A Randomized Phase III Trial Comparing 16 to 18 Weeks of Neoadjuvant Exemestane (25 mg Daily), Letrozole (2.5 mg), or Anastrozole (1 mg) in Postmenopausal Women With Clinical Stage II and III Estrogen Receptor Positive Breast Cancer

Trial Information

A Randomized Phase III Trial Comparing 16 to 18 Weeks of Neoadjuvant Exemestane (25 mg Daily), Letrozole (2.5 mg), or Anastrozole (1 mg) in Postmenopausal Women With Clinical Stage II and III Estrogen Receptor Positive Breast Cancer

OBJECTIVES:

Primary

- Determine whether anastrozole, exemestane, or letrozole administered for 16 to 18 weeks
as neoadjuvant endocrine treatment for postmenopausal patients with stage II or stage
III estrogen receptor (ER)-positive breast cancer should be chosen as the aromatase
inhibitor arm of a future study that will compare neoadjuvant aromatase inhibitor (AI)
treatment with neoadjuvant chemotherapy. (Cohort A)

- To determine whether patients who have a high Ki-67 value (> 10%) after 2 weeks of
neoadjuvant AI treatment experience a higher than expected pathological response rate
to neoadjuvant chemotherapy (20%) than would be typically observed for postmenopausal
patients with unselected ER+ rich tumors (estimated to be 5%), indicating that an early
assessment of proliferation is a useful approach to the identification of a
chemotherapy sensitive subgroup of ER+ tumors. (Cohort B [patients enrolled after the
375th patient])

Secondary

- Compare the neoadjuvant treatment regimens relative to the rates of improvement in
surgical outcome: mastectomy with primary skin closure and negative surgical margins
(for T4 a, b, c tumors); breast conserving surgery with negative final margins (for T3
tumors and T2 tumors classified as requiring mastectomy at baseline); breast conserving
surgery at first attempt (for T2 tumors classified as potential candidates for breast
conservation).

- Compare and confirm the radiological response rates (mammography and ultrasound by
central radiological analysis) between these three neoadjuvant treatment regimens.

- Compare the relative safety of the neoadjuvant treatment regimens in terms of reported
adverse events.

- To compare the tumor pathologic size between the neoadjuvant treatment regimens, to
compare the rates of pathological complete response, and to compare down-staging to
stage I.

- To compare the rate of complete cell cycle response between the three treatment
regimens (Ki67 staining of 1% or less in the post treatment sample).

- Compare the incidence of metastatic lymph node involvement on the three arms of the
study in patients who have a lymph node dissection at the end of neoadjuvant treatment.

- Determine the 10-year incidence of local recurrence in patients treated with these
regimens.

- To collect tumor tissue, serum specimens, and plasma specimens to develop predictive
biomarkers that can be used to select tumors for neoadjuvant AI therapy.

- Collect surgical specimens post-AI neoadjuvant therapy to identify markers of de novo
resistence to AI therapy.

OUTLINE: This is a multicenter study comprising cohort A (phase III study) and cohort B
(phase II study). Once cohort A accrual is met (375 patients), subsequent patients are
enrolled to cohort B. Patients in both cohorts are stratified according to T stage (T2 vs T3
vs T4), and randomized to 1 of 3 aromatase inhibition (AI) treatment arms.

- Arm I: Patients receive oral exemestane once daily for 16-18 weeks.

- Arm II: Patients receive oral letrozole once daily for 16-18 weeks.

- Arm III: Patients receive oral anastrozole once daily for 16-18 weeks. Patients in
cohort B undergo breast biopsy after 2-4 weeks of AI treatment for analysis of Ki-67
levels. Patients with Ki-67 level ≤ 10% continue AI treatment. Patients with Ki-67
level > 10% (high) are given the option to switch to neoadjuvant chemotherapy or
undergo immediate breast surgery.

After completion of AI therapy, all patients undergo partial or radical mastectomy or
lumpectomy with or without lymph node dissection.

After surgery, patients are followed up periodically for 10 years.

PROJECTED ACCRUAL: A total of 610 patients (375 for cohort A and 235 for cohort B) will be
accrued for this study.