Introduction
Breast cancer (ICD-10 C.50), with an annual incidence of 70,000 new cases, is the most common type of cancer among women in Germany, and the second most frequent cause of cancer-related death in women [
1]. Implementation of screening procedures and development of new therapies revealed constant mortality rates despite increasing incidence; age-standardised mortality declined slightly and the 5-year relative survival increased over the past decade [
2,
3].
Standard therapy of patients with early breast cancer consists of surgery, radiation and adjuvant systemic therapy. However, as breast cancer is highly heterogeneous, the selection of adjuvant systemic therapy depends on stage, histology and on molecular subtypes of the tumour [
4,
5]. Current adjuvant systemic therapy options include chemotherapy, endocrine therapy for hormone receptor (HR)-positive tumours, and targeted biological agents such as trastuzumab for human epidermal growth factor receptor (HER2)-positive tumours. The treatment decision is based on multiple factors and includes—in addition to tumour biology and the predicted sensitivity to particular treatment types—the patients’ physical constitution, biological age and comorbidities as well as patients’ preferences.
The current St. Gallen international experts consensus recommends endocrine therapy alone for adjuvant systemic therapy of luminal A-like breast cancer subtypes with low risk in the majority of cases, while additional chemotherapy should be considered in patients with four or more lymph nodes involved [
6]. For luminal B-like HER2-negative subtypes, endocrine therapy and chemotherapy is recommended in the majority of cases, while for luminal B-like HER2-positive subtypes, chemotherapy, anti-HER2 targeted therapy and endocrine therapy are recommended for all patients. For the triple-negative subtype [oestrogen-receptor (ER)-negative, progesterone receptor (PR)-negative and HER2-negative], chemotherapy should include an anthracycline and a taxane [
6].
Despite these recommendations and other clinical practice guidelines, the extent to which these recommendations are incorporated into routine clinical practice is only partially known [
7‐
9] and often limited by retrospective data collection [
10].
There are several regimens for which treatment efficacy has been established in randomized controlled trials (RCTs) [
11]. These regimens differ in duration, the combination and the dosages of drugs given. Research in clinical trials usually focuses on treatment with one regimen or drug of interest and also applies stringent selection criteria. As more treatment options have become available, a wider variety of treatments was applied to individual patients in everyday routine care. However, knowledge on the use of existing chemotherapeutic agents and combinations in routine care is limited. Clinical cohort studies like ours can help to fill this gap and provide insight into treatment and outcome of patients in routine care [
12].
Today, most systemic breast cancer treatments can be delivered on an outpatient basis. In Germany, ambulatory care is predominantly provided by office-based specialists and hospital outpatient centres. This paper focuses on the cytotoxic treatment of patients with early breast cancer in daily routine practice as well as changes in treatment over time from 2007 to 2014.
Discussion
The aim of this analysis was to describe the different adjuvant cytotoxic treatment approaches for patients with early breast cancer in daily routine practice in Germany as well as to identify changes in treatment strategies since 2007. The strengths of our study are the use of prospectively collected clinical data on systemic therapy in unselected primary breast cancer patients throughout Germany. The TMK is not limited to patients treated with a particular substance, thus providing a unique assessment of the different systemic treatment strategies applied outside of clinical trials, reflecting the “real-world” setting. Furthermore, we show a comprehensive analysis of treatment patterns of the breast cancer subtypes. Outcome data from the TMK will be analysed after an adequate follow-up time is reached, addressing the key question as to how the clinical efficacy shown in RCTs translates into clinical effectiveness in daily routine practice.
Our data show a shift from F + A-based to anthracycline/taxane-based regimens as preferred adjuvant treatment. While the increase of taxane use was apparent in all patient subgroups, the greatest increase was among node-negative patients. Our regression model examined the association of patient and tumour characteristics with receipt of taxane-free adjuvant chemotherapy. Positive nodal stage, triple negative or HER2-positive tumours, tumour grading, year of therapy and tumour size were significantly associated with decreasing odds for taxane-free therapy. Older age was the only factor associated with increasing odds for taxane-free therapy.
