Favorable outcome of immunotherapy use in metastatic HR+/HER2- breast cancer: a population-based cohort study.

Introduction
Metastatic HR+/HER2− breast cancer accounts for approximately 73% of all metastatic breast cancer cases[1]. The current standard treatments primarily include endocrine therapy, CDK4/6 inhibitors, and chemotherapy[2]. These therapies have significantly improved survival of patients, with the median OS extending to 2–5 years[3]. However, most patients ultimately experience disease progression and lack further effective treatment options. Consequently, new treatment strategies are urgently needed to improve survival in patients with advanced HR+/HER2− breast cancer.
Recently, immunotherapy, particularly immune checkpoint inhibitors (ICIs), has achieved significant breakthroughs in treating metastatic triple-negative breast cancer (mTNBC), substantially altering the treatment paradigm for this subtype[4]. In the IMpassion130 trial[5], atezolizumab combined with nab-paclitaxel significantly extended progression-free survival (PFS) in programmed cell death 1 ligand 1 (PD-L1)-positive mTNBC patients, while in the KEYNOTE-355 trial[6], pembrolizumab combined with various chemotherapy regimens significantly improved OS in PD-L1-high expressing mTNBC patients. These studies have led to the widespread application of immunotherapy in mTNBC, sparking interest in its potential application in HR+/HER2− breast cancer.
Although HR+/HER2− breast cancer typically exhibits low PD-L1 expression and low tumor-infiltrating lymphocytes (TILs) level, which are generally linked to limited response to immunotherapy, several exploratory studies suggest that some HR+/HER2− patients may still benefit from immunotherapy, either as monotherapy or in combination therapy[4]. For example, the KEYNOTE-028 study demonstrated modest antitumor activity with pembrolizumab monotherapy in metastatic HR+/HER2− breast cancer patients[7]. Similarly, JAVELIN Solid Tumor trial[8] reported similar results with avelumab. Furthermore, the KELLY trial evaluated the combination of pembrolizumab and ibrutinib, showing positive antitumor activity in advanced HR+/HER2− breast cancer patients[9]. In the NCT02778685 trial, palbociclib in combination with letrozole and pembrolizumab as a first-line treatment showed a 31% complete response rate and good tolerability[10]. Additionally, research exploring the combination of immunotherapy with antibody-drug conjugates, such as Sacituzumab govitecan, which targets Trop-2 and releases a topoisomerase I inhibitor payload, has indicated encouraging results in enhancing the antitumor effects of immunotherapy[11,12]. Despite these studies being in early stages with small sample sizes and limited follow-up, they provide important insights into the potential application of immunotherapy in mHR+/HER2− breast cancer.
The National Cancer Database (NCDB) provides a powerful platform for analyzing real-world data from a large cohort of patients in the US[13]. By utilizing this data, we aim to study the clinical and sociodemographic factors influencing the application of immunotherapy in metastatic HR+/HER2− breast cancer and assess its impact on patient survival. Our study will offer valuable insights into the role of immunotherapy in advanced HR+/HER2− breast cancer and provide guidance for its integration into future clinical treatment strategies. Through this analysis, we hope to provide strong support for improving treatment plans for metastatic HR+/HER2− breast cancer patients and optimize future immunotherapy strategies to improve patient outcomes.

Patients and methods

Patient cohort and data source
We analyzed data from the NCDB, a comprehensive cancer registry that encompasses approximately 70% of newly diagnosed malignancies across the United States. Our analysis focused on individuals with an initial diagnosis of metastatic HR+/HER2− breast cancer between 2013 and 2021, specifically those who did not undergo surgical intervention for breast cancer.
Exclusion criteria applied in this investigation comprised: (1) secondary or recurrent tumors; (2) male breast cancer cases; (3) incomplete documentation of AJCC T and N staging, radiotherapy, cytotoxic therapy, hormone therapy, or survival times. Based on these criteria, a total of 17 211 eligible patients were included in the study, of whom 15 638 did not receive immunotherapy and 1573 received immunotherapy. Subsequently, 1571 patients receiving immunotherapy were matched with 1571 patients not receiving immunotherapy by using the propensity score matching with a ratio of 1:1. The detailed flowchart for the patients’ selection is shown in Figure 1. This cohort study has been reported in line with the STROCSS guidelines[14].

Propensity score-matching analysis
In order to reduce possible selection bias for survival comparison, a one-to-one PSM between patients with or without immunotherapy was conducted by utilizing the nearest-neighbor matching method with a caliper of 0.05. We included all the covariates in patients’ baseline characteristics into the propensity score-matching (PSM) algorithm, including age, race, income, location, diagnosis year, T and N stage, treatment modalities (endocrine therapy, chemotherapy, and radiotherapy), and metastasis sites (bone, brain, liver, and lung). A histogram of standardized differences before and after PSM was plotted to visually exhibit the properties of the matching procedure (Supplemental Digital Content Figure 1, available at: http://links.lww.com/JS9/F208). The PSM procedure and the calculation of standardized differences were performed using R packages of MatchIt.

