Phase III randomized trial of doxorubicin + cisplatin versus doxorubicin + 24-h paclitaxel + filgrastim in endometrial carcinoma: a Gynecologic Oncology Group study

G. F. Fleming1,*, V. L. Filiaci2, R. C. Bentley3, T. Herzog4, J. Sorosky5, L. Vaccarello6 and H. Gallion7,{dagger}

1 Section of Hematology/Oncology, Department of Medicine, University of Chicago, Chicago, IL; 2 Gynecologic Oncology Group Statistical and Data Center, Roswell Park Cancer Institute, Buffalo, NY; 3 Pathology Department, Duke University Medical Center, Durham, NC; 4 Department of Obstetrics & Gynecology, Division of Gynecologic Oncology, Washington University School of Medicine, St Louis, MO; 5 Division of Gynecologic Oncology, University of Iowa Hospitals and Clinics, Iowa City, IA; 6 Division of Gynecologic Oncology, Department of Obstetrics & Gynecology, Arthur James Cancer Hospital and Research Institute, Ohio State University, Columbus, OH; 7 Division of Gynecologic Oncology, University of Kentucky, Lexington, KY, USA

* Correspondence to: Dr G. F. Fleming, Section of Medical Oncology (MC2115), University of Chicago, 5841 S. Maryland Avenue (Room I-211), Chicago, IL 60637, USA. Tel: +1-773-702-6712; Fax: +1-773-702-0963; Email: gfleming{at}medicine.bsd.uchicago.edu


    Abstract
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Background: This study was performed to determine whether 24-h paclitaxel plus doxorubicin and filgrastim was superior to cisplatin plus doxorubicin in patients with endometrial cancer with respect to response, progression-free survival (PFS) and overall survival (OS).

Patients and methods: Eligible chemotherapy-naïve patients were randomly assigned to doxorubicin 60 mg/m2 intravenously (i.v.) followed by cisplatin 50 mg/m2 i.v. (arm 1, n=157) or doxorubicin 50 mg/m2 i.v. followed 4 h later by paclitaxel 150 mg/m2 i.v. over 24 h plus filgrastim 5 µg/kg on days 3–12 (arm 2, n=160). Starting doses were reduced for prior pelvic radiotherapy and age >65 years. Both regimens were to be repeated every 3 weeks for a maximum of seven cycles.

Results: There was no significant difference in response rate (40% versus 43%), PFS (median 7.2 versus 6 months) or OS (median 12.6 versus 13.6 months) for arm 1 and arm 2, respectively. Toxicities were primarily hematological, with 54% (arm 1) and 50% (arm 2) of patients experiencing grade 4 granulocytopenia. Gastrointestinal toxicities were similar in both arms.

Conclusions: Doxorubicin and 24-h paclitaxel plus filgrastim was not superior to doxorubicin and cisplatin in terms of response, PFS or survival in advanced endometrial cancer.

Key words: chemotherapy, cisplatin, doxorubicin, endometrial cancer, paclitaxel


    Introduction
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Most patients with endometrial cancer present with early-stage disease and are cured with local therapy. However, the prognosis for patients with inoperable stage III or IV, or recurrent, endometrial cancer remains poor. A small subset of these patients benefits from hormonal manipulation, but other active systemic therapies are still needed. Prior to the availability of taxanes, anthracyclines and platinum compounds represented the most active cytotoxic agents, each producing single-agent responses in the range of 20–30% [1Go, 2Go]. In 1993, the Gynecologic Oncology Group (GOG) reported 45% of patients responding (including 22% complete responses) [3Go] to the combination of 60 mg/m2 doxorubicin plus 50 mg/m2 cisplatin in advanced or recurrent endometrial carcinoma. This was significantly superior in a randomized trial to the 27% response obtained from doxorubicin alone. Overall survival (OS), however, was not significantly better in the combination arm, with a median of ~9 months [1Go]. Similarly, a randomized phase II/III trial conducted by the European Organization for the Research and Treatment of Cancer–Gynecologic Cancer Cooperative Group reported 43% of patients responding using a doxorubicin–cisplatin doublet versus 17% responding using doxorubicin only. Survival was not significantly superior in the combination arm (9 versus 7 months; P=0.0654) [4Go]. Results achieved with a doxorubicin–cisplatin combination have been confirmed by a GOG trial randomly allocating treatment with doxorubicin plus cisplatin to either standard or circadian-timed schedules. There was no difference between the two arms; overall response rate was 47% with a median survival of 12 months [5Go].

