Bone Marrow Transplantation (2001) 28, 859–863  2001 Nature Publishing Group All rights reserved 0268–3369/01 $15.00 www.nature.com/bmt Ovarian cancer Phase II trial of high-dose intravenous doxorubicin, etoposide, and cyclophosphamide with autologous stem cell support in patients with residual or responding recurrent ovarian cancer

RJ Morgan1, JH Doroshow1, L Leong1, J Schriber4, S Shibata1, S Forman1, V Hamasaki1, K Margolin1, G Somlo1, J Alvarnas4, M McNamara1, J Longmate3, J Raschko1, W Chow1, S Vasilev2, K McGonigle2 and Y Yen1

1Department of Medical Oncology and Therapeutics Research, City of Hope National Medical Center, Duarte, CA, USA; 2Department of Gynecologic Oncology, City of Hope National Medical Center, Duarte, CA, USA; 3Department of Biostatistics, City of Hope National Medical Center, Duarte, CA, USA; and 4City of Hope/Samaritan Bone Marrow Transplant Program, Phoenix, AZ, USA

Summary: when patients are treated with aggressive debulking surgery followed by platinum-based chemotherapy.1–3 At present, This study was performed in order to evaluate the tox- however, there is no proven curative therapy for patients icities, progression-free and overall survival of patients having persistent disease following initial therapy, or for with responsive residual or recurrent ovarian cancer those whose disease recurs. treated with high-dose chemotherapy. Twenty-seven Steep dose–response relationships for anthracyclines and patients were treated. Doxorubicin, 165 mg/m2 over 96 h alkylating agents have been described for ovarian cancer.4 (days −12 to −8), etoposide 700 mg/m2 every day ×3 Efforts to increase dose intensity and decrease toxicity (days −6to−4), and cyclophosphamide 4.2 g/m2 on d include the use of bone marrow and/or peripheral blood −3 was followed by stem cells and granulocyte colony- stem cells for hematopoietic support.5 The use of this stimulating factor. The median days of granulocyte approach in patients with residual or recurrent disease has Ͻ ␮ count 500/ l was 14 (range 10–42) and platelets resulted in long-term disease-free survivals of 15–25% in Ͻ ␮ 20 000/ l was 13 (range 2–80). Median numbers of multiple phase II trials of high-dose chemotherapy, with red cell and platelet transfusions were 15 (5–16) and 14 plateaus in survival curves appearing at approximately the (4–103). Toxicity included mucositis requiring narcotic 3-year post-treatment time point.6,7 analgesia in all patients. Asymptomatic decreases in Ͻ Multiple combinations of chemotherapeutic agents have ejection fraction to values 50% were observed in four been utilized as preparative regimens including platinum patients. No clinical congestive heart failure was and non-platinum-based designs. Dottino et al8 examined observed. One death due to sepsis was observed. Median the effectiveness of etoposide in combination with cisplatin progression-free survival is 7.5 months (1.0–56 months); in previously treated patients and described a threshhold five patients remain alive, two of whom remain pro- dose for etoposide, suggesting a dose–response relation- gression-free at 19.5 and 24.5 months post transplant. ship. Mulder et al6 reported complete responses in six of Median overall survival is 14.0 months (1–68 months). Ͻ We conclude that high-dose anthracyclines may be eight patients with microscopic or 2 cm residual disease in patients treated with a combination of a dose-intense safely administered to ovarian cancer patients. The 9 short overall and progression-free survivals observed in etoposide/cyclophosphamide combination. Jacobs et al our population suggest that this combination is not opti- have analyzed the effect of dose intensity on survival in mal. Bone Marrow Transplantation (2001) 28, 859–863. previously untreated ovarian cancer patients. Those who Keywords: high-dose chemotherapy; ovarian cancer; received high-dose intensities of cisplatin, doxorubicin, or stem cell transplantation all drugs combined survived significantly longer than those who received lower dose intensities. Based on these and other published series we were encouraged to undertake a phase II trial of intravenous doxorubicin, etoposide and Clinical trials in advanced ovarian cancer patients have cyclophosphamide utilizing stem cell support in patients documented long-term disease-free survivals of 15–25% with clinically persistent or recurrent responsive ovarian carcinoma. The chemotherapy doses were chosen based on phase I experience at our institution which documented that Correspondence: Dr RJ Morgan Jr, Department of Medical Oncology and high-dose doxorubicin can be safely administered without Therapeutics Research, City of Hope National Medical Center, 1500 E Duarte Rd, Duarte, CA 91010, USA significant cardiac toxicity when administered as a 96 h Received 11 May 2001; accepted 9 August 2001 continuous infusion.10 We report the results of that trial. High-dose chemotherapy in ovarian cancer RJ Morgan et al 860 Patients and methods amended, and the remaining 19 patients had reinfusions of peripheral blood mononuclear cells alone (mean number Patient selection 10.4 × 108/kg) for hematopoietic support, which were mob- ilized, harvested, and stored using previously published Between April 1992 and August 1998, 27 patients with per- methods.10 sistent or responding recurrent, biopsy-proven coelomic epithelial cancers (epithelial ovarian, Fallopian tube or pri- Chemotherapy administration: Doxorubicin 165 mg/m2 mary peritoneal carcinoma, all of which display similar was administered by continuous intravenous infusion over clinical characteristics and have similar recommendations 96 h via a dual lumen central venous (Hickman) catheter for management) were eligible for this trial. Newly diag- during the first 4 days of therapy (day −12 to day −8). On nosed patients must have been previously treated with days −6to−4, etoposide, 700 mg/m2/day was infused over debulking surgery followed by platinum-based chemo- 2 h. On day −3, cyclophosphamide 4.2 g/m2 was adminis- therapy, and have sustained at least a partial response; those tered over 2 h using MESNA as a uroprotectant. At least with recurrent ovarian cancer must have sustained at least 2 × 108 mononuclear cells/kg (including both bone marrow a partial response following second-line chemotherapy and peripheral blood mononuclear cells) were rapidly documented by second-look surgery or radiographic or thawed and reinfused on day 0 by rapid intravenous clinical evaluation. Patients were required to have a Karnof- infusion. G-CSF 5 ␮g/kg was administered intravenously р sky performance status of 80%, and physiologic age 55 daily starting on day 0 until the total white count was Ͼ years. Cardiac ejection fraction must have been 50% as Ͼ1000/␮l for 3 consecutive days. measured by MUGA radionuclide scan. Adequate renal, hematopoietic, and hepatic function was defined as: serum creatinine р1.5 mg/dl or 24 h creatinine clearance of Statistical methods ␮ 50 ml/min; total white blood cell count 4000/ l, platelet Toxicities have been categorized using the Common Tox- ␮ Ͼ count 150 000/ l, hemoglobin 10 g/dl; SGOT and SGPT icity Criteria for Autologous Bone Marrow Transplantation within two times normal limits and serum bilirubin of the National Cancer Institute. Overall and progression- р 1.5 mg/dl. All patients gave their voluntary informed con- free survivals were calculated according to the method of sent and signed a consent document that had been reviewed Kaplan and Meier.11 and approved by the Institutional Review Boards of the City of Hope National Medical Center and Samaritan Hospital. Results

