OBJECTIVES:

• Compare the progression-free survival and response of patients with stage III or IV or recurrent endometrial cancer treated with doxorubicin, cisplatin, paclitaxel, and filgrastim (G-CSF) vs tamoxifen and megestrol.
• Compare the survival of patients treated with these regimens.
• Determine if progesterone receptor status provides information on whether patients are more likely to benefit from chemotherapy.
• Compare the toxicity profiles of these treatment regimens in these patients.
• Compare the quality of life of patients treated with these regimens.

OUTLINE: This is a randomized, cross-over, multicenter study. Patients are stratified according to progesterone receptor status (negative vs positive). Patients are randomized to 1 of 2 treatment arms.

• Arm I:Patients receive chemotherapy comprising doxorubicin IV over 15-30 minutes followed by cisplatin IV over 1 hour on day 1; paclitaxel IV over 3 hours on day 2; and filgrastim (G-CSF) subcutaneously beginning on day 3 and continuing for 10 days. Chemotherapy repeats every 21 days for up to 7 courses in the absence of disease progression or unacceptable toxicity.
• At time of disease progression, patients cross-over to hormonal therapy as in arm II.
• Arm II: Patients receive hormonal therapy comprising oral megestrol twice daily on weeks 1-3 followed by oral tamoxifen twice daily on weeks 4-6. Hormonal therapy repeats every 6 weeks in the absence of disease progression or unacceptable toxicity.

At time of disease progression, if patients have not previously been enrolled on arm I, patients cross-over to receive chemotherapy as in arm I.

Quality of life is assessed at baseline, 6 weeks, time of progression, and then after 6 weeks on cross-over therapy.

Patients are followed every 3 months for 2 years, every 6 months for 3 years, and then annually thereafter.

PROJECTED ACCRUAL: Approximately 630 patients will be accrued for this study within 42 months.

DISEASE CHARACTERISTICS:

• Histologically confirmed primary stage III or IV or recurrent endometrial cancer
• Poor curative potential with radiotherapy or surgery (alone or in combination)

• Measurable disease

  • At least one lesion accurately measured in at least one dimension

    • At least 20 mm by conventional techniques, including palpation, x-ray, CT scan, or MRI OR
    • At least 10 mm by spiral CT scan
  • Disease in a previously irradiated field as sole site of measurable disease allowed only if clear progression after completion of radiotherapy

• Estrogen receptor(ER)/progesterone receptor (PR) status of primary tumor required

  • ER/PR status of measurable tumor optional

PATIENT CHARACTERISTICS:

Age:

• Not specified

Performance status:

• GOG 0-2

Life expectancy:

• Not specified

Hematopoietic:

• Platelet count at least 100,000/mm^3
• Granulocyte count at least 1,500/mm^3

Hepatic:

• Bilirubin normal
• SGPT no greater than 3 times upper limit of normal

Renal:

• Creatinine no greater than 1.6 mg/dL

Cardiovascular:

• LVEF at least 50%
• No third-degree or complete heart block, unless pacemaker is in place
• Other conduction abnormalities or cardiac dysfunction allowed at the investigator's discretion
• No history of deep venous thrombosis
• No uncontrolled angina

Pulmonary:

• No history of pulmonary embolus

Other:

• No other malignancy within the past 5 years except nonmelanoma skin cancer
• No concurrent medical illness that would preclude study
• No serious uncontrolled infection
• No serious peripheral neuropathy
• No circumstances that would preclude study compliance
• No sensitivity to E. coli-derived drug preparations

PRIOR CONCURRENT THERAPY:

Biologic therapy:

• Prior biologic therapy allowed

Chemotherapy:

• No prior cytotoxic chemotherapy, including chemotherapy for radiosensitization

Endocrine therapy:

• No prior hormonal therapy for endometrial cancer

Radiotherapy:

• See Disease Characteristics
• At least 4 weeks since prior radiotherapy involving the whole pelvis or more than 50% of the spine

Surgery:

• See Disease Characteristics

Other:

• Concurrent cardiac conduction-altering medications such as digitalis, beta blockers, or calcium channel blockers allowed at the investigator's discretion