|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2001 American Society for Clinical Oncology Adjuvant Clodronate Treatment Does Not Reduce the Frequency of Skeletal Metastases in Node-Positive Breast Cancer Patients: 5-Year Results of a Randomized Controlled TrialFrom the Departments of Oncology and Radiology, Helsinki University Central Hospital, Helsinki, Finland; and Department of Oncology, Uppsala University, Uppsala, Sweden. Address reprint requests to Inkeri Elomaa, MD, Department of Oncology, Helsinki University Central Hospital, Haartmaninkatu 4, FIN-00290 Helsinki, Finland; email inkeri.elomaa{at}helsinki.fi
PURPOSE: Bisphosphonates have effectively reduced the development and progression of bone metastases in advanced breast cancer. The aim of this study was to determine whether bone metastases could be prevented by adjuvant clodronate treatment in patients with primary breast cancer. PATIENTS AND METHODS: Between 1990 and 1993, 299 women with primary node-positive breast cancer were randomized to clodronate (n = 149) or control groups (n = 150). Clodronate 1,600 mg daily was given orally for 3 years. All patients received adjuvant therapy: premenopausal six cycles of CMF chemotherapy and postmenopausal antiestrogens (randomized to tamoxifen 20 mg or toremifene 60 mg/d for 3 years). Seventeen patients were excluded from the analyses because of major protocol violations. The final population was 282 patients. Intent-to-treat analyses were also performed for all major end points. The follow-up time was 5 years for all patients. RESULTS: Bone metastases were detected equally often in the clodronate and control groups: 29 patients (21%) versus 24 patients (17%) (P = .27). The development of nonskeletal recurrence was significantly higher in the clodronate group compared with controls: 60 patients (43%) versus 36 patients (25%) (P = .0007). The overall survival (OS) and disease-free survival (DFS) rates were also significantly lower in the clodronate group than in the controls (OS, 70% v 83%, P = .009; DFS, 56% v 71%, P = .007, respectively). In multivariate analyses, clodronate remained significantly associated with DFS (P = .009). CONCLUSION: Adjuvant clodronate treatment does not prevent the development of bone metastases in node-positive breast cancer patients. However, clodronate seems to have a negative effect on DFS by increasing the development of nonskeletal metastases.
BREAST CANCER IS one of the bone-seeking malignancies. One of three first relapses occurs in the skeleton. Approximately 70% of the patients with metastatic breast cancer develop bone metastases during their remaining life time.1 The mode of action of bisphosphonates has not been completely elucidated. They decrease bone resorption by inhibiting development of osteoclast from monocytes, by diminishing the recruitment and activity of osteoclasts, or by stimulating apoptosis of osteoclasts, either by direct effects on the osteoclast or indirectly through the osteoblast.2,3 Bisphosphonates also inactivate macrophages and are found to induce apoptosis of phagocytosing cells in vitro4-6 and in vivo.4,7 Because macrophages produce a variety of bone-resorbing cytokines, and because the osteoclasts originate from the monocyte phagocyte system, an effect through this pathway may also be possible.8 Bisphosphonates do not seem to exert their inhibitory effect on bone resorption through a direct effect on cancer cells, although some evidence of their antitumor activity has been observed against myeloma cells.9 Bisphosphonates have successfully been used in the treatment of malignant hypercalcemia and skeletal metastases. In breast cancer patients, eight placebo-controlled trials with clodronate or pamidronate have shown retarded progression of lytic bone lesions, reduced appearance of new bone metastases, and decreased frequency of hypercalcemia, pathologic fractures, and bone pain.10-17 So far, only one study is available on the impact of adjuvant bisphosphonate treatment on survival in primary operable breast cancer. In that study, clodronate reduced not only the development of bone metastases but, surprisingly, also visceral metastases in patients with bone marrow micrometastases.18 We have previously demonstrated that adjuvant clodronate treatment prevents bone loss in postmenopausal breast cancer patients and reduces chemotherapy-induced bone loss in premenopausal patients.19,20 The present study aimed to examine the effect of adjuvant clodronate treatment on the development of bone metastases in patients with node-positive breast cancer in a prospective randomized study with a minimum of 5 years of follow-up.