More recently, the GOG evaluated single-agent paclitaxel 250 mg/m2 as a 24-h continuous infusion with granulocyte colony-stimulating factor (G-CSF) support in chemotherapy-naïve patients with advanced endometrial carcinoma. Overall, 36% of patients responded, including 14% who experienced complete responses [6Go]. At the time this trial was planned, little information regarding activity using paclitaxel infusions of shorter duration in endometrial cancer was available. The goal of this study was to determine whether replacing cisplatin with a 24-h paclitaxel infusion would increase OS, progression-free survival (PFS) or objective response rate in patients with advanced endometrial cancer. The regimen combining doxorubicin with a 24-h paclitaxel infusion was developed by the Eastern Cooperative Oncology Group and tested in a large multicenter study of patients with metastatic breast cancer [7Go]. Because prolonged infusion of paclitaxel results in greater myelosuppression, growth factor support was required on the experimental arm.


    Materials and methods
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Eligibility criteria
Chemotherapy-naïve women with histologically documented measurable stage III, stage IV or recurrent endometrial carcinoma and a GOG performance status (PS) of 0–2 were eligible. Prior therapy with hormones was permitted, but was to be discontinued prior to study entry. At least 4 weeks were to have elapsed since completion of radiotherapy involving the whole pelvis or over 50% of the spine. Adequate hematological and end organ function was required, as evidenced by a white blood cell count (WBC) of at least 3000/µl, an absolute granulocyte count of at least 1500/µl, serum creatinine no more than 1.6 mg/dl, alanine aminotransferase no more than three times upper limits of normal, bilirubin within institutional normal limits, and a left ventricular ejection fraction (LVEF) at ≥50%. Patients with third-degree or complete heart block were eligible only if a pacemaker was in place. Patients with concomitant medical illness such as uncontrolled infection, uncontrolled angina or serious peripheral neuropathy were ineligible. Patients with a prior or concomitant malignancy (other than non-melanoma skin cancer) or hypersensitivity to Escherichia coli-derived drug preparations, were ineligible. The Institutional Review Boards of all participating institutions approved the protocol prior to patient enrollment, and all patients provided written informed consent in accordance with federal, state and local institutional guidelines.

Patient evaluation and treatment
Pretreatment evaluation included a history and physical examination, complete blood cell count (CBC) with differential and platelet count, serum biochemistry, CA-125, baseline assessment of PS, tumor measurements, chest X-ray, electrocardiogram (EKG), measurement of LVEF and an audiogram, if clinically indicated. CBC, differential and platelet count was to be repeated weekly for all patients, and also on day 12 for patients on arm 2 receiving filgrastim. Physical examination, assessment of toxicity, serum biochemistry and CA-125 were to be repeated each cycle. Tumor measurements were to be repeated each cycle if obtained by physical exam, and every other cycle if by radiological evaluation. LVEF measurement was to be repeated every three cycles.

Treatment on arm 1, the standard treatment arm, consisted of doxorubicin 60 mg/m2 intravenously (i.v.) followed immediately by cisplatin 50 mg/m2 i.v. Cisplatin was diluted in 250 cm3 0.9% sodium chloride and infused over 1 h. Patients who had received prior pelvic radiotherapy or who were over the age of 65 years were to receive reduced starting doses of doxorubicin 45 mg/m2 and cisplatin 40 mg/m2.

