HER2-targeted therapy influences CTC status in metastatic breast cancer

Purpose As an independent, negative-prognostic biomarker for progression-free survival (PFS) and overall survival (OS), circulating tumor cells (CTCs) constitute a promising component for developing a liquid biopsy for patients with metastatic breast cancer (MBC). The effects of HER2-targeted therapy such as trastuzumab, pertuzumab, T-DM1, and lapatinib on CTC status and longitudinal enumeration were assessed in this trial. Methods CTC status of 264 patients with MBC was analyzed prior to and after 4 weeks of a new line of palliative systemic therapy. CTCs were assessed using CellSearch®. Three groups were compared: patients with HER2-positive MBC receiving ongoing HER2-targeted therapy (n = 28), patients with de novo HER2-positive MBC and no HER2-targeted therapy in the last 12 months prior to enrollment and start of HER2-targeted therapy (n = 15), and patients with HER2-nonamplified disease and no HER2-targeted therapy (n = 212). Results Positive CTC status (≥ 5 CTC/7.5 ml blood) at enrollment was observed in the 3 groups for 17.9, 46.7, and 46.2% (p = 0.02) of patients, respectively. At least one CTC/7.5 ml was seen in 28.6, 53.3, and 67.0% (p < 0.001) of these patients. Furthermore, 3.6, 40.0, and 3.3% (p < 0.001) of the patients had at least one HER2-positive CTC. After 4 weeks of therapy 7.1, 0.0, and 31.1% (p = 0.001) of patients had still a positive CTC status (≥ 5 CTC/7.5 ml blood). At least one CTC/7.5 ml was still observed in 25.0, 20.0, and 50.5% (p = 0.004) of the patients. Furthermore, 7.1, 0.0, and 1.9% (p = 0.187) had at least one HER2-positive CTC. After 3 months of therapy, 35.7, 20.0, and 28.3% (p = 0.536) showed disease progression. Conclusions HER2-targeted therapy seems to reduce the overall CTC count in patients with MBC. This should be taken into account when CTC status is used as an indicator for aggressive or indolent metastatic tumor disease.


3
Background Worldwide, metastatic breast cancer (MBC) is a major cause of cancer-related death in women [1][2][3]. MBC, also called stage IV breast cancer, is considered to be not curable [4]. The intent of therapies is therefore palliative and meant to stabilize the disease with tolerable side effects for the patients. The human epidermal growth factor receptor 2 (HER2) gene is overexpressed in about 10-30% of patients with invasive breast cancer [5][6][7]. In the past, HER2 gene expression was correlated with poor clinical outcome in early breast cancer and metastatic disease alike [8][9][10]. Since the development of HER2-targeted therapies such as trastuzumab, pertuzumab, T-DM1, and lapatinib, prognosis has changed dramatically. In hormone-receptor positive primary breast cancer, HER2 positivity now even represents a favorable predictor for overall survival (OS) [5]. Newfound HER2 overexpression in biopsies of metastases allow additional, well-tolerated therapy options in the metastasized situation.
Apart from solid metastases, a liquid biopsy as detector and surrogate of the systemic tumor burden is needed to take into account the heterogeneity of the disease [11]. As an independent, negative-prognostic biomarker for progression-free survival (PFS) and OS, circulating tumor cells (CTCs) constitute a promising component for developing a liquid biopsy for patients with MBC [12][13][14]. Indeed, CTCs are a versatile tool in clinical therapy management and can distinguish between aggressive and indolent metastatic tumor disease [15][16][17]. CTC monitoring as a prognostic tool in MBC has therefore been introduced in several guidelines such as ASCO and AGO [18,19]. However, CTCs could not yet show in clinical trials to be predictive for clinical benefit when used to guide decisions on systemic therapy [18]. Independent of clinical and molecular variables, ≥ 5 CTCs per 7.5 ml blood are regarded as the threshold for stratification [20].
As CTCs reflect a subpopulation of the total tumor cell population, characterization and treatment of CTC might be a promising tool for optimizing therapy [21]. Meng et al. demonstrated the presence of HER2-positive CTCs in patients with HER2-negative primary tumors [22]. HER2-positive CTC even proved to be a relevant prognostic factor, independent of primary tumor and metastatic phenotype [23]. However, the therapeutic and predictive relevance of HER2-positive CTC phenotypes is still the subject of controversial discussion [24], and more clinical trials are needed to evaluate their clinical significance.
Retrospective studies and xenograft models suggest that HER2-targeted monoclonal antibody therapies might even be able to target both HER2-nonamplified cancer cells and cancer stem cell populations via antibody-dependent, cell-mediated cytotoxicity (ADCC) [25]. In a phase-II trial, trastuzumab was able to eliminate CTCs independent of HER2 status and decreased the incidence of clinical relapses [26]. On the other hand, another phase-II trial with the intention to treat CTCs in HER2-nonamplified, nonmetastatic breast cancer had to be stopped and showed that trastuzumab does not decrease the detection rate of CTCs [27].
To further investigate these controversial findings, this retrospective study was conducted to demonstrate the immediate effect of HER2-targeted therapies on CTCs in the metastatic setting.

