The trial was conducted in accordance with the Helsinki Declaration and the International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) Harmonized Tripartite Guidelines for Good Clinical Practice. The study protocol was approved by the Ethics Committee of the Medical Faculty, Friedrich-Alexander University Erlangen Nuremberg (Location of the Principal Investigator), and in addition by all relevant ethics committees at each study site, and all of the patients provided written informed consent.
This report describes a phase II, multicenter, prospective, randomized, and controlled open-label trial. Patients had to be over the age of 18 and had to be postmenopausal as defined by age over 55; age ≤ 55 years but without a menstrual period for more than 1 year; or with luteinizing hormone and follicle-stimulating hormone levels > 40 IU/L or estradiol levels < 5 ng/dL; or had to have undergone bilateral oophorectomy before the diagnosis of breast cancer. The invasive breast cancer must have been histologically confirmed by core needle biopsy and had to have either a positive estrogen receptor (ER) and/or progesterone receptor (PgR) status, defined by core biopsy immunohistochemistry with >10% positive malignant epithelial cells. Clinical Stage had to be ≥ cT1c (Size ≥1.5 cm) without distant metastases (M0). Tumors with a size of ≥1.5 cm were allowed as target lesions, although this size is not considered as a target lesion according to RECIST criteria . However tumor size assessments with mammography or breast ultrasound seem accurate enough to allow tumors of this size for evaluation .
Other inclusion criteria were: adequate renal function (creatinine clearance >30 mL/minute calculated using the Cockcroft-Gault equation), adequate bone-marrow function, adequate hepatic function, life expectancy of at least 12 months, and Eastern Cooperative Oncology Group (ECOG) status ≤ 2.
Exclusion criteria were: inflammatory breast cancer, prior letrozole or bisphosphonate treatment; patients with unstable angina or other uncontrolled cardiac disease; inflammatory breast cancer, evidence of distant metastases, other concurrent malignant disease, or current dental problems; and a history of diseases affecting bone metabolism.
Treatment and tumor assessment plan
Patients were randomly assigned to one of the two neoadjuvant treatment groups — either letrozole 2.5 mg/day plus a total of seven infusions of zoledronic acid 4 mg every 4 weeks (LET + ZOL); or letrozole 2.5 mg/day only (LET) — at a ratio of 1 : 1. Surgery for the breast tumor and axillary lymph nodes was scheduled 6.5 months after the patient had received her first dose of the study treatment. Sentinel biopsies were permitted, but complete axillary dissection was mandatory if there were positive findings. If the tumor progressed during chemotherapy, the study treatment was discontinued and further treatment was at the discretion of the investigator, who was the attending physician.
The size of the tumor was assessed before the start of therapy, after 4 months of treatment, and after 6 months of treatment, in accordance with our modified response evaluation criteria (see above). All of the images were read locally. In order to avoid unblinded reader bias in this open-label study, a central review of all mammograms was additionally performed by an independent blinded radiologist in order to assess the response status of the target lesion using standardized criteria, as described below. Principally MRI and ultrasound were allowed assessment methods as per study protocol but no study site chose those imaging methods.
The primary objective of the study was explore whether the combination of letrozole (2.5 mg/day) and zoledronic acid (4 mg q4w, or dose-adjusted based on renal function) is superior to letrozole (2.5 mg/day) monotherapy in relation to the tumor response after 6 months of preoperative treatment in postmenopausal patients with primary breast cancer. “Response” was defined as a complete response (CR: complete disappearance of all target lesions) or a partial response (PR: at least a 30% decrease in the sum of the largest diameter of all target lesions), based on MRI or mammography and/or sonography in accordance with the modified RECIST criteria . Tumor response evaluation was performed similar to RECIST, however lesions ≥ 1.5 cm as assessed by mammography or ultrasound were allowed as target lesions. Radiological tumor size assessment, either by ultrasound or mammography, has been shown to be very reliable in this range of tumor sizes [18–20].
Safety was primarily assessed by documentation of adverse events (AEs). The severity of AEs was classified in accordance with the National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events (NCI-CTCAE version 3.0). AEs were documented during the period from first exposure to the study drug to 30 days after last exposure to it. Adverse events, whether reported by the patient, discovered during general questioning by the investigator, or detected through physical examination, laboratory tests, or other means, were recorded on the adverse event log in the study case report form and followed carefully until resolution. It was not mandatory for abnormal laboratory values or test results to be considered AEs unless they induced clinical signs and symptoms or required therapeutic interventions. Adverse events were described by duration (start and end dates, or at the final examination if continuing), severity (NCI-CTCAE grades 1–5), relationship to study drug (not suspected, suspected with letrozole, suspected with zoledronic acid, suspected with both), and action taken.
Quality of life was assessed using the FACT-B questionnaire (Functional Assessment of Cancer Therapy- Breast Cancer). The Quality of life questionnaires should be completed at each visit (baseline, month 1-6) by the patient upon arrival at the clinic and before the patient has either been interviewed by the physician or received study medication. The FACT-B was analysed in the ITT population using the ‘data as observed’ by visit as total score and separated by the sub-scores physical well-being, social/family well-being, emotional well-being, functional well-being and breast cancer subscale.
Recruitment of a total of 850 patients was planned in order to reach a power of 80% to demonstrate the superiority of the combination therapy (LET + ZOL) in comparison with letrozole alone (LET) with regard to the response rate after 6 months. A response rate of 35% and an increase in the response by 10% with the addition of zoledronic acid was assumed for this calculation. Randomization was stratified by nodal status (N-negative vs. N-positive), tumor grading (G1 vs. G2–3) and center, and performed within strata in a 1 : 1 ratio.
A total of 168 patients were enrolled before the study was terminated due to low recruitment. Given the original assumptions, a figure of 200 patients would result in a power of only 25%, instead of the required 80%. The study was therefore grossly underpowered, and the results of the primary analysis must be regarded as exploratory rather than confirmatory.
In general, all summary statistics are presented by treatment group (LET vs. LET + ZOL). Categorical variables are summarized by absolute and relative frequencies. Continuous variables are summarized by descriptive statistics of mean, standard deviation, minimum, median, and maximum. Time-to-event data, including rates of affected patients, were assessed using Kaplan–Meier statistics.
A logistic regression model was fitted for the response, including the dichotomous factors “treatment” (LET vs. LET + ZOL), “nodal status” (N+ vs. N–), “tumor grading” (G1 vs. G2 and G3), “tumor size at baseline” (largest diameter ≤ 2 cm vs. > 2 cm) and “age” (< 65 years vs. ≥ 65 years).
The odds ratio (OR) for the combination therapy relative to the monotherapy was estimated. The difference between the treatment groups was tested using the likelihood ratio test with a two-sided significance level of 5%. P values are presented together with the two-sided 95% confidence intervals (CIs) for the OR.
For six patients in the LET + ZOL arm and two patients in the LET arm, a response assessment was not available after 6 months but only after 4 months. For these patients, it was assumed that the treatment response at that time would carry forward to the final assessment time point at 6 months (last observation carried forward, LOCF).
All of the statistical analyses were carried out using SAS version 8.2.