The treatment plan for arm 2, the experimental treatment arm, consisted of 50 mg/m2 of doxorubicin on day 1 administered by i.v. bolus or brief infusion followed 4 h later by paclitaxel at a dose of 150 mg/m2 administered as a 24-h infusion. Filgrastim was to be given subcutaneously at a dose of 5 µg/kg on a daily basis, starting on day 3 of each cycle and continuing until day 12 or until the WBC was at least 10 000/mm3 (post nadir). Patients who had received prior pelvic radiotherapy or who were over the age of 65 years were to receive reduced starting doses of doxorubicin 40 mg/m2 and paclitaxel 120 mg/m2.

All patients with a body surface area (BSA) >2 m2 were dosed as though their BSA was 2 m2. Treatment was continued for seven cycles, or until disease progression or unacceptable toxicity necessitated therapy discontinuation. Doses were reduced for grade 3 or 4 WBC, granulocyte or platelet toxicity, and could be re-escalated if subsequent cycles produced toxicity of no more than grade 1. Filgrastim was to be added for patients on arm 1 who had persistent grade 4 hematological toxicity or treatment delays despite dose reduction. Grade 3 peripheral neuropathy required discontinuation of therapy until the severity of symptoms was no worse than grade 1, with the paclitaxel or cisplatin dose reduced for subsequent cycles. Grade 3 or 4 mucositis or diarrhea required dose reduction. If creatinine was >1.6 mg/dl on the day of therapy, cisplatin was discontinued. Protocol prescribed therapy was discontinued in patients experiencing a drop in LVEF to <50%, or a 20% decrease from baseline value, or for any toxicity-related delay in therapy of >3 weeks.

Response was measured according to standard GOG criteria. A complete response was defined as the disappearance of all gross evidence of disease for at least 4 weeks. Partial response was a ≥50% reduction in the product of perpendicular diameters obtained from measurement of each lesion for at least 4 weeks. Increasing disease was defined as a ≥50% increase in the product of perpendicular diameters of any lesion or the appearance of any new lesion within 8 weeks of study entry. Stable disease was disease meeting none of the above criteria. Survival was defined as observed length of life from entry on study to death or date of last contact. PFS was defined as date from entry on study to date of reappearance or increasing parameters of disease, death or date of last contact. Standard GOG common toxicity criteria were used to grade the severity of adverse events.

Pharmacokinetics
During cycle 1 only, blood samples for paclitaxel levels were taken at 0 (pre-infusion), 3, 22 and 27 h after the start of the paclitaxel infusion. Ten milliliters of blood were drawn in a heparinized tube and centrifuged within 60 min of collection. The plasma was then separated out and frozen at –20°C until shipping to a central laboratory for analysis. One hundred and twenty-nine women had at least one sample drawn after start of infusion, which could be used for pharmacokinetic analysis. Full results of a population pharmacokinetic analysis will be published elsewhere. Results were consistent with published literature on paclitaxel pharmacokinetics for total paclitaxel following long infusion, and demonstrated decreased total paclitaxel clearance with increased age, increased aspartate aminotransferase, or lower weight (David Spriggs, personal communication).

Statistical methods
The primary end point of the study was frequency of objective response (complete and partial). Secondary end points include duration of PFS and OS. A total sample size of 290 patients was targeted to detect an overall relative odds of response of 1.9 with type I error set at 0.05 in one tail and type II error of 0.20 [8Go]. This sample size would also allow an 83% chance of detecting a 40% increase in survival duration with 12 months of follow-up (240 deaths) and type I error of 0.05 in one tail [9Go]. Additionally, this sample size would permit an 88% chance of detecting a similar increase in duration of PFS with 12 months of follow-up (280 failures) [9Go].

The GOG Statistical and Data Center (SDC) carried out randomization with equal probabilities balancing the sequence of assigned regimens within institutions and strata. The stratification classified the patients into two groups: patients ≤65 years old and no history of radiation therapy, and patients >65 years old or with a history of radiation therapy. The sequence of treatment assignments was concealed. After patient eligibility was verified by the SDC, random treatment assignment was provided. Data were collected and reviewed centrally at the SDC. The study chair reviewed patient data to assess protocol compliance, toxicity, eligibility and end points. Pathology materials were collected at the SDC and reviewed by the GOG Pathology Committee, including slides documenting the primary and metastatic disease. The Gynecologic Oncology Committee centrally reviewed all patient eligibility data.