Patients and Treatments The study population consisted of 299 patients with primary node-positive breast cancer. Eligible for the trial were pre- and postmenopausal women with operable breast cancer and histologically proven axillary metastases (T1 to T3, N1/2, M0) treated between October 1990 and July 1993 at Helsinki University Hospital, Department of Oncology. Exclusion criteria included the following: (1) age greater than 75 years; (2) Karnofsky performance status of less than 70%; (3) presence of other malignancies; (4) peptic ulcer or its symptoms; (5) pregnancy; (6) serum creatinine concentration greater than 150 µmol/L. Informed consent was obtained from all participants. The study was approved by the local ethics committee. The planned treatment involved (1) surgery with axillary evacuation and total mastectomy or breast-conserving resection, (2) postoperative radiotherapy with megavoltage irradiation in a total dose of 50 Gy in 25 fractions to regional lymph nodes and the operation scar after mastectomy or residual breast tissue after breast-conserving resection, and (3) simultaneous adjuvant chemotherapy or endocrine therapy. Planned adjuvant treatment for premenopausal patients was chemotherapy, which consisted of six cycles of cyclophosphamide 600 mg/m2, methotrexate 40 mg/m2, and fluorouracil 600 mg/m2 intravenously on day 1 and at successive 3-week intervals (CMF). The first 45 postmenopausal patients were treated with adjuvant antiestrogen therapy (tamoxifen 20 mg/d) for 3 years. Later postmenopausal women were randomly allocated to receive either adjuvant tamoxifen 20 mg or toremifene 60 mg/d for 3 years. All patients were randomized to either receive or not receive oral clodronate (Bonefos; Leiras Pharmaceutical Company, Helsinki, Finland) 1,600 mg daily for 3 years. Both randomizations were performed at the beginning of the adjuvant treatment, and the clodronate treatment was started simultaneously with the adjuvant therapy at a median of 35 days (range, 14 to 80 days) after operation, except for the first nine patients, who began clodronate treatment an average of 6 months later than adjuvant treatment ( Table 1).
Of the 299 randomized patients, 21 did not fulfill the inclusion criteria; of them, 17 patients were excluded from the analyses because of major protocol violations (second primaries and hematogenous metastases at randomization). The final population was 282 patients ( Table 2). All of these patients underwent surgery as planned. Four patients refused radiotherapy (one in the clodronate group and three in the control group). One hundred fifty patients were treated with chemotherapy and 129 with antiestrogens (81 with tamoxifen and 48 with toremifene); two control patients refused adjuvant chemotherapy and one refused antiestrogen therapy. Seven patients interrupted antiestrogen therapy after a median of 14 months, mainly because of side effects (three in the clodronate group and four in the control group). Six patients interrupted chemotherapy because of side effects after a median of three cycles (range, two to five cycles), two in the clodronate group and four in the control group. Two premenopausal patients in both groups were treated with anthracycline-based chemotherapy instead of CMF, and nine patients were treated with chemoendocrine therapy (CMF and antiestrogen therapy; seven in the clodronate and two in the control group). Nineteen patients interrupted clodronate treatment after a median of 15 months: 16 patients because of gastrointestinal side effects, two patients because of refusal to continue therapy, and one because of misunderstanding. Three patients had dose reduction because of gastrointestinal side effects. The data of all these patients were included in the analyses in their planned treatment groups.
Follow-Up Staging investigations for breast cancer included clinical examination, liver ultrasound, chest x-ray, and bone scintigraphy. Basic laboratory tests before the randomization included a complete blood cell count and sedimentation rate, liver enzymes (transaminase, alkaline phosphatase, 5-nucleotidase), serum creatinine, calcium, and electrolytes. Bone scintigraphy was performed before the treatment and after 1, 2, 3, and 5 years. If there was evidence of bone metastases, x-rays were obtained of the affected areas. Clinical investigation and basic laboratory tests were repeated every 4 to 6 months with a radiologic examination if necessary. Investigators performing bone scans and radiologic examinations were blinded to treatment allocation. The planned 5-year observation time point was reached in all patients who did not die before 5 years of observation. Two cases of death without breast cancer recurrence have been censored before 5 years, one in each group.
