Genetic alterations in plasma DNA such as mutations in proto-oncogenes, tumor suppressor genes, or aberrantly methylated DNA can be surrogate markers as at least some of the DNA could have derived from the primary tumor. For example, greater amounts of amplified MYC-N DNA were detected in patients with neuroblastoma compared with healthy individuals.(8) However, the detection of tumor-associated DNA in the plasma could indicate the presence only of circulating nucleic acids, not necessarily CTCs. It is not possible to determine whether the DNA came from CTCs or if it is simply being shed as free DNA from primary tumors. In addition, DNA has a much longer plasma half-life, the presence of circulating free DNA may reflect merely the presence of nucleic acids, not tumor cells. A further difficulty is that because changes such as DNA methylation occur early in tumorigenesis, such abnormalities might be detected in DNA shed from dysplastic lesions in patients who do not have a neoplastic lesion.(9) Although DNA-based CTCs detection is promising, there are many challenges to be overcome before it can be used in clinical practice.

The presence of CTCs can be indirectly estimated by the detection of tumor-associated mRNA using RT-PCR.(10) As RNA disappears quickly from the blood after cell death, the detection of RNA likely indicates the presence of a whole tumor cell.(11) Cytokeratin 19 seemed to be a good surrogate marker of circulating epithelial tumor cells.(12) Although RT-PCR has high sensitivity, the specificity is generally low, resulting in higher false positive rates due to contamination and expression of markers in normal cells.(13-15) Quantitative RT-PCR with multiple specific markers might resolve the issue of the specificity, but the target mRNA transcript needs to be carefully selected.(9,16) Other limitations of RT-PCR based assays include inability for further evaluation of morphological analysis and molecular profiling assays of CTCs.(17)

Cytometric approaches isolate individual CTCs, allowing further morphological and molecular characterization of the cells, including steroid receptor or HER2 status. Enrichment of epithelial tumor cells is generally needed to increase the sensitivity to an acceptable level because CTCs are rare event. In recent years, there have been many efforts to devise automated technologies.

To improve stratification of breast cancer patient, many issues remain unresolved, and new predictive and prognostic factors are needed in clinical practice. About 30% of node negative patients will present distant metastases, whereas 40% of the patients with involved axillary lymph nodes will survive for 10 yr without clinical recurrence.(28) The initial CTCs status in the patient's peripheral blood might be an important marker for the estimation of prognosis, and a negative prognostic impact associated with the detection of CTCs during the course of breast cancer is generally accepted. By stratifying the prognosis using CTCs, systemic treatment could be better targeted to patients who are most likely to relapse, thereby avoiding unnecessary toxic therapy for lowest risk patients.

CTCs detected by RT-PCR for CK19 mRNA have an independent prognostic value as a marker for poor clinical outcome in patients with breast cancers.(28-30) Pachmann et al.(31) reported a correlation between distant relapse and an increase in the number of CTCs. The presence of CTCs 3-4 weeks after surgery and before the initiation of chemotherapy was associated with shorter disease free survival and overall survival in patients with node negative breast cancer.(28)

In metastatic setting, Cristofanilli et al.(18) demonstrated that CTCs are highly prognostic. More than five CTCs per 7.5 mL of whole blood were identified as the threshold. Elevated CTCs at baseline predicted extremely short median progression free survival and overall survival. Even more interesting was that patients whose CTCs level was high at baseline and then was reduced to below five CTCs/7.5 mL of blood at the first follow up had similar progression free and overall survival compared to the group that had low levels of CTCs at baseline.(4,18,32) This result suggested that CTCs provided independent prognostic information beyond tumor burden. Furthermore, assessment of CTCs was suggested to be a superior surrogate endpoint than current radiology imaging studies for predicting overall survival in patients with metastatic breast cancer.(33) Other researchers also found that a detectable level of CTCs appeared to correlate with the worse clinical course in patients with metastatic disease.(20,34)

However, although a potential role of CTCs as an independent prognostic factor is generally accepted, the future success of CTCs test competing with conventional prognostic markers or growing data of multigene assay might be controversial.

