A Phase I Study of Estramustine, Weekly Docetaxel, and Carboplatin Chemotherapy in Patients with Hormone-Refractory Prostate Cancer Free

Prostate cancer remains the second leading cause of cancer death in men in the United States (1). Although initially responsive to androgen deprivation therapy, prostate cancer almost always becomes refractory to hormonal manipulation (2) Mitoxantrone chemotherapy combined with corticosteroids provides palliative benefit in the management of symptomatic disease, but no improvement in survival (3, 4). Phase I and II trials have shown that the combination of docetaxel and estramustine is effective and well tolerated in patients with hormone refractory prostate cancer HRPC; refs. (5–9). Two recently reported large phase III trials showed that docetaxel combined with either prednisone or estramustine was associated with a survival benefit compared with mitoxantrone and prednisone in the treatment of HRPC (10, 11).

Platinum compounds have been studied both as single agents and in combination regimens in the treatment of hormone refractory prostate cancer (12–22). Yagoda et al. (12) reported a 12% partial remission and 24% stable disease rate in 25 patients with measurable HRPC treated with cisplatin. Merrin (13) noted partial or complete responses in 17 of 54 (31%) patients treated with weekly cisplatin, whereas Moore et al. (14) found a 10% response rate in 29 patients treated with a similar regimen. Carboplatin has a distinct toxicity profile from cisplatin that allows for potentially equivalent antitumor effect with less toxicity (15). Miglietta et al. (16) treated 35 patients with weekly carboplatin at a dose of 150 mg/m². Prostate-specific antigen (PSA) declines of ≥50% were seen in 28%, with a mean duration of about 6 months. Also, 27% with measurable disease had a partial response. Jungi et al. (17) treated 27 patients with carboplatin 400 mg/m² every 28 days and noted decreased pain and/or an improved performance status in 48%.

Platinum compounds also have been studied in combination regimens. Cisplatin or carboplatin was combined with etoposide and estramustine in men with poorly differentiated HRPC (18). Fifteen of 30 patients experienced tumor response by PSA declines or decrease in measurable disease. Several trials have evaluated the efficacy of estramustine, paclitaxel, and carboplatin (19–21). In one multicenter trial, 56 patients were treated with estramustine, weekly paclitaxel, and carboplatin every 4 weeks. PSA declines of ≥50% were seen in 67%, and measurable responses were noted in 15 of 33 patients (45%; ref. 19). The Cancer and Leukemia Group B completed a phase II triall of estramustine, docetaxel and carboplatin given on an every 3 week schedule and showed PSA declines >50% in 68% of patients and measurable responses in 52% (22).

In summary, platinum compounds have modest but distinct single agent activity in HRPC. However, combined with taxanes and estramustine, carboplatin may offer the greatest potential for benefit. In addition, since weekly docetaxel chemotherapy is associated with less myelotoxicity than every 3 week docetaxel, particularly in older patients who may have received prior radiotherapy, we undertook this phase I trial to determine the maximum tolerated dose as well as the safety and efficacy of weekly docetaxel in combination with estramustine and carboplatin.

Eligibility. Patients had histologically confirmed prostate cancer and showed progression despite androgen deprivation therapy. Progression was defined as either an increase in the product of bidimensional diameters of one or more radiographically documented sites of measurable disease or by PSA progression according to the PSA Working Group criteria (23). Patients had Eastern Cooperative Oncology Group performance statuses of 0 to 2. Required laboratory values included a WBC count ≥3,000/mL, platelet count ≥100,000/mL, serum creatinine ≤2.0 mg/dL, bilirubin less than or equal the upper limit of normal, AST ≤1.5 times the upper limit of normal, and testosterone levels ≤50 ng/mL. Prior chemotherapy was allowed if it did not include taxanes or platinum derivatives. Patients with significant peripheral neuropathy and current congestive heart failure (New York Heart Association Class 2 or higher) were excluded. At least 4 weeks must have elapsed from radiation or antiandrogen therapies. No patients were allowed with myocardial infarction or deep venous thrombosis within 1 year or significant change in anginal pattern within 6 months of study entry. All patients signed an informed consent form approved by the institutional review board of the Dana-Farber/Partners Cancer Care before participating in this trial.

Treatment Plan Patients were treated with one of five escalating doses of docetaxel for three consecutive weeks followed by a rest week. Docetaxel was given i.v. on days 2, 9, and 16 of a 28-day cycle. Patients received the same dose of treatment to which they were assigned for each cycle of chemotherapy, unless a dose-reduction was instituted secondary to toxicity. Dexamethasone was given orally at a dose of 8 mg, ∼12 hours and 1 hour prior to and again 12 hours after docetaxel. Carboplatin was given i.v. at a doseAUC (6) or (5), over 60 minutes, on day 2 of each cycle (24). Creatinine clearance was estimated using the Cockcroft-Gault formula (25). Estramustine was given orally at a dose of 140 mg three times daily on days 1 to 5, 8 to 12, and 15 to 19 of each cycle. Treatment was discontinued at any time if patients had unacceptable toxicity, evidence of progressive disease, occurrence of dose-limiting toxicity, or withdraw consent.