Statistical Analysis
Pretreatment characteristics of the subjects in the two study groups are given in Table 3. The two groups were balanced with respect to pretreatment characteristics, except for PgR status. In the clodronate group, there were more PgR-negative patients (P = .03).
Bone Metastases and Morbidity Bone metastases were detected similarly often in the clodronate group and in the controls: 29 (21%) versus 24 (17%), respectively ( Table 4). There were no significant differences in 5-year skeletal disease-free survival (DFS) between the study groups: 79% in the clodronate group and 83% in the controls (P = .27) ( Fig 1).
The extent of metastases in bone scan at the time of their appearance and at the end of the follow-up, the radiologic pattern (lytic, sclerotic, or both) of the skeletal metastases, pathologic fractures, palliative radiotherapy to skeleton, and hypercalcemic episodes were documented. In both groups, skeletal metastases progressed. No significant differences were found between the study groups in any skeletal events, although there was a trend toward fewer pathologic fractures and need of palliative radiotherapy for bone in the clodronate group ( Table 5).21
Survival Analyses and Nonskeletal Recurrence Sixty-one patients (44%) in the clodronate group and 42 control patients (29%) developed metastatic disease (Table 4). Five-year DFS was significantly lower in the clodronate group: 56% versus 71% (P = .007) ( Fig 2). Five-year distant disease-free survival (DDFS, skeletal and visceral metastases) was 63% in the clodronate group and 73% in the controls (P = .061), and 5-year local control rates were 75% and 87%, respectively (P = .002).
Visceral metastases were detected in 51 patients (37%) in the clodronate group and in 32 patients (22%) in the control group (Table 4). Five-year visceral DFS was significantly lower in the clodronate group (63%) as compared with the controls (78%; P = .007). Nonskeletal recurrences (visceral and local) were detected more frequently in the clodronate-treated patients (60 patients, or 43%) than in the controls (36 patients, or 25%) (Table 4). Nonskeletal DFS rates were 57% and 75% for patients in the clodronate and control groups, respectively (P = .0007) (Fig 1). Forty-two clodronate-treated patients (30%) and 24 control patients (17%) died (Table 4). Five-year OS rates were 70% and 83% for clodronate-treated patients and control patients, respectively (P = .009) (Fig 2).
Uni- and Multivariate Analyses and Relative Risks of Recurrence and Death
In the univariate analyses for OS, significant prognostic factors were nodal status, ER and PgR status, primary distant metastases, tumor size, and study treatment group. In the multivariate analyse, nodal status, PgR status, and study treatment group remained statistically significant for OS (Table 6).
The negative impact of clodronate on DFS was more prominent in patients with larger tumors (tumor size
Intent-to-Treat Analyses The main results were unchanged in the intent-to-treat analyses. Five-year DFS, DDFS, local control, visceral DFS, nonskeletal DFS, and OS were still significantly lower in the clodronate group as compared with the controls (P = .004, .042, .001, .018, .001, and .024, respectively), and no significant differences were found between the study treatment groups according to bone DFS (P = .11). In the multivariate analyses, nodal status, PgR status, and study treatment group maintained a statistically significant association with DFS (P < .00001, .007, and .006, respectively). However, in the multivariate analyses of OS, nodal status, PgR status, and ER status remained statistically significant (P = .001, .005, and .03, respectively), but the study treatment group lost its significance (P = .19).
