Differential impact of hormone receptor status on survival and recurrence for HER2 receptor-positive breast cancers treated with Trastuzumab

Introduction Hormone receptor status has major implications for treatment and survival of breast cancer. Yet the impact of hormone receptor status on outcome after Trastuzumab has received little attention. The objective here was to explore any differential effects of Trastuzumab treatment (Trast +ve) on Luminal B HER2 or HER2+(ER−) breast cancer subtypes. Methods A cohort of 469 HER2 receptor-positive breast cancers was categorised by molecular subtype and Trastuzumab treatment. Effects of Trastuzumab treatment on survival, locoregional recurrence and distant metastasis were investigated by subtype, using univariate and multivariate analysis. Results Trast +ve Luminal B HER2 patients had significant improvements in 5-year DFS (p < 0.001) and OS (p < 0.001), while Trast +ve HER2+(ER−) patients had significant improvements in 5-year DFS (p = 0.012) alone. Only Trast +ve Luminal B HER2 cancers displayed a significant reduction in LRR rates (p < 0.001). A significant reduction in distant metastasis rates was seen in Trast +ve Luminal B HER2 (p < 0.001) and HER2+(ER−) (p = 0.009) cancers. Interestingly, bone metastasis rates in Trast +ve Luminal B HER2 cancers demonstrated the greatest reduction (36.2–6.7%). Multivariate analysis of Trast +ve patients found no difference in distant metastasis rates (p = 0.96) between subtypes. Significantly, lower LRR rates were seen in Trast +ve Luminal B HER2 cancers, compared to Trast +ve HER2+(ER−) (p = 0.018). Conclusion An enhanced response to Trastuzumab was seen in Luminal B HER2 cancers. We highlight how Trastuzumab treatment changed the natural history of the HER2 receptor-positive breast cancer, demonstrating improved efficacy in changing the outcome of hormone receptor-positive patients. Electronic supplementary material The online version of this article (doi:10.1007/s10549-017-4225-5) contains supplementary material, which is available to authorized users.


Introduction
Advances in molecular profiling have allowed breast cancer to be categorised into clinically relevant molecular subtypes. [1][2][3]. In approximately 20-30% of breast cancers, the HER2 receptor is over expressed [4,5], resulting in increased cell signalling, uncontrolled cellular proliferation and poor clinical prognosis. Half of these are hormone receptor-positive Luminal B HER2 and half are hormone receptor-negative HER2?(ER-). Differences in survival and outcome occur between the two HER2 receptor-positive subtypes [6,7]. However, some studies have shown no significant difference when assessing longterm survival between the two subtypes [8][9][10]. Clinically, the two subtypes present with distinct patterns of recurrence. In Luminal B HER2 subtype bone is the most common distant metastasis site, similar to Luminal A breast cancers [8,11]. HER2?(ER-) cancers have the highest rates of locoregional recurrence (LLR) overall and tend to initially metastasise to visceral organs, such as the lung [8,12]. Trastuzumab is a monoclonal antibody that binds to the HER2 receptor and interferes with the HER2-mediated signalling cascade, preventing proliferation and eventually leading to cell death [13]. Trastuzumab was originally used to treat metastatic breast cancers and was shown to significantly improve median survival from 20.3 to 25.1 months [14]. Multiple studies have shown that Trastuzumab used in the adjuvant setting reduces recurrences and can increase survival in HER2-positive patients by up to 33-39% [15][16][17][18]. More recently, several studies have shown that when used in the neo-adjuvant setting Trastuzumab significantly increases pathological complete response (pCR) rates [19,20].
While improvements in survival in HER2 receptorpositive cancers have been demonstrated in multiple studies, few studies have examined if Trastuzumab treatment has a varied response between Luminal B HER2 and HER2?(ER-) breast cancers. The aim of this study was to assess the impact of Trastuzumab therapy on the survival or outcome on the two HER2 receptor-positive breast cancer subtypes, before and after the introduction of Trastuzumab as part of the adjuvant treatment regime.

Patient cohort
This study group consists of all patients with HER2 receptorpositive breast cancers treated at a tertiary referral unit entered into a prospectively maintained database from 1991 to 2014. Only patients with a definitive HER2 receptorpositive subtype were included. All clinical pathological details and treatment regimes were analysed. Hormone receptor-positive patients received hormone therapy, as per standard clinical treatment protocols at the time of diagnosis. Clinically, testing for HER2 receptor status began at our centre in 1999. In order to find HER2 receptor-positive patients not treated with Trastuzumab, we used a cohort of patients identified using retrospective testing of pathology samples. A cohort of HER2 receptor-positive patients who received no Trastuzumab treatment was provided by retrospective testing of HER2 receptor status on patients included on a prospectively collected tissue microarray (samples from 1994 to 2001). Historically in our program, Trastuzumab therapy was introduced as adjuvant therapy in 2006, prior to this it was available only to patients recruited on clinical trials. Patients were categorised as received adjuvant/neoadjuvant Trastuzumab (Trast ?ve) or no Trastuzumab treatment (Trast -ve).