Pre-treatment evaluation Patient characteristics All patients had a complete history and physical examin- Twenty-seven patients were enrolled on this study, of ation including documentation of weight, Karnofsky per- whom 26 received the entire planned course of treatment. formance status, presence of measurable or evaluable dis- The patient characteristics are summarized in Tables 1 and ease, a complete blood count with differential, platelet 2. All patients were eligible and evaluable for toxicity and count, 18-channel blood chemistry analysis, serum CA-125 are reported for progression-free and overall survival. The level, serum magnesium level, chest X-ray, electrocardio- median age was 48 years (range 28–63), and the median gram, radionuclide MUGA scan, urinalysis, and computed Karnofsky performance status was 90% (range 80–100%). tomographic scans of the chest, abdomen and pelvis. A Twenty-two patients were Caucasian and five were His- complete blood count with differential, and platelet count, panic. Twenty-five of the patients were platinum-sensitive, and electrolyte evaluation were repeated daily while receiv- ing chemotherapy and while myelosuppressed. The CA-125 Table 1 Patient characteristics level and MUGA scans were repeated 30 to 45 days after- hospital discharge, and thereafter at least every 3 months No. of patients (more often at the discretion of the referring physician). Gender: Female 27 Treatment plan Median age in years (range) 48 (28–63) Histology Stem cell harvest: The first eight patients received bone Papillary serous 19 marrow mononuclear cells (mean 2.71 × 108/kg) sup- Undifferentiated 4 Endometrioid 2 plemented by three collections of peripheral blood stem Mucinous 1 8 cells (mean 5.3 × 10 /kg). Granulocyte colony-stimulating Clear cell 1 factor (G-CSF) 5 ␮g/kg was administered intravenously Karnofsky performance status daily for 5 days; bone marrow stem cells were then har- 100% 14 vested from the posterior iliac crest using standard tech- 90% 11 10 80% 2 niques. All patients then continued G-CSF for successive Race days during peripheral stem cell apheresis. After favorable Caucasian 22 re-engraftment data for peripheral blood stem cells alone Hispanic 5 became available, the stem cell support regimen was

