Liquid biopsy to test new treatment strategies in breast cancer: are we there yet?

Breast cancer is the most common cancer in women in Europe [1]. Recommendations for systemic adjuvant therapy are currently based on the analysis of primary tumor characteristics [2]. For example, the administration of trastuzumab in women with breast cancer relies on the analysis of HER2 protein overexpression or HER2 gene amplification in the primary tumor [3]. However, the target of adjuvant systemic therapy is thought to be the minimal residual disease (MRD), which is present after primary surgery but undetectable by currently used conventional imaging approaches [4]. Recent technological advances have enabled the detection and characterization of bone marrow disseminated tumor cells (DTCs) and peripheral blood circulating tumor cells (CTCs), which are thought to be surrogate markers of MRD. Clinical trials studying the effect of different treatments on DTCs or CTCs have been recently reported. In two studies, the effect of a bisphosphonate, zoledronic acid, on bone marrow DTCs, was examined [5, 6]. The investigators used novel end points such as the difference in the detection rate of DTCs after a predefined treatment period between a control arm of (neo) adjuvant chemotherapy and an experimental arm of the same chemotherapy plus zoledronic acid. Interestingly, both studies showed that the detection rate of DTCs was lower after treatment with zoledronic acid, supporting the hypothesis that zoledronic acid antitumor efficacy might partly depend on DTC elimination. Both clinicians and patients more easily accept a blood draw than a bone marrow aspiration. Therefore, CTCs could potentially be used as a liquid biopsy to assess new treatment strategies. This innovative approach was used in the study by Georgoulias et al. [7], in which the decision about secondary adjuvant treatment was based on the results of a liquid biopsy rather than primary tumor analysis. The authors should receive credit for initiating such a study in 2003. In the trial, they used a RT-PCR to detect cytokeratin (CK) 19 messenger RNA (mRNA) in blood samples from patients with early breast cancer who were receiving adjuvant chemotherapy [7]. They screened 378 patients to identify 100 (26%) with detectable CK19 mRNA both before and after they received chemotherapy. Only patients whose blood was CK19 mRNA positive at both timepoints were randomized between six cycles of trastuzumab versus observation. The authors reported the analysis for patients with HER2negative primary tumors (36 randomized to trastuzumab and 39 to observation). Of the women who received trastuzumab, 27 (75%) tested CK19 mRNA negative, compared with only 7 (18%) in the observation arm. At a median follow-up of 67.2 months, patients on the observation arm had an almost four times higher risk of relapse than the patients who had received treatment [7]. The authors also carried out double immunofluorescence staining for CK and HER2 in peripheral blood mononuclear cells (PMBCs) cytospins from a subset of the patients showing CK19 mRNA positivity. They found that 51 (89%) of those 57 patients had CK-positive/HER2-positive cells, suggesting that the trastuzumab benefit was associated with their presence. Although the results of this trial are intriguing, it should be noted that only 75 patients were randomized, and independent validation is needed. Moreover, there are concerns about the methodology used for CTC detection. No independent laboratories have validated the sensitivity and specificity of the method for CTC detection in early breast cancer using the same procedure to identify CK19 mRNA in peripheral blood. Another group of investigators have shown that CK19 mRNA can be expressed in normal PBMCs, particularly in the lymphocyte population. By using immunomagnetic enrichment for epithelial cells, these investigators were able to reduce the level of background signals to <5% in the PBMCs of healthy donors [8]. However, in the study by Georgoulias et al., no such immunomagnetic enrichment for epithelial cells was carried out. Instead, CK19 mRNA was detected by RT-PCR in RNA extracted from PBMCs, hence raising concerns about the specificity of this method for CTC detection. The only technology that has received Food and Drug Administration approval for CTC detection as an aid in monitoring women with metastatic breast cancer is CellSearch® (Veridex, Warren, NJ) [9]. This is a semi-automated system that first uses immunomagnetic separation to isolate epithelial cell adhesion molecule (EpCAM)-positive cells from whole blood followed by an immunofluorescent staining of captured cells with antibodies specific for CKs 8, 18, 19 (pan-CK) and CD45 (specific for leucocytes). The same cells are also stained with 4⍰,6-diamidino-2-phenylindole-2 (DAPI; to confirm the presence of a cell nucleus). A CTC is defined as an EpCAMpositive cell staining for pan-CK and DAPI but not for CD45. Two independent studies have validated the analytical performance of CellSearch® for clinical use in metastatic breast cancer [10, 11]. Data on the role of CTC detection and characterization using CellSearch in nonmetastatic cancer are only recently emerging [12–15]. An international collaborative study involving 30 readers from 15 academic and 2 industry labs is currently being carried out to harmonize CTC ed ito ria ls editorials Annals of Oncology 23: 1653–1655, 2012 doi:10.1093/annonc/mds111 Published online 19 April 2012