Abstract
Background:
Breast cancer is the most prevalent cancer and second leading cause of cancer-related mortality among Canadian women. Most of cases belong to the HR+/HER2− subtype, representing approximately two-thirds of all instances.
Objectives:
This real-world evidence study aims to comprehensively analyze the treatment pattern, and clinical outcomes of Canadian patients diagnosed with early-stage HR+/HER2− breast cancer.
Design:
This retrospective, longitudinal cohort study involved 541 patients enrolled in the pan-Canadian cancer patient registry PMT (Personalize My Treatment).
Methods:
The cohort included patients with newly diagnosed or recurrent stage II or III HR+/HER2− breast cancer between January 1st, 1992, and May 31st, 2022. Summary statistic to describe treatment pattern and Kaplan Meir analysis for clinical outcome were used.
Results:
In the adjuvant setting, our study found that ET was administered to 75.6% of the cohort, with a significant preference for combining ET with cytotoxic agents and, particularly in stage III patients. In addition, neoadjuvant therapy, primarily using cytotoxic agents, was higher in stage III patients, and those receiving neoadjuvant therapy were more likely to either continue with ET as adjuvant treatment. The median duration of adjuvant ET was 4.5 years. In the adjuvant patient population, recurrence rates progressively increased over time from 13.2% after 2 years, 21.4% after 3 years, 30.3% after 5 years, and peaking at 58.4% after 10 years. Median time to recurrence for the patient population on ET was 7.76 years. OS rate for patients on ET was 94.6% at 5 years and 78.3% at 10 years.
Conclusions:
This study highlights the high unmet need in stage II and stage III breast cancer, with 1 patient out of 3 recurring after 5 years, and more than half recurring after 10 years despite adjuvant treatment with ET alone. This highlights the need for more effective and tolerable treatment options to address disease recurrence in both the short and long term for eBC HR+/HER2− patients in Canada.
Keywords
Introduction
In the landscape of oncological diseases afflicting women, breast cancer emerges as the foremost adversary, occupying the position of the most diagnosed cancer among females and the second highest contributor to female mortality rates. 1 In Canada, women face a lifetime risk of developing breast cancer of 12.8% (one in 8) and a lifetime probability of dying from it of 2.8% (one in 36). 2 In 2024, it is projected that 30,500 Canadian women will be diagnosed with breast cancer, with an estimated 5,500 succumbing to the disease. 3 Amongst its early-stage manifestations, it constitutes a considerable fraction of initial diagnoses. 4 The therapeutic crusade against early-stage breast cancer is rooted in a curative intent and is strategically multifaceted, involving surgical, radiological, and medical oncology disciplines. Delving into the molecular stratification, breast cancer is classified based on the presence of hormone receptors (HR) and the human epidermal growth factor receptor 2 (HER2), discerned through immunohistochemical methodologies. Patients are categorized into 4 subtypes: Luminal A, Luminal B, HER2-enriched, and basal-like, each with its unique prognosis and therapeutic roadmap. 5 Commonly, breast cancer is categorized into 3 primary subtypes: HER2-positive (regardless of hormone receptor status), HR+/HER2− (hormone receptor-positive/HER2-negative), and triple-negative breast cancer.5 -8 Each subtype carries its own unique prognosis and set of treatment approaches. 9
The predominance of HR+/HER2− breast cancer is notable, representing approximately two-thirds of all cases. 10 The care strategy for this subtype is elaborate, including surgical excision of the malignancy and nodal assessment.11,12 Neoadjuvant chemotherapy could facilitate breast-conserving treatment and minimize the surgery impact. Neoadjuvant chemotherapy is particularly advised for patients presenting with sizable tumors, extensive involvement of axillary lymph nodes, and for those with advanced stages or aggressive cancer subtypes, notably triple-negative breast cancers.11,12 Following surgical excision, adjuvant endocrine intervention targeting HR to mitigate recurrence risks is recommended as standard of care over the course of 5-10 years, with radiation and chemotherapy reserved for high-risk profiles.13,14 The therapy selection and its temporal span are tailored, considering recurrence risk, patient predilections, and other determinants such as tolerance and adherence to treatment. 15 In addition, cyclin-dependent kinase 4/6 inhibitors (CDK4/6i) have been integrated into treatment protocols for HR+/HER2− breast cancer. Abemaciclib is approved worldwide for the adjuvant treatment of node-positive patients at high risk of recurrence,16 -18 while ribociclib has received approval in the United States and in Europe for use in combination with an aromatase inhibitor for the adjuvant treatment of early-stage patients at high risk of recurrence. In Canada, ribociclib is approved for advanced or metastatic HR+/HER2− breast cancer, with its use for early-stage disease currently under review. 19
Despite this comprehensive treatment approach, disease recurrence 20 occurs in approximately 13% to 41% of patients,21 -24 depending on factors nearly 30% of patients diagnosed will experience a recurrence, 20 leading many of them to late-stage disease. Among the clinical features associated with higher risk such as large primary tumor, high tumor grade, advanced stage, extent of axillary lymph node involvement, Ki67 index and other multi gene assays-based proliferative scores, these factors collectively contribute to prognostic and therapeutic decision-making.25 -28 Defining high recurrence risk in early breast cancer is inherently complex due to interpatient heterogeneity and the variable accessibility of prognostic tools across clinical settings. While some guidelines incorporate specific risk thresholds to inform therapeutic decisions, their implementation is not universal, often constrained by differences in resource availability and access to these tools. In addition, real-world practice may diverge from guideline recommendations due to variability in healthcare systems, clinician preferences, and patient factors. It is also important to recognize that reliance on specific thresholds or prognostic methods may occasionally result in false-negative classifications, with the potential to under-identify patients who could benefit from more intensive treatment.25 -29
This study catalogs the clinical outcomes of Canadian patients battling early-stage breast cancer scrutinizing the recurrence incidence post-endocrine therapy initiation. It is anticipated that this endeavor will enrich the knowledge surrounding the management of early breast cancer, fostering refined treatment strategies and elevating patient prognoses in the Canadian healthcare environment.
