Our work describes RWD from a large Brazilian cohort of patients with HR+/HER2- EBC treated in routine care in the private healthcare setting. We describe clinical characteristics, patterns of systemic therapy, and survival outcomes. Among the main findings are the distribution of patients into clinicopathological risk categories. These data are critical considering incorporating therapeutic agents, such as adjuvant abemaciclib for selected high-risk patients with HR+/HER2- EBC disease. Furthermore, systemic treatment patterns are described, and trends toward possible underuse of chemotherapy and OFS in premenopausal patients with intermediate or high-risk disease can be observed, which may negatively impact patients’ outcomes treated in routine practice.
Real-world data (RWD) on patients with BC holds paramount importance in shaping evidence-based clinical decisions and optimizing patient outcomes. Unlike data from controlled clinical trials, RWD provides insights into the broader patient population encountered in routine clinical practice, offering a comprehensive understanding of treatment patterns, effectiveness, and gaps in access or utilization of therapies in diverse patient demographics and healthcare settings(12). Additionally, RWD is a cornerstone for pharmacoeconomic studies, facilitating cost-effectiveness analyses, resource allocation decisions, and healthcare policy development in public and private settings(13).
The multidisciplinary management of HR+/HER2- EBC has seen significant advances with the incorporation of personalized systemic treatment strategies based on the classification of patients into low-, intermediate-, and high-recurrence risk groups. This classification is made according to clinicopathological risk factors, such as tumor size, lymph node involvement, and histological grade. Furthermore, biomarkers such as Ki67 expression and gene expression signatures such as OncotypeDX and MammaPrint are complementary. Patients with low-risk tumors, characterized by having a small, low-grade tumor and no lymph node involvement, can be treated with de-escalation strategies, especially if postmenopausal. On the other hand, patients with intermediate-risk and high-risk tumors should be considered for treatment escalation strategies with the use of chemotherapy, OFS in premenopausal women and combination endocrine therapy (ET) with CDK4/6i or PARP inhibitors (in patients with BRCA1/2 germline pathogenic variants).
We acknowledge that the definitions for low, intermediate, and high-risk categories are relatively arbitrary and there is no consensus in the main guidelines for breast oncology. Nevertheless, it is important to highlight that we adopted the criteria used in the main phase III randomized clinical trials (RCTs) that evaluated the use of CDK4/6 inhibitors in the adjuvant setting, the monarchE study(14) (which included only clinically high-risk patients) and the NATALEE study (with broader criteria for intermediate- and high-risk disease) (10). These recent RCTs have demonstrated a benefit in iDFS outcomes by incorporating abemaciclib and ribociclib into standard endocrine therapy. In the monarchE study, patients with high-risk clinicopathological characteristics were randomized to use ET with or without abemaciclib. The abemaciclib group had improved iDFS versus ET alone (HR 0.68, 95%CI 0.57–0.80, p < .0001). The 3-year iDFS rate in the experimental arm was 88.6% vs 82.9% in the control arm, translating into an absolute risk reduction of 5.7% with abemaciclib (14). Adjuvant ribociclib was evaluated in the phase III RCT NATALLEE, which randomized patients with intermediate and high-risk characteristics and demonstrated a gain in iDFS favoring the group that used ribociclib with an HR of 0.74 (95% CI 0.62–0.89, p = 0,0006) and 3-year iDFS of 90.7% in the ribociclib arm and 87.6% in the control arm(10).
The main findings of our RWD study can be divided into three themes. Firstly, we provide epidemiological data related to stratification into risk groups. This information has several potential implications, such as estimating potential patients eligible for systemic treatment (de)escalation strategies in our routine clinical practice. Second, we provide data on treatment patterns in routine clinical practice. Lastly, we describe the survival outcomes, primarily 3-year iDFS rates in cohorts grouped according to menopausal status and risk categorization. In our cohort, patterns of ET and CT use are described and call attention to the potential underuse of adjuvant systemic treatment strategies such as OFS and CT in premenopausal patients, which have robust evidence of clinical benefit in patients with intermediate and high-risk disease. Such findings have also been reported in other cohorts, such as the recent report by Heredia A et al. that showed that in a retrospective cohort of patients younger than 35 with EBC HR + from Chile, less than 20% received OFS (15). And even in the RxPONDER Trial that included 12.7% of premenopausal women, only 6.3% in the chemoendocrine group and 19.0% in the endocrine-only group had received OFS (16). This information impacts the development of medical education initiatives and the standardization of institutional pathways of care.
We have also found a numeric difference in 3-year iDFS rates when indirectly comparing our real-world population to patients enrolled in the control arms of RCTs. For example, in our cohort’s high-risk group, the 3-year iDFS was 72% (versus 84% 3-year iDFS in the control arm of the monarchE trial)(14). Also, the intermediate- and high-risk populations in our RWD study had 3-year iDFS of 83% (versus 87% 3-year iDFS in the control arm of the NATALEE trial)(10). These observations underscore a concerning disparity between real-world outcomes and those observed in controlled trial settings. Several factors could contribute to this difference, including the definitions used for patient eligibility in the studies, variations in patient demographics, comorbidities, treatment adherence, and access to supportive care services (17). Additionally, the heterogeneity of patient populations in real-world settings, which often include individuals with more complex health profiles than those enrolled in clinical trials, may also play a significant role (18). For this reason, any formal comparison would have been biased, and inferences could be inaccurate.
However, the implications of these findings are profound, suggesting the need for closer examination of real-world treatment outcomes and the development of strategies to address potential disparities to ensure that patients in clinical practice receive optimal care and support to improve their outcomes. Moreover, this information may be crucial in pharmacoeconomics, as incorporating new agents, such as CDK4/6 inhibitors in the adjuvant setting, may be associated with a higher absolute risk reduction given the worse prognosis observed in real-world settings.
While our study provides valuable insights into the risk profile of patients with HR+/HER2- EBC attending tertiary oncology services, several limitations should be acknowledged. Firstly, the study's retrospective nature introduces inherent biases and limitations, including potential selection bias and incomplete data capture. Secondly, the reliance on electronic health records may result in inaccuracies or missing data, which could impact the validity of our findings. Unfortunately, we did not have access to important information related to adherence to treatment, causes of interruption or changes in treatment, and information about therapies and events after progression. Furthermore, our study was conducted within a single healthcare network that treats Brazilian women with access to the private health system, limiting the generalizability of our results to other populations or healthcare settings, especially the public health setting.
However, our study has several strengths that confer robustness and clinical relevance to our findings. Firstly, we analyzed a large cohort of patients treated during a period of uniform perioperative standard-of-care, supporting the generalizability of our findings. Secondly, our detailed examination of clinicopathological characteristics and treatment strategies enables an in-depth analysis of factors affecting patient outcomes. Furthermore, our data underwent a rigorous analysis and review, incorporating specialized data curation to maximize data accuracy.