Circulating Tumor Cells as Predictors of Response and Failure in Breast Cancer Patients Treated with Preoperative Chemotherapy

Abstract

Aim

To explore the significance of circulating tumor cells (CTCs) detection in the course of preoperative chemotherapy (PC) and their effect on the outcomes.

Methods

Fifty-five patients with stage II/III invasive breast cancer were enrolled into a preoperative clinical trial. Patients were given PC with sequential single-agent doxorubicin and paclitaxel vs paclitaxel followed by doxorubicin. Blood samples (8 mL) were collected from patients before PC, after each phase, and at 6 months intervals during follow-up. Peripheral blood mononuclear cells were isolated and enriched for epithelial cells. Quantitative RT-PCR was used to determine the presence of cytokeratin 19 (CK19) mRNA. Samples were considered positive when the PCR curve crossed the standard threshold curve.

Results

After the first phase of chemotherapy, there was a 59% overall reduction in the median tumor volume. The percentage of volume reduction did not differ between patients who presented with detectable CTCs at baseline and those who did not (p=0.89). After the second phase of chemotherapy, there was a further decrease in the median tumor volume to 93% from baseline. There was no correlation between the lack of response and the presence of CTCs either after the first (p=0.36) or second (p=0.5391) phases of PC. The presence of CTCs was a predictor of local or distant relapse (p=0.0411). The detection of CTCs did not affect overall survival (p=0.2569).

Conclusion

CTCs can be used as predictors of relapse after definitive treatment of locally advanced breast cancer; however, CTCs detection in peripheral blood during the course of PC does not implicate a particular pattern of response to treatment.

Keywords

Breast cancer Circulating tumor cells Preoperative chemotherapy Reverse transcription polymerase chain reaction

Introduction

Circulating tumor cells (CTCs) can be detected in the peripheral blood of 10%-40% of patients with breast cancer, depending on the clinical setting and the method of detection (1 -5). However, the clinical significance and utility of CTCs remain unclear. Studies of CTCs in patients with metastatic breast cancer suggest that the presence and quantity of CTCs can predict disease progression and survival; however little is known about the relevance of CTCs in patients with earlier stages of the disease in the preoperative setting (6 -9).

Preoperative chemotherapy (PC) is a generally accepted standard-of-care treatment for locally advanced breast cancer and has become an acceptable alternative for operable breast cancer (10, 11). Predicting patients' responses to a PC regimen through the detection of CTCs in peripheral blood seems to be a very exciting application that could allow changing the line of treatment without the need for repeated biopsies.

The aim of this study was to determine the ability of CTCs to predict the response to PC or failure after treatment in patients with stage II/III breast cancer.

Materials and Methods

Patient cohort

From June 2000 to May 2004, 62 patients with stage II/III invasive breast cancer were enrolled into a phase II randomized trial of PC with sequential single-agent doxorubicin (4 cycles at 60 mg/m² every 2 weeks) and paclitaxel (9 cycles at 80 mg/m² weekly). Patients were randomized to receive doxorubicin followed by paclitaxel (group 1) or paclitaxel followed by doxorubicin (group 2). The study was approved by the institutional review board of the Massachusetts General Hospital and the Dana-Farber Cancer Institute. Written consent was obtained from each patient (Clinicaltrials/gov NCT00096291).

Per protocol, MRI and ultrasound measurements of breast tumor size were performed before and after each treatment phase. The MRI-measured tumor volume was used as a surrogate for clinical response because it has been shown to be reliably and accurately correlated with the histologically measured tumor size (12, 13) and the final histopathologic response (14, 15). In addition, blood samples were collected from all patients for detection of CTCs as a pre-planned correlative analysis.

Circulating tumor cell collection and measurement

Several methods have been developed for the detection of CTCs, among these the most widely used include the reverse transcription-polymerase chain reaction (RT-PCR), which seems to be more sensitive, and the immunomagnetic/fluorescent approach, which seems to be more specific (2).