Methods
Study design
This study employs a retrospective, longitudinal cohort design, utilizing real-world secondary data from the pan-Canadian Personalize My Treatment (PMT) cancer registry, 30 managed by Exactis Innovation. The cohort study was chosen to examine changes in treatment patterns over time and longitudinal outcomes, such as overall survival (OS) and recurrence rate. This longitudinal oncology registry is a Research Ethic Board (REB)-approved initiative in which cancer patients consent to provide access to their medical records for data collection and are prospectively followed throughout their cancer trajectory (NCT02355171). The PMT registry currently compiles anonymized clinical and molecular data from over 2644 breast cancer Canadian participants, with the number continuously increasing. This dataset includes a comprehensive range of information, primarily encompassing patient and tumor characteristics, outcomes, and treatment patterns.
The study includes all HR+/HER2− early breast cancer (eBC) patients present in the PMT registry who were diagnosed with stage II or III disease between January 1, 1992, and May 31, 2022, followed up for at least 1 year. The index date of this study was the date of first diagnosis for HR+/HER2− early breast cancer and patients were followed from the index date until the end of the study’s designated follow-up period (May 31, 2023), until death, or until they were lost to follow-up in the PMT registry (Supplementary Figure 1). If patients were lost to follow-up, they were censored at the time of their last available data, which could occur if they changed residences or transferred to treatment centers outside of the PMT network. The PMT registry data is updated annually; however, for this research project, the cohort’s data was updated to the end of the follow-up period by dedicated PMT coordinators at the respective centers. The rationale of capturing 30 years of data is to acquire a larger sample size to maximize the chance to understand the evolution of Canadian therapeutic landscape on risk of recurrence over 20 years. This is also aligned with a similar analysis from a US database analysis from O’Shaughnessy et al. 24
Study population and data collection
The cohort consisted of all patients diagnosed with HR+/HER2− early breast cancer in the PMT registry between January 1, 1992, and May 31, 2022, who met the following eligibility criteria. Patients were over the age of 18 at the time of their early breast cancer diagnosis. They were diagnosed with stage II or stage III breast cancer within the specified period and had received adjuvant and/or neoadjuvant therapy. Patient stage is recorded in the PMT database based on the stage explicitly written in the patient medical chart. Patients needed to be HR+/HER2− as defined based on the pathology report, where HER2− was indicated by IHC 0 or 1 + or IHC2+/ISH negative, and HR+ was indicated by ER-positive or PR-positive status. Patients with missing or unavailable HER2 testing results at primary diagnosis were excluded from the study. Only patients with confirmed HER2-negative status were included. In cases where multiple tests for hormone receptor expression were performed, the tests conducted closest to the diagnosis date were considered. Patients with concurrent primary cancer at the time of the initial early breast cancer diagnosis were excluded from the study.
In May 2023, out of 2,644 breast cancer patients included in the PMT registry, 551 met the specific study selection criteria outlined above. During the data monitoring phase, an additional 10 patients were found not to fulfill the eligibility criteria and were subsequently excluded. As a result, the final analysis was conducted on a carefully curated cohort of 541 patients (Supplementary Figure 2).
Patient demographics, tumor characteristics, treatment regimens, and outcomes were gathered retrospectively from a combination of electronic health records, article files, and inpatient hospitalization documents. All the variables were collected as explicitly recorded in the patient’s medical file. If the variable was missing, it was collected as unknown. This collected information was systematically coded and securely stored in our cloud-based PMT database. At the end of the follow-up period, all data entries underwent a rigorous centralized monitoring process. Inconsistent data were flagged and subjected to data queries, which were then addressed and verified by the PMT coordinators for accuracy and confirmation.
Statistical analysis
Data imputation was applied to incomplete dates for the following variables: birth, death, diagnosis, disease progression, surgery, radiation, and systemic therapy. The imputation rules are detailed in the supplementary information file. In addition, definitions of the variables, along with the classifications of therapy groups and subgroups, are also provided in the supplementary information file.
This study is primarily descriptive in nature, and therefore no formal hypothesis testing is specified. Continuous variables were described using the number of observations (N) and the median when applicable. Categorical variables were described by the number of observations (N) and the frequency (%) of each category. No statistical tests were performed on the descriptive analysis on treatments received in this cohort.
The recurrence rate at 2, 3, 5, and 10 years following the initiation of ET was determined using the Kaplan Meir method. The date of recurrence was defined as the first date of progression following a report of no evidence of disease or as any date of progression after the first surgery at the initial stage of diagnosis. The results were presented as rates in percentages with corresponding 95% confidence intervals (CIs). Only the patients who received ET as adjuvant therapy were included in the recurrence analysis.
For treatment patterns, clinical, and demographic characteristics of HR+/HER2− eBC patients in the registry, descriptive statistics were used. Categorical variables were presented as the total number (n) and proportion (%), where applicable. Numerical variables were described using medians with their respective ranges and/or means with standard deviations. Time on ET treatment, time to recurrence, recurrence rate, and OS from diagnosis were analyzed using unadjusted Kaplan-Meier (K-M), with medians and 95% CIs presented alongside K-M curves or cumulative incidence curves or as the event rate (percent) at 2, 3, 5, and 10 years with 95% CI.
For all time to event analysis, lost of follow up was handle as follow: patient were censored at the date of last recorded activity in their medical chart.
To protect patient confidentiality when using secondary health data, we chose to censor the data for counts below 5 in all our results.
