Phase II evaluation of docetaxel plus one-day oral estramustine phosphate in the treatment of patients with androgen independent prostate carcinoma


  • Victoria J. Sinibaldi M.S., C.R.N.P.,

    Corresponding author
    1. Department of Oncology, Johns Hopkins Medical Institutions, Baltimore, Maryland
    • Johns Hopkins Medical Institutions, 550 North Broadway, Suite 304, Baltimore, MD 21205
    Search for more papers by this author
    • Fax: (410) 614-7273

    • Victoria J. Sinibaldi and Michael A. Carducci are members of the Advisory Board and Speaker's Bureau for Aventis Pharmaceuticals.

  • Michael A. Carducci M.D.,

    1. Department of Oncology, Johns Hopkins Medical Institutions, Baltimore, Maryland
    Search for more papers by this author
    • Victoria J. Sinibaldi and Michael A. Carducci are members of the Advisory Board and Speaker's Bureau for Aventis Pharmaceuticals.

  • Sandra Moore-Cooper C.R.A.,

    1. Department of Oncology, Johns Hopkins Medical Institutions, Baltimore, Maryland
    Search for more papers by this author
  • Menachem Laufer M.D.,

    1. Department of Oncology, Johns Hopkins Medical Institutions, Baltimore, Maryland
    Search for more papers by this author
  • Marianna Zahurak Ph.D.,

    1. Department of Oncology, Johns Hopkins Medical Institutions, Baltimore, Maryland
    Search for more papers by this author
  • Mario A. Eisenberger M.D.

    1. Department of Oncology, Johns Hopkins Medical Institutions, Baltimore, Maryland
    Search for more papers by this author
    • Mario A. Eisenberger is a consultant, member of the Advisory Board, and member of the Speaker's Bureau for Aventis Pharmaceuticals.



Recent clinical trials have shown antitumor activity with the combination of docetaxel plus estramustine phosphate (EMP) in the treatment of patients with androgen independent prostate carcinoma (AIPC). However, the most commonly employed treatment schedules with EMP have been associated with significant gastrointestinal, cardiovascular, and thromboembolic toxicity. The authors hypothesized that the therapeutic index of the combination of docetaxel plus EMP for patients with prostate carcinoma could be enhanced by reducing the incidence and severity of EMP-associated toxicity, which could be accomplished by shortening the duration of exposure to EMP. To preserve the therapeutic synergism between docetaxel and EMP, they designed a regimen employing higher doses of oral EMP administered on the day of the docetaxel infusion.


From June 1, 1998 through September 28, 2000, 42 patients with AIPC were registered to receive docetaxel (70 mg/m2 intravenously over 1 hour) and EMP (280 mg orally every 6 hours × 5 doses) every 21 days, up to a maximum of 6 cycles. Dexamethasone was administered prior to docetaxel and coumadin 2 mg orally every day was taken during the study treatment period. Patient characteristics included a median age of 68 years, a median Eastern Cooperative Oncology Group performance status of 1, a median prostate specific antigen (PSA) level at study entry of 110.5 ng/mL, and a median of 2 prior hormonal manipulations. Ten patients (25%) had received prior chemotherapy, and 14 patients (33%) had received prior palliative radiation therapy.


Forty patients were evaluable for response and toxicity. Eighteen patients (45%; 95% confidence interval, 29–62%) had a decline > 50% in PSA level that lasted > 4 weeks with a median time to PSA progression and a median duration of PSA response of approximately 4.0 months. Four of 20 patients (20%) had partial soft tissue responses. Ten of 17 symptomatic patients (59%) had improvement in pain. The median survival for all patients was 13.5 months. The most prominent Grade 3 and 4 toxicities were reversible myelosuppression and fatigue. Nausea, emesis, diarrhea, and peripheral edema were minimal. No thromboembolic or hepatic complications were seen.


Docetaxel plus 1 multidose day of oral EMP was active in patients with AIPC and was associated with an acceptable toxicity profile. Overall, the therapeutic index of this regimen compared favorably with regimens that employed a longer administration of EMP. Cancer 2002;94:1457–65. © 2002 American Cancer Society.