An interim futility analysis of response was planned and carried out after a minimum of 75 patients had been evaluated in each arm. Accrual termination was to be considered if the odds of response in the experimental arm were less than that in the standard arm [10Go]. The GOG Data Monitoring Committee opted to continue accrual after reviewing response and toxicity data ~22 months after study activation.

A log rank test stratified by PS was used to test the independence of treatment with survival and PFS [11Go]. The Kaplan–Meier method was used to obtain estimates of PFS and survival [12Go]. A proportional hazards model was used to provide an estimate of the relative hazard stratified by PS for both PFS and survival [13Go]. An exact test stratified by PS [14Go] was used to test the benefit of the experimental arm with respect to response, and the conditional maximum likelihood estimate of the common relative odds ratio is reported with exact 95% confidence bounds [15Go]. The intention-to-treat principle was applied in the analysis of response, survival and PFS. Tabulation of adverse effects includes only treated patients.


    Results
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Three hundred and twenty-eight patients with primary stage III, stage IV or recurrent endometrial carcinoma were enrolled on this study between 12 August 1996 and 30 November 1998. Eleven patients (three on the standard arm, eight on the experimental arm) were ineligible due to: withdrawn consent (one), second primary cancer (three), wrong histology (one), wrong primary (two), inadequately documented primary pathology (two), elevated creatinine (one) and failure to have required testing/no measured disease (one). Three hundred and seventeen patients are included in the survival analysis (157 on arm 1 and 160 on arm 2). Three hundred and thirteen patients received at least one cycle of therapy and were evaluable for toxicity (156 on arm 1 and 157 on arm 2). Patient characteristics, with the exception of age at entry, were well balanced between the arms and are shown in Table 1. Sixty-nine percent of patients on arm 1 and 64% of patients on arm 2 received an initial dose reduction because of prior pelvic radiation and/or age >65 years, and 5% of all patients failed to receive required reductions. Seventy patients on arm 1 (45%) and 78 patients on arm 2 (49%) received all seven cycles of therapy. Median follow-up for those alive at last contact was 61 months. Two hundred and sixty-six deaths have been reported as of December 2003, with 130 on arm 1 and 136 on arm 2.


View this table:
[in this window]
[in a new window]
 
Table 1. Patient characteristics

 
Toxicity
Toxicities are shown in Table 2. Study therapy was discontinued because of toxicity in 25 (16%) patients on arm 1 and in 23 (14%) patients on arm 2. The major toxicity on both arms was granulocytopenia. Eighty-five patients (54%) on arm 1 and 78 patients (50%) on arm 2 experienced grade 4 granulocytopenia. However, only 25 patients on arm 1 (16%) and 35 patients on arm 2 (22%) experienced fever of any grade or from any cause. Thirty-four percent of patients who required and received initial dose reduction, 35% of patients who required but did not receive initial dose reduction and 42% of patients who did not require initial dose reduction experienced grade 4 granulocytopenia in cycle 1. Thrombocytopenia was rare, with only 1% of patients in either arm experiencing a grade 4 decrease in platelets.


View this table:
[in this window]
[in a new window]
 
Table 2. Toxicity in treated patients, all grades

 
Gastrointestinal symptoms, primarily nausea and vomiting, were the most common non-hematological toxicities. Cardiac toxicity was uncommon in both arms, with study therapy discontinued in 23 patients because of a decline in ejection fraction. Treatment may have contributed to the death of eight patients. One patient on arm 1 and two on arm 2 died of neutropenic sepsis. For one of the patients on arm 2, it was unclear whether she was taking her filgrastim as directed. In addition, one patient on arm 1 died of sepsis not related to neutropenia, one died of disease plus herpes zoster infection and one died of renal failure. One patient on arm 2 died at home without explanation, and one died of acute myelogenous leukemia (AML). The patient with AML had been treated with seven cycles of doxorubicin and paclitaxel, and achieved a complete response. Twenty-two months after initiation of chemotherapy she was diagnosed with AML. Her karyotype revealed an 11q23 translocation, as has been reported in chemotherapy (topoisomerase II)-associated leukemias [16Go].