The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Statement: guidelines for reporting observational studies was followed, and the completed checklist is provided in supplementary Table 5. 31
Results
Demographics and clinicopathological characteristics of the cohort
Our study included patients undergoing adjuvant or neoadjuvant therapy for newly diagnosed or recurrent stage II or III HR+ HER2− breast cancer between January 1, 1992, and May 31, 2022; there were a total of 541 patients in our PMT registry at the time the study was initiated and the median follow up time of the cohort was 8.1 years. The Table 1 shows a significant concentration of patients from Quebec, accounting for 84.8% of the total 541 patients in the study, with major contributions from the following sites: CHUQ, CHUM, and JGH, while other regions like Ontario, New Brunswick, and Alberta have smaller representations.
Study population distribution by sites and provinces.
TBCC—Tom Baker Cancer Center, TOH—The Ottawa Hospital; JGH—Jewish General Hospital; CHUM—Center hospitalier de l’Université de Montréal; CIUSSS MCQ -Center intégré universitaire de santé et de services sociaux de la Mauricie-et-du-Center-du-Québec; CHUQ—Center hospitalier de Québec; TMH—The Moncton Hospital; DGLDUHC—Dr. Georges-L.-Dumont University Hospital Center; CHUS—Center hospitalier universitaire de Sherbrooke, SHSC—Sunnybrook Health Sciences Center.
This study’s cohort consisted of 307 stage II and 234 stage III breast cancer patients, with a median age at diagnosis of 53 years (interquartile range [IQR] = 44, 61) (Table 2). Tumor morphology analysis revealed that infiltrating duct carcinoma was the predominant type, representing 68% of the overall cases. Lobular carcinoma was observed in 15% of patients. High-grade disease was present in 31% of patients overall, with intermediate grade tumors being prevalent in 50% of patients. Low-grade tumors were less common, seen in 13% of patients, with a higher proportion in stage II (15%) than in stage III (9.4%).
Patient demographic and clinical characteristics.
Median (IQR); n (%).
Nodal status was positive in 79% of patients overall, with a higher percentage in stage III (93%) compared with stage II (68%). This high rate of positive nodal status (N+) compared with negative nodal status (N0) prevented further stratification into N0 and N+ groups in subsequent treatment pattern and outcome analyses. The Ki67 proliferation index, a marker of cell proliferation, was categorized into 2 groups: <20% and >=20%. The majority of patients (425, 79%) were not tested for Ki67 status. In the tested population, 42 patients (7.8%) had a Ki67 index of <20%, while 74 patients (14%) had a Ki67 index of >= 20%. BRCA1 and BRCA2 testing were performed on 30% across the overall population. The presence of BRCA mutations was identified in 35 patients (6.5% of all patients).
Treatment landscape
The treatment patterns received by the patients of the cohort were comprehensively analyzed, with the results summarized in Table 3. This table highlights the distribution of neoadjuvant, surgical, and first line adjuvant therapies among patients categorized by disease stage (stage II and stage III). Supplementary Tables 1 to 4 provide further granularity on the specific modalities and drugs utilized within each therapy group. The neoadjuvant treatments observed in the study were categorized into 3 distinct groups for analysis: “cytotoxic,” “other,” and “no” neoadjuvant therapy. The “cytotoxic” group included patients who received cytotoxic therapies either as monotherapy or in combination with other cytotoxic therapies. The “other” group encompassed all other combinations and types of neoadjuvant treatments, including cytotoxic combined with endocrine therapy, radiation, clinical trials, VEGFi, anti-HER2 agents, CDK4/6 inhibitors, and various combinations thereof. Patients who did not receive any neoadjuvant therapy were categorized in the “no” group. This classification allows for a streamlined comparative analysis of the various treatment strategies within the patient cohort. Most patients did not receive neoadjuvant therapy, with 63.2% of stage II patients and 51.7% of stage III patients forgoing this option. Among patients who received neoadjuvant therapy, cytotoxic agents were the most frequently used, with 23.5% of stage II and 35.5% of stage III patients receiving these treatments. Of the 155 patients who received cytotoxic agents, cyclophosphamide was the most commonly administered, used by 148 patients (95.5%), followed by paclitaxel and doxorubicin, which were used by 91 and 90 patients (58.7% and 58.1%) respectively, as detailed in Supplementary Table 4. Surgical intervention was nearly universal, with 97.7% of stage II and 91.0% of stage III patients undergoing primary tumor and/or local lymph node surgery. This underscores surgery as a cornerstone of treatment for early-stage HR+/HER2− breast cancer across both stages.
Summary of treatment patterns received by HR+/HER2− early breast cancer patient population.
Include cytotoxic therapy combined with endocrine therapy, radiation, clinical trials, targeted therapies, and various combinations of these treatments.
Other therapies include CDK4/6, radiation, anti-HER2 therapy, and various combination of these treatments.
Other therapies include CDK4/6, radiation, anti-HER2, VEGFi therapies, and various combination of these treatments.
In the adjuvant setting, the first line treatments observed were categorized as follows: “cytotoxic + ET ± other therapies,” “ET ± other therapies,” “clinical trial ± any other therapy,” “cytotoxic ± other therapies,” “radiation,” and “no” adjuvant therapy. The “cytotoxic + ET ± other therapies” group included any treatment involving cytotoxic therapy combined with endocrine therapy but excluding those involving clinical trials (Supplementary Tables 3 and 4). The “ET ± other therapies” group comprised treatments involving endocrine therapy, excluding those combined with cytotoxic agents or clinical trials. The “clinical trial ± any other therapy” group encompassed all treatments that included participation in clinical trials, regardless of other therapies involved. The “cytotoxic ± other therapies” group included any treatment that involved cytotoxic therapy, excluding those combined with ET or clinical trials. The “radiation” group was designated for treatments that involved radiation therapy alone. Finally, the “no” group consisted of patients who did not receive any adjuvant therapy.
In the adjuvant setting, a considerable proportion of patients received cytotoxic therapy combined with ET associated or not with other therapies (36.2% of stage II and 43.2% of stage III patients). Among these, the majority (174 out of 212, 82.1%) received a combination of cytotoxic, endocrine therapy and radiation (Supplementary Table 5). Cyclophosphamide remained a prominent cytotoxic agent in this adjuvant therapy group, administered to 199 out of 212 patients (93.9%) (Supplementary Table 4).