Longitudinal monitoring of the CTCs status could be an effective strategy for measuring the therapy efficacy and resistance and monitoring treatment response. Early indicators of response or resistance to treatment are an important issue in clinical practice. In addition to being a sign of tumor cell dissemination, increases and decreases in their number may also serve as a surrogate marker for response to therapy.(35) Furthermore, monitoring CTCs in blood would be an ideal strategy because its noninvasive sampling procedure allows repeated analyses.

For example, an increase by 10-fold or more at the end of therapy by MAINTRAC® analysis is a strong predictor for relapse.(36) Although patients could be monitored for therapy response by longitudinal CTCs assay, the ASCO group concluded that the measurement of CTCs was not to be used to influence treatment decisions in patients with breast cancer in the 2007 update of their recommendations.(37)

The Southwest Oncology Group and the Breast Cancer Intergroup of North America are now conducting a prospective trial in which patients with metastatic breast cancer who have a positive CTC count using the CellSearch system after 1 cycle of first-line chemotherapy will be randomly assigned to a group that will continue with the current therapy or to a group that will switch the current therapy to a different chemotherapeutic agent (Southwest Oncology Group protocol S0500). This prospective trial could solve the uncertainty in the possibility of treatment monitoring by CTCs assay.

The courses of neoadjuvant chemotherapy could influence the number of CTCs. In one neoadjuvant trial, the reduction in the number of CTCs with chemotherapy strongly predicted for the final tumor size, suggesting that CTCs could potentially be used to monitor the response to primary chemotherapy.(38) The correlation between the reduction in CTCs and the final reduction in tumor size indicates that monitoring of CTCs could predict tumor responsiveness of particular therapy and avoid unnecessary toxicity during neoadjuvant therapy by discontinuing an ineffective therapy.

However, the role of CTCs in neoadjuvant setting is unclear. For clinical use of CTCs in neoadjuvant setting, we need long term follow up data about the correlation between a change in the number of CTCs and patients prognosis.

Recently, many investigators have reported CTCs detection methods in breast cancer patients, including one that uses different volumes of peripheral blood and others that use different enrichment procedures (density centrifugation, filtration, immunomagnetic beads), different methods (RT-PCR, immunocytochemistry, immunofluoresence) or epithelial tumor markers (cytokeratins, mucins, mammaglobin, etc.) Thus, it is extremely difficult to compare one study with others.

In CTCs research, the main challenge lies in development of an optimized and standard assay platform with a reliable sensitivity, specificity and reproducibility. Once the optimum assays are developed, techniques validated in the laboratory should be taken into clinical setting in the context of meticulously designed trials.

Ascertainment of the phenotype of CTCs also has a potential role in tailored treatment. CTCs can exhibit different characteristics compared to the primary tumors, and molecular characterization of CTCs might contribute to improving more individualized therapies.(5,39)

Many investigators are developing methods to evaluate biologically important markers in CTCs. treatment-related markers such as estrogen receptor (ER) and HER2 can also be detected in CTCs, and these phenotypic markers are now being investigated as possible biological indicators of target modulation. For example, HER2 status of CTCs might be different from that of the corresponding primary tumors.(40-42) The fact that trastuzumab is known to have a substantial positive impact on the survival of patients with HER2 positive disease, monitoring of CTCs for HER2 could help identify patients who would not otherwise receive trastuzumab due to a negative HER2 status on the primary tumor to benefit from this therapy. In terms of ER status, there has been very limited data about comparison of the expression between primary tumor and CTCs, which might be useful in predicting the benefit of endocrine therapy.(43) in addition, gene-expression profiling of ctcs might be feasible in decision making of tailored individual treatment strategies in the near future.

Finding tumor cells in peripheral blood raises questions about the possibility of the existence of subset of cells sharing many characteristics with cancer stem cells (CSCs). According to the CSC hypothesis, at least some of the CTCs should have properties of metastatic potential.(44) The majority of CTCs lack the ability to metastasize. It appears that the potential to grow into metastases may be restricted to a small fraction of stem cells.(45) Being a heterogeneous population, some of these cells exhibit a more aggressive behavior by generating early metastases, and CTC subpopulations may be identified and used as predictive tools to measure patient response to targeted agents.(46)