DOI 10.1002/cncr.10350

In the year 2001, there were approximately 198,100 patients with newly diagnosed prostate carcinoma, and there were an estimated 31,500 deaths from this disease.1 Androgen deprivation treatment is the most effective systemic approach for patients with metastatic disease. Although most patients respond favorably to this treatment, evidence of disease progression to an androgen independent state occurs within a predictable time frame. Data from large, prospective, randomized clinical trials evaluating androgen deprivation in men with metastatic disease have shown a median time to disease progression between 12 months and 18 months and a median survival of 24–36 months.2–7 Current data on chemotherapy in patients with androgen independent disease show a consistent and reproducible (between studies) rate of benefit manifested by symptomatic and objective responses (declines in serum prostate specific antigen [PSA] levels and tumor shrinkage) and similar ranges in the duration of responses and survival. The spectrum of potential toxicities associated with most agents cautions their use in elderly men with concurrent comorbidities. The evaluation of new drug regimens is of paramount importance to improved therapy for patients with this disease.

Estramustine phosphate (EMP) is a nitrogen mustard derivative of estradiol-17-β phosphate that has been approved by the U.S. Food and Drug Administration for the management of patients with prostate carcinoma. This agent is unique because, in addition to a hormonal mechanistic action (estrogenic), it exerts cytotoxic activity through microtubular inhibition8 and binding to the nuclear matrix.9 EMP as a single agent has shown limited activity in patients with androgen independent prostate carcinoma (AIPC), with objective response rates ranging from approximately 16% to 20%.10 At high doses (i.e., 12–14 mg/kg per day), significant toxicities (i.e., nausea, emesis, diarrhea, and cardiovascular and thromboembolic complications) have been documented.11 In an effort to achieve greater cytotoxicity, EMP has been combined with agents like paclitaxel, vinblastine, and etoposide. The response rates reported with various combinations have ranged from 40% to 65%, as measured by decreases in measurable soft tissue disease and serum PSA levels.8, 12–15

More recently, enthusiasm has been generated by the antitumor activity observed with docetaxel (Taxotere; Aventis Pharmaceuticals, Bridgewater, NJ) in patients with AIPC. Preclinical data suggest that docetaxel blocks cell division in the G2/M phase of the cell cycle, and its cytotoxicity may be triggered by bcl-2 phosphorylation with ensuing apoptosis and antimicrotubular action. Encouraging preclinical antitumor activity in AIPC cell lines16 prompted evaluation of docetaxel in patients with prostate carcinoma. Given as a single agent to patients with AIPC at a dose of 75 mg/m2 every 3 weeks,17, 18 or weekly at a dose of 36 mg/m2,19 or in combination with continuous daily oral EMP at a dose of 12–14 mg/kg,11, 20 PSA responses of 38–82% were reported. Toxicities were more pronounced with docetaxel given every 3 weeks alone or in combination with EMP. Approximately 10% of patients who received the docetaxel plus EMP combination reportedly developed thromboembolic complications.16, 20–22

Petrylak et al.21, 23 conducted a Phase I–II evaluation of the combination of docetaxel and a 5-day course of EMP. Twenty of 32 evaluable patients (62.5%) had a PSA decline of at least 50%, 5 of 18 patients (27.7%) had measurable soft tissue responses, and 8 of 15 patients (53%) had subjective responses (improvement in pain).21, 23 Grade 3–4 fatigue (20.5%) and neutropenia (52.9%) were major side effects. Three of 34 patients (8.8%) developed thromboembolic complications. Savarese et al. conducted a cooperative group trial (i.e., Cancer and Leukemia Group B Trial 9780) evaluating docetaxel and EMP as given by Petrylak et al.21, 23 with the addition of hydrocortisone.20 Twenty-seven of 39 patients (69%) had a PSA decline of at least 50%, and 4 of 21 patients (19%) had measurable soft tissue responses. Toxicity included Grade 3–4 leukopenia and/or neutropenia (50%), fatigue (16%), and thromboembolic complications in 4 of 41 patients (9.7%).22

In view of the reproducible level of activity observed with the docetaxel and EMP combination and the consistent association with a relatively high incidence of significant, potentially lethal thromboembolic complications, we piloted the combination of docetaxel plus a shorter course of EMP in an attempt to reduce the incidence of serious toxicity (unpublished data). Eight patients underwent treatment with oral EMP at a dose of 280 mg three times per day on Days 1–3 and docetaxel on Day 2 at doses ranging from 60 mg/m2 to 80 mg/m2 repeated every 21 days. Subjective improvement in pain and urinary obstructive symptomatology was observed. Four patients had objective responses: Three had major regressions of bidimensionally soft tissue metastases. Three of the 8 patients developed deep vein thrombosis (two of whom also developed pulmonary emboli). Other toxicities included mild nausea, anorexia, edema, fatigue, mild hoarseness, and peripheral neuropathy.