Bone Marrow Transplantation High-dose chemotherapy in ovarian cancer RJ Morgan et al 861 Table 2 Prognostic factors relating to high-dose chemotherapy 1.00

No. of patients 0.75 Platinum sensitive 25 Platinum resistant 2 Residual disease following initial chemotherapy 0.50 Stage 3 13 Stage 4 2 Second-look surgery 12 No second-look surgery 3 0.25 Microscopic disease 9 Progression-free survival Ͻ1 cm residual 6 Relapsed disease responding to chemotherapy 0.00 Initial progression-free survival (median (range) 12 months (1–42) 0 10 20 30 40 50 60 70 Time (months)

Figure 1 Progression-free surivial of the 27 patients entered on the study. while two progressed within 6 months of a platinum-con- taining regimen and were resistant. Disease status at entry on this study included: 15 patients with residual disease 1.00 following debulking surgery and platinum-based chemo- therapy; 13 of whom had stage 3 and two had stage 4 dis- ease at the time of their initial diagnosis. Twelve patients 0.75 had a recurrence of ovarian cancer that was responding to second-line chemotherapy. All of these patients had pre- treatment CT scans and were judged to have recurrent dis- 0.50 ease of Ͻ1 cm. Nineteen patients had serous papillary adenocarcinomas, four had undifferentiated adenocarci- Overall survival nomas, one had clear cell carcinoma, two had endometrioid 0.25 carcinomas, and one had a mucinous adenocarcinoma. All patients had undergone prior surgery and platinum-based 0.00 chemotherapy; no patient had received prior radiation 0 10 20 30 40 50 60 70 therapy. Time (months) Of the patients with residual disease following initial Figure 2 Overall survival of the 27 patients entered on the study. platinum-based chemotherapy, 12 were documented by a second-look surgical procedure and three by clinical status. Nine had only microscopic residual disease and six had planned doses of etoposide, at which time the chemo- Ͻ1 cm residual disease at this time. The median initial pro- therapy was stopped. She subsequently received her stem gression-free survival of those relapsed patients who cell reinfusion and recovered without sequelae. Her CA- responded to reinduction chemotherapy was 12 months 125 level rose 18 months following treatment; however, (range 3–42). In the recurrent group the median number the patient remains alive and without clinical symptoms of of chemotherapy courses administered prior to high-dose disease 24.5 months following treatment. chemotherapy was two (range 1–4). Disease progression was documented by radiographic criteria in 18 patients, and clinically with accompanying Progression-free and overall survival data increase or a new elevation of CA-125 levels in seven patients. One patient died while on treatment, and one Progression-free and overall survival curves are shown in patient remains without evidence of disease progression. Figures 1 and 2. The median progression-free and overall survivals are 7.5 months (range 1.0–56 months) and 14.0 Pre- and post-treatment CA-125 levels and therapeutic months (range 1–68 months), respectively (8.3 and 14 outcomes months for patients with residual disease and 8 and 21 months for patients with recurrent disease (P values, log All patients were judged to be in response to induction or rank statistic, Ͼ0.44 for progression-free survival and Ͼ0.3 reinduction chemotherapy at the time that they were entered for overall survival)). on this protocol and began stem cell collection. CA-125 Four of 27 patients remain alive, one of whom remains levels (summarized in Table 3) were examined on day 1 of clinically relapse-free at 42 months post transplant. One treatment and repeated at 30–45 days post transplant. Of patient died of complications related to sepsis due to Enter- the 27 patients who entered the trial, 22 had CA-125 levels obacter cloacae while on treatment, which developed on within the normal limit (Ͻ35) on day 1 of chemotherapy. day +4. One patient who entered the trial with well-con- Of these, 16 remained within normal limits following the trolled asthma did not complete the planned course of completion of treatment. Two patients who had CA-125 chemotherapy due to the development of bronchospasm fol- levels Ͻ20 pre-treatment increased to 49 and 459 post treat- lowing the doxorubicin infusion and one of the three ment. Of the four remaining patients (out of 22) beginning

