Cohort Profile: The Colon Cancer Family Registry Cohort (CCFRC)

Colorectal cancer has long been one of the most frequently diagnosed cancers in the world, with an estimated 1.4 million new cases diagnosed each year (9.8% of worldwide cancer diagnoses) and the cause of 694 000 deaths (8.5% of all worldwide cancer deaths) in 2012.1 In 1996, as a commitment to reduce morbidity and mortality from this disease, the National Cancer Institute (NCI) of the U.S. National Institutes of Health invited investigators to apply for funding to establish a ‘Cooperative Family Registry for Colorectal Cancer Studies’ (RFA: CA-96-011). The main NIH stated aims were: to collect pedigree information, epidemiological data and related biological specimens from participants with and without colorectal cancer and with and without a family history of the disease, as a resource for interdisciplinary studies on the aetiology of colorectal cancer; and to identify a population at high risk of colorectal cancer that could benefit from preventive strategies. This cohort profile provides an update of the Colon Cancer Family Registry, described in detail in Newcomb et al.2 The basic premise of this initiative is that family-based designs across the spectrum of risk, in which cases, controls and their relatives are all recruited into a single research infrastructure, would enable the study of genetic aetiology, gene penetrance, gene-gene interaction and interaction with lifestyle factors. Thus, in 1997, the Colon Cancer Family Registry was established with funding support from the NCI. For Phase I (1998–2002), 5 years of funding was awarded to six Colon Cancer Family Registry sites: Cancer Care Ontario (Toronto, ON, Canada); Fred Hutchinson Cancer Research Center (Seattle, WA, USA); Mayo Clinic (Rochester, MN, USA); University of Hawaii (Honolulu, Hawaii, USA); University of Southern California Consortium (comprising the Universities of Southern California, Minnesota, North Carolina, Colorado and Arizona, Dartmouth University and the Cleveland Clinic Foundation, USA); University of Queensland (Brisbane, QLD, Australia). The Colon Cancer Family Registry received funding renewals for Phase II (2003–07) and Phase III (2008–12) with the addition of: University of Melbourne (Melbourne, VIC, Australia) substituting for the University of Queensland; Memorial University (Newfoundland, Canada) as a collaborative site within the Cancer Care Ontario site. In 2004–11, the ethnic/racial minority component of the Colon Cancer Family Registry was expanded through the recruitment of additional African American and Japanese American families with a separate NCI grant that included the University of Hawaii, the University of Southern California, the University of North Carolina, the Fred Hutchinson Cancer Research Center and the Cancer Prevention Institute of California. Phase IV (2013–18) of the Colon Cancer Family Registry was funded by the NCI as a Cancer Epidemiology Cohort, and consequently renamed the Colon Cancer Family Registry Cohort (CCFRC). This phase saw the addition of: Stanford University (CA, USA) as the administering site for the Colon Cancer Family Registry, and Mayo Clinic (Scottsdale, AZ, USA) as the administering site for the Mayo Clinic CCFRC site. Recruitment sampling schemes and inclusion and exclusion criteria varied by CCFRC site and funding phase. Details of the recruitment methods at each CCFRC site have been published previously.2 Briefly, recruitment protocols fall broadly into two main categories: population-based and clinic-based. Population-based probands were either people with a diagnosis of recently diagnosed colorectal cancer (case-probands) identified from cancer registries, or people without a prior diagnosis of colorectal cancer (control-probands) randomly sampled from the general population living in the relevant recruitment area using Medicare and Driver’s License files, telephone subscribers lists or electoral rolls, who were frequency-matched for age to the case-probands. Clinic-based probands were people with or without colorectal cancer who were attendees at a family cancer clinic or genetics clinic. Cases with known familial adenomatous polyposis were excluded. Once recruited, probands were asked for permission to contact their relatives for recruitment. The CCFRC recruited 42 489 participants—from 15 049 families—who completed a baseline questionnaire between 1998 and 2012 (Table 1). Recruitment numbers within clinic-based families was, on average, twice that for population-based families (5.3 vs. 2.6 relatives per family, respectively). The majority of participants self-reported as Caucasian/White followed by Asian ethnicities and African American/Black (Table 2). Number of families and participants of the Colon Cancer Family Registry Cohort by sex and colorectal cancer (CRC) status at baseline recruitment Probands recruited from a population-based source. Only the University of Melbourne recruited relatives of control-probands. Probands recruited from a family cancer clinic source. Number of families and participants of the Colon Cancer Family Registry Cohort by sex and colorectal cancer (CRC) status at baseline