Our preliminary observations in this small pilot study indicated that, even with a 3-day course of oral EMP, significant toxicity occurs, perhaps comparable to longer administration schedules, possibly related to the slow elimination phase of EMP, which binds tightly to serum albumin. Preclinical data has shown that the synergism between taxanes and EMP is not strongly dependent on EMP concentrations, and enhanced cytotoxic effects can be seen at very low EMP concentrations.24, 25 We chose to evaluate a shorter schedule, hypothesizing that toxicities commonly seen with EMP (especially thromboembolic complications) could be minimized without compromising efficacy of this potentially active drug combination. The results of our study evaluating a short schedule of EMP in combination with docetaxel given on a 21-day cycle are presented below.


Patient Eligibility

Eligible patients were required to have histologically proven adenocarcinoma of the prostate that progressed to an androgen independent state. In the presence of castrate levels of testosterone (< 20 ng/mL), patients had to have evidence of disease progression, as defined by rising PSA levels on two consecutive measurements at least 2 weeks apart in addition to 1) either a new lesion on bone scan or 2) an increase in the size of a measurable lesion on a computed tomographic (CT) scan of the abdomen/pelvis or chest. Patients who received secondary hormonal therapies, such as megestrol acetate, aminoglutethemide, corticosteroids, flutamide, and nilutamide, had to demonstrate rising PSA levels at 2 weeks and 4 weeks after the cessation of therapy. Those patients who received bicalutamide had to demonstrate rising PSA levels at 2 weeks and 6 weeks after the cessation of therapy. Patients who had not undergone orchiectomy were required to continue medical means of gonadal ablation with a luteinizing hormone-releasing hormone analogue. Previous exposure to chemotherapy or biologic treatments was limited to two or fewer regimens; and no previous exposure to EMP, taxanes, or vinca alkaloids was permitted. Further inclusion criteria included an Eastern Cooperative Oncology Group (ECOG) performance status of 0–2; a testosterone level < 20 mg/L; and appropriate renal, hepatic, and hematologic function at baseline. All patients were required to give written, informed consent.


Pretreatment evaluation included a complete medical history, physical examination, chest X-ray, bone scan, CT scan of the abdomen and pelvis (and CT scan of the chest only if clinically indicated), and laboratory studies (including a complete blood count with differential, comprehensive chemistry profile, testosterone level, prothrombin time [PT], partial thromboplastin time [PTT], international normalizing ratio [INR], PSA, and acid phosphatase level [PAP]).

Follow-up evaluation included weekly interval histories (with toxicity assessments) and laboratory tests (complete blood count with differential, chemistry profiles, PT, PTT, INR, urinalysis, PAP, and PSA). Bone scans, CT scans of the abdomen/pelvis, and chest X-rays were obtained after every three cycles and then every 3 months until the patient experienced disease progression. A pain medication diary was maintained on a daily basis while patients were on the study. The Functional Assessment of Cancer Therapy-Pain (FACT-P) quality-of-life (QOL) questionnaire was administered pretreatment and just before the initiation of each treatment cycle.

Toxicity and Response Criteria

The National Cancer Institute Common Toxicity Criteria (version I) was used to evaluate patients for toxicity during each cycle. The requirements for Phase II adverse drug reactions were followed strictly. If Grade 3 or 4 hematologic/nonhematologic toxicities (excluding neurotoxicity) were observed during any cycle, then the patient's dose was deescalated to 60 mg/m2 for subsequent treatment cycles. Treatment continued as long as the Grade 4 hematologic toxicity recovered to Grade 1 within 1 week and nonhematologic toxicity recovered to Grade 1 within 3 weeks of occurrence. Patients who developed Grade 3 neurotoxicity at any dose level or Grade 4 nonhematologic toxicity at the 60 mg/m2 dose level also were removed from study.

