z, Wang X, et al. Neoadjuvant endocrine therapy for strongly hormone receptor-positive and HER2-negative early breast cancer: results of a prospective multi-center study. Breast Cancer Res Treat 2022;195:301-10. [Crossref] [PubMed]
42. Mukhtar RA, Hoskin TL, Habermann EB, et al. Changes in Management Strategy and Impact of Neoadjuvant Therapy on Extent of Surgery in Invasive Lobular Carcinoma of the Breast: Analysis of the National Cancer Database (NCDB). Ann Surg Oncol 2021;28:5867-77. [Crossref] [PubMed]
43. Taylor N, Healey E, Morrow A, et al. Aligning intuition and theory: enhancing the replicability of behaviour change interventions in cancer genetics. Implement Sci Commun 2020;1:90. [Crossref] [PubMed]
44. Davidoff F, Dixon-Woods M, Leviton L, et al. Demystifying theory and its use in improvement. BMJ Qual Saf 2015;24:228-38. [Crossref] [PubMed]
45. Jayasinghe K, Biros E, Harris T, et al. Implementation and Evaluation of a National Multidisciplinary Kidney Genetics Clinic Network Over 10 Years. Kidney Int Rep 2024;9:2372-85. [Crossref] [PubMed]
46. Jayasinghe K, Quinlan C, Mallett AJ, et al. Attitudes and Practices of Australian Nephrologists Toward Implementation of Clinical Genomics. Kidney Int Rep 2021;6:272-83. [Crossref] [PubMed]
47. Cheng L, Kugenthiran N, Quinlan C, et al. Optimising the mainstreaming of renal genomics: Complementing empirical and theoretical strategies for implementation. Eur J Hum Genet 2025;33:351-9. [Crossref] [PubMed]
48. Zhang J, Yu Y, Lin Y, et al. Efficacy and safety of neoadjuvant therapy for HER2-positive early breast cancer: a network meta-analysis. Ther Adv Med Oncol 2021;13:17588359211006948. [Crossref] [PubMed]
49. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007;370:1453-7. [Crossref] [PubMed]
50. Chen H, Cantor A, Meyer J, et al. Can older cancer patients tolerate chemotherapy? Cancer 2003;97:1107-14. [Crossref] [PubMed]
51. Pan H, Gray R, Braybrooke J, et al. 20-Year Risks of Breast-Cancer Recurrence after Stopping Endocrine Therapy at 5 Years. N Engl J Med 2017;377:1836-46. [Crossref] [PubMed]
52. Smith I, Robertson J, Kilburn L, et al. Long-term outcome and prognostic value of Ki67 after perioperative endocrine therapy in postmenopausal women with hormone-sensitive early breast cancer (POETIC): an open-label, multicentre, parallel-group, randomised, phase 3 trial. Lancet Oncol 2020;21:1443-54. [Crossref] [PubMed]
53. Arima N, Nishimura R, Osako T, et al. The importance of tissue handling of surgically removed breast cancer for an accurate assessment of the Ki-67 index. J Clin Pathol 2016;69:255-9. [Crossref] [PubMed]
54. Nishimura R, Osako T, Okumura Y, et al. Ki-67 as a prognostic marker according to breast cancer subtype and a predictor of recurrence time in primary breast cancer. Exp Ther Med 2010;1:747-54. [Crossref] [PubMed]
55. Dy A, Nguyen NJ, Meyer J, et al. AI improves accuracy, agreement and efficiency of pathologists for Ki67 assessments in breast cancer. Sci Rep 2024;14:1283. [Crossref] [PubMed]
56. Nielsen TO, Leung SCY, Rimm DL, et al. Assessment of Ki67 in Breast Cancer: Updated Recommendations From the International Ki67 in Breast Cancer Working Group. J Natl Cancer Inst 2021;113:808-19. [Crossref] [PubMed]
57. Hammond ME, Hayes DF, Dowsett M, et al. American Society of Clinical Oncology/College of American Pathologists guideline recommendations for immunohistochemical testing of estrogen and progesterone receptors in breast cancer. Arch Pathol Lab Med 2010;134:907-22. [Crossref] [PubMed]
58. Wolff AC, Hammond MEH, Allison KH, 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. J Clin Oncol 2018;36:2105-22. [Crossref] [PubMed]
59. Dowsett M, Hayes D, McShane L et al. Train Yourself: International Ki67 Breast Cancer Working Group [Internet]. IKWG; 2022 [cited 2025 Jul 3]. Available online: https://www.ki67inbreastcancerwg.org
60. Fan L, Liu J, Ju B, et al. A deep learning based holistic diagnosis system for immunohistochemistry interpretation and molecular subtyping. Neoplasia 2024;50:100976. [Crossref] [PubMed]
61. Sato N, Tsujimoto M, Nakatsuji M, et al. Flow cytometric analysis for Ki67 assessment in formalin-fixed paraffin-embedded breast cancer tissue. BMC Biol 2024;22:181. [Crossref] [PubMed]
62. Goldbach MM, Burkbauer L, Bharani T, et al. Effectiveness of a Short Duration of Neoadjuvant Endocrine Therapy in Patients with HR+ Breast Cancer-An NCDB Analysis (2004-2016). Ann Surg Oncol 2021;28:8651-62. [Crossref] [PubMed]
63. Martins-Branco D, Nader-Marta G, Molinelli C, et al. Ki-67 index after neoadjuvant endocrine therapy as a prognostic biomarker in patients with ER-positive/HER2-negative early breast cancer: a systematic review and meta-analysis. Eur J Cancer 2023;194:113358. [Crossref] [PubMed]
64. Lerebours F, Pulido M, Fourme E, et al. 5-years relapse-free survival and predictive factors identification in HRpositive HER2-negative breast cancer patients treated with anastrozole or fulvestrant neoadjuvant endocrine therapy: pooled analysis of the CARMINA02 and HORGEN phase II trials. European Journal of Cancer 2018;92:S19.
65. Semiglazov VF, Semiglazov VV, Dashyan GA, et al. Phase 2 randomized trial of primary endocrine therapy versus chemotherapy in postmenopausal patients with estrogen receptor-positive breast cancer. Cancer 2007;110:244-54. [Crossref] [PubMed]
66. Dixon JM. Role of endocrine therapy in the neoadjuvant surgical setting. Ann Surg Oncol 2004;11:18S-23S. [Crossref] [PubMed]
67. Fontein DB, Charehbili A, Nortier JW, et al. Efficacy of six month neoadjuvant endocrine therapy in postmenopausal, hormone receptor-positive breast cancer patients--a phase II trial. Eur J Cancer 2014;50:2190-200. [Crossref] [PubMed]
68. Walsh, N. 60% of women didn’t have chemotherapy thanks to groundbreaking new multi-gene test. Extra.ie [Internet]. 2023 Dec 10; Available online: https://extra.ie/2023/12/10/news/chemotherapy-multigene
69. Kalinsky K, Barlow WE, Gralow JR, et al. 21-Gene Assay to Inform Chemotherapy Benefit in Node-Positive Breast Cancer. N Engl J Med 2021;385:2336-47. [Crossref] [PubMed]