Analyses Adjusted by PgR Status
A recent study from Heidelberg18 has indicated that adjuvant clodronate treatment prevents not only bone but also extraskeletal metastases in early-stage breast cancer patients. We have not been able to verify these findings. In the present study, no significant differences were found in the incidence of skeletal metastases between the study groups. However, the incidence of nonskeletal metastases, both visceral and local, was significantly higher in the clodronate group, which led to a lower DFS in the study group. The negative impact of clodronate was more prominent in patients with more advanced and aggressive disease (large tumors, higher number of metastatic lymph nodes, ER-negative disease). Especially in ER-negative patients, the difference in DFS was large (29% in favor of the control group). In ER-positive patients, no significant difference was found. This interaction was not seen with PgR status. The study arms were well balanced except for hormone receptor status, which may have exaggerated the adverse effect of clodronate in the present study. The adverse effect of clodronate on OS therefore lost its significance in the multivariate analyses and after adjusting the data by PgR status, but it still had a significant adverse effect on DFS. Although our results differ from those of the Heidelberg study,18 there are some major differences between these two studies that partly might explain the discrepancy between the findings. First, even though the study populations are of the same size, they differed from each other by stage of the disease. In our study, all patients had axillary lymph node metastases. The Heidelberg study18 included all patients with tumor cells in bone marrow, and among those, 48% were node-negative patients with better prognosis. Furthermore, the impact of clodronate might be different in bone marrow–negative patients. Second, although in both studies clodronate was given orally 1,600 mg/d, the duration of the treatment was longer in our study than in the Heidelberg trial (3 v 2 years). Third, bone mineral density measured from our patients remained significantly higher in the clodronate-treated patients as compared with the controls,22,23 reflecting a good compliance, effective absorption of the drug, and a clear bone effect of clodronate. No compliance measurements were given in the report of the Heidelberg study. Fourth, the follow-up time was 2 years longer in our study (5 v 3 years), even though the difference between the study groups in the present study was seen already during the first year. Fifth, the number of events in our study was twice as high as in the Heidelberg study. Finally, both the intent-to-treat analysis and analysis of only eligible patients in our study gave the same results, whereas in the Heidelberg study, patients with poor compliance seem to have been excluded from the analyses. Likewise, the multivariate analyses in our study, which are not reported from the Heidelberg study, support an adverse effect of clodronate on DFS. Actually, in line with the previous studies, our results do not exclude a partial positive effect of clodronate on skeletal invasion. Even though clodronate did not prevent the development of bone metastases in the present study, bone as a first site of relapse was less frequent in the clodronate group than in the controls (11% v 33%, respectively). In the only trial in which prophylactic clodronate treatment was studied in women with advanced breast cancer but without manifest bone metastases, the total number of skeletal events was reduced significantly, but the number of patients who developed them was not reduced and no survival benefit was demonstrated.17 In all randomized studies of patients with advanced breast cancer, bisphosphonates have retarded, delayed, or postponed progression of skeletal events, but they have never totally inhibited their development.10-16 Recently, there has been a preliminary report of another adjuvant clodronate study22 that consists of 1,079 node-negative and node-positive primary breast cancer patients who were treated with oral clodronate with the same schedule as in the present study. After the treatment period of 2 years, fewer patients developed bone metastases in the clodronate group in postmenopausal women, but not in premenopausal women. However, the follow-up time is still short and the number of events relatively low (13%). No data on nonskeletal metastases are available. The influence of bisphosphonates on extraskeletal metastases has been poorly documented in the previous studies of advanced breast cancer. Only two studies touch the subject of extraskeletal metastases. In neither of these were extraskeletal metastases influenced by bisphosphonates, but the follow-up time has been relatively short and the patients have been under the anticancer treatment at the same time.14,23 Despite the strong statistical evidence in favor of a deleterious effect of clodronate on breast cancer progression in the present trial, the lack of a plausible mechanism for such an effect still requires a cautious attitude toward our findings. However, it is surprising that bisphosphonates would act almost exclusively on calcified tissues in view of the large array of their effects on mononuclear phagocyte lineage.8 Bisphosphonates induce apoptosis in osteoclasts and macrophage-like cells, and they can suppress the activity of phagocytotic cells.4-7 Depression of monocyte-macrophage antitumor activity may lead to a consequent increase of extraskeletal metastases. Although there are little data supporting the importance of immunologic factors in breast cancer, a study of low-dose continuous prednisone added to adjuvant CMF chemotherapy increased the risk for relapse and second malignancies in premenopausal breast cancer patients.24 Interestingly, there are some studies of bisphosphonate-treated mice in which new potent amino bisphosphonates inhibit osteolytic bone metastases through promoting the apoptosis of osteoclasts and breast cancer cells in bone, but the agents have no effect on the apoptosis in breast cancer cells at the primary site. However, single use of bisphosphonate without simultaneous anticancer therapy, despite profound suppression of bone metastases, increased adrenal metastases depending on administration protocol.25,26 We conclude that adjuvant clodronate treatment does not prevent the development of bone metastases in patients with node-positive breast cancer. Instead, the patients treated with clodronate seemed to be at excessive risk for the development of nonskeletal metastases and local recurrence. Because the number of skeletal recurrences was not increased, a partial inhibitive effect on skeletal recurrence cannot be excluded. More studies are needed to verify our results and elucidate the mechanism of this effect.