Endocrine therapy, either alone or combined with other non-cytotoxic treatments, was employed in 34.5% of stage II and 26.1% of stage III patients. Notably, the majority of these patients (123 out of 167, 73.6%) received endocrine therapy in conjunction with radiation (Supplementary Table 3). Tamoxifen and anastrozole were the primary endocrine agents used in this group, administered to 81 and 66 patients, respectively, out of the 167 (Supplementary Table 4). A minority of patients did not receive any adjuvant therapy, with 14.7% of stage II and 16.2% of stage III patients falling into this category. In addition, participation in clinical trials was documented in 4.9% of stage II and 8.1% of stage III patients, indicating a major level of engagement with investigational treatments. Radiation therapy as a standalone adjuvant treatment was utilized in a small percentage of patients (3.9% of stage II and 2.6% of stage III). This treatment pattern analysis highlights that cytotoxic agents combined with endocrine therapy was the primary therapeutic option for stage III patients, while stage II patients receive endocrine therapy combined or not with cytotoxic agents in approximately equal proportions. The adjuvant treatment profile for stage III patients aligns with more aggressive treatment, combining cytotoxic agents with endocrine therapy for those at higher risk. The rationale for the 50/50 distribution of adjuvant treatments in stage II patients remains unclear, as other high-risk factors—such as lymph node involvement, Ki-67 staining, tumor size, and gene panel results—were not adequate or available in this cohort.
We also analyzed the treatment sequence patterns among the patients. Among the 315 patients (58.2%) who did not receive neoadjuvant therapy, nearly all proceed to surgery (312 patients, 99%), with the majority (198, 62.9%), subsequently receiving endocrine therapy combined with cytotoxic agents as adjuvant therapy (Table 4). Most of those who received neoadjuvant therapy, particularly cytotoxic types, also proceeded to surgery (96.1%). However, their subsequent adjuvant therapy patterns showed a higher rate of endocrine therapy without cytotoxic agents (54.4% vs 22.1%), and higher rates of patients without adjuvant therapy (21.5 versus ⩽ 1%). Lower rates of cytotoxic agents combined with ET (6% vs 63.5%) were observed. Patients receiving other forms of neoadjuvant therapy were less likely to undergo surgery (73.2%, n = 52) and exhibited a more diverse distribution of adjuvant treatments, with 38.5% (n = 20) receiving no adjuvant therapy, 32.7% (n = 17) receiving endocrine therapy without cytotoxic agents, 11.5% (n = 6) receiving radiation alone, and 9.6% (n = 5) receiving ET combined with cytotoxic therapy (Table 4 and Figure 1).
Distribution of patients by treatment categories and subsequent therapies.
The proportion is calculated from the parent group based on this sequence of event: neoadjuvant, surgery and adjuvant therapy.
Sankey diagram for treatment sequencing in HR+/HER2− early BC patients.
Adjuvant endocrine therapy treatment exposure and discontinuation rate
We decided to investigate further the modality of endocrine therapy in the adjuvant setting, regardless of other treatments, including examining the overall exposure, treatment duration, and discontinuation rate.
Our results showed a similar overall proportion of patients in both stages received endocrine therapy in the adjuvant setting, with 75.2% in stage II and 76.1% in stage III, leading to an overall rate of 75.6% across all patients (Table 5). The median duration of adjuvant ET treatment was also consistent between the groups, with stage II patients having a median of 4.5 years (IQR = 3.8-5.0 years) and stage III patients having a median of 4.6 years (IQR = 3.8-5.2 years). Overall, the median duration for the entire cohort was 4.5 years (IQR = 4.0-5.0 years). These findings highlight that both stage II and stage III breast cancer patients have a similar likelihood of receiving and continuing adjuvant ET, with a comparable median duration of therapy.
Endocrine treatment exposure in patient cohort.
Median (95% CI), n (%).
We also explored the distribution of patients in our cohort by duration of endocrine therapy and found that nearly half of the patients (48.3%) discontinued adjuvant ET within 3 years, while (21.3%) continued therapy between 3 and 5 years. Only 19.6% of patients remained on therapy between 5 and 7 years, and a small percentage continued beyond 7 years, with just 3.1% persisting over 10 years (Table 6). These findings highlight the challenge of maintaining long-term adherence to adjuvant ET and the evolution of treatment paradigms over time with recommendation to extend adjuvant therapy duration.
Distribution of patients by duration of adjuvant endocrine therapy.
Patients with missing ET treatment end date and patients with still ongoing adjuvant ET were excluded from the analysis.
Figure 2 displays Kaplan-Meier curves illustrating the duration of adjuvant endocrine therapy, overall (Figure 2A) and stratified by stage (Figure 2B) among patients who received this therapy type. The curves showed a distinct inflection point around the 5-year mark, indicating a significant number of treatment discontinuations at that time. This is consistent with current clinical guidelines which recommend endocrine therapy for at least 5 years. 32 This point marks a critical phase in treatment adherence in our cohort, as notable differences in discontinuation rates between stage II and stage III subgroups become apparent after this period. Specifically, Figure 2B reveals that patients with stage III cancer tend to discontinue adjuvant ET later than those with stage II cancer.
Kaplan-Meier curves for time on adjuvant ET from date of surgery in cohort patients.
We then analyzed the discontinuation rate of adjuvant endocrine therapy among cohort patients (Table 7). We arbitrarily set a 5-year window to define ET adjuvant completion. Patients who received less than 5 years of ET adjuvant therapy were assigned to the discontinuation group. For stage II patients, 30.7% (51 out of 166) completed ET, while 69.3% (115 out of 166) discontinued it. Similarly, for stage III patients, 30% (36 out of 120) completed ET, and 70% (84 out of 120) discontinued it. Overall, across both stages, 30.4% (87 out of 286) of patients completed ET, whereas 69.6% (199 out of 286) discontinued therapy. These findings indicate that a significant majority of patients in both stage II and stage III cohorts did not complete their prescribed course of adjuvant ET, highlighting a substantial challenge in maintaining long-term adherence to endocrine therapy in breast cancer treatment.
