Abstract
Background: There is strong evidence for the isolated tumour cells (ITCs) in the bone marrow of breast cancer patients having prognostic impact both at primary diagnosis and during recurrence-free follow-up. The goal of this study was to investigate the therapeutic efficacy of zoledronate on the persistence of ITC. Patients and Methods: A total of 172 primary breast cancer patients without evidence of distant recurrence but detection of ITC in bone marrow were followed up. Zoledronate was administered every 4 weeks for 6 months to 31 patients who had completed surgery and adjuvant chemotherapy. In a matched-pair analysis, these patients were compared to 141 patients who did not receive additional zoledronate treatment. The bone marrow was re-examined after a median of 7.9 months (SD 0.89) and 11.5 months (SD 12.41; p=0.11), respectively. Patients were followed-up prospectively for a median of 39 months after the first aspiration. Results: While ITCs were detected in all 172 patients at the time of first bone marrow aspiration, ITCs were detected in four patients (13%) following 6 months of zoledronate therapy in contrast to 38 patients (27%) of the control group (p=0.099). The reduction in cell numbers between the first and second aspiration reached statistical significance in the zoledronate group (p=0.02 vs. p=0.14). Persistent ITCs at the follow-up aspiration were associated with reduced recurrence-free survival (p=0.05). Conclusion: These results indicate a potential antineoplastic effect of the cell cycle-independent agent zoledronate on persisting ITCs in a dormant state.
- Breast cancer
- recurrence-free follow-up
- isolated tumour cells
- minimal residual disease
- zoledronate treatment
Recently, the highest level of evidence (LOE 1) was reached for the prognostic impact of isolated tumour cells (ITCs) in bone marrow of breast cancer patients both at primary diagnosis and during recurrence-free follow-up (1, 2). A pooled analysis, comprising 4,703 patients confirmed poor outcome in patients with ITCs before initiation of primary therapy (2), whereas 726 patients with persisting ITCs during recurrence-free follow-up had an increased risk of distant relapse and shortened survival (2). Moreover, ITCs have been shown to persist in a dormant state for years before they start to proliferate, stimulate angiogenesis and form new metastatic tumours (3, 4). ITCs therefore can be used not only to detect patients with an increased risk for disease recurrence, but may also serve as a therapeutic target for individualized treatment interventions after phenotyping these tumour cells.
Few clinical trial results demonstrating the efficacy of a therapeutic intervention on ITC are currently available. Low expression of proliferation markers, such as Ki-67 and p120, indicates that most of these cells are resting in the G0 phase of the cell cycle in a dormant state (3). This dormancy may explain the limited effect of adjuvant cytostatic treatment (5). On the contrary, cell cycle-independent bisphosphonates have not only demonstrated good efficacy in the treatment of bone metastases, but have also shown inhibition of intracellular pathways, resulting in direct antitumour activity in vitro (6, 7). Moreover, breast cancer patients receiving bisphosphonates as part of their adjuvant systemic treatment have prolonged survival, pointing also to a clinical benefit of bisphosphonates in the primary setting (8, 9, 28).
Patients and Methods
Study design. This was a non-randomized phase II trial evaluating the efficacy of zoledronate applied at 4 mg q4w for 6 months with loading dose of 8 mg for the elimination of persistent ITCs in the bone marrow of 31 women with primary breast cancer during recurrence-free follow-up. A matched-pair analysis was performed retrospectively to compare these patients to 141 historic controls with evidence of ITCs in bone marrow but with no additional zoledronate treatment. All patients had their bone marrow re-evaluated after a median of 7.9 months (SD 0.89) in the treatment group and 11.5 months (SD 12.41) in the control group (p=0.11).
Patients. Eligible patients for the treatment intervention study had histologically confirmed, completely excised primary breast cancer pT1-4 pN0-3 pM0, with evidence of persistent ITCs in their bone marrow at least 6 months after completion of surgery and adjuvant chemotherapy, if so indicated. Patients were also excluded if they had received prior treatment with bisphosphonates within the last 6 months. Adjuvant endocrine treatment with tamoxifen or aromatase inhibitors was the only concomitant antitumour treatment allowed in the study protocol. Patients from the treatment intervention study were compared to 141 consecutive control patients followed up before the initiation of this treatment study. The median observation time after primary diagnosis was 40.6 months (SD 26.9 monthly).
