Trial Information

Aromatase Inhibitors: Skeletal Effects and the Role of CYP19 Gene Polymorphisms

A family history of osteoporosis will be noted. Data on the factors that affect estrogen
exposure such as age at menarche, average number of periods per year during the reproductive
years, number of years of birth control pill use, total number of pregnancies, number of
pregnancies to term, months of lactation and age at menopause will be recorded. These data
will be collected at baseline. Current medications and other medical problems will be
recorded at baseline and during follow-up visits. New medical problems and new medications
will be noted on follow-up.

Physical examination will be done at baseline and during follow-up visits. Body mass index
calculated as weight (kg) divided by the square of the height (m2) will be obtained every
visit. Height will be obtained using a standard stadiometer and weight will be taken using a
standard weighing scale.

Dietary calcium and Vitamin D intake will be estimated from a 7-day dietary recall at
baseline and adjusted to a daily intake of 1200-1500 mg of calcium per day. Vitamin D intake
will also be adjusted to 800 IU per day.

Biochemical Analysis. Routine serum comprehensive metabolic profile (CMP) will be measured
by standard autoanalyzer technique, serum 25-hydoxyvitamin D by double-antibody
radioimmunoassay (Incstar Corp., Stillwater, Minnesota), and FSH by chemiluminescent assay
(Bayer, NY) in those cases when the last menstrual period occurred less than 12 months
before the signing of the informed consent. These measurements will be done through the
Barnes-Jewish Hospital Laboratory. Serum estradiol will be measured by radioimmunoassay
(Diagnostic Systems Laboratories Inc.), and sex hormone binding globulin by double antibody
radioimmunoassay (Diagnostic Laboratories Inc). Serum estradiol measurements will be done at
the Core Laboratory of the General Clinical Research Center (GCRC) of Washington University
School of Medicine while the sex hormone binding globulin assay will be done at Dr.
Villareal's laboratory. The inter- and intra-assay coefficients of variation for these
assays are all < 10%. In order to improve accuracy of the analysis the biochemical tests
(except for 25-OH Vitamin D) will be performed at the end of the study and will be run by
the same operator and plate (run in batch).

Bone Mineral Density (BMD) BMD of the total body, lumbar spine and the proximal left femur
will be measured by dual energy X-ray absorptiometry (DEXA) using Hologic QDR 4500 (Hologic
Inc., Waltham Ma, U.S.A.). Measurements will be done at month 0, 6 and 12 (see Visits and
Testing Schedule). To guarantee the necessary flexibility for the patients who can not be
seen at the exact date for the follow up a window of 1 month earlier and 2 months of late
will be considered. BMD of the lumbar spine will be measured from L1 to L4. The non-dominant
hip will be used for studies on the proximal femur. The main regions of interest on the
proximal femur will be the total femur, femoral neck and femoral trochanter. The coefficient
of variability for these repeated measurements in our institution is1.09% for the lumbar
spine and 1.2% for the total femur in our center.

Bone Turnover Markers will be obtained at month 0, 6 and 12. We will follow a marker of bone
resorption (urine cross-linked N- telopeptide of type I collagen [urine NTx] or, when a
proper 24 hour urine collection is not done, serum cross-linked telopeptide of type I
collagen [serum CTx) will be analyzed) and one marker of bone formation (serum bone specific
alkaline phosphatase {BSAP). Due to the diurnal pattern of bone markers, all serum specimens
will be collected in the fasting state while urinary NTx will measured on 24-hour urine
collection. These markers will be measured by ELISA at the Core laboratory for Clinical
Studies at Washington University as follows: Urine NTX (Osteomark, Ostex International,
Inc., Seattle, WA), serum CTX (serum crosslaps, Osteometer, Hawthorne, CA) and serum BSAP
(Alkphase-B, Metra Biosystems, Inc., Mountain View, CA). Intra-assay coefficients of
variations are in the range of 4-7% (32;33). In order to improve accuracy of the analysis
biochemical tests will be performed at the end of the study and will be run by the same
operator and plate (conducted in batch).