recruitment Probands recruited from a population-based source. Only the University of Melbourne recruited relatives of control-probands. Probands recruited from a family cancer clinic source. Distribution of participants by race Self-reported by questionnaire. Distribution of participants by race Self-reported by questionnaire. We have used both active and passive follow-up methods to update the cohort, where active follow-up includes direct contact with participants and passive follow-up includes indirect methods; details as follows. Approximately every 4–5 years after completing their baseline questionnaire, all participants of population-based case-families (but not control-families) and clinic-based families were asked, either by telephone interview or by self-completed questionnaire (mailed or online), for updates on their personal and family history of cancer as well as history of surgery, cancer screening and some risk factors. Of the 37 436 participants who completed baseline questionnaires and were approached for follow-up, 27 918 completed the first follow-up questionnaire [response proportion (or response ‘rate’ of those alive) 83%], 3549 died before being approached for the first follow-up and 5969 could not be contacted or refused follow-up. Of the 27 918 participants who had completed the first follow-up, 18 958 completed their second follow-up questionnaire (response rate 87%), 1934 had died, 2824 were either uncontactable or refused, and 4202 are still in process. Of the 18 958 participants who had completed the second follow-up, 8371 had completed their third follow-up questionnaire (response rate 95%), 1536 had died, 368 were either uncontactable or refused and 8683 are still in process (Figure 1). Progress of follow-up of participants of the Colon Cancer Family Registry Cohort (as of June 2017). Participation is defined as the percentage of those who were alive at contact attempt who completed the questionnaire. The total number of person-years of follow-up by participants who completed a follow-up questionnaire is 276 762 person-years. As this is a family study, the vital status and cancer diagnoses of participants were also ascertained, even if they did not participate in the follow-up themselves, based on interviews of any relatives who were also participants. Including the reports by relatives, the total number of person-years of follow-up of all participants who completed a baseline questionnaire was 338 970 person-years, an average of 9.1 years per participant. These comprise approximately: 49 000 person-years for those recruited within 2 years after colorectal cancer diagnosis (thus relevant for studies of colorectal cancer survival and risk of metachronous cancer); 39 000 person-years for relatives with CRC and probands recruited more than 2 years after colorectal cancer diagnosis (thus relevant for studies of survivors of colorectal cancer); and 251 000 person-years for those with no previous diagnosis of colorectal cancer (thus relevant for studies of colorectal cancer risk and aetiology)—Table 3. Numbers of incident colorectal cancer diagnosis and deaths occurring in study participants (except controls) of the Colon Cancer Family Registry since baseline recruitment by different cohort types, as of June 2017 Cohort useful for studies of colorectal cancer survival and risk of metachronous cancer. Cohort useful for studies of survivors of colorectal cancer. Cohort useful for studies of colorectal cancer risk and aetiology. Based on follow-up interview or report from participating relative. Numbers of incident colorectal cancer diagnosis and deaths occurring in study participants (except controls) of the Colon Cancer Family Registry since baseline recruitment by different cohort types, as of June 2017 Cohort useful for studies of colorectal cancer survival and risk of metachronous cancer. Cohort useful for studies of survivors of colorectal cancer. Cohort useful for studies of colorectal cancer risk and aetiology. Based on follow-up interview or report from participating relative. One or more of the following passive follow-up activities have been conducted at each site of the CCFRC: data linkage with local and national death files, population-based cancer registries and electoral rolls; annual newsletters; reviews by genetic counsellors; and other mailings to participants. Passive follow-up was regularly conducted on all participants—at intervals that varied by site, type of follow-up activity and cost—to obtain information on new cancers, vital status and cause of death, and to update contact information. During active and passive follow-up, all new reports of colorectal polyps and all cancers were recorded. Attempts were made to verify cancers using medical records, cancer registry data and confirmatory reports from relatives. To date, 824 (2.2%) participants have been diagnosed with an incident colorectal cancer since baseline; of those, 170 were diagnosed before the age of 50 years (Table 3); and 3582 (9.5%) participants have been diagnosed with an incident non-colorectal cancer since baseline. The total 4164 incident non-colorectal cancers were as follows: 772 skin, 568 breast, 599 