Patients and methods
The CTC study of the National Center for Tumor Diseases (NCT) enrolls patients before administering a new line of systemic therapy when MBC or progressive disease (PD) of MBC is diagnosed. Blood draws are performed at enrollment and 4 weeks after starting a new line of systemic therapy. Since March 2010, 505 patients were enrolled (Fig. 1). In this retrospective trial, all patients for whom CTC data were recorded at baseline and after 4 weeks of therapy were included (n = 264). Exclusion criteria were: no blood draw after 4 weeks of therapy (n = 207), no blood draw at enrollment (n = 16), patients who were already included in the trial (n = 13), no follow-up information (n = 9), no metastatic disease (n = 3), or withdrawal of patient's consent to participate in the study (n = 2). Furthermore, ≥ 5 CTCs per 7.5 ml peripheral blood was defined as CTC-positive [20]. Criteria for inclusion were measurable progressive metastatic disease according to the Response Evaluation Criteria in Solid Tumors (RECIST) criteria [28], age > 18 years, and written informed consent to participate in the study. Ethical approval was obtained from the Ethics Committee of the Medical Faculty of the University of Heidelberg, approval no. S-295/2009. The study population was divided into 3 groups: patients with HER2-positive MBC with HER2-targeted therapy in the new and previous therapy line ("HER2 therapy", n = 28); patients with HER2-positive MBC and no HER2-targeted therapy in the last 12 months prior to enrollment and start of HER2-targeted therapy ("New HER2 therapy", n = 15); and patients with HER2-nonamplified disease and no HER2targeted therapy ("No HER2 therapy", n = 212). The "New HER2 therapy" group included patients that were pretreated with chemotherapy and/or endocrine therapy for MBC, since 5-20% of patients show newfound HER2 overexpression in biopsies of metastases [29][30][31].
Therapy response was evaluated every 3 months via CT and/or MRI scan and categorized according to the Response Evaluation Criteria in Solid Tumors (RECIST) as progressive disease (PD), stable disease (SD), complete remission (CR), or partial response (PR) [28].
Clinical characteristics of the cohort were described as absolute and relative frequencies for binary and ordinal variables, as well as mean and 95% confidence intervals for continuous variables. Mean PFS and OS times were estimated counting from the timepoint of study enrollment. Differences between groups were analyzed by Chi-squared tests (categorical data) and t tests (continuous variables). Statistical analyses were performed using R (version 3.6.0) [41]. Figures were generated using Microsoft Office Version 16.30. Since this is an exploratory study, p values should be interpreted in a descriptive sense. p values smaller than 0.05 were defined as significant.

