Overall Study Design
The Pathways Study is a prospective cohort study, with recruitment of women from the Kaiser Permanente Northern California (KPNC) patient population as soon after diagnosis of invasive breast cancer as possible. Women are typically enrolled within two months post-diagnosis during an in-person baseline interview. Blood specimens and saliva samples are also collected at baseline to enable investigation of molecular and other factors. Among all participants, follow-up questionnaires are mailed at 6 months and 24 months after baseline to assess updates in lifestyle factors while telephone interviews occur every 12 months after baseline to identify breast cancer outcomes. The study was approved by the institutional review board of all collaborating institutions. Written informed consent is obtained from all participating subjects.
Identification of Women Eligible for Recruitment
Since January 9, 2006, women diagnosed with invasive breast cancer are being recruited from the KPNC patient population into the Pathways Study. To be eligible, women have to be at least 21 years of age at diagnosis and a current KP member, have a recent (usually within two months) diagnosis of invasive breast cancer (all stages), have no previous history of malignant cancer, speak English, Spanish, Cantonese, or Mandarin, and live within a 65-mile radius of a field interviewer.
The study is designed for rapid case ascertainment in order to enroll women prior to start of adjuvant therapy. The KP Laboratory Utilization and Reporting System (LURS) database, which exists in a mainframe environment, is automatically accessed on a daily basis for pathology reports and is the method by which pathologists report their findings to clinicians. This automated process involves scanning all records for any new patients with Systematized Nomenclature of Human and Veterinary Medicine (SNOMED) Topography and Morphology Codes indicating malignant breast cancer. For those records found to be eligible, demographic, pathology, and physician-related data are collected and uploaded to a Structured Query Language (SQL) server to begin the tracking process. This procedure enables the identification of women diagnosed with breast cancer within one day of notification by the pathologist.
Breast cancer diagnosis and patient notification of the incident diagnosis is verified in the KP Clinical Information Presentation System (CIPS). In addition, a passive consent is obtained from the patient's physician of record by an email notification stating our intention to contact his or her patient for study recruitment. If the physician has not objected within one week after notification of intent to contact the patient, recruitment procedures are initiated.
Recruitment of Participants
Recruitment, enrollment, and baseline data collection are conducted by 14 field interviewers (two Spanish-speaking and one Chinese-speaking) based throughout the major geographic regions served by KPNC, including the East Bay, Marin, San Francisco, San Jose, Fremont, Sacramento, and Stockton. Women who are diagnosed with breast cancer are assigned to interviewers according to geographic location. Interviewers are not expected to travel further than 65 miles from their home. Thus, some women are not eligible for recruitment because of their distance from an interviewer.
Prior to recruitment, and after confirmation of patient notification of her breast cancer diagnosis and passive physician consent, an invitation letter, reply postal card, and brochure describing the study is sent to the potential participant. Concurrently, potential participants are assigned to interviewers based on geographic location as noted above. After two weeks, or if we are notified of interest by the potential participant, the interviewer will contact her by telephone. If the woman agrees to participate, an in-person interview is arranged at a location of convenience for the woman, usually in her home. Interviewers are expected to make no more than ten telephone calls to recruit each potential participant.
All relevant recruitment information is documented through a secure study website, including assignment of participants to interviewers, and information on contacts between the interviewer and the potential participants. Patients are excluded who are contacted more than two months after diagnosis and if they have begun chemotherapy, radiation treatment, or hormone therapy. Interviewers are responsible for recording special needs or disabilities of the participant which might affect future study retention and for completing an evaluation of the overall quality of the interview.
For quality control purposes, yearly re-certifications are conducted among all interviewers in which the field staff coordinator reviews the baseline interview protocols and procedures with the interviewer at the study coordinating center and then observes a live baseline interview in the field. Appropriate feedback and corrective action is taken as necessary to ensure standardization of the interview process among all field staff. Furthermore, quality assurance procedures, including review of baseline files by two separate staff members, are conducted at the study coordinating center to provide an additional check of data collection consistency. Participants are called back for clarification if necessary.