In our cohort study, 81% of the patients with HR-positive tumours were treated with endocrine therapy and additional HER2-inhibitors were documented for 81% of the patients with HER2-positive tumours. Our cohort did not include patients treated exclusively with endocrine therapy. In addition, although HR- and HER2-status were documented for 98% of our patients, we cannot distinguish between luminal A and luminal B subtypes because data on Ki-67 status were not collected prior to 2011. Both limitations have to be taken into account when comparing our data with other published studies.
Compared to the patients from different European and Californian registry cohorts, our patients are similar with regard to the proportion of HR- and HER2-receptor subtypes [
14‐
17], as well as distribution of age and CCI [
17‐
20], if known restrictions of these registries (e.g. an age limit of 75) and our primary focus on patients receiving cytotoxic treatment are accounted for. The proportion of patients with CCI ≥ 1 (12%) is also similar to patients in other real-life settings [
17‐
19]. When compared to patients in RCTs, those treated in routine care are considerably different with regard to demographic and clinical characteristics [
21]. In our TMK cohort, median age at start of adjuvant systemic treatment is 56 years, compared to approximately 50 years in RCTs [
22‐
24]. Furthermore, the TMK includes patients who would have been excluded from RCTs because of the severity of comorbidities [
22‐
24]. This indicates that the general state of health of our patients is less favourable, compared to patients in prospective clinical trials.
Positive nodal stage is a known predictor of relapse [
25,
26] and pivotal studies [
27,
28] and meta-analyses [
29,
30] have shown a survival benefit of about 3% (5-year survival-rate) by adding a taxane to an anthracycline-based chemotherapy for node-positive patients. Therefore, as expected, the highest rate of combined anthracycline/taxane regimen in our cohort is seen in the node-positive patient population. Nevertheless, the highest increase in taxane therapy can be seen in node-negative patients across all subtypes. The increase in taxane-based chemotherapy since 2008 over all patient subgroups in our cohort reflects changes in the treatment guidelines: while taxanes were not recommended for node-negative patients in 2008–2009, the use of taxanes for all patients receiving chemotherapy has been highly recommended since 2012 by the German AGO-guidelines [
31]. However, the benefit of taxane-based chemotherapy for all node-negative patients has yet to be proven. Trials including node-negative patients that showed a survival benefit, either only included high risk patients [
23] or showed only a significant benefit for the node-positive subgroup [
25]. Node-negative high-risk patients are probably the reason why the EBCTCG meta analysis in 2012 [
32] also reported a small but significant reduction of the relative recurrence risk. This means, that some node-negative patients might benefit from taxane-based therapies, especially if other risk factors are present [
23]. However, there is also evidence that as many as 70% of node-negative patients could be treated effectively with surgery, radiotherapy and endocrine therapy alone [
33,
34]. On the other hand, due to the cardiotoxic side effects of anthracyclines, adjuvant combinations of docetaxel and cyclophosphamide (TC) are recommended for node-negative or low-risk node-positive breast cancer as an alternative to anthracycline-based therapies [
35]. Nevertheless, only 10% of the patients of our cohort received anthracycline-free regimens.
Currently it is not possible to reliably identify node-negative patients, who will benefit from anthracycline/taxane-combination therapies. This is probably the main reason why an increasing proportion of node-negative patients in the TMK received such a combination therapy. However, if the overall risk of recurrence is small, the treatment choice has to be counterbalanced with treatment-related acute and long-term toxicities [
36‐
38]. Especially in the light of recent findings regarding treatment choice based on gene-expression profiles, the decision for adjuvant chemotherapy treatment should not be taken lightly: the PlanB trial showed that patients with early HR-positive breast cancer and enhanced risk (assessed by the 21-gene recurrence score) had excellent 3-year survival rates despite omitted chemotherapy [
39]. Similarly, 5-year survival rates of patients with high clinical, yet low genomic risk (assessed with the 70-gene signature test), were comparable regardless of the receipt of chemotherapy [
40]. On the other hand, it has been shown that the patients’ perceived estimation of increased risk of relapse is a major determinant for using adjuvant chemotherapy despite uncertainties regarding the degree of benefit when added to endocrine therapy in the low-risk HR-positive population. Thus the rise of taxane-use in node-negative patients in the TMK might also be partially attributed to patients’ personal preference.