Maximum tolerated dose was defined as one dose level below that at which two or more patients experienced dose-limiting toxicity. Dose-limiting toxicity was defined as an absolute neutrophil count of <500 mm³, a platelet count of <100,000 mm³ for >7 days, or nonrecovery of absolute neutrophil count to >1,500 mm³ or platelets to >100,000 mm³ within 14 days of a given dose of docetaxel. Dose-limiting toxicity also included any irreversible grade 3 or 4 nonhematologic toxicity, excluding nausea, vomiting, alopecia, and hypersensitivity reactions.

Toxicity and Response Evaluation Patients were followed weekly. Patients were restaged every 2 months, including computed tomography scans of the abdomen and pelvis, bone scans, and chest X-ray s. Treatment toxicity was graded according to the National Cancer Institute Common Toxicity Criteria. All patients with evidence of measurable disease underwent tumor response evaluated by the Response Evaluation Criteria in Solid Tumors (26). PSA response was defined according to the PSA Working Group and required a ≥50% reduction in PSA sustained for at least 4 weeks (23).

Statistical Methods. Descriptive statistics were used to characterize patients at baseline. Exact binomial confidence intervals were provided for response rates. The method of Kaplan-Meier (27) was used to estimate overall survival.

Thirty-two patients were registered for this trial. Two patients withdrew prior to starting therapy; one who was ineligible before treatment because of a deep venous thrombosis diagnosed within 1 year and a second who began treatment for brain metastases. Therefore, 30 patients are included in this analysis. Baseline patient characteristics are listed in Table 1. Most patients had received prior secondary hormonal therapies (53% ketoconazole, 33% PC-SPES) and 10% of patients had received prior chemotherapy with mitoxantrone. All patients had a performance status of 0 or 1.

This study was initiated with a carboplatin dose of AUC (6). Five patients were treated at this dose of carboplatin, three at the first dose level, and two at the second dose level. Two of three patients in the first dose level experienced significant dose delays and were removed from study after two and four cycles of treatment, respectively, because of delayed thrombocytopenia. One of two patients at the second dose level similarly experienced prolonged thrombocytopenia after his second cycle and was removed from study. Table 2 shows the toxicity seen in these five patients. In general, few grade 3 or 4 nonhematologic toxicities were seen. In addition, only one patient experienced grade 3 or 4 neutropenia and thrombocytopenia.

Because of concerns regarding significant dose delays and need for dose reductions using the carboplatin dose of AUC (6), the trial was reinitiated with a carboplatin dose of AUC (5). Table 3 shows the toxicities seen for 25 patients treated with five dose levels of docetaxel with carboplatin AUC (5). The decision was made to stop escalation at dose level 5 (docetaxel dose 43 mg/m²), as evidence from other studies suggested that the maximum tolerated dose of single agent docetaxel was 36 to 40 mg/m² and because of the significant activity seen at this dose level (28, 29). Therefore, an additional nine patients were accrued at dose level 5 to acquire more data about toxicity at this dose level. No dose-limiting toxicities were seen at all dose levels tested. No grade 3 or 4 myelosuppression was seen until dose level 5, when one patient each was noted to have grade 3 anemia, neutropenia, and thrombocytopenia. One patient developed grade 4 thrombocytopenia and two patients developed grade 4 neutropenia. Of 12 patients treated at dose level 5, four patients had a maximum grade 4 toxicity and seven patients had a maximum grade 3 toxicity.

Table 4 shows the antitumor efficacy noted among 30 treated patients. Among all patients, a PSA decline of ≥50% was seen in 63.3% [95% confidence interval (95% CI), 43.9-80.1]. At the recommended phase II dose (dose level 5), PSA declines of ≥50% were seen in 75.0% (95% CI, 42.8-94.5). Of 17 patients with measurable disease, four (23.5%; 95% CI, 6.8-49.9%) had at least a partial response, whereas 10 (58.8%) had a ≥50% reduction in PSA after treatment.

Five patients were not evaluable for objective response, four because of progression prior to a second cycle of treatment, and one because toxicity. In addition, two patients were unevaluable for PSA response, both because of progression. Eight patients are known to have died, all of prostate cancer. Median survival is 14.6 months. Figure. 1 shows overall survival and 95% CI on the survival estimate, calculated using the Kaplan-Meier method.