Baseline and Follow-up Data Collection
During the baseline interview, which averages three hours in length, interviewers first consent the study participants, and then administer detailed questionnaires on diet, exercise, use of CAM, and psychosocial and quality-of-life measures. Anthropometry measures are taken, including arm, waist, and hip measurements. A saliva sample is obtained, and if the interviewer is a certified phlebotomist, a blood sample is also collected. Otherwise, arrangements are made to draw blood at a KPNC facility for a courtesy research blood draw or at home by an outside vendor (Examination Management Services, Inc. [EMSI], Irving, TX). A $20 gift certificate is mailed to the participant upon completion of the interview. Completed questionnaires are sent every two weeks to the study coordinating center. Blood samples are shipped to Roswell Park Cancer Institute on the same day as collection.
The baseline data collection consists of interviewer- and self-administered questionnaires. Interviewer-administered questionnaires collect information on demographics, family health history, prenatal exposures, pregnancy health history, developmental history, menstrual history, child health history, history of breast care screening procedures, smoking history, hormone use, medication history, use of vitamins and minerals, and consumption of organic foods. Self-administered questionnaires collect information on diet (food frequency questionnaire [FFQ] and three-day food record), physical activity, CAM use, lymphedema, and psychosocial and quality-of-life measures.
A packet of follow-up materials is mailed to the participant's home at 6 and 24 months after baseline. Follow-up materials include a primary follow-up questionnaire, FFQ, three-day food record, and a form and tape measure for anthropometric measurements. The primary follow-up questionnaire is similar to the baseline interview in asking for updates on menstrual history, smoking, medication use, vitamin use, organic food consumption, CAM use, physical activity, lymphedema, and psychosocial measures. Interviewer assistance is offered to participants to complete the follow-up packet. shows the timeline of data collection in the Pathways Study.
Timeline of Data Collection in the Pathways Study
Dietary history is collected using two assessment methods: a 139-item modified version of the Block 2005 FFQ (NutritionQuest, Berkeley, CA) and an undocumented three-day food record (Nutritional Assessment Shared Resource [NASR], Fred Hutchinson Cancer Research Center, Seattle, WA).
For the FFQ, food items were selected by identifying the top population contributors of each nutrient among Whites, African Americans, and Hispanics in the National Health and Nutrition Examination Survey (1999−2002) [13
]. The 139 food items and additional questions were selected to be representative of a wide range of dietary factors, as well as to capture foods that are popular in Hispanic and Asian populations. Questionnaires are sent to NutritionQuest for scanning using a nutrient database developed primarily from the USDA Food and Nutrient Database for Dietary Studies [15
A three-day food record is left with participants to complete over the next two weeks after the baseline interview. They are instructed by the interviewer to record three non-consecutive days (two weekdays and one weekend day) of all the foods consumed being as descriptive as possible for brand names and preparation methods as well as including nutrition facts labels and recipes. The booklet also contains 12 questions on food-use patterns such as “When you eat bread or rolls, how often do you add butter or margarine?” to collect information typically obtained during review of completed food records. The participants are given a postage-paid return envelope to mail the food record to the study coordinating center. Completed food records are sent to NASR, and trained staff enter information into the Nutrition Data System for Research (NDS-R) software (version 4.04, Food and Nutrient Database 32) [16
], using a set of established rules to standardize entry of foods with incomplete information.