Detection of isolated tumour cells in bone marrow. Bone marrow evaluation was performed according to a previously published and unmodified standardized protocol (10-12). In summary, bone marrow samples were collected in EDTA-treated tubes and centrifugated using Ficoll-Hypaque (Pharmacia, Freiburg, Germany) density gradient (density 1.077 g/mol) at 900 ×g for 30 minutes. Mononucleated interface cells were then washed, and 106 cells were centrifuged onto each glass slide at 150 ×g for 5 minutes. The monoclonal antibody A45-B/B3 (Micromet, Munich, Germany), directed against a common epitope of cytokeratin polypeptides, including the cytokeratin (CK) heterodimers 8/18 and 8/19 (13), was used to detect tumour cells. Isotype control was achieved by using an appropriate dilution of unrelated mouse-myeloma antibody (Sigma, Deisenhofen, Germany) on the patients' bone marrow specimens. The specific reaction of the primary antibody was developed with the alkaline phosphatase anti-alkaline phosphatase (APAAP) technique and stained with new fuchsin (14). A semi-automated screening system (MDS, Applied Biosystems, Newcastle, Great Britain) was used to screen 2×106 bone marrow cells per patient. All slides were examined by two independent observers, who agreed on the result in over 95% of the specimens. Additionally, a control group of 203 patients with benign disease was examined to verify the specificity of the method (12).
Statistical analysis. Data quality was controlled by verifying all event reports during follow-up by re-examining the original data files. Kaplan-Meier life-table curves were compared using the log-rank test to estimate survival (15), Cox's regression analysis was used for multivariate analyses of survival. The variables were entered forward stepwise (16). Only cancer-associated deaths were considered events. To compare categorical variables, the χ2 test was used. The two-tailed t-test was used to calculate the differences of the mean of the independent samples that had continuous variables. p-values of less than 0.05 were considered significant in two-sided tests. No adjustment of the error probability for multiple testing was performed. The Statistical Package for the Social Sciences 14.0 (SPSS Inc., Chicago, Illinois, USA) was used for analyses.
Patient characteristics at the time of primary diagnosis.
Results
Patient characteristics. Patient characteristics of the patients are shown in Table I. All factors were well balanced between the two patient groups. There was a trend towards an increased number of T1 tumours in the zoledronate group (74% vs. 52%), but this difference was not significant.
Acute toxicity associated with zoledronate treatment.
In the zoledronate group, all patients had completed primary surgical and chemotherapy treatment for at least six months. Seventeen hormone receptor-positive patients (54.8%) continued their adjuvant endocrine treatment with 20 mg/day tamoxifen simultaneously to zoledronate, 2 patients received a gonadotrophin-releasing hormone analogue additionally (6.5%) and 1 patient was on letrozole. As all historic patients with recurrence-free breast cancer were included in the control group, 95 patients underwent their first bone marrow aspiration at primary diagnosis (67.4%), 12 patients were examined during adjuvant cytostatic treatment (8.5%) and 34 patients during follow-up (24.1%). As all patients were to receive standard antitumour therapy, 75 control patients (53.2%) received adjuvant cytostatic treatment, and 22 patients were on tamoxifen, between first and second aspiration (15.6%).
Overall, zoledronate was well tolerated, with bone pain as the most frequent side-effect, followed by headache and increased temperature, which occurred after the first application in 4 patients (12.9%). All side-effects were observed most frequently after the first application of zoledronate, were only moderate and were easily controlled by paracetamol. No patient discontinued treatment prematurely (Table II).
Fate of ITCs in bone marrow. The initial bone marrow aspiration was performed at a mean of 29.9 months (SD 20.18) after primary diagnosis in the zoledronate group, while the control patients underwent the procedure 10 months after primary diagnosis (SD 27.75), which was a statistically significant difference (p<0.0001) (Table III). The bone marrow was re-examined after a median of 7.9 months (SD 0.89) in the treatment group and 11.5 months (SD 12.41) in the control group (p=0.11, t-test). ITCs were detected in all 172 patients at the initial bone marrow aspiration. In the follow-up aspiration, 4 patients (13%) showed persistent ITC in bone marrow after 6 months of zoledronate therapy, whereas 38 patients (27%) in the control group were diagnosed with persisting ITCs (p=0.099) (Figure 1). In the intitial aspiration, the median number of isolated tumour cells was 2 for both the zoledronate (range 1-11) and the control group (range 1-269; p=0.46). Both groups showed lower cell counts in the follow-up aspiration, with a median of 0 cells (range 0-4 and 0-132 in the zoledronate group and in the control group, respectively, p=0.12, χ2-test). The reduction in cell numbers between the first and second aspiration reached statistical significance in the zoledronate group (p=0.02) in contrast to control group (p=0.14, t-test).
ITC positivity rate in initial and follow-up aspiration for patients receiving zoledronate and control patients.