In the current study we used a mixture of both approaches through an initial immunoseparation step followed by the RT-PCR step. In a trial to achieve a reasonable balance between sensitivity and specificity, we used 2 types of markers for the detection of CTCs. The first was cytokeratin 19 (CK19), which is considered a marker of high sensitivity but low specificity for breast cancer cells (1 -5), the second marker was BS106, also known as Small Breast Epithelial Mucin (SBEM), which exhibits a high specificity for breast cancer cells (16). It has been previously shown that a panel of markers, rather than a single marker gene, decreases the likelihood of false positive results (17).

Blood samples were collected from 55 patients prior to starting chemotherapy, after the first chemotherapy drug (doxorubicin or paclitaxel), after the second chemotherapy drug or prior to surgery, and every 6 months during follow-up. The mononuclear cell fraction from 8 ml of blood was isolated from CPT tubes using gradient centrifugation, it was washed, counted, and frozen down in 10% DMSO using a rate-controlled freezer. Cells were stored at −140°C. Circulating tumor cells were enriched by immunomagnetic selection using BerEP4 antibody (anti-EpCAM) attached to Dynal magnetic beads (Dynabeads^®). RNA was purified using Qiagen RNeasy kits. RNA was amplified by RT-PCR specific for CK19, BS106 and beta-2-microglobulin (B2M). Amplification was detected in real time using Beacon probes. Blood samples were considered positive if the PCR curve crossed the threshold before 45 cycles.

Statistical analysis

The study of CTCs measurements, their changes and correlations with tumor response and clinical outcomes was conceived as a secondary pre-planned correlative analysis of clinical trial data; therefore the sample size of the clinical trial (originally planned as n=100, but accomplished as n=62, the protocol was stopped when new types of chemotherapy became available) was motivated not by this analysis but rather by the trial primary aim. Nevertheless, the sample size of the present secondary study (n=55, the number of patients with available blood samples) is large enough to detect moderate quantitative changes in CTCs levels over time and moderate correlation of CTCs with quantitative tumor response endpoints, such as measurements of tumor volume. For example, detection of a correlation corresponding to a Kendall's tau correlation coefficient of 0.3 (a moderate effect) requires, with power of 80% and type 1 error of 5%, a sample size of just n=39 (18). Similarly, a paired Wilcoxon test requires n=42 subjects to detect a change, assuming power of 80% and type 1 error of 5%, as to detect the probability of 75% for a negative sum of 2 CTCs changes (18).

Results

Patients, Treatment and Tumor Characteristics

Fifty-five patients had samples collected (29 in group 1 and 26 in group 2) and processed for CK19, BS106 and B2M. Clinical characteristics of these patients are summarized in Table I. No significant differences in clinical variables were found between the study groups, either in our cohort or in the parent study.

Table I

PATIENTS, TREATMENT AND TUMOR CHARACTERISTICS ACCORDING TO THE TREATMENT ARM

	Group 1 Doxorubicin first (n=29)	Group 2 Paclitaxel first (n=26)	All	P value
Age
Median	50	47	48	0.1429
Range	47 - 54	41 - 51	41 - 54
Clinical T stage				0.7828
T2	19 (65.5%)	18 (69.2%)	37 (67.3%)
T3	10 (34.5%)	8 (30.8%)	18 (32.7%)
Clinical N stage				0.4483
N0	16 (55.2%)	17 (55.4%)	33 (60%)
N1	12 (41.4%)	9 (34.6%)	21 (38.2%)
N3	1 (3.4%)	0 (0%)	1 (1.8%)
MRI tumor volume				0.7762
Median (CC)	11. 5	13	12.6
Range	7.5 - 35.3	6.3 - 33.2	6.3 - 35.3
Estrogen receptor status				0.3387
Positive	21 (72.4%)	22 (84.6%)	43 (78.2%)
Progesterone receptor				0.2393
Positive	19 (65.5%)	21 (80.8%)	40 (72.7%)
Patients with CTCs detected by CK19	10 (34.5%)	8 (30.8%)	18 (32.7%)	1.000
Patients with CTCs detected by CK19 and BS106	5 (17.2%)	3 (11.5%)	8 (14.5%)	0.7084

The presence or absence of CTCs in peripheral blood was not significantly associated with age, tumor size, nodal status, MRI-measured tumor volume, ER or PR status (Tab. II).

