2175 Chemotherapy in Patients with Prostate Specific Antigen–Only Disease after Primary Therapy for Prostate Carcinoma A Phase II Trial of Oral Estramustine and Oral Etoposide Hidayatullah G. Munshi, M.D. BACKGROUND. A Phase II study was initiated to evaluate the effectiveness of an oral Kenneth J. Pienta, M.D. regimen of etoposide and estramustine in patients with early recurrent prostate David C. Smith, M.D. carcinoma. METHODS. Patients with early recurrent prostate carcinoma as indicated by an 1 Division of Hematology and Medical Oncology, increasing prostate specific antigen (PSA) level and without any evidence of met- Department of Internal Medicine, University of astatic disease were treated with oral etoposide 50 mg/m2/day and estramustine 15 Michigan School of Medicine, Ann Arbor, Michigan. mg/kg/day in divided doses for 21 days, followed by a 7-day rest period. Patients 2 University of Michigan Comprehensive Cancer received a maximum of four cycles. Center, Ann Arbor, Michigan. RESULTS. Eighteen patients were entered in this study. The median serum PSA was 3.1 (range, 0.3–30.3) at the time of entry into the trial. Sixteen patients were assessable for response. Serum PSA declined to undetectable levels in 13 patients with 2 additional patients meeting the criteria for partial response; the median duration of response was 8.5 months (range, 1–18 months). Most patients devel- oped gastrointestinal, cardiac, or hematologic complications. Grade 3 toxicities included neutropenia (one patient), deep venous thrombosis (three patients), and chest pain (one patient). One patient developed acute myelogenous leukemia (French–American–British, acute myelogenous leukemia M5) 23 months after ini- tiating the chemotherapy. CONCLUSIONS. The combination of oral etoposide and oral estramustine resulted in a high rate but only a short duration of response in patients with early recurrent prostate carcinoma. The regimen was poorly tolerated, and the toxicity was sig- nificant. This regimen should not be considered standard therapy for the treatment of early recurrent prostate carcinoma, but further exploration of treatment in this setting is warranted. Cancer 2001;91:2175–80. © 2001 American Cancer Society. KEYWORDS: prostate carcinoma, early recurrent, etoposide, estramustine, chemo- therapy, leukemia, thrombosis. rostate carcinoma is one of the most commonly diagnosed can- Pcers in American men. In 1999, 179,300 men were expected to receive a diagnosis of prostate carcinoma in the United States, and Support by SPORE Grant P50-CA069568 and Can- 1 cer Center Core Grant 5-P30-CA46592-06. 37,000 men were predicted to die of the disease. The widespread use of prostate specific antigen (PSA)–based testing has resulted in a Address for reprints: David C. Smith, M.D., 7-302 dramatic increase in the number of men diagnosed with localized CCGC 0946, 1500 E. Medical Center Drive, Ann disease.2 Many of these men will undergo either radical prostatec- Arbor, MI 48109-0946; Fax: (734) 647-9480; E- tomy or radiotherapy. Unfortunately, approximately 50% of the pa- mail: [email protected] tients undergoing prostatectomy3 and approximately 60% of men Received May 18, 2000; revision received Febru- treated with radiotherapy will have biochemical recurrence within 10 ary 5, 2001; accepted February 12, 2001. years.4 Recurrent prostate carcinoma is a significant problem. Some © 2001 American Cancer Society 2176 CANCER June 1, 2001 / Volume 91 / Number 11 of these patients may be offered additional local ther- treatment of metastatic disease before enrolling in this apy, and many will be treated with hormonal therapy.5 trial. All patients gave written informed consent in A significant proportion of these patients will develop accordance with federal, state, and institutional guide- metastatic disease. Unfortunately, the duration of re- lines. mission with hormonal therapy is limited in patients Pretreatment evaluations consisted of a history with metastatic disease, with androgen failure often and physical examination with assessment of perfor- occurring within 18–24 months of starting therapy.6 mance status, and laboratory studies including com- Consequently, the survival rates of patients with re- plete blood count, serum chemistry profile, prothrom- current disease have not increased over the past 5 bin time, PSA level, radionuclide bone scan, CT scan decades.7 of the abdomen and pelvis, and chest X-ray. Complete Prostate carcinoma consists of both androgen- blood counts were monitored weekly, and the serum dependent and androgen-independent clones. Tumor PSA was monitored before each cycle. After 4 cycles of progression after androgen ablation is due to prolifer- therapy, serum PSA was measured every 12 weeks. ation of androgen-independent cells.8 Chemotherapy Patients in this study were treated and followed in has been used to treat patients