Fifteen zoledronate-treated patients underwent additional follow-up bone marrow aspirations after a median of 19 months (range 4.7-38.7 months) following treatment. Among 12 patients without evidence of ITC in bone marrow immediately after treatment, 10 had a persistently negative bone marrow status. All 3 patients with evidence of ITCs immediately after zoledronate treatment had negative bone marrow status on longer follow-up. Two of these patients showed elimination of tumour cells without further therapeutic intervention, while one was negative after treatment with trastuzumab, which was initiated after the detection of HER2/neu-positive ITCs.
Detection of ITCs in bone marrow in comparison of the zoledronate and control groups.
ITCs in bone marrow and survival. With a median follow-up of 41 months after primary diagnosis (range 2-112) the disease had recurred in 34 patients (19.8%), 31 in the control group (22.0%) and 3 in the zoledronate group (9.7%). Fourteen patients (41.2%) presented with locoregional relapse, whereas 20 patients (48.8%) had distant metastasis as the first site of recurrence (namely 11 patients with visceral metastasis, 3 patients with skeletal metastasis, 6 patients with combined skeletal and visceral disease). During the follow-up period, 17 patients (9.8%) had died of breast cancer; 6.5% (n=2) of patients receiving zoledronate had died compared to 10.6% (n=15) in the control group. Neither recurrence-free (p=0.44; Figure 2) nor overall survival (p=0.99, log-rank test) were statistically different between the treatment and control groups.
When the bone marrow status was analysed at follow-up aspiration regardless of the treatment group, persistent ITC were associated with decreased recurrence-free survival and overall survival. Mean recurrence-free survival was 112 months [97-127, 95% confidence interval (CI)] in patients with negative bone marrow status compared to 87 months (59-115, 95% CI) in patients with positive bone marrow status (p=0.05; Figure 3). Furthermore, patients with persistent ITCs died earlier, with a mean survival time of 68 months (59-77, 95% CI) versus 86 months in patients with negative bone marrow status (77-94, 95% CI; p=0.07).
Discussion
This study evaluates the efficacy of zoledronate on persisting ITCs in bone marrow of breast cancer patients prior to recurrence of their disease. The results indicate that a considerable number of patients (87%) present with elimination of ITCs after 6 months of zoledronate treatment. When compared to control patients receiving no additional zoledronate treatment, treated patients showed a trend towards increased tumour cell elimination in (ITC positivity 13% vs. 27%; p=0.099). The reduction in ITC counts reached statistical significance in the zoledronate group (p=0.02) in contrast to control patients (p=0.14); 77.7% (n=7) of bone marrow negative patients remained free of ITC on longer follow-up. Regardless of the patient group, persistent ITCs at follow-up were associated with reduced recurrence-free survival (p=0.05)
Recent data have confirmed the relevance of ITCs in bone marrow as an indicator of decreased survival, both at primary diagnosis and during recurrence-free follow-up (17, 18), independently of established prognostic factors. A pooled analysis of 4,703 patients with primary breast cancer demonstrated that the presence of ITCs was a significant prognostic parameter for poor overall, as well as disease-free, survival during the entire 10-year observation period (p<0.001) (2). Three independent European groups have also collected bone marrow aspirates during recurrence-free follow-up at a mean interval of 31.7 months (SD 19.4 months) after primary diagnosis of breast cancer and followed up the patients for a median of 54.5 months (SD 25.1 months) (1). Persistent ITCs were detected in 15.4% of patients during recurrence-free follow-up and were also associated with significantly shorter disease-free and overall survival (each p <0.0001), both in univariate and multivariate analysis. Therefore, this study used the detection of ITCs in bone marrow to select patients with increased risk of relapse, who might benefit from additional treatment approaches. ITCs were used both as therapeutic target and as a early marker for treatment response.
Kaplan-Meier estimates of recurrence-free survival in patients with zoledronate treatment and control patients.
While all patients in the present analysis had evidence of ITCs, the comparison of zoledronate-treated patients with our control patients is subject to several limitations. While all patients were recruited into the zoledronate trial during follow-up without evidence of distant metastasis, the majority of control patients were examined earlier during the disease. Additionally, more patients received simultaneous endocrine treatment in the zoledronate than in the control group (58.1 vs. 15.6%). Whether these differences between groups influenced the elimination of ITCs in bone marrow remains unclear. Furthermore, a non-significant trend was found for an increased number of T1 tumours in the zoledronate group (74.2% vs. 51.8%; p=0.09). Yet risk selection in this study was based on bone marrow positivity and patients underwent bone marrow aspirations at a mean of 30 months in the zoledronate group and 10 months in the control group, thus limiting the influence of tumour size at primary diagnosis. All other primary tumour characteristics were well balanced.