Table II

PATIENT AND TUMOR CHARACTERISTICS ACCORDING TO BASELINE CTCs

	CTCs present at baseline (n=37)	CTCs absent at baseline (n=18)	All (n=55)	P value
Age				0.1586
Median	48	51	48
Range	41 - 53	46 - 54	41 - 54
Clinical T stage				0.7613
T2	24 (64.9%)	13 (72.2%)	37 (67.3%)
T3	13 (35.1%)	5 (27.8%)	18 (32.7%)
Clinical N stage				1.000
N0	22 (59.5)	11 (61.1%)	33 (60%)
N1	15 (40.5%)	6 (33.3%)	21 (38.2%)
N3	0 (0%)	1 (5.6%)	1 (1.8%)
MRI-measured tumor volume				0.7156
Median	12.6	13	12.6
Range	7 - 35.3	6.3 - 28.6	6.3 - 35.3
Estrogen receptor status				0.2983
Positive	27 (73%)	16 (88.9%)	43 (72.7%)
Progestrone receptor status				0.7492
Positive	26 (70.3%)	14 (77.7%)	40 (72.7%)

At baseline, 18 (32.7%) out of the 55 patients had CTCs detected in their peripheral blood. There were no differences in the rate of baseline CTCs detection between treatment groups (p=1.000). Using a pre-established cutoff for CTCs detection, 1 (3.3%) out of 30 healthy donors had a positive blood sample.

Results after the First Phase of Chemotherapy

After the first phase of chemotherapy the percentage of patients with CTCs detected by CK19 decreased from 32.7% to 26% in the entire cohort; however this was not statistically significant (p=0.4692). In group 1, CTCs were detected in the peripheral blood of 23% (n=6) of patients, with only 8% (n=2) of those patients having CTCs detected at baseline and 15% (n=4) with newly detected CTCs after the end of the first phase of chemotherapy. In group 2, CTCs were detected in 30% (n=6) of the cases after the end of the first phase of chemotherapy, with 15% (n=3) of the patients being CTCs-positive at baseline and 15% (n=3) with newly detected CTCs.

There was no statistically significant difference between treatment groups with regard to the total number of patients with detected CTCs after the first phase of treatment (p=0.7379).

There was an overall reduction of the MRI-measured tumor volume as calculated from the MRI-measured tumor dimensions. The median tumor volume decreased from 12.6 cc (range 6.7-34.8 cc) to 4.1 cc (range 1.9-9.6 cc), corresponding to a 67% reduction in tumor volume. The percentage of volume reduction did not differ between patients who had detectable CTCs at baseline compared to those without (p=0.89).

No difference in the magnitude of clinical response was found when patients were further stratified according to treatment arm (p=0.2391). The odds ratio (OR) for lack of clinical response and the presence of CTCs after the first phase of chemotherapy was 1.81 (95% CI 0.50-6.58, p=0.36), however this result was not statistically significant (Tab. III).

Table III

CORRELATION BETWEEN LACK OF CLINICAL RESPONSE AFTER FIRST CHEMOTHERAPY AND PRESENCE OF CTCs AT BASELINE OR AFTER THE FIRST CHEMOTHERAPY

	OR for lack of response	95% CI	P value
At baseline
CK19 mRNA +	0.84	0.29 - 2.24	0.7470
CK19 mRNA + & BS106 +	0.54	0.13 - 2.24	0.3992
After 1^st chemotherapy
CK19 mRNA +	1.81	0.50 - 6.58	0.3673
CK19 mRNA + & BS106 +	1.31	0.30 - 5.68	0.7206