Supported by Leiras Pharmaceutical Company, Helsinki, Finland. We thank study nurse Inga Skog for her excellent care of the patients in the study as well as for the handling of the patients records and case report forms.
1. Coleman RE, Rubens RD: The clinical course of bone metastases from breast cancer. Br J Cancer 55: 61-66, 1987[Medline] 2. Sahni M, Guenther HL, Fleisch H, et al: Bisphosphonates act on rat bone resorption through the mediation of osteoblasts. J Clin Invest 91: 2004-2011, 1993 3. Nishikawa M, Akatsu T, Katayama Y, et al: Bisphosphonates act on osteoblastic cells and inhibit osteoclast formation in mouse marrow cultures. Bone 18: 9-14, 1996[Medline] 4. Hughes DE, Wright KN, Uy HL, et al: Bisphosphonates promote apoptosis in murine osteoclasts in vitro and in vivo. J Bone Miner Res 10: 1478-1487, 1995[Medline] 5. Selander KS, Mönkkönen J, Karhukorpi E-K, et al: Characteristic of clodronate-induced apoptosis in osteoclasts and macrophages. Mol Pharmacol 50: 1127-1138, 1996[Abstract] 6. Rogers MJ, Chilton KM, Coxon FP, et al: Bisphosphonates induce apoptosis in mouse macrophage-like cells in vitro by a nitric oxide independent mechanism. J Bone Miner Res 11: 1482-1491, 1996[Medline] 7. Espenes A, Press C, van Rooijen N, et al: Apoptosis in phagocytotic cells of lymphoid tissue in rainbow trout (Oncorhynchus mykiss) following administration of clodronate liposomes. Cell Tissue Res 289: 323-331, 1997[Medline] 8. Fleisch H: Bisphosphonates in Bone Disease: From the Laboratory to the Patient, ed 2. New York, NY, Parthenon Publishing Group, 1995, pp 46-50 9. Shipman CM, Rogers MJ, Apperley JF, et al: Bisphosphonates induce apoptosis in human myeloma cell lines: A novel anti-tumor activity. Br J Hematolog 98: 665-672, 1997 10. Elomaa I, Blomqvist C, Gröhn P, et al: Long-term controlled trial with diphosphonate in patients with osteolytic bone metastases. Lancet 1: 146-149, 1983[Medline] 11. Martoni A, Guaraldi M, Camera P, et al: Controlled clinical study on the use of dichloromethylene diphosphonate in patients with breast carcinoma metastasizing to the skeleton. Oncology 48: 97-101, 1991[Medline] 12. Paterson AH, Powles TJ, Kanis JA, et al: Double-blind controlled trial of oral clodronate in patients with bone metastases from breast cancer. J Clin Oncol 11: 59-65, 1993[Abstract]
13.
van Holten-Verzantvoort ATM, Kroon HM, Bijvoet OL, et al: Palliative pamidronate treatment in patients with bone metastases from breast cancer. J Clin Oncol 11: 491-498, 1993 14. Conte PF, Latreille J, Mauriac L, et al: Delay in progression of bone metastases in breast cancer patients treated with intravenous pamidronate: Results from a multinational randomized controlled trial. J Clin Oncol 14: 2552-2559, 1996[Abstract] 15. Hortobagyi GN, Theriault RL, Lipton A, et al: Long-term prevention of skeletal complications of metastatic breast cancer with pamidronate. J Clin Oncol 16: 2038-2044, 1998[Abstract]
16.