Few trials have as yet been published using ITCs either as a marker of treatment efficacy or as a therapeutic target in breast cancer patients. In one trial, 59 newly diagnosed patients with advanced or inflammatory disease underwent bone marrow aspirations before and after anthracycline and taxane-based chemotherapy (5). While cytostatic treatment did not significantly reduce the number of ITCs, a positive bone marrow result following treatment was associated with lower overall survival (p=0.01). In another trial, ITCs in bone marrow were detected in 53% of patients after neoadjuvant chemotherapy and correlated with the pathological therapy response (19). This lack of effect of adjuvant chemotherapy in a relevant number of patients may be explained by the low expression of proliferation markers as shown both in genomic analyses and immunocytochemical phenotyping suggesting a state of dormancy in these cells (4, 20).
Kaplan-Meier estimates of recurrence-free survival in patients with negative and positive bone marrow status at the follow-up examination.
Consequently, cell cycle-independent agents directed against specific characteristics of ITC may be more beneficial than chemotherapeutics. Three non-randomized phase II trials demonstrated good tumour cell elimination in patients receiving the monoclonal antibody 17-1A (edrecolomab) directed against EpCAM (21-23), while Bozionellou et al. has shown the elimination of CK-19 mRNA in 93% of patients treated with trastuzumab (24). These trials, however, included only small patient numbers, required the presence of the specific target either on the primary tumour or on ITCs and have not been able to show a clinical benefit.
Alternatively, bisphosphonates, which have been established as a standard of care in metastatic disease, may be a non cell cycle-dependent treatment option. Based on the prevention of tumour cell adhesion, induction of apoptosis, antagonism of growth factors and antiangiogenic effects, antitumour efficacy in primary breast cancer has been discussed. Four trials have been published on the role of bisphosphonates in the adjuvant setting to date, suggesting that their use may be beneficial (8, 25-28). All but one trial, including the largest set with 1,803 premenopausal breast cancer patients, reported a reduced incidence of distant metastasis and improved survival for patients treated with bisphosphonates, even after a follow-up of up to 10 years (8, 9, 26-29). A recent study describes the efficacy of zoledronic acid in postmenopausal women with early breast cancer receiving adjuvant letrozole (30).
In this study, zoledronate treatment for six months resulted in an elimination of ITCs in 87% of treated patients, which is comparable to therapeutic efficacy that has been demonstrated for edrecolomab and trastuzumab (21, 24). Moreover, the continuous absence of ITCs on longer follow-up in 82% of bone marrow-negative patients was demonstrated. In contrast to other trials published so far, a comparison was made between treated patients and control patients, who received standard anticancer therapy, but no additional zoledronate treatment. Surprisingly, an elimination of ITCs in a relevant number of control patients (73%) was also observed, which might be attributable to standard chemotherapy and endocrine treatment, as well as to immunological processes. This unexpectedly high clearing rate in the control patients hampers the estimate of zoledronate efficacy on ITCs, but also puts elimination rates found by other groups into perspective. Comparing the zoledronate-treated patients to the controls, a non-significant trend towards an increased tumour cell elimination by adding zoledronate to standard anticancer therapy was observed (p=0.099). Furthermore, persistent ITCs at follow-up were associated with worse outcome, confirming the role of ITCs as a valuable tool for treatment monitoring.
In conclusion, persistent ITCs in bone marrow predict an increased risk of subsequent relapse and reduced survival. Bone marrow aspirations represent an intriguing option not only for selecting patients at risk of pending recurrence, but also for early assessment of treatment efficacy. The results of this study indicate a potential antineoplastic effect of the cell cycle-independent agent zoledronate on persisting ITCs in a dormant state, and thus form the basis for future prospectively randomised trials that are needed to determine the actual clinical benefit from secondary adjuvant treatment intervention.
Acknowledgements
Supported by a grant from the Friedrich-Baur-Stiftung, Muenchen, Germany.
The Authors are sincerely grateful to Heidi Coleman for her editorial support.
Footnotes
- Received August 19, 2009.
- Revision received March 31, 2010.
- Accepted April 6, 2010.
- Copyright© 2010 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- Prognostic Value of RANKL/OPG Serum Levels and Disseminated Tumor Cells in Nonmetastatic Breast Cancer
- The Key Role of Bisphosphonates in the Supportive Care of Cancer Patients
- Prognostic Relevance of Disseminated Tumor Cells in the Bone Marrow of Patients with Primary Breast Cancer - Results of a Standardized Follow-Up
- Role of IL-1 receptor-associated kinase-M (IRAK-M) in priming of immune and inflammatory responses by nitrogen bisphosphonates