Endocrine treatment discontinuation rate in adjuvant setting.
Patients with missing ET treatment end date and patients with still ongoing adjuvant ET were excluded from the analysis.
Recurrence rate and time to recurrence
After analyzing the overall treatment patterns and the overall adjuvant ET treatment characteristics in our patient cohort, we then examined the impact of these treatment modalities on patient outcomes. Due to the small sample size of other groups, we assessed recurrence rates in patients receiving chemotherapy combined with ET, those receiving ET without cytotoxic agents and the overall recurrence rates in patients receiving ET regardless of other treatments.
We report these recurrence rates at various intervals following the initiation of adjuvant therapy in our population, stratified by the therapy group received and by stage (Tables 8 and 9). At all reported timepoints, patients who received chemotherapy combined with ET ± other therapy had lower recurrence rates than those who received ET alone or combined with non-cytotoxic agents (2 years: 10.4% vs 16.8%; 3 years: 18.4% vs 25.3%; 5 years: 23.1% vs 39.6%).
Recurrence rate stratified by group of therapy received in the adjuvant setting.
Recurrence rate after initiation of endocrine therapy.
The overall recurrence rates following the initiation of ET ranged from 13.2% at the 2-year mark to 58.4% at the 10-year mark. The sample size permitting, for stage II patients, the recurrence rate was 11.3% at 2 years, increasing to 19.6% at 3 years, and 29.8% at 5 years, indicating a steady climb in recurrence risk over time. In stage III patients, we observed slightly higher recurrence rates at the 2-, 3-, and 5-year marks than stage II patients, with 15.7%, 23.7%, and 31%, respectively.
The median times to recurrence from date of surgery by therapy group during adjuvant therapy and after the initiation of ET are summarized in Table 10. The overall median time to recurrence from date of surgery was 7.76 years (CI = 7.14 to 9.61 years). Both stage II and stage III patients had similar median times to recurrence, with 7.7 years (CI [6.93-9.76 years]) for stage II and 7.78 years (CI = 7.21 to 11.45 years) for stage III. When stratified by adjuvant therapy group, patients who received chemotherapy combined with ET had a median time to recurrence of 9.51 years (95% CI [7.76-11.45]), patients who received ET ± other therapy had a median of 7.01 years (95% CI [5.72-7.84]), and patients enrolled in clinical trials had a median of 6.34 years (95% CI [5.29-NA]). Figure 3 graphically represents the cumulative incidence for recurrence overall (Figure 3A), by stage (Figure 3B), and by therapy group in the adjuvant setting (Figure 3 C).
Time to recurrence from date of surgery stratified by stage and adjuvant therapy.
Cumulative incidence Curves for Time to Recurrence in cohort patients. The panel A shows the overall recurrence over time with a 95% confidence interval. The panel B compares recurrence between stage II (red curve) and stage III (blue curve) patients, highlighting the censoring events on the x-axis. The panel C depicts recurrence for different treatment groups: ET ± other therapies (blue), Cytotoxic ± other therapies (green), and patients in clinical trials (red), with censoring events indicated on the x-axis. The x-axis represents time to recurrence in years, and the y-axis shows the cumulative incidence of recurrence.
OS from early-stage diagnosis
Finally, we evaluated the OS of the patient cohort from early-stage diagnosis, finding a median OS of 17.77 years. Figure 4 illustrates Kaplan-Meier survival curves for the entire cohort (Figure 4A) and stratified by stage (Figure 4B), providing a visual representation of the OS data.
Kaplan-Meier Curves for OS from early-stage diagnosis by Various Factors.
Table 11 delineates the 5- and 10-year survival rates for the study cohort. The overall 5-year survival rate was 91.4% (95% CI [89.0%-93.9%]), decreasing to 75.3% (95% CI [70.8%-80.1%]) at the 10-year mark. When disaggregated by cancer stage, stage II patients had a higher 5-year survival rate of 93.6% (95% CI [90.9%-96.4%]) compared with stage III patients, who had a rate of 88.5% (95% CI [84.3%-92.9%]). However, the 10-year survival rates for stage II (75.1%, 95% CI [69.0%-81.8%]) and stage III (75.3%, 95% CI [68.9%-82.4%]) patients converged, showing similar long-term outcomes.
5-year and 10-year survival rates by various factors.
For further analysis, we focused on the treatments patients received in the neoadjuvant and adjuvant settings (Table 11). Approximately, 40% of the patients received neoadjuvant treatment. These patients had a 5-year survival rate of 89.5% (95% CI [85.0%-93.8%], n = 167), whereas patients who did not receive neoadjuvant treatment had a higher 5-year survival rate of 92.8% (95% CI [89.9%-95.7%], n = 267). The difference between the 2 groups became more pronounced at the 10-year mark, with neoadjuvant patients showing a survival rate of 68.4% (95% CI [60.2%-77.7%], n = 48) compared with 79.2% (95% CI [74.0%-84.7%], n = 110) for those who did not receive neoadjuvant treatment (Kaplan-Meier curves for time to death available in Figure 4C). This difference could be explained by a potential selection bias, as patients who received neoadjuvant therapy had an equal mix of stage II and stage III cancers (50% each), while those who did not receive neoadjuvant therapy were more likely to have stage II cancer (61.6% stage II and 38.4% stage III). This suggests that neoadjuvant therapy was more often given to patients with more advanced disease, which could explain their poorer long-term survival outcomes compared with those who did not receive neoadjuvant therapy.