Statistical analysis
We used descriptive statistics to present all demographic and clinicopathological variables. Associations between clinicopathological parameters were examined using the Pearson χ2 test. For survival outcomes, we constructed Kaplan–Meier curves and Cox regression modeling to assess differences between groups. Variables with P < 0.10 on univariate analysis were included in the multivariate analysis. To identify variables that influenced immunotherapy administration, we conducted logistic regression analysis. Odds ratios (ORs) and hazard ratios (HRs) with corresponding 95% confidence intervals (CIs) were derived through logistic regression modeling and Cox regression modeling, respectively.
The statistical computations for descriptive analyses, Pearson χ2 testing, logistic regression, and Cox proportional hazards modeling were executed with SPSS software version 24.0 (IBM Corp). R software (version 4.2.2) was utilized to perform the Kaplan–Meier survival analyses. Statistical significance was established at a two-sided P value below 0.05.

Results

Baseline characteristics
A total of 17 211 metastatic HR+/HER2− breast cancer patients were included in our final analysis, with 1573 individuals (9.1%) having received immunotherapy and 15 638 (90.9%) not receiving it. Baseline characteristics were compared between groups using the Pearson χ2 test (Table 1). The age distribution among patients receiving immunotherapy skewed younger, with a higher proportion in the 35–50 (19.6% vs 15.9%, P < 0.001) age range, and a smaller percentage in the 70+ age group (27.7% vs 33.9%, P < 0.001). Immunotherapy use was more common among white patients (79.0% vs 76.8%, P= 0.030). Patients receiving immunotherapy had higher rates of advanced AJCC N stage (N3; 15.3% vs 13.3%, P= 0.012), bone (87.2% vs 77.9%, P < 0.001), and lung (24.3% vs 20.2%, P < 0.001) metastases. Additionally, patients receiving immunotherapy were more likely to receive hormone therapy (84.5% vs 80.2%, P < 0.001), chemotherapy (64.1% vs 59.8%, P < 0.001), and radiotherapy (33.6% vs 28.9%, P < 0.001).

Factors associated with immunotherapy use
To identify factors related to immunotherapy use, a multivariate logistic regression analysis was carried out (Table 2). Patients of the black race were less likely to receive immunotherapy compared to the white race (OR = 0.83; 95% CI: [0.71–0.97]; P = 0.022). Individuals residing in urban settings had a reduced probability of immunotherapy treatment relative to those in metropolitan regions (OR = 0.82; 95% CI: [0.69–0.97]; P = 0.022). In terms of clinical characteristics, later AJCC N stage (N3 vs N0: OR = 1.35; 95% CI: [1.13–1.62]; P = 0.001), bone (OR = 1.81; 95% CI: [1.54–2.11]; P < 0.001) and lung (OR = 1.15; 95% CI: [1.02–1.30]; P = 0.026) metastases was associated with increased odds of immunotherapy use. Moreover, treatment modalities emerged as significant predictors. Hormone therapy (OR = 1.21; 95% CI: [1.04–1.39]; P = 0.012) and radiotherapy (OR = 1.19; 95% CI: [1.06–1.33]; P = 0.002) displayed strong associations with immunotherapy use. However, despite showing correlation in univariate analysis, chemotherapy did not maintain statistical significance in the multivariate model (OR = 1.08; 95% CI: [0.96–1.21]; P = 0.184).

Survival analyses
The overall survival (OS) of patients who had received immunotherapy or not was compared by the Kaplan–Meier method. As exhibited in Figure 2A, patients who had undergone immunotherapy had a superior OS compared with patients who had not (median 46.5 vs 35.5 months, P < 0.001). Moreover, Cox regression models were also constructed to further assess the survival difference. Both univariate and multivariate analyses indicated that the immunotherapy group had a significantly better clinical outcome compared to the no immunotherapy group (multivariate: HR = 0.75; 95% CI: [0.70–0.80]; P < 0.001) (Supplemental Digital Content Table 1, available at: http://links.lww.com/JS9/F209). Following propensity score matching at a 1:1 ratio, 3142 patients were paired according to their immunotherapy status. As shown in Table 1, no statistical difference was observed for all covariates after PSM with P values > 0.05. The Kaplan–Meier curve showed that patients who received immunotherapy had a better prognosis than those who had not (Fig. 2B). The multivariate Cox regression analyses further confirmed this survival benefit of immunotherapy (HR = 0.77; 95% CI: [0.70–0.85]; P < 0.001) (Table 3). Additional factors were identified to be associated with inferior survival, including older age (70+ vs age < 35; HR = 1.56; 95% CI: [1.15–2.12]; P = 0.004), black race (vs white; HR = 1.21; 95% CI: [1.05–1.39]; P = 0.008), brain (HR = 1.47; 95% CI: [1.22–1.78]; P < 0.001), liver (HR = 1.63; 95% CI: [1.45–1.83]; P < 0.001) and lung (HR = 1.23; 95% CI: [1.10–1.37]; P < 0.001) metastases, and received radiotherapy (HR = 1.15; 95% CI: [1.04–1.27]; P = 0.008). Conversely, undergoing hormone therapy (HR = 0.46; 95% CI: [0.41–0.53]; P < 0.001) and chemotherapy (HR = 0.62; 95% CI: [0.56–0.68]; P < 0.001) was associated with better outcomes.