Results after the Second Phase of Chemotherapy

The percentage of patients with detectable CTCs in their peripheral blood declined to 12.2% (n=5) after the second phase of chemotherapy and, when compared to the number of patients with detectable CTCs at baseline, this reduction was statistically significant (p=0.0129). The distribution of these patients between the treatment arms did not differ significantly (p=0.18). The median tumor volume was 0.9 cc (range 0-4.6 cc), corresponding to a 93% volume reduction from the baseline MRI-measured tumor volume. There was no association between lack of response to chemotherapy and detection of CTCs in peripheral blood after the second phase of chemotherapy (OR=1.72, p=0.54) (Tab. IV). These results did not change when the data were reanalyzed using both the CK19 mRNA and the BS106 values.

Table IV

CORRELATION BETWEEN THE NUMBER OF CTCs-POSITIVE PATIENTS AND THE MRI-MEASURED TUMOR VOLUME

	Doxorubicin first arm	Paclitaxel first arm	All
After 1^st phase of chemotherapy	ρ = 0.191	ρ = 0.232	ρ = 0.039
	P = 0.3488	P = 0.3253	P = 0.7966
Change after 1^st phase of chemotherapy	ρ = 0.012	ρ = 0.149	ρ = 0.029
	P = 0.9548	P = 0.5317	P = 0.5453
After 2^nd phase of chemotherapy	ρ = 0.141	ρ = 0.020	ρ = 0.060
	P = 0.5529	P = 0.9327	P = 0.7127
Change from baseline after 2^nd phase of chemotherapy	ρ = 0.131	ρ = 0.439	ρ = 0.287
	P = 0.5833	P = 0.0529	P = 0.0727

CTCs detection after surgery

The median follow-up was 64.3 months (range 16-90 months). Of the 17 patients who developed either local or distant relapsed disease, 14 had detectable CTCs in their peripheral blood during follow-up and prior to the development of the relapsed disease. In a Cox proportional-hazard model with time-dependent covariates, the presence of CTCs was a predictor of imminent relapse, HR 3.28 (95% CI 0.98-10.99, p=0.04). When the model was restricted to patients who had no previously detectable CTCs and developed CTCs only during follow-up, the presence of CTCs was significantly associated with relapse, HR 1.50 (95% CI 1.05-2.15, p=0.04). No association could be found between the presence of CTCs and a worse overall survival, HR 3.46 (95% CI 0.31-38.71, p=0.25).

The table presents Spearman's rho (ρ) corrections coefficients and p values for testing whether ρ=0 for correlations between CTCs detected by CK19 and MRI-measured tumor volume. There was no statistically significant correlation.

Discussion

The prognostic significance and the clinical implications obtained from detecting CTCs in the peripheral blood of breast cancer patients have been questioned in many studies, most of which have been performed in the metastatic (6 -9) and adjuvant settings (19, 20). Few studies explored the prognostic significance of CTCs detection in the preoperative setting (21 -23).

The results of the current study suggest that the detection of CTCs in the peripheral blood of women with locally advanced breast cancer before the initiation of the PC is not a useful tool to predict which patients are likely to respond to chemotherapy. We were not able to demonstrate a correlation between the lack of clinical response and the presence of CTCs after the first and the second phase of chemotherapy (p=0.36 and p=0.53, respectively). It should be noted that the number of patients with detectable CTCs dropped from 32.7% to 26% after the first phase of chemotherapy and to 12.2% after the second phase. These results did not differ when we used the data obtained for both markers (CK19 mRNA and BS106) in the analysis.