Another critical observation was the impact of adjuvant ET on survival outcomes. Patients who received ET had a significantly higher 5-year survival rate of 94.6% (95% CI [92.4%-96.9%]) compared with those who did not receive ET, who had a rate of 81.1% (95% CI [74.4%-88.4%]). This survival advantage for ET recipients persisted at the 10-year mark, with a survival rate of 78.3% (95% CI [73.4%-83.6%]), although data for non-ET patients were unavailable due to limited sample size (Table 11 and Figure 4D).
Finally, we assessed the survival rates stratified by therapy groups received during the adjuvant setting. Only 3 groups had enough patients for analysis: those who received ET ± other therapy, those who received chemotherapy combined with ET ± other therapy, and those who did not receive adjuvant therapy. At the 5-year mark, patients who did not receive any adjuvant treatment had the lowest survival rate (82.8%) compared with the other 2 groups, which had similar survival rates (ET ± other therapy: 93.7%; chemotherapy + ET ± other therapy: 92.2%). At the 10-year mark, only 2 groups had enough patients to be reported, showing a noticeable difference between the chemotherapy combined with ET ± other therapy group (79.1%) and the ET ± other therapy group (70.6%) (Table 11 and Figure 4E). This suggests that combining chemotherapy with ET in the adjuvant setting may provide a more effective long-term treatment strategy, leading to better survival outcomes and reduced recurrence rates compared with ET alone or with other non-cytotoxic therapies. However, this conclusion should be interpreted with caution, as we did not explore confounding factors such as high-risk profiles, ET adherence rates within the 2 groups, demographic factors, and patients treated with CDK4/6i. There was an extremely limited number of patients exposed to use of CDK4/6i in adjuvant with ribociclib, palbociclib, and abemaciclib accounting for 1.2%, 4.1%, and less than 1% of treatments. Further research accounting for these variables is necessary to validate the observed benefits of combining chemotherapy with ET.
Discussion
The primary goal of this study was to delineate the treatment patterns, recurrence risk, and the OS for HR+/HER2− early breast cancer patients using real-world data from 4 provinces in Canada. For HR+/HER2− early breast cancer patients, adjuvant endocrine therapy has been the standard of care since the early 1990s.18 -20 Our study evaluates the overall treatment approaches for HR+/HER2− early breast cancer patients in the adjuvant setting, particularly those receiving endocrine therapy and its impact on disease recurrence.
The analysis of treatment patterns revealed several insights into the real-world therapeutic approaches and their adherence to clinical guidelines. First, the findings indicate that a substantial proportion of patients forgo neoadjuvant therapy, with 63.2% of stage II and 51.7 % of stage III patients not receiving any presurgical treatment. This could reflect decision making based on stage reduction objective, tumor aggressiveness, lymph node engagement and patient preference. In addition, the benefit of neoadjuvant therapy has been controversial in HR+/HER2− early breast cancer patients.33,34 Among those receiving neoadjuvant therapy, chemotherapy was predominant, particularly in stage III, suggesting a more aggressive approach in advanced stages. In the adjuvant setting, the combination of cytotoxic therapy and endocrine therapy represents the primary treatment option and is commonly paired with radiation. This approach was more prevalent in stage III cases than in stage II, highlighting its significance in managing advanced stages of the disease. Conversely, endocrine therapy, alone or combined with non-cytotoxic treatments was employed more commonly in stage II than stage III. Our findings revealed that a significantly higher proportion of patients who received neoadjuvant therapy, particularly those treated with cytotoxic agents, subsequently received endocrine therapy as adjuvant therapy or did not receive any adjuvant therapy compared with patients who did not receive neoadjuvant therapy. Specifically, 54.4% of patients who received neoadjuvant therapy continued with endocrine therapy as adjuvant therapy, in contrast to only 22.1% of those who did not receive neoadjuvant therapy.
This disparity suggests that the initial use of cytotoxic agents in the neoadjuvant setting, likely influenced by the aggressiveness of the tumor at diagnosis, may play a role in shaping subsequent treatment decisions. An important aspect not addressed in this manuscript is the use of prognostic tools such as Oncotype DX, which have significantly influenced risk stratification and treatment decision-making in HR+ breast cancer. This trend underscores the importance of individualized treatment plans based on patient response to neoadjuvant therapy and highlights the need for careful monitoring and tailored therapeutic strategies to optimize patient outcomes. A very small proportion of patients in our study received CDK4/6 inhibitors as part of their adjuvant therapy, with ribociclib, palbociclib, and abemaciclib accounting for 1.2%, 4.1%, and less than 1% of treatments, respectively (Supplementary data). It is important to note that at the time of this analysis, ribociclib, abemaciclib, and palbociclib are currently approved in Canada for use in advanced breast cancer, palbociclib reported negative results in their adjuvant trial and as a result does not have this indication, and abemaciclib was approved for adjuvant use in a fraction of the population covered in this analysis, more specifically node-positive, early breast cancer at high risk of disease recurrence based on clinicopathological features. Based on current findings, future treatment strategies in HR+ breast cancer are increasingly focusing on optimizing endocrine therapy, either as monotherapy or in combination with CDK4/6 inhibitors in the adjuvant setting. This evolving approach reflects ongoing efforts to balance efficacy and toxicity, with several prospective trials currently evaluating the role of CDK4/6 inhibitors in reducing recurrence risk among high-risk patients.