In this phase I dose escalation trial, estramustine, weekly docetaxel, and monthly carboplatin showed significant activity and an acceptable toxicity profile. Weekly docetaxel has been noted to be less myelosuppressive than every 3-week dosing (28, 29). In addition, estramustine may have a myeloprotective effect when combined with cytotoxic chemotherapy such as vinblastine (30). On the other hand, carboplatin given every 4 weeks can clearly cause neutropenia and thrombocytopenia. Indeed, in this trial, we found excessive delayed myelosuppression with a carboplatin dose of AUC (6), which required a dose adjustment of carboplatin to continue this study. However, when the carboplatin dose was decreased to AUC (5), the regimen was better tolerated and a maximum tolerated dose was not reached, although myelosuppression represented the most common grade 3 and 4 toxicities. Nonetheless, it is also important to note that, although this regimen had an acceptable toxicity profile, the median age of patients on trial was 64 years and the performance status was either 0 or 1. Proper patient selection remains an important consideration for clinicians considering the use of such regimens.

Although this was a phase I trial, data across all dose levels showed that the combination was active even at the lower doses of docetaxel, when combined with estramustine and carboplatin. At the recommended phase II dose, 9 of 12 (75%) of patients experienced a PSA decline of at least 50% and 2 of 5 (40%) patients with measurable disease showed at least a partial response. The degree to which carboplatin enhances clinical response and survival in patients with HRPC treated with estramustine plus docetaxel is uncertain. Five recent phase I or II trials involving 183 patients, including this report, have been published using estramustine, carboplatin and a taxane (19–22; Table 5). PSA declines of ≥50% have been seen in 60% to 100% of patients enrolled on these trials. Measurable responses were seen in 24% to 65% of patients with measurable disease, although if only patients who received the recommended phase II dose are included, the range is 40% to 65% across these trials.

The transition of prostate cancer from a hormone sensitive to hormone refractory state is associated with increased neuroendocrine differentiation (31). Markers of neuroendocrine differentiation, such as chromogranin A and neuron-specific enolase, are seen more frequently as cancers progress (32, 33). Since platinum compounds have activity against cancers with neuroendocrine differentiation, their incorporation into regimens for HRPC may represent an additional option for some patients.

Recent data presented at the 2004 Annual Meeting of the American Society of Clinical Oncology showed a survival benefit for docetaxel plus prednisone compared with mitoxantrone plus prednisone (11). In this trial, 1,006 patients were randomized to docetaxel 75 mg/m² every 3 weeks, docetaxel 30 mg/m² weekly for 5 of 6 weeks, and mitoxantrone 12 mg/m² every 3 weeks. All patients received prednisone 5 mg p.o. twice daily. Median survival was 18.9, 17.4, and 16.5 months, respectively. The hazard ratio was 0.76 (95% CI, 0.62-0.94; P = 0.0009) in the every 3-week docetaxel arm compared with the mitoxantrone arm. The weekly docetaxel arm did not show a statistically significant improvement in survival compared with mitoxantrone. As a result of this trial, docetaxel every 3 weeks plus prednisone has been approved by the Food and Drug Administration for the treatment of hormone refractory prostate cancer. In the same meeting, another large randomized trial of 770 patients randomized to estramustine plus docetaxel or mitoxantrone plus prednisone was also presented (10). Median survival was 18 and 15 months, respectively. The hazard ratio was 0.77 (95% CI, 0.64-0.94; P = 0.008). Grade 3 and 4 toxicities were more frequent in the estramustine plus docetaxel arm as a result of more gastrointestinal and cardiovascular events. These two trials have put into question both the need for estramustine as well as the relative efficacy of weekly chemotherapy.

Since this clinical trial utilized both estramustine and weekly docetaxel chemotherapy, what is the potential role of this regimen in the first-line management of HRPC? This is a reasonable question in light of the recent phase III trials noted above. There are several possible clinical scenarios where such combinations could be useful. For instance, the measurable response rates in the two randomized trials ranged from 8% to 17% in the docetaxel arms, not significantly higher than with mitoxantrone plus prednisone (10, 11). Patients with measurable soft tissue disease represent an opportunity to test new and potentially more active chemotherapy combinations. Another opportunity exists in the setting of patients who progress after docetaxel plus prednisone, a group for whom few options exist. Carboplatin may enhance response to docetaxel in patients who previously have progressed on taxanes (34). Therefore, another potential trial would be to examine this regimen as a second-line combination in patients who have progressed after initial taxanes.

In summary, the recommended phase II dose for this combination is estramustine 140 mg p.o. three times daily on days 1 to 5, 8 to 12, and 15-19; docetaxel 43 mg/m² on days 2, 9, 16; and carboplatin at dose [AUC (5)] on day 2, repeated every 28 days. Further research is needed to characterize patients for whom carboplatin may provide clinical benefit beyond the effects of estramustine and docetaxel.