Subtypes definitions
Breast cancer subtypes were defined using ER, PR and HER2 receptor status. Luminal B HER2 was defined as (ER and/or PR ?ve, HER2 ?ve) and HER2?(ER-) as (ER and PR -ve, HER2 ?ve) according to standard clinical pathological guidelines. The ER and PR receptor status were determined independently by clinical pathologists using immunohistochemistry as per ASCO guidelines (ALLRED score [2 or more than 1% stain positive). The HER2 receptor status was identified by Herceptest, as part of the routine clinical evaluation, with a score of 3? considered positive. Any ?2 inconclusive results were confirmed using a FISH testing as per ASCO guidelines, with a HER2/CEP17 ratio greater than two considered amplified.

Survival
Overall survival (OS), disease free survival (DFS), and patterns of recurrence were determined. The 5-year DFS & OS were determined and only patients who had completed 5 years of follow-up were included in the analysis.

Recurrence
Breast cancer recurrence was defined as a return of cancer after treatment and after a disease free period. Only stage I-III breast cancers were included for this section of the analysis. Recurrence was divided into LRR and distant metastasis. LRR is defined as recurrence at the same site of the primary cancer or the regional lymph nodes, while distant metastasis is recurrence at a distant site from the primary cancer.

Statistics
Statistical analysis was performed using R statistical software version 3.2.3. A p value of less than 0.05 was considered statistically significant. The Kaplan-Meier method was used to determine survival distributions. The log rank was used to determine any statistically significant differences in survival between the indicated groups. Cox regression was used for multivariate analysis, with logistic regression used to analyse categorical data.

Ethics, consent and permissions
This study was conducted in accordance with the granted National University of Ireland Galway and University College Hospital Galway ethical approval. Informed consent was obtained from all patients. All patients had histologically confirmed breast cancer and all relevant clinicopathological and demographic data were obtained from a prospective breast cancer database.

Cohort description
The study consisted of 468 HER2 receptor-positive patients eligible for this study, treated in our institute between 1991 and 2014. From these, 287 (61%) were found to be Luminal B HER2, with the remaining 181 (39%) patients HER2?(ER-). The median age of patients was 63 and the median follow-up was 49 months. The majority of the overall cohort was recruited after Trastuzumab received approval for adjuvant treatment in 2006 (Table 1). For the purpose of this analysis, patients were categorised as either Trast ?ve (received adjuvant/neo-adjuvant Trastuzumab) or Trast -ve (no Trastuzumab treatment).
Overall, 299 (63.9%) patients were treated with Trastuzumab (Trast ?ve). The clinical pathological details of the cohort are listed in Table 1, demonstrating the two subtypes are relatively matched for age, stage and treatment. The only statistical significant difference between the Luminal B HER2 and HER2?(ER-) was observed in the grade category, where the HER2?(ER-) cohort had a higher proportion of grade 3 cancers (49.8 vs. 79.3% p \ 0.001). Furthermore, 263 (91.6%) of the Luminal B HER2 cancers received adjuvant hormone therapy. In the series recurrence occurred in 94 (20.1%) patients, of which 15 (3.2%) had LRR alone. 54 (11.5%) patients had distant metastasis alone and 25 (5.3%) patients had both LRR and distant metastasis. There was no significant difference in the distribution of age, stage or treatment of cancers between the two subtypes (Table 1).

Univariate analysis of survival
Survival was similar in Luminal B HER2 compared to HER2?(ER-) subtypes ( Figure S1) and Trastuzumab treatment significantly improved overall survival in both subtypes ( Figure S2). No difference was seen in survival between the two subtypes in either the Trast -ve or Trast ?ve patients. Next to assess the impact of Trastuzumab introduction on each subtype, the 5-year DFS and OS were compared between the Trast -ve and Trast ?ve groups in both subtypes. Analysing the DFS and OS by subtype, an increased survival rate is seen for Trast ?ve patients in both Luminal B HER2 and HER2?(ER-) patients. However, a greater improvement was seen in Luminal B HER2 patients. Luminal B HER2 cancers had a statistically significant improvement in both 5-year DFS (p \ 0.001) and OS (p \ 0.001) (Fig. 1a, b), while the HER2?(ER-) only had a significant improvement in DFS (p = 0.012) but not OS (p = 0.135) (Fig. 1c, d). Analysis of the effects of Trastuzumab treatment on distant metastasis by site of recurrence and subtype was performed (Table 3). For Trast -ve patients, bone was the most common site of metastasis for Luminal B HER2 cancers, while lung was the most common site in HER2?(ER-). Following Trastuzumab treatment in the Luminal B HER2 Bold values indicate significant p value cancers, a significant reduction was seen for all distant sites of metastasis (except brain). The site with the greatest reduction in metastasis, due to Trastuzumab treatment, was in bone (22.9 vs. 3.8%, p \ 0.001). HER2?(ER-) breast cancers did show decreases in all metastatic sites except for brain in Trast ?ve patients; however, these reductions only approached significance in bone (p = 0.075), lung (p = 0.086) and liver (p = 0.075).