whose cancers have the outpatient clinic. Estramustine was provided by become refractory to androgen therapy. Treatment Pharmacia & Upjohn (Kalamazoo, MI) and etoposide with an oral regimen of etoposide and estramustine was supplied by Bristol-Myers Squibb (Nutley, NJ). can achieve a significant response in these patients.9,10 Oral estramustine was given at a dose of 10 mg/kg/day Several additional regimens also have been demon- rounded to the nearest multiple of 140 (maximum 2 strated to have activity in patients with advanced pills 3 times a day) and oral etoposide 50 mg/m2/day prostate carcinoma.11 Unfortunately, the median sur- rounded to the nearest multiple of 50 (maximum 2 vival in these Phase III trials of these regimens is only pills per day) for 21 days, repeated every 28 days. The 11–12 months despite the use of chemotherapy.12,13 dose of etoposide was decreased to 1 tablet a day Patients with advanced prostate carcinoma have a alternating with 1 tablet twice a day for patients with significant tumor burden, which may limit the effec- a body surface area of less than 1.75 m2. tiveness of the chemotherapy in these trials. There Patients were treated with a maximum of four have been, to our knowledge, no trials reporting treat- cycles of therapy. Retreatment required that the ANC ment of patients with early recurrent prostate carci- was greater than 1500/mm3, and the platelet count noma with chemotherapy. Here, we report the results was greater than 75,000/mm3 on Day 1 of each cycle. of a trial in patients with minimal disease, as demon- Dose modification of etoposide was based on Day 21 strated by increasing PSA and without any evidence of ANC and platelet count of the preceding cycle for the metastatic disease, who were treated with an oral reg- next and additional cycles. Etoposide was decreased imen of etoposide and estramustine. The objective of to 1 tablet per day alternating with 1 tablet twice a day this Phase II study was to assess the toxicity of this for ANC between 1000/mm3 and 1499/mm3 and/or for regimen in this population and the response of mini- platelets between 50,000/mm3 and 74,999/mm3. Eto- mal disease to therapy as measured by the serum PSA poside was further decreased to 1 tablet per day for level. ANC less than 1000/mm3 and/or for platelets less than 50,000/mm3. Etoposide was held for the remainder of MATERIALS AND METHODS the cycle if patient had an ANC less than 1500/mm3 or Patients were eligible if they had histologically con- platelet count less than 75,000/mm3 on Days 7 or 14. firmed adenocarcinoma of the prostate associated Patients were transfused for hemoglobin less than 8.0 with elevated PSA that had returned to normal after g/dL or for symptomatic anemia as evidenced by primary therapy (Ͻ 1.0 ng/mL for prostatectomy pa- shortness of breath or severe fatigue. Patients with tients or Ͻ 4.0 ng/mL for radiation therapy patients) Grades 1 and 2 nausea were treated with antiemetics. and had an increasing PSA level that had at least Patients with Grade 3 nausea had their drugs discon- doubled from their lowest PSA level after the primary tinued and were taken off study secondary to toxicity. therapy. Patients were required to have a performance After the tenth patient had been enrolled, the subse- status of 3 or better, with a life expectancy of 12 weeks quent patients on this protocol were maintained on 1 or greater, and an adequate bone marrow reserve with mg/day of warfarin for prophylaxis against deep ve- an absolute neutrophil count (ANC) greater than nous thrombosis. 1500/mm3 and a platelet count greater than 100,000/ The study initially was designed to assess re- mm3. Patients were excluded if computed tomography sponse as measured by PSA criteria. Complete re- (CT) scan or bone scan demonstrated metastatic dis- sponse was defined as a decrease in PSA to undetect- ease, or if they had received hormonal therapy for the able levels for at least 4 weeks with maintenance of Chemotherapy for Early Recurrent Prostate Carcinoma/Munshi et al. 2177 performance status and without the appearance of TABLE 1 new lesions. Partial response was defined as a de- Patient Characteristics crease in PSA of greater than or equal to 50% lasting at Characteristic n least 4 weeks, again with maintenance of performance status and without the appearance of new lesions. No. enrolled 18 Disease progression was defined as increase in the Age (yrs) serum PSA level of 100% over baseline or the appear- Median 62 ance of new lesions. The protocol used a standard Range 48–75 Initial Gleason score two-stage Phase II design to limit accrual if extreme Median 7 results were demonstrated. Twenty evaluable subjects Range 5–9 were to be enrolled in the first phase, followed by PSA at time of diagnosis enrollment of an additional 20 subjects if more than Median 18 one but less than six subjects had response.