Fax: (212) 305-0178
A Phase II trial of docetaxel and estramustine in patients with refractory metastatic breast carcinoma
Article first published online: 17 JAN 2003
Copyright © 2003 American Cancer Society
Volume 97, Issue 3, pages 537–544, 1 February 2003
How to Cite
Tiersten, A. D., Nelsen, C., Talbot, S., Vahdat, L., Fine, R., Troxel, A., Brafman, L., Shriberg, L., Antman, K. and Petrylak, D. P. (2003), A Phase II trial of docetaxel and estramustine in patients with refractory metastatic breast carcinoma. Cancer, 97: 537–544. doi: 10.1002/cncr.11082
- Issue published online: 17 JAN 2003
- Article first published online: 17 JAN 2003
- Manuscript Accepted: 11 SEP 2003
- Manuscript Revised: 6 SEP 2002
- Manuscript Received: 27 JUN 2002
- Aventis Pharmaceuticals
- breast carcinoma
The similarity between the mechanism of action between docetaxel and estramustine generated the hypothesis of synergistic antimicrotubule effects and cytotoxicity when the two agents are combined. In addition, it has been demonstrated that estramustine binds P-glycoprotein in vitro and, thus, may prevent the efflux of taxanes in tumors that over-express P-glycoprotein. To further evaluate the combinations clinical efficacy and safety, a trial was performed in heavily pretreated patients with metastatic breast carcinoma (MBC).
Thirty-six patients with MBC were treated with estramustine 900 mg/m2 per day divided into 3 doses given on Days 1–3 and docetaxel 70 mg/m2 given by intravenous administration over 1 hour on Day 3 after the first dose of estramustine, every 21 days. Patients may have received any number of prior chemotherapy regimens for MBC.
Nine partial responses were observed in 31 assessable patients, for an objective response rate of 29% (95% confidence interval, 14–48%). The median progression free survival was 4 months (range, 1–41 months), and the median overall survival was 17 months (range, 2–45 months). Severe toxicities (Grade 3 or 4) were neutropenia, hypophosphatemia, and thrombosis. Seventy-five percent of patients experienced either an improvement or no change in quality of life.
The combination of docetaxel and estramustine produced responses in heavily pretreated women with MBC while maintaining quality of life. Cancer 2003;97:537–44. © 2003 American Cancer Society.
In 2001, 192,200 women will be newly diagnosed with breast carcinoma, and it will lead to 40,800 deaths.1 Initial standard chemotherapy for patients with metastatic breast carcinoma (MBC) produces complete response (CR) rates and partial response (PR) rates of 50–65% without a major prolongation of survival.2 Combination chemotherapy for patients with refractory, advanced breast carcinoma yields a significantly lower response rate.3 Complete responses are rare for this group of patients. The response to endocrine therapy for advanced disease is in the range of 30–40% for patients with hormone receptor positive tumors.2 Improvement in these response rates has been the subject of intense investigation, and new agents and/or strategies clearly are needed.
The introduction of the taxanes (docetaxel and paclitaxel) offered hope in this regard. Docetaxel is a semisynthetic taxoid that was derived originally from the European yew tree. It inhibits tubulin depolymerization, thereby disrupting mitosis and causing cell death.4 Docetaxel and paclitaxel compete for a common taxane-binding site on the β-tubulin protein, for which docetaxel has approximately a two times higher affinity compared with paclitaxel.5 Not surprisingly, then, it has been shown that docetaxel is almost twice as potent as paclitaxel in its antimitotic activity.4 In a Phase II study investigating the effects of this agent in heavily pretreated and anthracycline-resistant patients with MBC, the overall response rate was 40%.6 Several Phase III randomized trials also have supported the use of docetaxel in the treatment of patients with MBC. In one study that evaluated doxorubicin 75 mg/m2 and docetaxel 100 mg/m2 every 3 weeks, patients with MBC demonstrated a greater response rate and time to progression in the docetaxel arm (33.3% vs. 47.8%; P = 0.008; 21 weeks vs. 26 weeks, respectively).7 In addition; compared with mitomycin C and vinblastine in patients who previously failed an anthracycline-containing regimen, improvements in response rate, time to disease progression, and overall survival were seen with docetaxel (30% vs. 11.6%, P < 0.0001; 19 weeks vs. 11 weeks, P = 0.001; and 11.4 months vs. 8.7 months, P = 0.0097).8 Thus, docetaxel has proven promising in the treatment of patients with MBC.
