Introduction
With 2.1 million new diagnoses predicted for 2018, breast cancer represents the most prevalent type of cancer in women worldwide [
1]. Despite new therapeutic agents and improved survival, (metastatic) breast cancer is considered a leading cause of death among women, with about 700,000 deaths in 2018 [
1‐
3]. Inevitably, therefore, the tumor burden must be closely monitored and targeted systemic therapy (chemotherapy, endocrine therapy, targeted therapy, and biological therapy) or local therapy (surgery and radiation therapy) adjusted.
The estrogen receptor (ER), progesterone receptor (PR), and human epidermal growth factor receptor 2 (HER2) statuses of primary breast cancer tissue are used clinically to approximate biological subtypes, to predict outcome, and to guide therapy decisions, especially for endocrine and HER2-targeted regimens [
4,
5]. However, numerous studies have shown substantial discordance rates in ER, PR, and HER2 receptor profiles between primary and metastatic tumors. Based on a meta-analysis of 39 studies, Schrijver et al. [
6] reported conversion rates of 19.3% for ER, 30.9% for PR, and 10.3% for HER2, respectively. Yeung et al. [
7] showed similar findings in a meta-analysis of 47 studies including 3,384 matched primary and metastatic breast cancer cases with median discordance rates of ER, PR, and HER2-expression of 14%, 21%, and 10%, respectively. Furthermore, loss of receptor positivity is associated with poorer prognosis [
8,
9]. Indeed, the meta-analysis by Schrijver et al. showed that therapy was changed due to receptor discordance in 14% for ER, 62% for PR, and 67% for HER2, respectively, assuming that patients gaining hormone receptor positivity qualify for endocrine therapy and those gaining HER2 positivity are eligible for HER2-directed therapy [
6].
Given the broad evidence for receptor status conversion during tumor progression, NCCN, ESMO, EGTM, and ASCO guidelines congruently recommend re-testing hormone receptors for metastatic lesions where feasible [
4,
10‐
12]. However, it is not known whether tissue sampling of metastatic sites has any significant effect on patient survival or quality of life [
13,
14] and whether the biology of the primary or the metastatic lesion(s) should guide therapeutic decisions [
11].
Our study retrospectively compared ER, PR, and HER2 receptor profiles in biopsies of primary breast cancer and corresponding metastatic lesions in a large study population to assess individual changes throughout tumor progression and location-specific discordance rates.
Discussion
The choice of targeted therapy for patients with metastatic breast cancer is guided by ER, PR, and HER2 status of the primary tumor, despite evidence indicating that hormone receptor status may change during tumor progression [
4,
6,
11]. Discordant receptor expression between primary tumor and metastatic site is prognostic [
16‐
18] and might be clinically relevant if changes in hormone receptor status go unnoticed. Antiendocrine or HER2-directed therapy might be ineffective, with the additional cost of side effects if the metastatic tumor loses its receptor positivity, whereas potential effective targeted therapy might be withheld if gain of receptor positivity remains undetected. However, conversion rates vary to a great extent in the literature. Furthermore, there is disagreement concerning an association of such changes with outcome and clinical implications.
In our study, we found that clinically used biomarkers were highly unstable between the primary tumor and the metastatic lesion. ER, PR, and HER2 status changed in 14%, 32%, and 15%, respectively. Although the change of HER2 status was not statistically significant, the percentage of discordant patients is clinically meaningful. These discordance rates are similar to the pooled random effects percentages of 19%, 31%, and 10% that Schrijver et al. reported [
6]. However, in contrast to previous reports [
6], we were unable to see any significant difference between site-specific discordance rates.
Patients who lost HR positivity had a significantly poorer prognosis than concordantly receptor-positive patients. Gain of HER2 positivity was associated with a significantly more favorable prognosis than concordantly negative receptor status which may reflect benefit from adjustment of therapy. This finding is in line with prior studies, underlining the importance of reassessing receptor status in the metastatic setting [
8,
9,
19‐
21].
There are several conceivable explanations for a discordant receptor status in breast tumors, including technical and analytical variability in immunohistochemical analysis, a heterogeneous tumor biology, and biological evolution.
Several studies reported varying concordance of results between the surgical specimen and core needle biopsy of ER (77.8 to 99%), PR (69 to 97%), and HER2 (64 to 97%), respectively, with particular emphasis placed on a general tendency for higher discrepancies in PR status [
22‐
27]. In addition to substantial improvement in receptor measurement accuracy, interlaboratory variation still exists in the assignment of receptor status [
28,
29]. The choice of method (such as IHC and RT-PCR), the assay method (dual-antibody vs single-antibody ER assay), as well as sample processing (e.g., decalcification reduced staining intensity especially in bone metastasis) may yield to discordant results [
29‐
31]. Moreover, the cut-off for ER/PR positivity may vary when histology results from different pathologists are compared and have changed over the last decades. Nevertheless, it is unlikely that receptor status conversion in our cohort is solely attributable to technical issues, as differences in OS depending on the ER and PR are biologically meaningful observations given the superior prognosis of hormone receptor-positive breast cancers [
32].
Breast cancers are genomically and transcriptomically heterogeneous tumors [
33‐
36]. This intratumoral heterogeneity is also observed for ER expression [
16]. Additionally, different metastatic locations seem to activate different gene expression profiles within metastatic cells [
37].
As tumors are heterogeneous and sequential biopsies are invasive procedures, increasing focus is being put on liquid biopsies. Here, sequential sample collection is not difficult and technological advances such as the circulating tumor cell detection or the characterization of cell-free tumor DNA in blood are promising approaches [
38,
39]. With continuous measurement, these biomarkers could represent current systemic tumor burden, monitor evolving tumor biology in real time, assess treatment efficacy, and, thus, guide therapy more comprehensively [
40‐
42].
In our collective, antiendocrine treatment was subject to change in 20% of patients and HER2-directed treatment in 14% of patients before and after metastasis biopsy. This is in line with previously reported fractions of 18–57% and 7–50% in other cohorts ( [
43‐
47]). Concluding, it is clinically relevant to detect changes in receptor status in the metastatic setting in order to tailor therapeutic interventions according to present tumor manifestation.
The validity of our data is limited and should be interpreted with caution, as in some cases more than one metastasis and/or local recurrences were subjected to biopsy. Moreover, in patients where treatment was carried out in closer to home clinics, therapeutic measures recorded here were mere recommendations, the adherence to which was not recorded. Therefore, causality of metastatic receptor status change and accordingly treatment change as documented in our dataset cannot be established. Moreover, Trastuzumab only received approval for treatment of metastatic breast cancer in 2000 and for adjuvant treatment in patients with early breast cancer in 2006 in Germany. Our dataset, however, also includes patients treated before those dates, which therefore introduces another confounding variable [
48].
Further limitations of our study are the retrospective design and the decentralized determination of receptor status. Thus, for some patients immunohistochemical analysis of the primary tumor and the metastasis was carried out in different laboratories. Furthermore, biopsies were collected and analyzed over a period of about 15 years, during which cut-off values for hormone receptor status were lowered. However, this approach reflects clinical reality, especially as endocrine or HER2-directed treatment was chosen due to the respective definition of ER-/PR and HER2 positivity.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.