Several authors have reached similar results using different methods for CTCs detection. Van der Auwera and colleagues compared 3 CTCs detection methods (Cell Search System, Adna Test and a multimarker quantitative RT-PCR). They found a significant concordance between the number of CTCs detected by the Cell Search System and the multimarker RT-PCR (24). In the “GEPARQuattro” trial, which included 213 non-metastatic breast cancer patients, the prevalence of CTCs-positive patients dropped from 22% to 11% after PC and there was no association with the primary tumor response (23). In a phase II trial on 118 patients, Pierga (21) and colleagues evaluated the prognostic significance of the detection of CTCs in the preoperative setting. The authors found that neither the detection of CTCs before or after PC, nor changes in CTCs count during the preoperative treatment was predictive of pathologic response. However, CTCs detection before and/or after PC was significantly associated with early metastatic relapse (p=0.013, median follow-up: 18 months). Pachman et al (22) evaluated a group of 30 patients receiving PC; the patients received 1 of 3 regimens: dose dense epirubicin, epirubicin/cyclophosphamide or epirubicin/cyclophosphamide/trastuzumab. The authors were able to observe a correlation between the decrease in the number of CTCs during the first cycles of treatment and the tumor response to chemotherapy. However the limited sample size of their study and the inconsistency of this finding among the 3 treatment arms (the trastuzumab arm did not show this relation between CTCs count and response to treatment) make the results of this study in need of more validation before drawing solid conclusions about the importance of CTCs quantification.

In the current study, the detection of CTCs in the postoperative setting increased the hazard of relapse, HR 1.50 (95% CI 1.05-2.15, p=0.04). This finding, however, was not translated into an overall survival difference in favor of those who did not have CTCs, HR 3.46 (95% CI 0.31-38.71, p=0.25). It is possible that such a difference exists (note that HR>1) but with a number of only 8 death events the power is too low to detect it. In another study of 41 patients undergoing surgery for breast cancer (with a mean follow-up of 8.6 months), Biggers (25) and colleagues were unable to show a difference in overall survival between patients with detectable CTCs and patients with no detectable CTCs. In a larger study of 167 patients conducted in the adjuvant setting, Xenidis (20) and associates were able to show a strong correlation between the detection of CTCs using CK19 mRNA and recurrence. The difference in recurrence rate between patients with persistently positive and those with persistently negative CK19 mRNA was highly significant (p=0.00026). The still ongoing German SUCCESS trial which enrolled non-metastatic lymph node-positive and high risk lymph node-negative patients before and after Taxane-based adjuvant therapy demonstrated that at least 1 CTC was detectable in the peripheral blood in 10% of the patients and that the patients with persistently negative CTCs had better prognosis (26). This strong correlation between CTCs detection in the post-operative setting and the development of local or distant relapse could be useful in designing a clinical trial to test the hypothesis that further systemic therapy might be helpful for patients with detectable CTCs after receiving neoadjuvant chemotherapy.

The results of the current study did not show a correlation between the detection of CTCs and other important tumor variables, such as nodal status and tumor size (Tab. II). These results mirror what was observed by Stathopoulou (27) and colleagues in their study of 142 patients with operable breast cancer patients. This dissociation between the tumor burden and the detection of CTCs may well be due to different biological phenotypes being responsible for locoregional vs hematogenous spread.

There were certain limitations to our study. First, due to the small number of patients enrolled (n=55) into the study, strong conclusions could not be drawn. Therefore, studies with a larger series of patients are needed, especially to address the question of the relationship between CTCs detection and response to PC. Second, quantitative analysis of the CTCs data in our study was not used as the clinically predictive thresholds for the non-metastatic patients are not exactly known and they are likely to lie within the range of 0-10 cell/ml (28, 29). Moreover, it has been shown that during the time of epithelial-mesenchymal transition, downregulation of the EpCAM expression might occur. This could have contributed to the false negative tests in the current study, especially since the entire study was conducted using an EpCAM-based assay. Nevertheless, the finding of a single CTC in 7.5 ml of peripheral blood was shown to be associated with an increased risk of metastases (30, 31). Third, the use of MRI to quantify the clinical response to PC might have led to either an underestimation of the actual tumor size, due to loss of tumor vasculature, or an overestimation (32, 33) of the size, due to the non-specific enhancement caused by the proliferative disease. Thus, a more robust method for quantifying tumor response in the preoperative setting needs to be established.