Discussion
In this analysis, the CTC status of patients with MBC, stratified by ongoing therapy and previous treatment, was investigated. The point prevalence of CTC at enrollment was highest in patients receiving ongoing palliative chemotherapy for HER2-nonamplified stage IV breast cancer. Throughout all therapy groups, a decrease in CTCs was observed after 4 weeks of a new line of palliative therapy (Figs. 2 and 3). Even for the patients in whom disease had progressed after 3 months of therapy, a tendency for CTCs to decrease is evident, as demonstrated by the gray bars in Figs. 2 and 3. As described in multiple studies previously, less reduction or persistence of CTCs under therapy correlates with PD [13,16,39,42,43]. Therefore, patients were not only divided in groups with < or ≥ 5 CTCs but also in < or ≥ 1 CTC at enrollment and after 4 weeks of therapy. Looking at the HER2-nonamplified patients, divided by palliative therapy just before enrollment, the subgroups with the highest rates of PD had also higher rates of CTCs at the time of enrollment (  [45]. These findings were also confirmed by our results. The impact of CTC status in de novo MBC and the correlation between first-line HER2-targeted therapy vs. first-line endocrine and chemotherapy is apparent when comparing the "New HER2 therapy" group with the "No therapy before enrollment" group of the HER2-nonamplified subgroup; the prevalence of ≥ 1 CTC at enrollment was similar (53.3 vs. 53.1%) but the rate of positive CTC status (≥ 5 CTC) at enrollment was higher in the "New HER2 therapy" group (46.7 vs. 30.6%). After 4 weeks of therapy, ≥ 1 CTC were found in 20.0 vs. 38.8% of patients and a positive CTC status was observed in 0 vs. 22.4% of patients. PD rate was similar at 20.0 vs. 20.4%. PFS and OS were better in the "New HER2 therapy" group (14.5 vs. 9.9 months, 42.7 vs. 31.0 months, respectively).
As other retrospective studies and xenograft models have shown that HER2-targeted monoclonal antibody therapies might even be able to target HER2-nonamplified cancer cells and cancer stem cell populations including CTCs, the indication for HER2-targeted therapies might need to be extended [25,26]. Antibody-dependent, cell-mediated cytotoxicity (ADCC) might impact a subgroup of patients showing heterogenic tumor spread, represented by CTCs. Nevertheless, these findings also demonstrate that low rates of CTC levels do not necessarily correlate with better prognosis in crosstherapy comparisons. Negative phase-II trials, such as the TREAT-CTC trial, with the intention to treat CTCs of HER2nonamplified patients, illustrate the complexity of the matter [27]. The clinical significance of the CTC phenotype for guiding therapeutic decisions is currently being investigated in the DETECT studies [46].
The clinical relevance of CTC status on prognosis, including HER2-positive patients, is undisputed [15]. These findings lead us to assume that the CTC status might be strongly affected by HER2-targeted therapies but does not necessarily correlate with prognosis in patients with HER2-positive de novo MBC as compared to HER2-nonamplified de novo MBC. Furthermore, the CTC status of HER2-positive patients needs to be evaluated according to the therapy situation (before vs. ongoing anti-HER2 therapy). This is underlined by the observation that patients receiving HER2-targeted therapy have, independent of the prognosis, significantly lower levels of CTCs than patients receiving chemotherapy, endocrine therapy, or no therapy ( Fig. 2 and 3). This overall decrease in CTC count must be considered when comparing different therapy groups.

Limitations
One main limitation lies in the fact that this is a retrospective analysis. Another limitation is the small number of analyzed patients in the "HER2 Therapy" and "New HER2 therapy" groups. Indeed, the small number of patients with PD highlights the implicit difficulties in analyzing the rate of CTCpositive cases. The subgroup receiving anti-HER2 therapy can therefore only be reported descriptively and used to generate new hypotheses about therapy effects on CTC levels.

Conclusions
First-line HER2-targeted therapy of metastatic breast cancer seems to reduce CTC levels greater than endocrine or chemotherapy. Ongoing anti-HER2 therapy seems to be associated with lower overall CTC levels.