Preclinical and clinical studies suggest that the effects of docetaxel in treating patients with other malignancies, like hormone-refractory prostate carcinoma (HRPC), are enhanced when used in combination with other antimitotic agents like estramustine.9, 10 The uptake of estramustine is mediated by an estramustine-binding protein (EMBP), a protein distinct from the estrogen receptor. EMBPs have been discovered on the surface of breast carcinomas and brain tumors as well.11–13 Once inside the cell, estramustine depolymerizes microtubules by binding to both microtubule associated proteins (MAPs) and tubulin.14–16
The similarity between the mechanism of action of the taxanes and estramustine engendered the hypothesis that synergistic antimicrotubule effects and cytotoxicity could be achieved by combining the two. Preclinical studies demonstrating the additive effects of estramustine and vinblastine, another antimicrotubule agent, on the suppression of microtubule dynamics lend support to this theory.16 It also has been shown that estramustine binds P-glycoprotein in vitro and, thus, may have the added effect of preventing the efflux of taxanes in P-glycoprotein over-expressing tumors.17
Phase I and II studies have investigated the combinations of docetaxel and estramustine in patients with HRPC. An analysis of two such Phase I studies suggested that the maximum tolerated dose of docetaxel is 70 mg/m2 for minimally pretreated patients and 60 mg/m2 for patients with extensive pretreatment.18, 19 In addition, there appears to be less toxicity with a 5-day schedule of estramustine administration (280 mg) compared with a 21-day regimen (14 mg/kg per day).18, 19 A Phase II study demonstrated that docetaxel (70 mg/m2), estramustine (10 mg/kg per day for 5 days), and low-dose oral hydrocortisone (30 mg in the morning and 10 mg in the evening) was an effective and tolerable treatment for men with HRPC.20 It is noteworthy that the response rates were higher in that study (50%) compared with the rates attained in studies that investigated estramustine in combination with other antimicrotubule agents, including vinblastine and paclitaxel. Furthermore, the median survival (20 months) was longer than that reported in other Phase II studies. Based on these encouraging findings, we undertook an investigation into the combined effects of docetaxel and estramustine in patients with MBC.
MATERIALS AND METHODS
Informed content was obtained from all patients prior to enrollment. Patients with histologically confirmed, measurable, or evaluable MBC were eligible. Measurable disease was defined as a lesion measuring at least 1 cm × 1 cm and bidimensionally measurable by physical examination or radiologic investigation. Evaluable disease was assessable, but not measurable, and was documentable on radiographs. Any number of prior chemotherapy or hormonal therapy regimens were allowed if 4 weeks had elapsed since the completion of chemotherapy (for weekly regimens, 2 weeks off therapy sufficed). Required clinical parameters included a life expectancy of 2 months, age > 18 years, and an Eastern Cooperative Oncology Group performance status of 0–2. Patients with stable brain metastases were allowed as well as patients with a single lesion that was amenable to resection by surgery. Required initial laboratory values included a granulocyte count ≥ 3000/μL; a platelet count ≥ 100,000/μL; and creatinine, bilirubin, blood urea nitrogen, and aspartate aminotransferase or alanine aminotransferase levels < 1.5 times the institutional upper limit of normal. Patients who were pregnant or lactating and patients with symptomatic ascites, pleural effusions, peripheral edema, and/or a history of thromboembolism (unless they currently were undergoing therapeutic anticoagulation) were excluded from the trial. All pathology was reviewed centrally.