Body measurements are taken by the field interviewer during the baseline interview and are also requested with the follow-up mailed questionnaire packets. For the mailed data collection, women are instructed to have a family member or friend assist with the measurements; if necessary, an interviewer is available to take the measurements. Waist, hips, and arm lengths and circumferences are measured using a specialized tape measure called the Figure Finder Tape Measure (Novel Health Products, Rockton, IL), which is designed to create a uniform amount of tension (four ounces). The waist is measured one inch above the navel in a horizontal plane while the hips are measured around the maximum extension of the buttocks in a horizontal plane. The measurements of both arms consist of serial circumference measurements at the wrist, elbow, and upper arm as well as the length between the three bounding circumferences. All measurements are taken twice, with a third measurement obtained if a one centimeter margin of error is exceeded.
Complementary and Alternative Medicine (CAM)
The use of CAM is assessed with a modified two-part self-administered questionnaire based on previous studies [18
]. Part 1 is a checklist asking for ever use of five main modalities: herbal and botanical supplements such as black cohosh, ginseng, and mistletoe (90 items); other natural products such as co-enzyme Q10, DHEA, and melatonin (26 items); special diets such as low-fat, macrobiotic, or vegan (12 items); mind-body healing such as hypnosis, support groups, or yoga (17 items); and body-based, energy-based, and other treatments such as acupuncture, naturopathy, or reiki (23 items). Space is available for women to report use of CAM therapies that are not specified on the questionnaire. Part 2 asks for more detailed information for each CAM supplement, approach, or treatment reported in Part 1, including total length of use, use in the five years before diagnosis and since diagnosis, and frequency, duration, and reason of pre-diagnosis use. Whether the CAM therapy was used during conventional breast cancer therapy is also determined in follow-up questionnaires.
Physical activity is assessed in Pathways with a questionnaire developed for another study of breast cancer survivors within KPNC (Life After Cancer Epidemiology Study [LACE] [22
]), which was modeled loosely after the Arizona Activity Frequency Questionnaire (AAFQ) [23
]. The questionnaire is divided into four main sections: job or work-related activities, activities not related to paid or volunteer work, recreational activities, and transportation. The activities not related to paid or volunteer work are further sub-divided into household chores (six items), caregiving (five items), and home maintenance and repairs (seven items). Recreational activities are sub-divided into sports, exercise, and dance (23 items) and sedentary recreational activities, such as reading or socializing (six items). Four items are included under transportation. For each activity that respondents did at least once a month during the past six months, they report the frequency, duration, and intensity. A validation study of the LACE questionnaire is currently underway using accelerometers and physical activity diaries as direct validation criteria and cardiorespiratory fitness and body composition as indirect criteria.
Psychosocial and Quality-of-Life Measures
A number of instruments are used in Pathways to measure various psychosocial issues important to women with breast cancer. Seven quality-of-life domains for patients with breast cancer are assessed using the Functional Assessment of Cancer Therapy questionnaire (FACT-B) [24
]. The Functional Assessment of Cancer Therapy-Taxane (FACT-Taxane) scale is used to measure neurotoxicity symptoms associated with some taxane agents [25
]. In addition, the Center for Epidemiological Studies Depression Scale (CES-D) [26
] as well as a 12-question scale developed by Scheier and Carver [27
] measure symptoms of depression during the past seven days and dispositional optimism, respectively. Questions were modified from the 1979 National Survey of Black Americans to assess the experience of stress during the past seven days in various areas including health, money, job, family, and crime [28
]. In addition, questions on racial discrimination [29
], health literacy [30
], aspects of doctor-patient communication [31
], preference for doctor-patient decision-making [32
], global trust in one's doctor [33
], and social support [34
] are asked. Finally, questions from the KP Breast Cancer Patient Survey are used to measure difficulty taking time off from work to attend medical appointments and keeping medical appointments due to distance or transportation.
Yearly health status update
Cohort members are contacted every year subsequent to their baseline interview to complete a health status update questionnaire asking about any new breast cancer events since the previous update, including recurrence and new breast cancer diagnosis, new diagnosis of another cancer, and hospitalizations or other conditions believed to be health consequences or late effects associated with breast cancer treatment. If an event is reported, KP electronic medical records are searched for confirmation of the event (see below). The first yearly health status update began in January 2007.