All patients were evaluated for PSA response, objective measurable disease response, subjective response, duration of response, time to disease progression, and survival. For the purpose of this article, the criteria for response were based on the guidelines from the PSA Working Group.26 A PSA decline ≥ 50%, confirmed by a second value at least 4 weeks later, was considered a PSA response. The reference PSA for these declines was a PSA value obtained within a 2-week period prior to starting study medication. The duration of PSA response was defined as the time from the first 50% decline in the PSA level to the time of a 50% increase from PSA nadir (the increase had to be at least 5 ng/mL) or the time to the first consistent rise in the PSA level. PSA progression was defined as a 25% increase over the baseline or nadir with an increase in the absolute value of the PSA level ≥ 5 ng/mL, confirmed by a second value.

Patients with measurable soft tissue disease had to meet traditional guidelines for response. A complete response was defined as the resolution of all measurable and evaluable disease lasting > 28 days and a decline in the PSA level to < 0.1 ng/mL in patients status postprostatectomy and ≤ 0.5 ng/mL in patients status post definitive radiation therapy lasting ≥ 28 days. A partial response was defined as a decrease in the sum of the products of all measurable lesions by ≥ 50% lasting ≥ 28 days. Patients with a 0% to < 50% decrease in the measurement of the sum of the products of all measurable lesions were considered to have stable disease. Bone scans were considered stable if there were no new lesions in two scans taken at least 2 months apart. Subjective responses were measured with pain medication diaries and the FACT-P QOL tool.

Treatment Regimen

Treatment consisted of docetaxel at a dose of 70 mg/m2 administered as a 1-hour infusion approximately 12 hours after the first dose of EMP of each cycle. This dose was decreased to 60 mg/m2 if significant toxicity occurred. Oral EMP was given at a dose of 280 mg every 6 hours for a total of 5 doses. Patients were instructed to take EMP on an empty stomach and to avoid dairy-containing products during the time they were taking EMP. No dose reductions were made for EMP.

Oral dexamethasone was given in an effort to reduce the incidence and severity of hypersensitivity reactions associated with docetaxel administration. Initially, oral dexamethasone was given at a dose of 8 mg twice daily for 3 days. In an attempt to ameliorate the significant fatigue observed immediately after the discontinuation of dexamethasone in the first 14 patients, the schedule and dose of the steroid were changed to 20 mg orally for 2 doses only (12 hours and 6 hours prior to docetaxel) for the remaining patients. Patients also were given oral coumadin (2 mg once daily) as prophylaxis for vascular events; however, dose adjustments were made only if INR levels were > 3.0. Cycles were repeated every 21 days up to a maximum of 6 cycles, with weekly assessments of toxicity, subjective clinical response (improvement in pain and/or urinary symptomatology), and PSA levels. Pain medication diaries were maintained daily. Patients were requested to document daily the date, time, name of medication, and dose taken. They also were asked to rate their daily pain on a scale of 0 to 10 at bedtime. QOL questionnaires (FACT-P) were obtained prior to each cycle.


A standard, two-stage, Phase II design was employed based on a response probability of ≥ 20%. Study termination would occur if the response probability was ≤ 5% in the first stage (i.e., no responses observed in the first 20 patients). If at least 1 of the 20 patients responded, then 20 additional patients would be evaluated in a second stage. Five or more responses (for the definition of response, see Toxicity and Response Criteria, above) observed in both stages would be considered evidence warranting further evaluation of the treatment regimen, given that toxicity and survival also looked favorable. The design had a significance level (probability of falsely declaring an agent with 5% response probability to warrant further study) of 0.05 and a power (probability of correctly declaring an agent with 20% response probability to warrant further study) of 92%. It was calculated that 40 adequately treated patients were needed to sufficiently estimate the probability of response or a particular toxicity to 16%. All registered patients were analyzed in an intention-to-treat analysis. Any toxicity that occurred with at least 10% probability was likely to be seen at least once (99% chance). Survival times were computed from the on-study date until the last follow-up date. The major statistical endpoints of this study were duration of survival and time to progression. Event time distributions were estimated with the method of Kaplan and Meier27 and were compared using the log-rank statistic.28 All statistical computations were performed using the SAS29 (SAS Institute, Cary, NC) or EGRET (Statistics and Epidemiologic Research Corp., Seattle, WA)30 software packages. All P values are two sided, and all confidence intervals are at the 95% level.