Treatment and Evaluation
Treatment consisted of estramustine 900 mg/m2 per day (divided into 3 doses administered on Days 1–3) and docetaxel 70 mg/m2 (by intravenous administration over 1 hour on Day 3 after the morning dose of estramustine) every 21 days. All patients received dexamethasone 8 mg twice daily for 5 doses starting 24 hours prior to docetaxel. The dose of docetaxel was reduced by 10% for patients with nadir absolute neutrophil counts < 500/μL only if the last course was complicated by documented neutropenic fever or if the platelet count was < 30,000/μL. Patients were not retreated unless their granulocyte count was > 3000/μL and their platelet count was > 100,000/μL within 2 days prior to starting estramustine. The dose of docetaxel also was reduced by 10% if treatment delay was > 1 week for hematologic toxicity. If a treatment delay was > 2 weeks (for hematologic toxicity), then the patient went off study.
At the outset of the study, no prophylactic anticoagulant was administered. After 7 thrombotic events were observed during 136 cycles of treatment, the protocol was amended to include prophylactic dalteparin (5000 UI injected subcutaneously daily). On this regimen, there were 2 additional clots during 26 cycles of treatment. Therefore, dalteparin was discontinued, and enteric-coated aspirin 325 mg and oral warfarin 2 mg daily starting on the first day of the estramustine dose were added.
Patients were evaluated at baseline by physical examination, chest X-ray, computed tomography (CT) scan of involved areas, head CT scan, bone scan, electrocardiogram, carcinoembryonic antigen (CEA) level, and CA15-3 level. Scans were repeated every three cycles while on study. The CEA and CA15-3 levels were measured after every cycle they were abnormal initially. Weekly complete blood counts were required during treatment. Physical examinations were conducted on Day 1 of each cycle of treatment. Tumor measurements were made then if they were assessable by physical examination or were made every three cycles if they were measurable only on X-ray or scan. Drug toxicity was evaluated on Day 1 of each cycle of treatment, and quality of life was assessed using the Manitoba Cancer Treatment and Research Foundation Functional Living Index: Cancer (FLIC) prior to starting treatment and on Day 1 of every third cycle.
Patients were treated until disease progression, the development of treatment-limiting toxicity, or withdrawal of consent. If a CR was achieved, then two additional cycles of chemotherapy were given. For patients with measurable disease, CR was defined by complete disappearance of all clinical evidence of tumor and all symptoms of disease. Patients with evaluable disease had to achieve the complete disappearance of all clinical evidence of disease and all bony lesions, and they needed to be free of all symptoms of disease to be considered a complete responder. A PR in patients with measurable disease and evaluable disease was defined as a decrease ≥ 50% in the sum of the products of the greatest perpendicular dimensions of all measured lesions compared with baseline measurements and a definite improvement in evaluable disease estimated to be in excess of 50%, respectively. In addition, for both measurable disease and evaluable disease, there must have been no simultaneous increase in the size of any lesion by ≥ 25% or appearance of a new lesion, and all nonmeasurable lesions must have remained stable or must have regressed. A PR in patients with evaluable, bone-only disease was defined by no new lesions, symptomatic improvement, and a decrease > 50% in tumor markers from baseline or normalized values. Stable disease required no significant change in disease status and no appearance of new lesions. Progressive disease for patients with measurable disease required a ≥ 25% increase in the area of any malignant lesion measuring > 2 cm2 in the sum of the products of the greatest perpendicular dimension of individual lesions in a given organ site. Tumor progression also was defined as significant clinical deterioration that could not be attributed to treatment or other medical conditions. For patients with evaluable disease, disease progression was defined by an unequivocal increase in area of any evaluable lesion by ≥ 25% or by the appearance of a new lesion. Time to treatment failure was measured from Day 1 of induction of chemotherapy to the day of death or disease progression.