Multivariate analysis of recurrence
Performing a multivariate analysis of recurrence risk by treatment, metastatic site and subtype (

Discussion
The impact of Trastuzumab treatment has been well established, greatly improving survival and significantly reducing recurrence in HER2 receptor-positive breast cancers [15,21,22]. It has also been shown to improve pathological complete response rates, and in our study pCR   Bold values indicate significant p value LRR Local Regional Recurrence, Distant distant metastasis metastasis at all sites, except for brain metastases. Furthermore, a greater reduction was seen in Luminal B HER2 breast cancers, with the largest reduction in overall metastasis rates seen in the bone metastasis. Few previous studies have compared survival or recurrence rates between the two HER2 receptor-positive breast cancers since the introduction of Trastuzumab. Romond et al. compared survival between the two HER2 receptorpositive breast cancers and found at 4-year follow-up that hormone receptor status minimally influenced the response to Trastuzumab, although hormone receptor status was reported as a significant predictor in DFS [15]. Our findings support previous findings where Trastuzumab treatment led to a reduction in LRR only in Luminal B HER2 cancers [24]. We demonstrate a statically significant reduction in LRR rates observed in Luminal B HER2 patients, while no significant difference was seen in the HER2?(ER-) subtype.
Previous studies demonstrated that bone was the most common recurrence site in Luminal B HER2 breast cancers and lung was most common for HER2?(ER-) [8,9]. We show Trastuzumab treatment resulted in a reduction in metastasis to all sites, except the brain which is explained by the fact that Trastuzumab does not cross the blood/brain barrier [25,26]. Surprisingly, Luminal B HER2 cancers treated with Trastuzumab showed the greatest reduction in distant metastasis to the bone. While Trastuzumab treatment did not lead to significant variations in brain  metastasis rates, it did result in the Luminal B HER2 cancers having a significantly reduced odds ratio of brain metastasis. This correlates with previous studies which show a higher brain metastasis rates in hormone receptornegative tumours [26]. A potential reason for variation between subtypes could be that HER2 receptor over expression reduces the response to hormone therapy. Studies have observed increased hormone resistance rates in Luminal B HER2 cancers compared to Luminal A cancers [27,28]. Cross talk between HER2 receptors and hormone receptors results in activation of the hormone receptor, even in the presence of hormone treatment [29]. Clinical trials have shown that the addition of Trastuzumab to hormone treatment improves survival in metastatic breast cancer [30]. In our study, the introduction of Trastuzumab significantly reduced distant metastasis, especially in bone metastasis rates. By reducing the activity of HER2 receptors, Trastuzumab may restore the response to hormone therapy in Luminal B HER2 cancers [31]. In the neo-adjuvant setting, this may provide an explanation as to why lower levels of pCR are seen in Luminal B HER2 cancers compared to HER2?(ER-) cancers [23].
Another potential difference between subtypes could be the increased number of grade 3 cancers seen in the HER2?(ER-) group. It has been shown that both HER2?(ER-) and triple-negative breast cancers have worse outcomes and present with higher-grade cancer than Luminal cancers [6]. Although, in a pooled analysis of neoadjuvant Trastuzumab trials, grade 3 cancers had a higher pCR rate than grade 1 and 2 cancers [23]. Importantly, our study revealed that on multivariate analysis of patients treated with Trastuzumab, grade 3 cancers did not have a lower DFS or OS risk when compared to grade 1 and 2 breast cancers.
In patients receiving neo-adjuvant Trastuzumab chemotherapy, around 40% of patients will have a complete pathological response [20,32]. This shows that a large proportion of patients only have a partial or no response to Trastuzumab treatment. This resulted in the development of new anti-HER2 receptor treatments targeting different pathways such as Pertuzumab, which has been shown to increase response [33]. Our study highlights the improved prognosis associated with anti-HER2 receptor therapy; it also demonstrates that a large proportion of patients survived despite not being treated with Trastuzumab. We believe this clearly indicates that Trastuzumab treatment is not required for all HER2 receptor-positive breast cancers. There is a clear need to develop a molecular or genetic scoring system to identify which patients will benefit from Trastuzumab treatment and those that will not.
This study once again shows the benefit of Trastuzumab treatment in HER2 receptor-positive breast cancers, demonstrating the effects on both survival and recurrence rates. It also highlights how a targeted therapy has altered responses in related breast cancer subtypes, emphasising their molecular differences. Demonstrating a more positive impact of Trastuzumab treatment on Luminal B HER2 cancers supports the need to further characterise the mechanism of action of Trastuzumab in each subtype, suggesting that key differences remain to be defined. This work highlights the need to fully understand the subtypespecific effects and mechanisms of action of Trastuzumab therapy, which will allow truly individualised breast cancer management regimes to be implemented.
Funding This work is supported by funding from Breast Cancer Research.
Author contributions AM, JALB and MJK performed data analysis and wrote the manuscript; AM, OK and EH performed the statistical analysis; CM, RM, AL, CC and MJK participated clinically in patient data provision and in manuscript preparation.

Compliance with ethical standards
Conflict of interest The authors declare that there are not conflicts of interest.
Competing interests The authors declare that they have no competing interests.
Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://crea tivecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made.