Subsequently, the subgroups that only received hormone therapy or chemotherapy were selected and subjected to survival analysis. Figure 3A and B showed that receiving immunotherapy is significantly related to improved clinical outcomes of patients who had only received hormone therapy both before and after PSM (P < 0.05). The multivariate Cox analyses further indicated use of immunotherapy as a protective factor for such patients after PSM (HR = 0.62; 95% CI: [0.51–0.76]; P < 0.001) (Table 4). In terms of subgroups that only received chemotherapy, the Kaplan–Meier curve showed a relatively obvious trend of survival benefit (Fig. 3C and D). While the P values obtained from the univariate Cox analysis and the Kaplan–Meier method following PSM were close to significance, at 0.058 and 0.057 respectively (Fig. 3D and Table 4), the multivariate Cox analysis demonstrated a statistically significant association between immunotherapy and enhanced survival prognosis (HR = 0.67; 95% CI: [0.38–0.92]; P = 0.015).

Discussion
This large-scale NCDB analysis evaluated the survival advantage of immunotherapy for patients with newly diagnosed metastatic HR+/HER2− breast cancer. To our knowledge, this represents one of the largest analyses using real-world data to evaluate immunotherapy-associated survival outcomes specifically in metastatic HR+/HER2− breast cancer, providing important insights into a clinical scenario where evidence is limited.
Analysis of demographic and clinical characteristics revealed several patterns in immunotherapy utilization. Immunotherapy was less likely to be used in older patients, particularly those aged 70 and above, which may reflect concerns about potential toxicities associated with immunotherapeutic agents in elderly populations, who often have reduced physiological reserve and have more comorbidities. We observed racial and geographic disparities in immunotherapy access, with white patients and those living in metropolitan areas more likely to receive immunotherapy compared to black patients (OR = 0.83; P = 0.022) and those in urban areas (OR = 0.82; P = 0.022), respectively. These findings are consistent with previous studies examining racial and socioeconomic disparities in access to immunotherapy for other forms of metastatic cancer[15,16]. These disparities may reflect broader social and economic factors as well as differences in treatment approaches between academic medical centers and community hospitals. In addition, the cost factor may play a very important role in the application of immunotherapy use in daily practice, especially in low- and middle-income countries. To alleviate this issue, we think the following strategies could be attempted: (1) Support initiatives like drug donation programs or financial aid schemes; (2) Choose less expensive PD-L1 inhibitors based on clinical efficacy and patient affordability, while ensuring safety and efficacy; (3) Expand insurance coverage or implement price controls.
Notably, immunotherapy use was more frequent in patients with extensive lymph node involvement (N3 vs N0: OR = 1.35; P = 0.001), suggesting that clinicians may preferentially use advanced treatments in patients with poor prognostic features to achieve better disease control. Our data also revealed that treatments were important predictors of immunotherapy use, with positive associations observed between immunotherapy utilization and hormone therapy (OR = 1.21; P = 0.012) and radiotherapy (OR = 1.19; P = 0.002), which might indicate potential synergistic effects. Research has shown that several chemotherapeutic agents can enhance antitumor immune responses through dual mechanisms: triggering the release of immune-activating molecules from cancer cells undergoing cell death or exerting secondary effects on immune cell subpopulations[17]. Preliminary results suggest that combining immunotherapeutic agents with hormonal treatments may potentially boost therapeutic efficacy and clinical outcomes[18]. Moreover, poor response to endocrine therapy has been associated with increased TILs[19]. CDK4/6 inhibitors, used as the initial endocrine treatment for metastatic HR+/HER2− breast cancer, have demonstrated an ability to boost tumor immunogenicity and work in conjunction with PD-1 blockade to improve antitumor efficacy[20,21]. A growing body of evidence also supports the idea that radiotherapy can modulate immune responses, either enhancing or inhibiting the immune system through a variety of mechanisms[22]. These stimulatory effects of radiotherapy include the activation of antigen-presenting cells, the release of chemokines that recruit T cells, and the activation of the STING pathway.