Statistical Design and Analyses
The objectives of this Phase II trial were to determine the efficacy and safety of the combination of estramustine and docetaxel in women with refractory MBC. We used the two-stage minimax design of Simon.21 In the first stage, we enrolled 27 women. If two or fewer patients responded, then the trial would be terminated due to lack of efficacy. If > 3 patients responded, then 13 additional patients were enrolled, for a total of 40 patients. If six or fewer patients had an objective response, then the regimen would be rejected for lack of efficacy. With this design, we had a 90% chance of rejecting the new regimen if its response rate was ≤ 10% and a 90% chance of accepting the regimen if its response rate was ≥ 25%. With 40 women in the trial, we had a 90% chance of detecting any unexpected toxicity that had a prevalence of at least 5.6%. A 95% confidence interval (95%CI) for the response rate extended no more than ± 16% around the observed response rate. We summarized survival and progression free survival by using Kaplan–Meier curves. Survival curves were compared using the log-rank test. Categoric variables were compared between groups using the Fisher exact test; all tests were two-sided.
The pretreatment characteristics of the women entered on the trial are listed in Table 1. Between March 1997 and January 2001, 36 eligible women with MBC were enrolled. The median age of patients was 56 years (range, 33–79 years).
|Characteristic||No. of patients||%|
|Patients on study||36||100|
|Performance status (ECOG)|
|< 2 regimens||10/21||48|
|≥ 2 regimens||11/21||52|
|High-dose chemotherapy with transplant|
|Stage of initial disease presentation|
|Disease free interval (yrs)|
|Sites of disease|
|Brain (treated, stable)||3||4|
|No. of organ sites involved|
|ER or PR status|
Hormonal therapy, chemotherapy, and radiation therapy were received by 19 women (53%), 20 women (56%), and 17 women (47%), respectively, in the adjuvant setting, and by 12 women (33%), 21 women (58%), and 11 women (31%), respectively, in the metastatic setting. Prior anthracycline treatment was received by 16 women (44%) in the adjuvant setting and by 9 women (25%) in the metastatic setting. In addition, 1 woman received adjuvant paclitaxel (3%), and 13 women (36%) received paclitaxel in the metastatic setting. The majority of women had one or two metastatic sites.
Treatment-related toxicities are listed in Table 2. The majority of treatment-related toxicities were of mild-to-moderate grade (78%). The most severe toxicities (Grade 3 or 4) were neutropenia and hypophosphatemia. There were 23 episodes of Grade 3 or 4 neutropenia. There were no treatment delays or dose reductions as a consequence of neutropenia. However, there were five episodes of neutropenic fever in three patients, requiring a 10% dose reduction of docetaxel, and five patients required granulocyte-colony stimulating factor.
|Toxicity||No. of episodes||Grade 1||Grade 2||Grade 3||Grade 4|
|Patients requiring G-CSF||5/36||—||—||—||—|
|Patients requiring blood transfusions||3/36||—||—||—||—|
|Alkaline phosphatase elevation||2||0||0||2||0|
|Total||368||216 (59%)||71 (19%)||61 (17%)||20 (5%)|
There were nine thrombotic events (one stroke and eight incidents of deep vein thrombosis [DVT]) in seven patients. Of the incidents of DVT, three were of the lower extremity, and five were of the upper extremity. Two of the five upper extremity DVTs were associated with central venous catheterization. With no prophylaxis, there was a total of seven thrombotic events during 136 cycles (5%). Prophylactic dalteparin was given for a total of 30 treatments but was discontinued after 2 additional thrombotic events were observed (8%). Dalteparin was replaced with warfarin and aspirin for the final 26 treatments of the study, and no thrombotic events were observed. The differences in the frequency of thrombosis seen with warfarin and dalteparin and with warfarin and no prophylaxis were analyzed for statistical significance. There were no statistically significant differences among the three types of prophylaxis (none, dalteparin, and warfarin) with respect to the rate of thrombosis. There were two episodes of gastrointestinal bleeding during the study, one of which occurred on the warfarin and aspirin regimen.