Vertebral Fracture Evaluation. We will evaluate vertebral fractures using the vertebral
fracture assessment (VFA) method performed by the same DEXA machine used to measure the bone
mineral density. Those patients whose VFA is suspicious for a fracture will undergo a
lateral spine X-ray. Vertebral fractures will be identified by visual inspection of lateral
spine radiographs of the thoracic and lumbar vertebrae, from T4 to L5. Interpretation of
radiographs will be done by the PI, who is an endocrinologist specializing in bone disease
and the Medical Director of the Bone Health Program in the Division of Bone and Mineral
Diseases at Washington University School of Medicine. If the PI will have some doubts, she
can easily ask the opinion from the bone radiologists at the Department of Radiology at
Washington University School of Medicine.

Menopausal symptom evaluation. Symptoms related to menopause will be evaluated using a
modified Leuven menopause questionnaire. This is a straightforward menopause-specific
questionnaire which covers vasomotor, gynecologic, psychological, cognitive and somatic
symptoms which was developed and validated in a previous study (Anti-Cancer Drugs 2004
15:753-760). The Leuven questionnaire includes definitions of each menopausal symptoms and
rated as: 1 (no symptom), 2 (mild) up to 5 (intolerable). This questionnaire will be
administered at month 0, 3, 6 and 12. This questionnaire will be administered during the
patient visits at month 0, 6 and 12, and it will be sent by mail approximately two weeks
prior to month 3. A phone call will also be made to the patients to prompt them to complete
the questionnaire approximately at the time they should receive it by mail (month 3). Those
who will fail to send the questionnaire within 1 week after the indicated dates will be
contacted by phone. A log will be kept to document phone call attempts and dates of mailings
to document persistence with compliance. A symptom score will be estimated for each
participant. The number of symptomatic subject in each genotype will also be estimated.

Psychometric testing: In the current project we will assess the cognitive impairment using
the mini mental state examination test (normal range of > 72 for women < 8 years of formal
education and > 76 for those who have > 9 years of formal education) and attention and
executive functions using Trail making tests A and B (the score is the number of seconds
spent in connecting 25 numbered circles in sequential order [trail A] or the seconds spent
in connecting numbered circles (1-13) alternately to letters of the alphabet (A-L) in
sequential order [Trail B]. In both tests a maximum of 180 seconds is allowed). A fluency
test will be performed asking to the subjects to name as many animals as they can during 1
minute. Additionally the Geriatric Depression Scale (GDS) will be performed in order to
exclude and to assess eventual depression. This test implies 30 questions, each one
corresponding to a score of 0 or 1. According to the final score a patient is classified
normal with a score ranging from 0 to 9, mildly depressed if from 10 to 20, and severely
depressed if the score is higher than 20. These tests will be administered during the
patient visits at month 0, 6 and 12.

Rationale. To date, 4 polymorphisms of the aromatase gene have been associated with
significant differences in BMD and fracture risk (16-18), age-related bone loss (18;29) a
few of which have also been found to be associated with differences in breast cancer risk
(30). Functional in-vitro analysis suggests an altered enzymatic activity in certain
variants for this gene perhaps accounting for differences in the risk of breast cancer and
osteoporosis. Variants with increased aromatase activity (18), as in those with higher
number of TTTA repeats, have higher BMD but higher risk for breast cancer (30) possibly a
consequence of higher estrogen levels. Because these variants have increased enzyme
activity, it is possible that this select group of women will have increased sensitivity to
aromatase inhibition and would experienced more bone loss than those with lower aromatase
activity. On the contrary, it is also possible that women with these variant alleles will be
relatively resistant to aromatase inhibition compared to variants with lower enzymatic
activity and will continue to synthesize estrogen from precursors and, therefore, will not
develop rapid bone loss. This proposal will address this important question by comparing the
variants of the different polymorphisms described below.