Circulating tumor cells in breast cancer: Advanced tools for “tailored” therapy?

Alterations in oncogenes and suppressor genes are critical steps in carcinogenesis and tumor progression in solid and hematological malignancies. The development of therapeutic strategies that interfere with the functional pathways regulated by these genes represents the most successful example of targeted therapy (e.g., imatinib in chronic myelogenous leukemia and trastuzumab in breast cancer) (1, 2). In this issue of PNAS, Meng et al. (3) raise interesting possibilities for exploring circulating tumor cells (CTCs) for personalizing targeted therapy of cancer in the future.

Alterations of members of the ErbB family of receptors are frequent in breast cancer and have been extensively investigated (4). HER-2 (also known as erbB-2 or neu) is a protooncogene that encodes a 185-kDa tyrosine kinase glycoprotein (4). Amplification of the HER-2 gene plays an important role in the pathogenesis of breast cancer and is associated with a poor prognosis in patients with axillary node-positive breast cancer (5). Treatment with trastuzumab (Herceptin), a humanized monoclonal antibody directed against the extracellular domain of HER-2/neu, has significantly improved clinical responses to chemotherapy and outcomes in primary and metastatic breast cancer (6, 7). Interestingly, several investigators have recently reported that the HER-2/neu status of tumor cells detected in metastatic sites, bone marrow, or peripheral blood of patients with advanced disease may differ from the original primary tumor, suggesting either a clonal selection or genetic instability (8, 9). Furthermore, the presence of concomitant alterations of other oncogenes [e.g., topoisomerase-2 α (TOPO-2α) and urokinase plasminogen activator (uPA)] or the truncated form of the receptor, p95HER2, appear to better define biological subsets of the disease and may contribute to decisions on treatment selection (10–13). Therefore, there is necessity for a “real time” testing of HER-2 status and other abnormally functioning genes in tumor cells, either from tissue or possibly from peripheral blood. These analyses also can contribute to identifying additional molecular markers that may aid in selection of therapeutic targets.

Discussion

CTCs can be detected and isolated from the peripheral blood of women with a diagnosis of breast malignancy, including women without current evidence of disease, a phenomenon called “tumor dormancy” (14). Their significance in predicting clinical outcome in the absence of disease is unknown.

CTCs detected in patients with both localized and metastatic breast cancer are significantly associated with worse outcome (15–17). In the study by Cristofanilli et al. (15), 177 patients with metastatic breast cancer were tested for the presence of CTCs. Sixty percent had at least two CTCs in 7.5 ml of whole blood, 49% had more than five, and 21% had more than 50. Thus, CTCs are a frequent phenomenon in these patients. The median survival was >18 months for patients with fewer than five CTCs and 10.1 months for those with at least five CTCs, with a P < 0.001. The cutoff point of five CTCs had been prospectively identified in patients in a training set and confirmed in different patients in a validation set. The presence of more than five CTCs was an independent predictor of survival.

The prognostic implication of detecting CTCs before initiating therapy for breast cancer raises important questions about the biological characteristics of these cells and the reasons for the reduced capacity of systemic treatments to arrest or eradicate the cancer. The report by Meng et al. (3) is the fourth in a series from this group exploring these questions and expands on their previous observations. The authors analyzed CTCs from the blood of patients with newly diagnosed, advanced breast cancer and from patients with recurrent breast cancer. The goals were to assess the value of measuring gene status in CTCs compared with cancer cells in the primary tumor tissue and to explore amplification of the uPA receptor gene.

The results were interesting and raise the possibility that circulating tumor cells may be representative of cells in tumor tissue, which of course would greatly enhance our capacity to monitor the presence of genetic abnormalities and changes that occur during the course of the disease. Gene expression was measured by immunofluorescence, and gene copy number was assessed by FISH analysis. The investigators first compared the status of HER-2 and uPA receptor genes in tumor tissue and tumor touch preps (TPs), finding excellent concordance. They conclude that TPs can be used to determine the status of these genes in patients, a useful finding that needs to be confirmed with other genes and in a larger series.

When the levels of HER-2 and uPA receptor gene copy number and expression were compared between CTCs and the primary breast cancer (determined on the tissue or on TPs from the tissue) there was very high concordance using the investigators' scales and definitions of positive versus negative, provided that acquisition of HER-2 gene amplification associated with cancer progression was taken into consideration. The results raise a number of interesting questions and possibilities.

First, heterogeneity of gene copy number and expression was observed among the CTCs from single patients (scoring at least 10 CTCs). This finding confirms heterogeneity among breast cancer cells from a single patient but cannot address the issue of whether the set of CTCs came from one or many sites in a single tumor mass or from one or many sites throughout the body. This question is critical because it is widely accepted that cancer cells that have spread to different metastatic sites express different genes and may harbor different genetic abnormalities.

Second, the CTCs studied by these investigators came from patients with recurrent disease and from patients with progressive advanced breast carcinoma who were in their initial episode. In all cases, the comparison was with the original primary cancer. It would be worthwhile to use this technology to compare the status of genes in CTCs with genes in the primary tumor in patients that initially have progressive advanced disease and, separately, in patients with recurrence of disease, to pursue the question of changes in gene expression between initial presentation and development of recurrence.

The results of the experiments reported by Meng et al. (3) suggest that clinical trials testing correlations between gene status data obtained from CTCs before treatment and the responses of patients to various therapeutic regimens might lead to diagnostic tests that can select the therapy most likely to be effective for an individual patient. Other interesting observations that deserve follow-up were that gene amplification/copy number correlated well with gene expression at the protein level in CTCs and TPs from tumor tissue, a finding that has not always been observed in studies performed on cells in the breast cancer tissue, and that coamplification of HER-2 and uPA was observed in the same CTCs and TP cells, whereas there was lack of amplification of both genes in other cells. The latter observation suggests that alterations in copy number of these two genes may be coregulated, which could have implications for treatment with a combination of targeted therapies.

CTCs provide unique access to samples of a patient's cancer for performing several biological analyses, including gene copy number by FISH, gene expression at the level of mRNA, and gene expression at the protein level by immunohistochemistry (9, 18–20).

The frequent clinical problems of resistance to standard chemotherapies and progression of disease indicate the importance of the opportunity to evaluate CTCs as potential noninvasive tools for improving selection of individualized cancer therapy. The current report from Meng et al. (3) has demonstrated the feasibility of evaluating HER-2 and uPAR status by analyzing single cells collected from peripheral blood of patients using enrichment methods. Their analyses, if confirmed for other genes and in a larger series, represent a significant refinement of traditional pathological diagnostics. Only prospective trials will determine whether CTCs from patients with either primary or recurrent disease are representative of the behavior and drug sensitivities of the cancer in the patient's tissues. To date, there only have been limited studies focusing on the possibility of selecting treatment on the basis of CTCs analysis (9).

Conclusion

In summary, we believe that a comprehensive analysis of the properties of CTCs is likely to provide new insights into the biology of breast cancer and contribute to defining novel treatments and better prediction of clinical benefit. Although further prospective validation will be necessary before practical applications of CTC analyses in the clinic can occur, the sophisticated analysis of single cells offered by application of novel technologies has opened the possibility of a new era in the development of more “tailored” and personalized therapies.