The nonhematologic toxicities are listed in Table 2. Grade 3–4 toxicities included edema (n = 6 patients), pulmonary toxicity (n = 3 patients), infusion reactions (n = 1 patient), myalgia (n = 1 patient), stomatitis (n = 1 patient), and unilateral deafness (n = 1 patient). Grade 3 hypophosphatemia occurred 14 times in 8 patients. Nausea, emesis, diarrhea, constipation, and fatigue were common but mild.
There were 21 treatment delays. Twelve delays were due to scheduling conflicts, and 9 delays were due to treatment-related toxicities (DVT, edema, pulmonary toxicity, nausea and emesis, and thrombocytopenia). Fifteen of 36 patients went off the study due to toxicity (42%). Another 15 patients eventually went off the study as a result of progressive disease (42%).
Response to Treatment
There were 192 cycles administered in total. The mean number of treatments per patient was 4 cycles (range, 1–16 cycles). Thirty-one of 36 patients who were eligible for the study were assessable for response (Table 3). There were no CRs. Nine of 31 patients were confirmed partial responders (29%; 95%CI, 14–48%). The median duration of response in partial responders was 12 months (range, 4–21 months). Thirteen patients had stable disease (42%; 95%CI, 25–61%), and 9 patients had progressive disease (29%; 95%CI, 14–48%). Four of 13 patients (31%) who had received prior paclitaxel achieved a PR. The response rate in patients who had previously experienced disease progression after receiving treatment with an anthracycline was 27%. The median time to response was 9 weeks. Table 4 lists the characteristics of responders.
|Measure||No. of patients||%|
|No. of patients assessable for response||31/36||86|
|No. of cycles administered|
|Time to progression (mos)|
|Overall survival (mos)|
|No. of responses|
|Overall response rate||29|
|Median time to response (wks)||9||—|
|Response duration (mos)|
|Response rate in paclitaxel failures||4/13||31|
|Response rate in anthracycline failures||6/22||27|
|Patient||Age (yrs)||ECOG performance status||No. of prior treatment regimens||Sites of disease||Time to progression (mos)||ER/PR status||HER-2 status|
|1||42||2||3||Lung, bone, soft tissue||9||P||N|
|4||62||0||2||Soft tissue, lung||21||P||N|
|6||56||0||2||Bone, soft tissue||14||P||N|
|7||79||1||1||Lymph node, lung||4||P||P (3+)|
|8||60||1||2||Bone, lymph node, Lung, liver||4||P||N|
Time to Disease Progression and Survival
The median follow-up was 18 months (range, 2–45 months). Four of 36 patients remained progression free at the time of this writing. The median progression free survival was 4 months (range, 1–41 months). Twenty-one patients have died. The median overall survival was 17 months (range, 2–45 months) (Fig. 1).
Quality of Life
Twenty-four patients completed the FLIC quality-of-life questionnaire at baseline and every 3 weeks, when applicable. Six of those patients experienced an increase > 10 points in their quality of life (25%), 12 patients reported no change (50%), and 6 patients reported a deterioration ≥ 10 points in their quality of life (25%).
This study showed that the combination of docetaxel and estramustine is an active chemotherapy regimen in patients with refractory MBC, with an objective response rate of 29% (95%CI, 14–48%). The median progression free survival and overall survival were 4 months and 17 months, respectively. The median response duration was 12 months. In addition, response rates of 31% and 27% were seen in patients who had progressed after prior paclitaxel therapy and anthracycline therapy, respectively.