Electronic medical records
Medical record abstraction from KPNC clinical and administrative electronic databases will be performed on all cohort members. Data of interest include breast tumor characteristics (including but not limited to hormone receptor status, HER-2/Neu expression, tumor size, number of positive nodes, and American Joint Committee on Cancer [AJCC] stage), chemotherapy, radiation therapy, and hormonal therapy regimen and associated adverse events, general medications, laboratory results, co-morbidities, recurrences, and new cancers. Data for pharmacy, laboratory, radiology, hospitalizations, emergency room visits, and other outpatient clinical visits can be obtained from the following databases: Outpatient Summary Clinical Record (OSCR), Pharmacy Information System (PIMS), Admission Discharge and Transfer (ADT), LURS, Case Management for Medical Oncology with Laboratory and Outcome Tracking (CAMMOLOT), Medical Record Management System – radiology component (MRMS), Transcription Results Reporting System - radiology component (TRRS), and Health Connect (KP's automated medical record system). Services provided by non-KP vendors can be collected from authorized outside medical services (AOMS) and claims, adjudication, and tracking system (CATS).
In addition, the KPNC Cancer Registry (KPNCCR) [35
] is accessed to obtain diagnostic tumor characteristic data. Data are collected, coded, and added to the KPNCCR approximately four months post-diagnosis to allow for the completion of treatment. Over 1,600 new cases are processed monthly. The registry covers 19 KPNC medical centers and includes over 322,000 cases of in situ
and invasive cancers diagnosed from 1947 to the present. Data in the KPNCCR are also reported to the State of California and Northern California (Bay Area) SEER Registries. Finally, computerized mortality files will be regularly searched for any cohort member whom we fail to contact.
Biospecimen Collection, Processing, and Storage
Blood and saliva
Blood is collected from the participant around the time of the baseline interview. Non-fasting blood is drawn into three 10 ml polystyrene tubes (lavender, green, and red tops) to maximize diversity in blood components available for later analyses (see below). When possible, the blood is drawn prior to initiation of adjuvant therapy to ensure integrity of biomarkers that may be assessed. The blood is shipped cold overnight to Roswell Park Cancer Institute (RPCI) for processing and storage. Protocols for shipment and handling were developed to standardize procedures to prevent variability that could affect sample degradation. In addition, 2 ml of saliva is collected using the Oragene™ DNA Self-Collection Kit (DNA Genotek Inc., Ottawa, Ontario, Canada) at the time of the baseline interview as a back-up source of DNA in the event a blood specimen is insufficient or not obtained. The saliva is mixed with preservative immediately after collection, kept at room temperature, and shipped to RPCI within three weeks of collection.
Upon receipt at RPCI, blood samples are processed and divided into multiple aliquots. The lavender top tube is the source of DNA for genotyping; the green top tube is the source of buffy coat, red blood cells (RBCs), and plasma; the red top tube is the source of serum. RBCs can be used for analysis of membrane-bound markers, as well as other factors that are best evaluated in erythrocytes, such as folate and catalase. Serum and plasma will be used for assessment of a number of circulating markers.
Approximately 4 ml of whole blood from the lavender tube is aliquoted into cryovials and stored at −80°C for DNA extraction at a later date. RBCs, buffy coat, plasma, and serum are aliquoted into 0.5 ml straws using the MAPI Cryobiosystem (IMV Technologies, Paris, France), slow frozen to −80°C, and then transferred into long term storage in liquid nitrogen vapor. All aliquots are imprinted with scannable barcodes and user readable characters. Their positions within the freezers, which are locked, temperature monitored, and alarmed, are tracked by a laboratory information management system (LIMS).
The Oragene™ saliva samples for future DNA extraction are received at RPCI and stored at room temperature in a locked storage device. The saliva samples are also labeled with scannable barcodes and user readable characters, and their positions within the storage system tracked by a LIMS.