Response
No improvement in response was observed for the experimental arm. The overall response rate was 40% in arm 1 and 43% in arm 2. The common odds of response ratio in arm 2 relative to arm 1 stratified by PS was 1.12 [95% confidence interval (CI) 0.69–1.79; P=0.36, one-tailed]. Including the 11 ineligible patients in the analysis did not appreciably change the results. Patients with prior pelvic radiotherapy were overall less likely to respond (35%, both arms combined) than were patients with no prior pelvic radiotherapy (48%, both arms combined). Statistically significant factors related to poorer response included PS of 2, prior radiotherapy, liver metastases, and disease that was recurrent after presentation at an earlier stage.

There were no significant differences in PFS (Figure 1) or OS (Figure 2) between the treatment arms. Median PFS was 7.2 months on arm 1 and 6 months on arm 2. Adjusting for PS, the hazard ratio relative to arm 1 is 1.01 (95% CI 0.80–1.28; P=0.46, one-tailed). Median OS was 12.6 months on arm 1 and 13.6 months on arm 2. Adjusting for PS, the death hazard ratio relative to arm 1 is 1.00 (95% CI 0.78–1.27; P=0.49, one-tailed). Statistically significant factors predicting for longer OS include good PS, grade 1 histology, extrapelvic disease other than lung and abdomen but including the liver, and months to first recurrence.



View larger version (19K):
[in this window]
[in a new window]
 
Figure 1. Progression-free survival (PFS) by randomized treatment.

 


View larger version (18K):
[in this window]
[in a new window]
 
Figure 2. Survival by randomized treatment.

 

    Discussion
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
Hormonal therapy can produce responses in a minority of patients with advanced endometrial cancer [17Go, 18Go]. However, most endometrial cancers will not respond to hormonal treatment, and those that do will eventually become refractory. The GOG has been systematically evaluating new cytotoxic agents for activity in endometrial cancer, and phase II data suggest that paclitaxel is among the most active agents ever tested. It was hoped that inclusion of paclitaxel in a first-line chemotherapy regimen would both improve response rates (and, by association, symptoms) [19Go] and prolong patient survival. However, substituting paclitaxel for cisplatin in a first-line chemotherapy regimen did not positively impact response rates, PFS or OS.

Toxicities in the two arms were similar, but the paclitaxel arm required the use of filgrastim. The rate of neuropathy was fairly low (8% and 9% grade 2 or higher in arm 1 and arm 2, respectively) and was similar to that seen in previous GOG endometrial cancer trials using doxorubicin–cisplatin [20Go]. These rates may increase in the adjuvant setting where a higher percentage of patients receive a full course of therapy. Of note is the single case of AML. Because relatively few patients with metastatic or recurrent endometrial cancer live long enough to develop AML, this raises concern about the potential frequency of such events when regimens are moved to the adjuvant setting. Estimates of the risk of secondary leukemia after adjuvant anthracycline-based therapy for breast cancer are in the range of 0.2% to 2.5% at 10 years, and may be increased in those women who also receive radiotherapy [21Go, 22Go]. While it is possible that G-CSF or taxane contributes to the risk of AML, one case of AML was also reported on the previous randomized GOG trial for advanced/recurrent endometrial cancer [20Go]. In that trial both arms received doxorubicin–cisplatin with no G-CSF. The patient reported here, as well as the patient treated on the previous randomized GOG trial, had received prior pelvic radiotherapy, and it is possible that this may contribute to the risk of development of AML. This will be of concern in adjuvant endometrial cancer trials using both pelvic radiotherapy and doxorubicin-containing regimens.