HR+/HER2− breast cancer has historically been considered immunologically “cold” and characterized by low immunogenicity, likely due to the scarcity of TILs, the prevalence of tumor-associated macrophages, and diminished HLA class I expression, all of which hinder its antitumor immune response[23]. Nevertheless, a recent study examining TILs in ER +/HER2− breast cancer indicated that certain high-risk subgroups may exhibit greater immunogenic potential, warranting further investigation of immunotherapy strategies[24]. While the prognosis data is not yet mature, recent trials such as CheckMate-7FL and KEYNOTE-75 have demonstrated that the addition of neoadjuvant immunotherapy to chemotherapy significantly enhanced pathological complete response rates in early-stage high-risk HR+/HER2− breast cancer[25,26]. These results highlight the importance of better understanding the heterogeneity of HR+/HER2− breast cancer and identifying patient subgroups that are more likely to respond to immunotherapy. Currently, data on the efficacy of immunotherapy in metastatic HR+/HER2− breast cancer and long-term clinical outcomes are limited. Early trials such as KEYNOTE-028 and JAVELIN indicated limited clinical benefit of monotherapy ICIs, pembrolizumab and avelumab, in treating metastatic HR+/HER2− breast cancer[7,8]. In pursuit of enhancing efficacy through combination strategies, the KELLY trial reported promising antitumor responses with pembrolizumab administered alongside eribulin in patients previously exposed to multiple lines of therapy[9]. However, findings from the NCT03051659 study revealed no statistically significant improvement in PFS or objective response rate when comparing eribulin plus pembrolizumab to eribulin monotherapy[27]. The ongoing phase 3 KEYNOTE-B49 trial is assessing the impact of adding pembrolizumab to chemotherapy in metastatic HR+/HER2− breast patients with PD-L1+ (CPS ≥1) who have progressed following previous endocrine therapy[28]. However, as of now, the results of this investigation have not been disclosed. In our retrospective study, immunotherapy was linked to reduced mortality risk (HR 0.77, P < 0.001). For comparison, the NCDB data reveal an HR value of 0.59 for immunotherapy in advanced triple-negative breast cancer[29]. While the survival benefit appears more modest than in TNBC, it remains clinically relevant given the limited options in metastatic HR+/HER2− disease. The survival benefit from immunotherapy usage was also statistically significant in subgroups that received hormone therapy (HR: 0.62, P < 0.001) or chemotherapy (HR: 0.67, P = 0.015) only, which further indicates that ICIs have shown promise as an adjunct to established treatment strategies for advanced HR+/HER2− breast cancer.
Our findings may have significant implications for clinical decision-making in patients with metastatic HR+/HER2− breast cancer. This consistent survival benefit observed with immunotherapy across multiple analyses suggests that this treatment modality deserves consideration in the therapeutic algorithm for these patients, especially for those with high-risk features such as extensive lymph node involvement. Our results indicating potential synergy between immunotherapy and both hormone therapy and chemotherapy may guide combination treatment strategies. Furthermore, our findings regarding demographic and geographic disparities in immunotherapy access highlight the need for efforts to ensure equitable distribution of novel cancer therapies. The data we have collected might also aid in shaping future clinical trial designs for immunotherapy targeting metastatic HR+/HER2− breast cancer, particularly regarding combination strategies with established treatments and patient selection based on disease burden or other clinical characteristics.
Several limitations of our study warrant consideration. First, as an inherent limitation of retrospective database analyses using the NCDB, we lacked information on details of immunotherapy regimen, dosing schedules, combination strategies, and duration of treatment. Second, the NCDB does not include data on important biomarkers such as PD-L1 status, TMB, or genomic signatures. Lack of PD-L1/TMB data precludes biomarker-driven subgroup analysis, which is critical for optimizing patient selection. Third, we did not have access to information on PFS, treatment-related toxicities, treatment completion rates, tolerance, or quality of life outcomes, which are important considerations in evaluating the overall benefit of immunotherapy.

Conclusion
In conclusion, our analysis of a large national database suggests that immunotherapy provides a significant survival benefit for metastatic HR+/HER2− breast cancer. The observed benefit was maintained across subgroups receiving different treatment modalities, suggesting potential versatility in combination strategies. These results address an important knowledge gap in breast cancer immunotherapy and have direct implications for clinical practice and upcoming research. Prospective clinical trials are warranted to further define optimal patient selection, timing, and combination strategies for immunotherapy in such types of breast cancer.

Supplementary Material