CONCLUSION

The results of this study have shown that the detection of CTCs in peripheral blood in the post-operative setting is correlated with local and/or distant relapse. We believe that from this point CTCs can be used as a prognostic indicator of a higher risk for relapse, a finding that could be used to develop further adjuvant therapies for patients in the high-risk group with detectable CTCs. However, the CTCs detection in peripheral blood in the course of PC does not correlate with a particular pattern of response to PC.

Footnotes

LIST OF ABBREVIATIONS

Financial Support: This study was supported in part by an Avon-NCI Progress for Patients Award on the Dana-Farber/Harvard Cancer Center SPORE in Breast Cancer CA089393. The study was approved by the institutional review board of the Massachusetts General Hospital and the Dana-Farber Cancer Institute.

Conflict of Interest Statement: Lisa Roberts and Natalie Solomon are employees of Abbott Molecular Inc., Des Plaines, IL. They both hold stock or other ownership to disclose. The rest of authors declare that they have no competing interests.

References

Stathopoulos

, Sanidas

, Kafousi

Detection of CK-19 mRNApositive cells in the peripheral blood of breast cancer patients with histologically and immunohistochemically negative axillary lymph nodes. Ann Oncol 2005; 16: 240–6.

Ring

, Zabaglo

, Ormerod

Detection of circulating epithelial cells in the blood of patients with breast cancer: comparison of three techniques. Br J Cancer 2005; 92: 906–12.

Gilbey

, Burnett

, Coleman

The detection of circulating breast cancer cells. J Clin Pathol 2004; 57: 903–11.

Ring

, Smith

, Dowsett

Circulating tumor cells in breast cancer. Lancet Oncol 2004; 5: 79–88.

Zach

, and Lutz

Tumor cell detection in peripheral blood and bone marrow. Curr Opin Oncol 2006; 18: 48–56.

Cristofanilli

, Budd

, Ellis

Circulating tumor cells, disease progression, and survival in metastatic breast cancer. N Engl J Med 2004; 351: 781–91.

Ellis

, Hayes

D.F.

, Lippman

M.E.

Treatment of metastatic disease. In: Harris

, Lippman

, Morrow

Diseases of the breast. 3rd ed. Philadelphia: Lippincott- Raven; 2004: 1101–59.

Dawood

, Broglio

, Valero

Circulating tumor cells in metastatic breast cancer. From prognostic stratification to modification of the staging system? Cancer 2008; 113: 2422–30.

Cristofanilli

, Broglio

, Guarneri

Circulating tumor cells in metastatic breast cancer: biologic staging beyond tumor burden. Clin Breast Cancer 2007; 6: 471–9.

10.

Buchholz

, Lehman

, Harris

Statement of the science concerning locoregional treatments after preoperative chemotherapy for breast cancer: a National Cancer Institute conference. J Clin Oncol 2008; 26: 791–7.

11.

Fisher

, Brown

, Mamounas

Effect of preoperative chemotherapy on local-regional disease in women with operable breast cancer: findings from National Surgical Adjuvant Breast and Bowel Project B-18. J Clin Oncol 1997; 15: 2483–93.

12.

Weatherall

, Evans

, Metzger

MRI vs. histologic measurement of breast cancer following chemotherapy: comparison with x-ray mammography and palpation. J Magn Reson Imaging 2001; 13: 868–75.

13.

Delille

, Slanetz

, Yeh

Invasive ductal breast carcinoma response to neoadjuvant chemotherapy: noninvasive monitoring with functional MR imaging—pilot study. Radiology 2003; 228: 63–9.

14.

Segara

, Krop

, Garber

Does MRI predict pathologic tumor response in women with breast cancer undergoing preoperative chemotherapy?

J Surg Oncol 2007; 96: 474–80.

15.

Yeh

, Slanetz

, Kopans

Prospective comparison of mammography, breast ultrasound and magnetic resonance imaging in patients undergoing neoadjuvant chemotherapy for palpable breast cancer. AJR Am J Roentgenol 2005; 184: 868–77.