Forty-two patients were entered on the study from June 1, 1998 through September 28, 2000. Forty patients were evaluable for toxicity and response. One patient took one dose of EMP and decided not to proceed with treatment. Another patient was found to have an elevated testosterone level after registration and never received study treatment. Patient characteristics for the 40 evaluable patients (Table 1) included a median age of 68 years (range, 47–80 years) and a median ECOG performance status of 1 (range, 0–1). All patients had androgen independent disease who failed a median of two prior hormonal therapies (range, one to four therapies). Nineteen patients had bone disease only, 5 patients had bone and visceral involvement, and 16 patients had bone and soft tissue and/or lymph node disease. The median pretreatment PSA level was 110.5 ng/mL (range, 5.02–2803.8 ng/mL). Twenty-four patients had received prior local therapy. Fourteen patients (35%) had received prior palliative radiation (1 patient had received external beam RT and strontium), whereas 10 patients (25%) had received prior chemotherapy (3 of 10 patients who received 2 prior regimens) prior to study entry. Two patients had received study treatments with allogeneic tumor vaccines, and one patient had received a 5 α-reductase inhibitor on a clinical trial.

Table 1. Patient Characteristics
  • ECOG: Eastern Cooperative Oncology Group; PSA: prostate specific antigen; RP: radical prostatectomy; RT: radiation therapy.

  • a

    Three patients had two prior regimens.

Total no. of patients entered42
Total no. evaluable patients40
Median no. of courses per patient (range)6 (1–6)
Median age in yrs (range)68 (47–80)
Median ECOG performance status (range)1 (0–1)
Sites of disease
 Bone only19
 Bone and visceral4
 Bone and soft tissue and/or lymph nodes17
Median (range) PSA at baseline (ng/mL)110.5 (5.02–2803.8)
Prior local therapy
 Prostatectomy plus RT4
 Definitive RT4
 Neoadjuvant hormonal therapy plus RT2
 Neoadjuvant hormonal therapy and RP2
RP and adjuvant hormonal therapy1
Prior systemic therapy
 First line hormonal therapy40
 Second line hormonal therapy28
 Third line hormonal therapy10
 Fourth line hormonal therapy2
Palliative RT
 External beam13
 External beam and strontium1
 Five-α reductase inhibitor1
 No. of patients with pain Grade ≥ 5 at baseline6

Forty evaluable patients received a total of 207 cycles of therapy (median, 6 cycles; range, 1–6 cycles). One hundred of the cycles included docetaxel at a dose of 70 mg/m2, and 107 cycles included docetaxel at a dose of 60 mg/m2. Twenty-nine of 40 patients completed 6 cycles of the planned therapy; however, only 7 patients received 6 total cycles with docetaxel at a dose of 70 mg/m2. Three of these patients had a PSA response; one patient also had a soft tissue response. Of 29 patients, 15 patients required a dose reduction to 60 mg/m2 after the first cycle of therapy. Eight of these patients had a PSA response; 1 also had a soft tissue response.


The toxicity observed with docetaxel plus 1-day EMP was mild to moderate in severity and quite manageable. Major hematologic and nonhematologic toxicities (Grade 3–4) are summarized in Tables 2 and 3. Neutropenia was the most prominent hematologic toxicity and required a dose reduction to 60 mg/m2 after the first cycle of therapy in 19 of 40 patients (48%). Neutropenia was reversible in all patients, with a nadir noted in approximately 8 days and median duration of approximately 7 days (i.e., the time to recovery with absolute neutrophil count ≥ 1500). Two patients experienced granulocytopenia with fever and required hospitalization: In both patients, there was complete recovery of the absolute neutrophil count.

Table 2. Hematologic Toxicity
ToxicityNo. of patients (%)
Grade 3Grade 4
Leukopenia16 of 40 (40)3 of 40 (7.5)
Neutropenia12 of 40 (30)18 of 40 (45)
Neutropenia/fever0 of 40 (0)2 of 40 (5)
Anemia1 of 40 (2.5)0 of 40 (0)
Thrombocytopenia0 of 40 (0)0 of 40 (0)
Table 3. Nonhematologic Toxicity
ToxicityNo. of patients (%)
Grade 3Grade 4
Fatigue28 of 40 (70)0 of 40 (0)
Diarrhea3 of 40 (7.5)0 of 40 (0)
Hyperglycemia4 of 40 (10)0 of 40 (0)
Local skin reaction3 of 40 (7.5)0 of 40 (0)
Nail changes3 of 40 (7.5)0 of 40 (0)
Congestive heart failure/atrial fibrillation0 of 40 (0)1 of 40 (2.5)

Acute fatigue was the most prominent nonhematologic toxicity, which generally occurred 2–3 days after the administered docetaxel and recovered to baseline or Grade 1 toxicity within 3 days after the initial occurrence. Significant extravasation reactions occurred in three patients. Pain, erythema, edema, paresthesias, and skin desquamation were noted in all three patients, with healing time ranging from 3 months to 6 months in duration. Nail changes were noted in three patients: In addition to linear banding and hyperpigmentation, exfoliation of the entire nail was noted. One patient developed atrial fibrillation with congestive heart failure. This was diagnosed on Day 22 of Cycle 6. The patient was treated conservatively with fluid restriction, diuretics, and cardiac medications. The patient had no documented history of prior cardiac comorbidities.