Subsequent to the design of this trial, a 3-h paclitaxel infusion has been shown to have substantial activity in endometrial cancer and shorter, less myelotoxic infusions have replaced more prolonged infusions in treating most tumor types [23Go–25Go]. However, shorter infusions may be more neurotoxic [26Go, 27Go] and may interact with doxorubicin to increase its cardiac toxicity [28Go]. The GOG has recently completed a trial randomizing therapy in women with advanced endometrial cancer between the doublet doxorubicin–cisplatin and the triplet doxorubicin–cisplatin–3-h paclitaxel [29Go]. Results from that trial, in which paclitaxel is separated from the doxorubicin and cisplatin by 24 h to minimize possible adverse cardiac and neurological interactions, show increased response rate and survival, but also increased neurotoxicity, with the triplet therapy. Other investigators have combined 3-h paclitaxel with carboplatin, which is less neurotoxic than cisplatin and also has activity in endometrial cancer [30Go]; preliminary results of that combination are also encouraging [31Go–33Go]. The GOG is currently comparing the combination of paclitaxel and carboplatin with the three-drug regimen of cisplatin, doxorubicin and paclitaxel.


    Acknowledgements
 
The authors wish to thank Drs David Spriggs and Diane Mould for their expert work in coordinating the pharmacokinetic testing and analysis in this study, the results of which will be reported elsewhere. This study was supported by National Cancer Institute grants to the Gynecologic Oncology Group (GOG) Administrative Office (CA 27469) and the Gynecologic Oncology Group Statistical and Data Center (CA 37517). The following GOG member institutions participated in this study: University of Alabama School of Medicine, Duke University Medical Center, Abington Memorial Hospital, Walter Reed Army Medical Center, Wayne State University, University of Minnesota Medical School, University of Mississippi Medical Center, Colorado Gynecologic Oncology Group P.C., University of California at Los Angeles, University of Washington, University of Pennsylvania, Milton S. Hershey Medical Center, University of Cincinnati, University of North Carolina School of Medicine, University of Iowa Hospitals and Clinics, University of Texas/Southwestern Medical Center at Dallas, Indiana University Medical Center, Wake Forest University School of Medicine, Albany Medical College, University of California Medical Center at Irvine, Tufts-New England Medical Center, Rush-Presbyterian-St Luke's Medical Center, SUNY Downstate Medical Center, University of Kentucky, Community Clinical Oncology Program, The Cleveland Clinic Foundation, Johns Hopkins Oncology Center, State University of New York at Stony Brook, Eastern Pennsylvania Gyn/Onc Center PC, Washington University School of Medicine, Memorial Sloan-Kettering Cancer Center, Cooper Hospital/University Medical Center, Columbus Cancer Council, MD Anderson Cancer Center, University of Massachusetts Medical Center, Fox Chase Cancer Center, Medical University of South Carolina, Women's Cancer Center, University of Oklahoma, University of Virginia, University of Chicago, Tacoma General Hospital, Thomas Jefferson University Hospital, Mayo Clinic, Case Western Reserve University, Tampa Bay Cancer Consortium, North Shore University Hospital, Brookview Research Inc. and Ellis Fischel Cancer Center.


    Notes
 
{dagger} Present address: Magee-Womens Hospital, University of Pittsburgh School of Medicine, Pittsburgh, PA, USA. Back

Received for publication April 6, 2004. Accepted for publication April 15, 2004.


    References
 Top
 Abstract
 Introduction
 Materials and methods
 Results
 Discussion
 References
 
1. Muss HB. Chemotherapy of metastatic endometrial cancer. Semin Oncol 1994; 21: 107–113.

2. Moore TD, Phillips PH, Nerenstone SR et al. Systemic treatment of advanced and recurrent endometrial carcinoma: current status and future directions. J Clin Oncol 1991; 9: 1071–1088.[Abstract]

3. Thigpen T, Blessing J, Homesley H et al. Phase III trial of doxorubicin +/– cisplatin in advanced or recurrent endometrial carcinoma: A Gynecologic Oncology Group (GOG) study. Proc Am Soc Clin Oncol199312261 (Abstr 830).