Overall, adjuvant ET was administered to 75.6% of the patient cohort, indicating a clear preference for this treatment modality over others. This trend aligns with international guidelines for the management of HR+/HER2− early breast cancer. The analysis of adjuvant ET exposure in this cohort provides several important insights into real-world adherence and its challenges. The median duration of ET was 4.5 years overall, aligning with the guideline recommendation of a 5-year course of therapy. Despite the high initiation rates, our study highlights significant challenges in maintaining long-term adherence to adjuvant ET. Nearly half of the patients (48.3%) discontinued therapy within 3 years. The completion rates of adjuvant ET revealed that only 30.4% of patients completed the prescribed course, highlighting a substantial challenge in long-term adherence and underscores the necessity for healthcare providers to address barriers to sustained ET use. A limitation of our study is that the reasons for discontinuation were not captured, preventing a comprehensive understanding or interpretation of this high discontinuation rate. Potential factors likely include treatment-related side effects, disease progression, patient education and preferences, financial constraints, and lack of support systems. The findings from our cohort emphasize the need for targeted interventions to improve adherence and support patients throughout their ET journey. Future research should focus on identifying and mitigating the factors contributing to early discontinuation to optimize treatment outcomes and reduce the risk of cancer recurrence.
The first outcome investigated was the real-world recurrence rate in the study cohort. We noted that the recurrence rates observed in that study were higher than reported in initial clinical trials (5-year recurrence rate: this study 30.3% vs 9%-12.1%). 35 The most recent findings from the MonarchE trial revealed a 5-year invasive disease-free survival (IDFS) rate of 83.6% in patients receiving adjuvant abemaciclib in combination with ET, compared with 76% in patients treated with ET alone. 36 Meanwhile, the NATALEE trial reported a 3-year IDFS rate of 90.4% for patients treated with ribociclib plus ET, versus 87.1% for those receiving ET alone.37,38
This discrepancy may stem from a broader definition of recurrence, which aligns with real-world data availability, as well as variations in the evaluation period for recurrence rates. These variations encompass changes in the therapeutic landscape over time and a less uniform treatment pattern compared with the standardized approach observed in randomized controlled trials (RCTs). However, these rates reflect real-world unmet needs as they encompass patients who were not cured or progressed rapidly after starting endocrine therapy in the adjuvant setting. This highlights the necessity for more effective therapeutic options to achieve better outcomes. Recurrence rates reported by another real-world study using the US-based Flatiron Health EHR-derived database and using a similar definition for recurrence than the present study also found a 2- and 5-year recurrence rates of 11.9% and 29.8%, respectively, which is in the same range that the ones reported in this study (2-year recurrence rate: 13.2%, 5-year recurrence rate: 30.3%). 39 The difference in recurrence rates observed in real-world settings as opposed to clinical trials might also be influenced by various factors, including treatment adherence issues such as early discontinuation of adjuvant ET, different patient population, variation in the intensity of surveillance and follow up in care between the 2 settings. Comparing the 10-year death rate in our study with a meta-analysis of trials on Tamoxifen and Aromatase Inhibitor (AI) use in adjuvant settings, we found similar outcomes (this study: 21.7% vs 21.3%-24%). 35 Patients who received chemotherapy combined with ET ± other therapy had a lower recurrence rate than the patients who did not received chemotherapy (5-year recurrence rate: 23.1% vs 39.6%, respectively). This difference is also reflected in the median time to recurrence of 9.51 versus 7.01 years in patients who received chemotherapy versus the ones who did not, highlighting that combining chemotherapy with ET in the adjuvant setting might offer a more effective long-term treatment strategy to some patients. However, this conclusion should be viewed with caution, as confounding factors were not considered.
Overall, the median time to recurrence was 7.76 years. Patients with more advanced disease tended to remain on adjuvant endocrine therapy beyond the recommended 5-year duration, which may explain the similar recurrence rates observed in both stage II and stage III subgroups. These findings are consistent with previous research suggesting that extending adjuvant ET from 5 to 10 years, as recommended by American Society of Clinical Oncology (ASCO) guidelines for high-risk patients, 32 significantly benefits HR+/HER2− early breast cancer patients with positive nodal status. Despite these guidelines and the high proportion of patients with positive nodal status present in the study cohort (86%), our study revealed a median adjuvant ET duration of 4.5 years, similar to the 4.8 years observed in another real-world study. 40 Although there’s a well-documented clinical advantage for HR+/HER2− early breast cancer patients who receive at least 5 years of adjuvant ET, several studies have noted a trend of early discontinuation in clinical practice.41 -44 A review evaluating endocrine therapy duration beyond clinical trials found that 31% to 73% of patients had ceased ET within 5 years. 45 Consistent with these publications, our findings showed that 69.6% of patients stopped adjuvant ET before completing the full 5-year treatment course. Multiple factors, including sociodemographic traits, concurrent medications, and adverse events could contribute to treatment discontinuation. Yet, there’s a critical need for increased efforts to support patients in completing their prescribed treatment, given the substantial impact on recurrence risk. In future studies, exploring the potential correlation between early discontinuation of adjuvant ET and its impact on OS would be compelling.
In our analysis of OS within the study cohort, minimal distinctions surfaced among subgroups based on stage. However, this observation might be attributed to the challenge of evaluating survival beyond the 10- to 12-year mark, given the overall small cohort as well as a limited number of patients with sufficiently extensive follow-up. on comparing patients who underwent adjuvant ET with those who did not, we reaffirmed the survival advantage associated with employing endocrine therapy in the adjuvant setting. Another key factor influencing OS, particularly for patients who progress to metastatic disease, is the introduction of new therapies in the metastatic setting, such as CDK4/6 inhibitors.
Conclusion
In summary, our findings reveal a gradual increase in recurrence rates over time, starting at 13.2% after 2 years, escalating to 21.4% after 3 years, reaching 30.3% after 5 years, and ultimately peaking at 58.4% after 10 years. This underscores the pressing need for improved treatment options for the overall population covered in this analysis.
Our exploration of adjuvant and neoadjuvant therapy patterns unveils the evolving treatment landscape for HR+/HER2− breast cancer. This underscores the significant unmet medical need in stage II and III breast cancer, where 1 in 3 patients’ experiences recurrence within 5 years, and over half face recurrence after 10 years, despite receiving adjuvant ET alone. It points to the urgent requirement for more effective and better-tolerated treatment alternatives that can diminish both short- and long-term recurrence rates and extend survival.