Other toxicities (Grade 1–2) included hypersensitivity reaction, alopecia, hypocalcemia, breast tenderness, rectal spasms, alteration in taste, paresthesias, myalgias, arthalgias, facial flushing, shortness of breath, anorexia, headache, and insomnia. Minimal nausea and fluid retention were noted. Hypersensitivity reactions were observed in seven patients. In all patients, the study drug was interrupted and restarted upon recovery without complications. No thromboembolic or bleeding complications were seen.

Therapeutic Activity

Of 40 evaluable patients (median follow-up, 12.4 months), 18 patients (45%; 95% confidence interval, 29–62%) had a decrease > 50% from baseline in PSA levels lasting > 4 weeks. The median time to PSA progression was approximately 4.0 months (Fig. 1), and the median duration of PSA response also was about 4.0 months. Four of 20 patients (20%; 1 patient with liver involvement and 3 patients with soft tissue/lymph node involvement) had partial responses (Fig. 2, Table 4).

Figure 1.

Time to prostate specific antigen (PSA) progression.

Figure 2.

Measurable tumor responses.

Table 4. Response
No. of evaluable patientsNo. of measurable disease responsesNo. of patients with a decrease in PSA ≥ 50% ≥ 4 weeks (95%CI)Median duration of PSA response in months (range)
  1. PSA: prostate specific antigen.

404 of 2018 of 40 (29–62%)4.1 (1.3–9.1)

Pain Response

Ten of 17 symptomatic patients had improvements in pain (59%), as measured by a decrease in baseline pain scores by 2 points (Table 5). However, there was no significant decrease in the amount of pain medications taken. It is noteworthy that 3 of 10 patients had an accompanying decline in PSA level by 50% lasting ≥ 4 weeks, and 1 of 10 patients had a PSA decline along with a partial soft tissue response.

Table 5. Pain Response
Baseline pain scoreNo. of evaluable patientsNo. of patients with a decrease in baseline pain score by 2 points ≥ 2 weeks (%)
2–5116 (55)
≥ 564 (67)


Consecutive FACT-P evaluations were available for 25 of 40 enrolled patients on the study. Of these, eight patients had only a PSA response, five patients had only a pain response, five patients were nonresponders, three patients had a PSA and partial soft tissue response, three patients had a PSA and pain response, and one patient had only a soft tissue response. Evaluations were not available for 15 patients due to 1) limited duration of treatment, 2) failure to complete the survey, and 3) loss of the survey after completion. Sixteen of the 25 evaluable patients (64%) had a decline in overall scores just prior to the second cycle; 11 of these patients (44%) had a decline > 10 points. Nine of the 25 evaluable patients (36%) had a decline > 10 points just prior to the sixth cycle of therapy.


Figure 3 depicts Kaplan–Meier survival curves for all patients, which were approximately 13.5 months. The median survival for PSA responders has not been reached (median follow-up, 15.2 months). With a median follow-up of 12.4 months for the entire patient population, 17 patients seventeen (43%) remain alive (median follow-up, 15 months), all of whom have progressed despite the current therapy. Eleven of these 17 patients were PSA responders; 3 of these 17 patients had measurable soft tissue responses.

Figure 3.

Median survival.


AIPC continues to present a challenge to health care providers. Historically, chemotherapeutic agents have shown minimal activity in patients with this disease; however, emerging data from trials evaluating the combination of agents targeting microtubules show promising antitumor activity. Docetaxel and EMP are antimicrotubular agents with different mechanisms of action and predominantly nonoverlapping toxicities. Although the antitumor activity observed with these two drugs combined has brought much enthusiasm among investigators, the toxicity profile has brought some concern. In previous studies with the two-drug combination, gastrointestinal toxicity (including nausea, emesis, and diarrhea) was moderate to severe, and the overall incidence of thromboembolic complications was approximately 10%.11, 16, 21, 22 These symptoms were believed to be associated with the EMP. Although shorter durations of EMP exposure have shown a decrease in gastrointestinal toxicity, the incidence of thromboembolic complications has remained static. It has been shown that fatigue and neutropenia are dose-limiting side effects associated with docetaxel.