Bone Marrow Transplantation High-dose chemotherapy in ovarian cancer RJ Morgan et al 862 Table 3 CA-125 (units/ml, N1 р35 units/ml) levels pre- and post treatment

Pre-treatment Pre-treatment Post treatment Post treatment Post treatment CA-125 level (No. patients) Ͻ20 units/ml 20–35 units/ml Ͼ35 units/ml

Ͻ20 units/ml 18 13 2 2 21–35 units/ml 4 1 2 Ͼ35 units/ml 5 2 3

therapy with levels 21–35, one patient died, one remained The role of anthracyclines in the chemotherapeutic man- normal, and two increased to 50. Of the five patients who agement of advanced ovarian cancer has been evolving were found to have levels Ͼ35 on day 1 of chemotherapy, over the past 20 years. Sequential clinical trials of single- two decreased to the normal range, while three remained agent and combination chemotherapy that included doxoru- elevated following treatment (one stable (67 → 68), and bicin initially suggested that anthracyclines improved over- two showing marked progression (310 → 2354 and all and progression-free survival; however, subsequent 79 → 187)]. phase III studies failed to confirm these findings.15,16 Thus, the current standard post-operative chemotherapy is six Hematologic recovery and toxicities of therapy cycles of a platinum analogue with paclitaxel. This results in median progression-free and overall survival rates of Hematologic recovery is summarized in Table 4. The approximately 18 and 38 months, respectively, in stage 3 median number of days of thrombocytopenia (platelets and 4 patients.17–19 A more recently published meta-analy- Ͻ20 000/␮l) was 13 (mean 19, range 2–80). Neutropenia sis,20 however, has reintroduced the question of the role of (500 neutrophils/␮l) persisted for a median of 14 days anthracyclines in the treatment of epithelial ovarian carci- (mean 16, range 10–42 days). The median number of red noma, and suggests that these agents may increase overall cell and platelet transfusions administered were 15 and 14, and progression-free survival. respectively, with means of 8 and 17. Despite the fact that first-line treatment of advanced Cardiac toxicity was mild. All patients except the patient ovarian cancer results in a 60–80% response rate with intra- who expired had follow-up MUGA scans within 1 to 3 venous chemotherapy, the majority of these patients die of months post treatment. Four patients had asymptomatic chemotherapy-resistant, recurrent disease.21,22 Patients with decreases in the ejection fraction from 57% (two recurrent or residual disease following initial therapy have patients) → 49% and 48%, and 59% (2 patients) → 48% median progression-free and overall survival rates of 3.6 and 45%. There was no observed clinical congestive heart and 9.5 months, which have improved only 0.2 months/year failure. Grade 4 mucositis was universal with all patients from 1980 to 1997.23 At present, there is no proven salvage requiring intravenous narcotic analgesics for pain control. treatment that results in long-term disease-free survivals in All patients developed neutropenic fever requiring intra- patients having residual disease following initial treatment, venous antibiotics. One patient developed systemic candidi- or in those patients with relapsed disease, irrespective of asis at the time of engraftment requiring prolonged treat- their response to second-line therapy. Because of these poor ment with amphotericin B. One patient died following results, the role of dose intensification utilizing bone mar- chemotherapy from complications of sepsis due to Entero- row stem cells for hematopoietic support has been investi- bacter cloacae. gated based on studies suggesting steep dose–response relationships in ovarian cancer patients.9,24 Retrospective and prospective non-randomized clinical trials of high-dose Discussion chemotherapy in epithelial ovarian cancer have confirmed high response rates and pathologic complete responses in Epithelial ovarian cancer is the leading cause of death from chemotherapy-resistant patients. Furthermore, early reports gynecologic malignancy and the fifth most common neo- of survival curves suggesting a plateau in progression-free plasm in women in the United States.12,13 At the time of and overall-survival at 24–48 months following treatment6,7 diagnosis, the majority of patients will have advanced dis- have been confirmed in more recent publications of Stiff et ease; optimal initial treatment consists of debulking surgery al,25 Legros et al26 and Shinozuka et al27 followed by chemotherapy administration and results in Prognostic factors have been examined by Stiff et al.25 14 approximately 15–25% long-term disease-free survival. They have shown that age, tumor bulk at the time of trans- plant, and platinum sensitivity, are predictive of longer- Table 4 Number of days required for hematologic recovery term survivals. Recent data from the European Intergroup (GINECO) study suggests a statistically significant median No. of days of platelets Ͻ20 000/␮l mean, 19, 13 (2–80) disease-free survival for patients receiving consolidation median (range) Ͻ ␮ high-dose chemotherapy compared to patients receiving No. of days of ANC 500/ l mean, median (range) 16, 14 (10–42) 28 No. of red blood cell transfusions mean, 8, 15 (5–16) conventional-dose chemotherapy. An analysis of predic- median (range) tive factors included a low tumor burden at the time of No. of platelet transfusions mean, median (range) 17, 14 (4–103) second-look surgery and the presence of paclitaxel in the initial chemotherapy regimen. The results of this multivari-