16.

Miksicek

, Myal

, Watson

, Walker

, Murphy

, and Leygue

Identification of a novel breast- and salivary gland-specific, mucin-like gene strongly expressed in normal and tumor human mammary epithelium. Cancer Res 2002; 62: 2736–40.

17.

Bosma

, Weigelt

, Lambrechts

Detection of circulating breast tumor cells by differential expression of marker genes. Clin Cancer Res 2002; 8: 1871–7.

18.

Noether , Gottfried

Sample Size Determination for Some Common Nonparametric Tests. J Am Statistical Association 1987; 398: 645–7.

19.

Pachmann

, Camara

, Kavallaris

Monitoring the response of circulating epithelial tumor cells to adjuvant chemotherapy in breast cancer allows detection of patients at risk of early relapse. J Clin Oncol 2008; 26: 1208–15.

20.

Xenidis

, Markos

, Apostolaki

Clinical relevance of circulating CK-19 mRNA-positive cells detected during the adjuvant tamoxifen treatment in patients with early breast cancer. Ann Oncol 2007; 18: 1623–31.

21.

Pierga

J.Y.

, Bidard

F.C.

, Mathiot

Circulating tumor cell detection predicts early metastatic relapse after neoadjuvant chemotherapy in large operable and locally advanced breast cancer in a phase II randomized trial. Clin Cancer Res 2008; 14: 7004–10.

22.

Pachmann

, Camara

, Kavallaris

Quantification of the response of circulating epithelial cells to neodadjuvant treatment for breast cancer: a new tool for therapy monitoring. Breast Cancer Research 2005; 7: R975–9.

23.

Riethdorf

, Müller

, Zhang

Detection and HER2 expression of circulating tumor cells: prospective monitoring in breast cancer patients treated in the neoadjuvant GeparQuattro trial. Clin Cancer Res 16; 2634–45.

24.

Van der Auwera

, Peeters

, Benoy

I.H.

Circulating tumour cell detection: a direct comparison between the CellSearch System, the AdnaTest and CK-19/mammaglobin RT–PCR in patients with metastatic breast cancer. Br J Cancer 2010; 102: 276–84.

25.

Biggers

, Knox

, Grant

Circulating tumor cells in patients undergoing surgery for primary breast cancer: preliminary results of a pilot study. Ann Surg Oncol 2009; 16: 969–71.

26.

SUCCESS (www.success-studie.de)

27.

Stathopoulou

, Vlachonikolis

, Mavroudis

Molecular detection of cytokeratin-19–positive cells in the peripheral blood of patients with operable breast cancer: evaluation of their prognostic significance. J Clin Oncol 2002; 20: 3404–12.

28.

Iakovlev

, Goswami

, Vecchiarelli

Quantitative detection of circulating epithelial cells by Q-RT-PCR. Breast Cancer Res Treat 2008; 107: 145–54.

29.

Pierga

, Bonneton

, Vincent-Salomon

Clinical significance of immunocytochemical detection of tumor cells using digital microscopy in peripheral blood and bone marrow of breast cancer patients. Clin Cancer Res 2004; 10: 1392–1400.

30.

Bidard

, Mathiot

, Delaloge

Single circulating tumor cell detection and overall survival in nonmetastatic breast cancer. Ann Oncol 2010; 21: 729–33.

31.

Krishnamurthy

, Cristofanilli

, Singh

Detection of minimal residual disease in blood and bone marrow in early stage breast cancer. Cancer 2010; 116: 3330–7.

32.

Tse

G.M.

, Chaiwun

, Wong

K.T.

Magnetic resonance imaging of breast lesions—a pathologic correlation. Breast Cancer Res Treat 2007; 103: 1–10.

33.

Knopp

M.V.

, Brix

, Junkermann

H.J.

MR mammography with pharmacokinetic mapping for monitoring of breast cancer treatment during neoadjuvant therapy. Magn Reson Imaging Clin N Am 1994; 2: 633–58.