This study regimen was generally tolerable for patients, with 29 of 40 patients (73%) receiving all 6 cycles as planned. Three patients received fewer than three cycles; however, two patients had rapid disease progression, and one patient was diagnosed with a second malignancy. No patients were taken off the study due to toxicity. Fatigue and neutropenia, as expected, were the most predominant side effects of this regimen, both of which are believed to be associated with docetaxel. In general, both side effects were of short duration and were quite manageable. No treatment-related deaths were observed. There were no episodes of thromboembolic complications seen in this study. Although the shorter duration of EMP exposure employed in this regimen is the most likely reason, prophylactic low-dose anticoagulants may have contributed to the absence of thromboembolic events. Petrylak et al.31 used a 5-day schedule of EMP and administered oral aspirin (81 mg daily) along with oral coumadin (2 mg daily) to his last 15 patients. Those authors observed no thromboembolic events in these later patients. Gastrointestinal side effects were minimal with our regimen, and no dose reduction was made for this toxicity.

Traditionally, chemotherapy for patients with AIPC has been associated with decreased QOL secondary to toxicity from therapy. The overall data from 25 patients who completed sequential QOL surveys suggest that this regimen did not have a significant impact on QOL; however, it did show a slight improvement (decline in scores) for 36% of patients. This conclusion is obscured by the fact that only patients who completed the FACT-P survey at the specified three intervals were included in this analysis. Furthermore, the absence of concurrent controls is a major limitation.

The PSA response rate observed in this trial appears comparable to the response rate observed in single-agent trials10, 18 and was only slightly lower than the response rate observed in combination trials that used longer administration schedules of EMP and a longer duration of treatment with docetaxel. Clearly, patient selection is a major factor influencing the outcome of Phase II trials with chemotherapy for patients with AIPC. It is noteworthy that the responses in this trial were observed in heavily pretreated patients, 25% of whom experienced disease progression after receiving third-line hormonal manipulations, 35% of whom received palliative radiation, and 25% of whom received prior chemotherapy. The response outcome in our study was comparable to the outcomes reported in all other studies,16, 20–27, 32–34 despite the fact that treatment was stopped after six cycles (18 weeks). Combination therapies with longer EMP administration schedules and schedules that offered continuous treatment until patients experienced dose-limiting toxicity or disease progression appear to demonstrate more favorable response rates; however, the median time to disease progression and the median duration of PSA response are quite comparable across all studies. It is possible that the somewhat higher incidence of PSA response reported with the EMP regimens is due to a hormonal effect on PSA expression of estrogens in the EMP moiety. The similarity in the distribution of time to disease progression in cohorts treated with or without EMP or treated with a short EMP course, like the current regimen, further supports this hypothesis. In the study reported by Petrylak et al.,23, 31 the investigators reported a median time to disease progression of about 4 months despite an encouraging survival figure of 22.8 months, which most likely was attributable to favorable patient selection. Our median survival of 13.5 months is comparable to the survival reported in other taxane-EMP trials, which have shown a median survival range of 12.8–15.9 months.33, 34

Taxane-EMP combinations have been the major focus in the treatment of patients with AIPC;, however, current data have shown that docetaxel is active as a single agent in this patients with disease. In view of the data from this study and those reported in single-agent docetaxel trials, the contribution of EMP needs to be explored further in adequately designed prospective, randomized studies. In a recently reported randomized Phase II study by Berry et al.,35 the combination of EMP and paclitaxel resulted in a higher PSA response rate than single-agent paclitaxel, with a similar distribution of time to disease progression and survival in treatment both arms. Although these data support our argument that EMP may influence primarily the response rate, issues related to study design prevent more reliable, definitive conclusions. Similarly, the relatively favorable tolerance with six cycles of therapy observed with our regimen suggests that the appropriate duration of treatment is a QOL issue worthy of further exploration in prospective, randomized trials. A number of patients with evidence of response at the end of the six-cycle period demonstrated subsequent responses with the same or similar docetaxel retreatment (unpublished data), an observation that has been confirmed by others using docetaxel regimens.31