Bone Marrow Transplantation High-dose chemotherapy in ovarian cancer RJ Morgan et al 863 ate analysis suggest that our patients should have been in fusion in patients with metastatic or high-risk primary breast a favorable group, however, the median progression-free cancer. Cancer 1994; 73: 1678–1685. and overall survivals are inferior to those in these reports 11 Kaplan EL, Meier P. Nonparametric estimation from incom- and by Legros et al.26 The reasons for this discrepancy are plete observations. J Am Stat Assoc 1958; 53: 457–481. not readily apparent. The preparative regimens used by the 12 Cannistra SA. Cancer of the ovary. New Engl J Med 1993; 329: 1550–1559. cited authors utilized platinum, or melphalan-based regi- 13 Ozols RF, Vermorken JB, Chemotherapy of advanced ovarian mens. It is possible that platinum, melphalan, or taxane- cancer: current status and future directions. Semin Oncol 1997; based regimens are more active and thus preferable for 24:S2–1–S2–9. superior long-term results in ovarian cancer patients. 14 Morgan Jr RJ, Copeland L, Gershenson D et al. NCCN Our study has shown that dose-intense doxorubicin can ovarian cancer practice guidelines. The National Comprehen- be safely administered to patients with ovarian cancer in a sive Cancer Network. Oncology (Huntingt) 1996; 10: 293– high-dose combination regimen that includes etoposide and 310. cyclophosphamide. However, the short overall and pro- 15 Neijt JP, ten Bokkel Huinink WW, van der Burg ME et al. gression-free survivals observed in our study population Randomized trial comparing two combination chemotherapy suggest that this non-platinum containing combination is regimens (CHAP-5 v CP) in advanced ovarian carcinoma. J Clin Oncol 1987; 5: 1157–1168. not optimal. 16 Omura GA, Bundy BN, Berek JS et al. Randomized trial of cyclophosphamide plus cisplatin with or without doxorubicin in ovarian carcinoma: a Gynecologic Oncology Group Study. Acknowledgements J Clin Oncol 1989; 7: 457–465. 17 Luck HJ, Meier W, Moebus V et al. Cisplatin/paclitaxel vs This work was supported in part by NCI Cancer, Center Support carboplatin/paclitaxel in ovarian cancer: update of an Arbeits- Grant CA 33572 and by CA 63265. gemeinschaft Gynaekologische Onkologie Study Group trial. Proc Am Soc Clin Oncol 1999; 18: 356 (Abstr. 1374). 18 McGuire WP, Hoskins WJ, Brady MF et al. Cyclophospham- ide and cisplatin compared with paclitaxel and cisplatin in References patients with stage III and stage IV ovarian cancer. New Engl JMed1996; 334:1–6. 1 Alberts DS, Green S, Hannigan EV et al. Improved thera- 19 Ozols RF, Bundy BN, Fowler J et al. Randomized phase III peutic index of carboplatin plus cyclophosphamide versus cis- study of cisplatin/paclitaxel versus carboplatin/paclitaxel in platin plus cyclophosphamide: final report by the Southwest optimal stage III epithelial ovarian cancer: a Gynecologic Oncology Group of a phase III randomized trial in stages III Oncology Group trial. Proc Am Soc Clin Oncol 1999; 18: 356 and IV ovarian cancer. J Clin Oncol 1992; 10: 706–717. (Abstr. 1373). 