Response rates from Phase II studies of estramustine monotherapy in patients with advanced breast carcinoma range from 27% to 39%.22, 23 The use of docetaxel alone in previously untreated patients with MBC has elicited response rates in the range of 40–68%; and, when combined with anthracyclines, the response rates have ranged from 53–57%.24–27 Considering our refractory population, we hypothesized at the outset of this study that a response rate of 25% would be worthy of consideration of further study. Our response rate exceeded that goal.
Our median progression free survival and overall survival of 4 months and 17 months, respectively, compare favorable with survival results reported when docetaxel has been administered as first-line therapy (5 months and 10 months, respectively).25 The median progression free survival ranged from 4 months to 5 months, and the median overall survival was approximately 9 months in an anthracycline-resistant population of patients who were treated with docetaxel.26, 27 This suggests that the combination of docetaxel and estramustine may confer a survival benefit compared with docetaxel alone.
The majority of treatment-related toxicities experienced with this regimen were expected and reversible. Notable toxicities included hypophosphatemia and thrombosis, both of which have been reported previously and were attributed to estramustine.
Although the hypophosphatemia was reversible and asymptomatic, the high incidence of thrombosis presents a serious limitation to using this estramustine-containing regimen in the treatment of patients MBC. There were 9 episodes of thrombosis on this study, 8 DVTs and 1 embolic stroke, which occurred in 7 patients over 188 cycles of treatment. The incidence of thrombosis per patient was 18%. The incidence of thrombosis in other studies using estramustine, although not quite as high, is significant. Although it was not statistically significant, there was a numeric difference between the frequency of thrombosis observed in patients on warfarin and aspirin (0%) and the frequency of thrombotic events observed in patients on dalteparin (8%) and no prophylaxis (5%). It should be noted that there is evidence to show that advanced malignancy, in itself, predisposes patients to thrombosis. Although it has been shown that antitumor drug therapy increases the risk of thrombosis in patients with malignant disease, patients with MBC are at a significantly higher risk for developing thromboses on a specific therapeutic regimen compared with patients who are treated adjuvantly with the same regimen.28 In a meta-analysis of 10 studies encompassing 6844 patients with Stage I–IV breast carcinoma, the incidence of thrombosis in patients with Stage I–II disease was 2.1% compared with 17.6% observed in patients with Stage IV disease.28 It is difficult to determine how many of the thrombotic events in our patients were due to estramustine and how many were due to the disease itself. Anticoagulation has proven successful in preventing thrombotic events and also may be effective in preventing tumor spread and metastasis, thus improving survival.29 Quality of life was maintained in most women who were treated with the combination of docetaxel and estramustine in this study. Seventy-five percent of these patients experienced either no change or an improvement in their quality of life. This is significant, because quality of life is an important end point when examining the efficacy of chemotherapy regimens for patients with MBC because of the palliative nature of the treatment for metastatic disease.
In conclusion, the combination of docetaxel and estramustine is an active regimen in patients with refractory MBC. Although response rates are lower compared with the rates seen in some studies that have investigated the efficacy of docetaxel or estramustine as monotherapy, they are encouraging considering the highly refractory and heavily pretreated nature of our patient population. Our median progression free survival, overall survival, and response duration rates compare favorably with the rates from other Phase II studies that examined the effects of docetaxel in similarly pretreated patient populations. Perhaps most importantly, this combination was effective in the maintenance and improvement of our patients' quality of life.
Supported in part by Aventis Pharmaceuticals. Aventis also provided honoraria for speaking about the study
- 2Treatment of metastatic breast cancer. In: HarrisJR, LippmanME, MorrowM, et al., editors. Diseases of the breast. Philadelphia: Lippincott Williams and Wilkins, 2000: 749., , .
- 13Estramustine binding protein in human brain tumor tissue. J Neurosurg. 1989; 14: 27–43., , , et al.
- 24Docetaxel as a single-agent therapy in metastatic breast cancer: clinical efficacy. Semin Oncol. 1997; 24(4 Suppl 13 ): S13-1–S13-18..