Theriault RL, Lipton A, Hortobagyi GN, et al: Pamidronate reduces skeletal morbidity in women with advanced breast cancer and lytic bone lesions: A randomized, placebo-controlled trial. J Clin Oncol 17: 846-854, 1999 17. Kanis JA, Powles T, Paterson AHG, et al: Clodronate decreases the frequency of skeletal metastases in women with breast cancer. Bone 19: 663-667, 1996[Medline]
18.
Diel I, Solomayer E, Costa SD, et al: Reduction in new metastases in breast cancer with adjuvant clodronate treatment. N Engl J Med 339: 357-363, 1998
19.
Saarto T, Blomqvist C, Välimäki M, et al: Chemical castration induced by adjuvant CMF therapy causes a rapid bone loss which is reduced by clodronate: A randomized study in premenopausal breast cancer women. J Clin Oncol 15: 1341-1347, 1997 20. Saarto T, Blomqvist C, Välimäki M, et al: Clodronate improves bone mineral density in postmenopausal breast cancer women treated with adjuvant antiestrogens. Br J Cancer 75: 602-605, 1997[Medline] 21. Soloway MS, Hardeman SW, Hickey O, et al: Stratification of patients with metastatic prostate cancer based on extent of disease on initial bone scan. Cancer 61: 195-202, 1998 22. Powles TJ, Paterson AHG, Nevantaus A, et al: Adjuvant clodronate reduces the incidence of bone metastases in patients with primary operable breast cancer. Proc Am Assoc Clin Oncol 17: 123a, 1998 (abstr 468) 23. Elomaa I, Blomqvist C, Porkka L, et al: Diphosphonates for osteolytic metastases. Lancet 1: 1155-1156, 1985
24.
Marini G, Murray S, Goldhirsch A, et al: The effect of adjuvant prednisone combined with CMF on patterns of relapse and occurrence of second malignancies in patients with breast cancer. Ann Oncol 7: 245-250, 1996 25. Yoneda T, Michigami T, Bing Yi, et al: Actions of bisphosphonates on bone metastasis in animal models of breast cancer. Cancer 88: 2979-2988, 2000[Medline]
26.
Sasaki A, Boyce BR, Story B, et al: Bisphosphonate risedronate reduces metastatic human breast cancer burden in bone in nude mice. Cancer Res 55: 3551-3557, 1995 Submitted August 4, 1999; accepted July 27, 2000. This article has been cited by other articles:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
2 test and Mann-Whitney U nonparametric test were used to assess differences in the frequency of individual prognostic factors between the groups. Statistical data processing was carried out with the SPSS computer software for Macintosh (SPSS, Inc, Chicago, IL). 
2 cm: relative risk [RR] = 1.48; 95% confidence interval [CI], 0.73 to 3.00; P = .28; tumor size > 2 cm: RR = 2.01; 95% CI, 1.25 to 3.23; P = .004) and higher numbers of nodal metastases (nodal number of one to three: RR = 1.74; 95% CI, 1.02 to 2.99; P = .04; nodal number > 3: RR = 2.19; 95% CI, 1.20 to 3.97; P = .01). In ER-positive patients, RR of disease relapse was 1.30 (95% CI, 0.77 to 2.20; P = .32), but in ER-negative patients, the RR was 2.29 (95% CI, 1.17 to 4.50; P = .016). No clear difference in the RR of recurrence was seen between PgR-positive (RR = 1.57; 95% CI, 0.84 to 2.96; P = .16) and PgR-negative cases (RR = 1.50; 95% CI, 0.87 to 2.56; P = .14). Five-year DFS in ER-positive patients was 66% in the clodronate group and 72% in the controls (P = .32), whereas in ER-negative patients, the DFS difference between the study groups was highly significant (35% v 64%, respectively; P = .013) (