2 Howell SB, Zimm S, Markman M et al. Long-term survival 20 A’Hern RP, Gore ME. Impact of doxorubicin on survival in of advanced refractory ovarian carcinoma patients with small- advanced ovarian cancer. J Clin Oncol 1995; 13: 726–732. volume disease treated with intraperitoneal chemotherapy. J 21 Alberts DS, Liu PY, Hannigan EV et al. Intraperitoneal cispla- Clin Oncol 1987; 5: 1607–1612. tin plus intravenous cyclophosphamide versus intravenous cis- 3 Swenerton K, Jeffrey J, Stuart G et al. Cisplatin-cyclophos- platin plus intravenous cyclophosphamide for stage III ovarian phamide versus carboplatin-cyclophosphamide in advanced cancer. New Engl J Med 1996; 335: 1950–1955. ovarian cancer: a randomized phase III study of the national 22 Thigpen T, Vance R, Khansur T, Malamud F. The role of cancer institute of Canada clinical trials group. J Clin Oncol chemotherapy in the management of coelomic epithelial carci- 1992; 10: 718–726. noma of the ovary. Cancer Invest 1997; 15: 277–287. 4 Hryniuk W. The importance of dose intensity in the outcome 23 Knopf K, Brown M, Kohn EC. Lack of improvement in sur- of chemotherapy. In: Hellman, S, DeVita V, Rosenberg S vival in patients with relapsed or refractory epithelial ovarian (eds). Important Advances in Oncology. Lippincott: Philadel- cancer, 1980–1997. Proc Am Soc Clin Oncol 1998; 17: 359. phia, 1988, pp 121–141. 24 Levin L, Hryniuk W. Dose intensity analysis of chemotherapy 5 Tepler I, Cannistra SA, Frei E et al. Use of peripheral-blood regimens in ovarian carcinoma. J Clin Oncol 1987; 5: 756– progenitor cells abrogates the myelotoxicity of repetitive out- 767. patient high-dose carboplatin and cyclophosphamide chemo- 25 Stiff PJ, Bayer R, Kerger C et al. High-dose chemotherapy therapy. J Clin Oncol 1993; 11: 1583–1591. with autologous transplantation for persistent/relapsed ovarian 6 Mulder POM, Willemse PHG, Aalders JG et al. High-dose cancer: a multivariate analysis of survival for 100 consecu- chemotherapy with autologous bone marrow transplantation in tively treated patients. J Clin Oncol 1997; 15: 1309–1317. patients with refractory ovarian cancer. Eur J Cancer Clin 26 Legros M, Dauplat J, Fleury J et al. High-dose chemotherapy Oncol 1989; 25: 645–649. with hematopoietic rescue in patients with stage III to IV 7 Viens P, Maraninchi D, Legros M et al. High dose melphalan ovarian cancer: long-term results. J Clin Oncol 1997; 15: and autologous marrow rescue in advanced epithelial ovarian 1302–1308. carcinomas: a retrospective analysis of 35 patients treated in 27 Shinozuka T, Miyamoto T, Muramatsu T et al. High dose France. Bone Marrow Transplant 1990; 5: 227–233. chemotherapy with autologous stem cell support in the treat- 8 Dottino PR, Goodman H, Kredentser D et al. Clinical trial of ment of patients with ovarian carcinoma. Cancer 1999; 85: etoposide and cisplatin as salvage therapy in advanced ovarian 1555–1564. carcinoma. Gynecol Oncol 1987; 27: 350–356. 28 Cure H, Battista C, Guastalla J et al. Phase III randomized 9 Jacobs AJ, Sommers GM, Homan SM et al. Therapy of trial of high dose chemotherapy (HDC) and peripheral blood ovarian carcinoma: the relationship of dose level and treatment stem cell (PBSC) support as consolidation in patients (pts) intensity to survival. Gynecol Oncol 1988; 31: 233–245. with responsive low-burden advanced ovarian cancer (AOC): 10 Somlo G, Doroshow JH, Forman S et al. High-dose doxorub- preliminary results of a GINECO/FNCLCC/SFGM-TC study. icin, etoposide, and cyclophosphamide with stem cell rein- Proc Am Soc Clin Oncol 2001; 20: 204a.

Bone Marrow Transplantation