4. Aapro MS, Van Wijk FH, Bolis G et al. Doxorubicin versus doxorubicin and cisplatin in endometrial carcinoma: definitive results of a randomized study (55872) by the EORTC Gynaecological Cancer Group. Ann Oncol 2003; 14: 441–448.[Abstract/Free Full Text]

5. Gallion HH, Brunetto VL, Cibull M et al. Randomized phase III trial of standard timed doxorubicin plus cisplatin versus circadian timed doxorubicin plus cisplatin in stage III and IV or recurrent endometrial carcinoma: a Gynecologic Oncology Group study. J Clin Oncol 2003; 21: 3808–3813.[Abstract/Free Full Text]

6. Ball HG, Blessing JA, Lentz SS et al. A phase II trial of paclitaxel in patients with advanced or recurrent adenocarcinoma of the endometrium: A Gynecologic Oncology Group study. Gynecol Oncol 1996; 62: 278–281.[CrossRef][ISI][Medline]

7. Sledge GW, Neuberg D, Bernardo P et al. Phase III trial of doxorubicin, paclitaxel, and the combination of doxorubicin and paclitaxel as front-line chemotherapy for metastatic breast cancer: an intergroup trial (E1193). J Clin Oncol 2003; 21: 588–592.[Abstract/Free Full Text]

8. Munoz A, Rosner B. Power and sample size for a collection of 2 x 2 tables. Biometrics 1984; 40: 995–1004.[ISI]

9. Rubinstein LV, Gail MH, Santner TJ. Planning and duration of a comparative clinical trial with loss to follow-up and a period of continued observation. J Chron Dis 1981; 34: 469–479.[ISI][Medline]

10. Weiand S, Therneau T. A two-stage design for randomized trials with binary outcomes. Control Clin Trials 1987; 8: 20–28.[CrossRef][ISI][Medline]

11. Mantel N. Evaluation of survival data and two new rank order statistics arising in its consideration. Cancer Chemother Rep 1996; 50: 163–170.

12. Kaplan EL, Meier P. Nonparametric estimation from incomplete observations. J Am Stat Assoc 1958; 53: 457–481.[ISI]

13. Cox DR. Regression model and life tables (with discussion). J R Stat Statistical Soc B 1972; 34: 187–219.

14. Cox DR. The Analysis of Binary Data. London, UK: Methuen & Co. (reprinted Chapman and Hall) 1977.

15. Mehta CR, Patel NR, Gray R. Computing an exact confidence interval for the common odds ratio in several 2 x 2 contingency tables. J Am Stat Assoc 1985; 80: 969–973.[ISI]

16. Pedersen-Bjergaard J, Sigsgaard T, Neilsen D et al. Acute monocytic or myelomonocytic leukemia with balanced chromosome translocations to band 11q23 after treatment with 4-epi-doxorubicin and cisplatin or cyclophosphamide for breast cancer. J Clin Oncol 1992; 10: 1444–1451.[Abstract]

17. Lentz SS, Brady MF, Major FJ et al. High-dose megestrol acetate in advanced or recurrent endometrial carcinoma: a Gynecologic Oncology Group study. J Clin Oncol 1996; 14: 357–361.[Abstract]

18. Thigpen JT, Brady MF, Alvarez RD et al. Oral medroxyprogesterone acetate in the treatment of advanced or recurrent endometrial carcinoma: a dose–response study by the Gynecologic Oncology Group. J Clin Oncol 1999; 17: 1736–1744.[Abstract/Free Full Text]

19. Geels P, Eisenhauer E, Bezjak A et al. Palliative effect of chemotherapy: objective tumor response is associated with symptom improvement in patients with metastatic breast cancer. J Clin Oncol 2000; 18: 2395–2405.[Abstract/Free Full Text]