Limitations
This study has several limitations that should be considered when interpreting the results. First, the generalizability of our findings is restricted due to the significant concentration of patients from Quebec, representing 84.8% of the total cohort. This regional skew may limit the applicability of the results to broader, more diverse populations across Canada and beyond. Second, our study did not classify patients based on their recurrence risk levels using genomic assays, which could impact treatment decisions and outcomes. In addition, menopausal status, an important factor influencing breast cancer treatment and outcomes, was not available for our cohort. This omission could lead to an incomplete understanding of the treatment’s patterns. In addition, a major limitation is the lack of outcome assessment based on the type of endocrine therapy received (tamoxifen vs aromatase inhibitors), which hinders a comprehensive understanding of how the evolution of treatment landscape over the last 30 years impacted patient outcomes. However, the distribution of diagnosis years was uneven (12 of 541 before 2000, 284 between 2000 and 2015, and 245 after 2015), which may impact the generalizability of results across diagnostic periods. Finally, the median follow-up time of 8.1 years, while substantial, may not be long enough to fully capture long-term outcomes and late recurrences, which are particularly relevant in hormone receptor-positive breast cancer.
Supplemental Material
sj-docx-1-bcb-10.1177_11782234251395892 – Supplemental material for Evaluation of Recurrence Rate in Canadian Patients With Stage II/III HR+/HER2− Early Breast Cancer in the Real-World Setting
Supplemental material, sj-docx-1-bcb-10.1177_11782234251395892 for Evaluation of Recurrence Rate in Canadian Patients With Stage II/III HR+/HER2− Early Breast Cancer in the Real-World Setting by Karen Gambaro, Kahina Rachedi, Mark Basik, Fred Saad, Anne-Marie Mes-Masson, Saima Hassan, Adriana Orimoto, Dominique Boudreau, François Vincent, Eve St-Hilaire, Helen Mackay, Mahmoud Abdelsalam, Steven M Yip, Arif Awan, Robert Hanel, Stéphanie Guillemette, Ricardo Leite, Marc-André Caron, Chandni Patel, Gerald Batist and Maud Marques in Breast Cancer: Basic and Clinical Research
Footnotes
Acknowledgements
We thank the 10 participating Canadian Cancer Centers: Jewish General Hospital, Montreal; Center hospitalier Universitaire de Sherbrooke, Sherbrooke; The Moncton Hospital, Moncton; Dr. Georges-L.-Dumont University Hospital Center, Moncton; The Ottawa Hospital, Ottawa; the Center hospitalier Université de Québec, Québec; the Center hospitalier de l’Université de Montréal, Montreal; CIUSSS de la Mauricie-et-du-Center-du Québec, Trois Riviere; and Sunnybrook Hospital, Toronto, Tom Baker Cancer Center, Calgary. We thank the operational team at each site: Bénédicte Foveau, Caterina Marcangione, Noemie Poirier, Suzanne Maltais, Laura Ross, Mona Mohammed, Ian Chute, Jamie Drapeau, Abdelalim Hamza, Nathalie Tremblay, Jean-Charles Hogue, Louise Rousseau, Alexandria Aubourg, Claudia Syed, Sophie Langevin, Genevieve Cormier, Julie Bilodeau, Nathalie Delvoye, Manon De Ladurantaye, Chantal Auger, Marie-Eve Caron, Julie Samson, Donna Morgan. We thank the operational team at Exactis Innovation: Rosa Garyfallia Christodoulopoulos. We thank the patients who enrolled in PMT and made this project possible.
Ethical considerations
Ethics approval for the PMT registry was provided by CIUSS West-Central Montreal Research Ethics Board (REB Number: MP-05-2016-321).
Consent to participate
All patients included in this study provided informed consent to participate in the PMT registry (NCT02355171).
Consent for publication
All patients included in this study provided informed consent to allow the publication of study results based on their data.
Author contributions
Karen Gambaro: Conceptualization, Data curation, Formal analysis, Methodology, Project administration, Writing—original draft, Writing—review and editing.
Kahina Rachedi: Data curation, Methodology, Writing—review and editing.
Mark Basik: Investigation, Resources, Writing—review and editing.
Fred Saad: Investigation, Resources, Writing—review and editing.
Anne-Marie Mes-Masson: Investigation, Resources, Writing—review and editing.
Saima Hassan: Investigation, Resources, Writing—review and editing.
Adriana Orimoto: Investigation, Resources, Writing—review and editing.
Dominique Boudreau: Investigation, Resources, Writing—review and editing.
François Vincent: Investigation, Resources, Writing—review and editing.
Eve St-Hilaire: Investigation, Resources, Writing—review and editing.
Helen MacKay: Investigation, Resources, Writing—review and editing.
Mahmoud Abdelsalam: Investigation, Resources, Writing—review and editing.
Steven Yip: Investigation, Resources, Writing—review and editing.
Arif Awan: Investigation, Resources, Writing—review and editing.
Robert Hanel: Investigation, Resources, Writing—review and editing.
Stéphanie Guillemette: Conceptualization, Writing—review and editing.
Ricardo Leite: Conceptualization, Writing—review and editing.
Marc-André Caron: Conceptualization, Writing—review and editing.
Chandni Patel: Conceptualization, Project administration, Writing—review and editing.
Gerald Batist: Investigation, Resources, Writing—review and editing.
Maud Marques: Conceptualization, Data curation, Formal analysis, Methodology, Project administration, Resources, Visualization, Writing—original draft, Writing—review and editing.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was financially supported by Novartis Pharmaceuticals Canada Inc.
Declaration of conflicting interests
The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: The authors declare that this study received funding from Novartis Pharmaceuticals Canada Inc. Novartis Pharmaceuticals Canada Inc had the following involvement with the study: study design and review of the manuscript.
Data availability statement
Individual-level data are not publicly available, as participants consented only to the sharing of aggregated results.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