20. Gallion HH, Brunetto VL, Cibull M et al. Randomized phase III trial of standard timed doxorubicin plus cisplatin versus circadian timed doxorubicin plus cisplatin in stage III and IV or recurrent endometrial carcinoma: a Gynecologic Oncology Group study. J Clin Oncol 2003; 21: 3808–3813.[Abstract/Free Full Text]

21. Diamandidou E, Buzdar A, Smith T et al. Treatment-related leukemia in breast cancer patients treated with fluorouracil–doxorubicin–cyclophosphamide combination adjuvant chemotherapy: the University of Texas M.D. Anderson Cancer Center experience. J Clin Oncol 1996; 14: 2722–2730.[Abstract]

22. Smith RE, Bryant J, DeCillis A et al. Acute myeloid leukemia and myelodysplastic syndrome after doxorubicin–cyclophosphamide adjuvant therapy for operable breast cancer: the National Surgical Adjuvant Breast and Bowel Project experience. J Clin Oncol 2003; 21: 1195–1204.[Abstract/Free Full Text]

23. Lissoni A, Zanetta G, Losa G et al. Phase II study of paclitaxel as salvage treatment in advanced endometrial cancer. Ann Oncol 1996; 7: 861–863.[Abstract]

24. Woo HL, Swenerton KD, Hoskins PJ. Taxol is active in platinum-resistant endometrial adenocarcinoma. Am J Clin Oncol 1996; 19: 290–291.[CrossRef][ISI][Medline]

25. Lincoln S, Blessing JA, Lee RB, Rocereto TF. Activity of paclitaxel as second-line chemotherapy in endometrial carcinoma: a Gynecologic Oncology Group study. Gynecol Oncol 2003; 88: 277–281.[CrossRef][ISI][Medline]

26. Smith RE, Brown AM, Mamounas EP et al. Randomized trial of 3-hour versus 24-hour infusion of high-dose paclitaxel in patients with metastatic or locally advanced breast cancer: National Surgical Adjuvant Breast and Bowel Project Protocol B-26. J Clin Oncol 1999; 17: 3403–3411.[Abstract/Free Full Text]

27. Connelly E, Markman M, Kennedy A et al. Paclitaxel delivered as a 3-hr infusion with cisplatin in patients with gynecologic cancers: unexpected incidence of neurotoxicity. Gynecol Oncol 1996; 62: 166–168.[CrossRef][ISI][Medline]

28. Gianni L, Munzone E, Capri G et al. Paclitaxel by 3-hour infusion in combination with bolus doxorubicin in women with untreated metastatic breast cancer: high antitumor efficacy and cardiac effects in a dose-finding and sequence-finding study. J Clin Oncol 1995; 13: 2688–2699.[Abstract]

29. Fleming GF, Brunetto VL, Cella D et al. Phase III trial of doxorubicin plus cisplatin with or without paclitaxel plus filgrastim in advanced endometrial carcinoma: a Gynecologic Oncology Group Study. J Clin Oncol 2004; 22: 2159–2166.[Abstract/Free Full Text]

30. Burke TW, Munkarah A, Kavanagh JJ et al. Treatment of advanced or recurrent endometrial carcinoma with single-agent carboplatin. Gynecol Oncol 1993; 51: 397–400.[CrossRef][ISI][Medline]

31. Dimopoulos MA, Papadimitrious CA, Georgoulias V et al. Paclitaxel and cisplatin in advanced or recurrent carcinoma of the endometrium: long-term results of a phase II multicenter study. Gynecol Oncol 2000; 78: 52–57.[CrossRef][ISI][Medline]

32. Zanotti KM, Belinson JL, Kennedy AW et al. The use of paclitaxel and platinum-based chemotherapy in uterine papillary serous carcinoma. Gynecol Oncol 1999; 74: 272–277.[CrossRef][ISI][Medline]

33. Weber B, Mayer F, Bougnoux P et al. What is the best chemotherapy regimen in recurrent or advanced endometrial carcinoma? Preliminary results. Proc Am Soc Clin Oncol 2003; 22: 453 (Abstr 1819).