prostate, 97 gastric, 52 small bowel, 103 hepatobiliary, 102 pancreas, 147 renal, 40 ureteric, 150 urinary bladder, 76 brain, 355 lung, 27 bone, 219 blood, 163 endometrial, 73 ovarian and 35 cervical cancers, and 586 in other organs. A total of 7019 (19%) participants (including those with and without colorectal cancer at baseline) are known to have died since baseline (Figure 1). At the baseline recruitment, CCFRC participants were asked to complete a detailed family history of cancer, a risk factor questionnaire, permission to access medical records pertaining to any colorectal cancer diagnoses, permission to access colorectal cancer tumours and, depending on the degree of relationship to the proband, to provide a blood (or buccal wash) sample—Table 4. Resources available from the Colon Cancer Family Registry Cohort, as of June 2017 Proband had a history of colorectal cancer (CRC) at baseline interview. Affected or unaffected with colorectal cancer at baseline interview. Spouse of proband, had no history of colorectal cancer at baseline interview. Proband had no history of colorectal cancer at baseline interview. Proband was recruited from a family cancer clinic. % of total items obtained. For example 20.6% of all baseline questionnaires completed by probands of population-based case families. Resources available from the Colon Cancer Family Registry Cohort, as of June 2017 Proband had a history of colorectal cancer (CRC) at baseline interview. Affected or unaffected with colorectal cancer at baseline interview. Spouse of proband, had no history of colorectal cancer at baseline interview. Proband had no history of colorectal cancer at baseline interview. Proband was recruited from a family cancer clinic. % of total items obtained. For example 20.6% of all baseline questionnaires completed by probands of population-based case families. All participants (probands and their participating relatives) were asked to complete the same detailed baseline risk factor survey using standardized questionnaires via personal or telephone interviews or mailed questionnaires. Items included demography, lifestyle factors, screening, medication and family history.2 Four CCFRC sites also asked participants to complete a self-administered food frequency dietary questionnaire. Three CCFRC sites (University of Hawaii, Cancer Care Ontario and University of Southern California consortium) used the questionnaire developed by the Multiethnic Cohort study in Hawaii and California.3 The University of Melbourne used the questionnaire developed by the Melbourne Collaborative Cohort Study.4 At each follow-up, participants were asked for the following events that might have occurred since the previous contact: cancer diagnoses; bowel and gynaecological surgery; screening for colorectal cancer; polyps; and cancer diagnoses and deaths in relatives. Some CCFRC sites opted to include additional questions pertaining to colorectal cancer risk factors. All baseline and follow-up questionnaires used by each CCFRC site can be accessed at: [http://www.coloncfr.org/questionnaires]. One or more participants from each family was asked to provide their family history of cancer by answering a standard set of questions for each of their relatives (irrespective of cancer history) including: sex and date of birth; cancer sites (except non-melanoma skin cancer), and ages or dates at diagnoses (for those with a cancer history); vital status and, if deceased, date of death. All CCFRC sites recorded detailed family history information for each first- and second-degree relative, and some sites expanded to third-degree relatives, depending on site-specific protocols (detail in Newcomb et al.2). Attempts were made to verify the anatomical site, extent of disease, age at diagnosis and pathology of tumours. Sources of verification used included pathology reports, medical and surgical records, cancer registry information and death certificates. Participants were asked to provide a blood or mouthwash sample. Of those who agreed, 93% provided a blood sample (Table 4).5 DNA was extracted from blood and mouthwash samples under CCFRC quality-control protocols, to maximize target DNA concentration and fragment size. To provide an unlimited supply of DNA and RNA for probands and selected relatives, lymphoblastoid cell lines of case-probands were immortalized using Epstein-Barr virus.6 Paraffin-embedded colorectal cancer tumours, as well as diagnostic pathology reports, were obtained from treating facilities with the consent of the participant or the next-of-kin if the participant was deceased. In addition, some sites also obtained polyps and non-colorectal tumours, especially cancers commonly identified as part of Lynch syndrome. Multiple sections were cut from each tumour and normal-tissue block, stained with haematoxylin and eosin (H&E) and reviewed by pathologists. For each colorectal cancer, a pathology review was completed (either by examination of the H&E slides or extraction of relevant data from available pathology reports) to obtain the following standardized set of tumour features: grade, histological type, stage (depth of infiltration in large bowel wall and spread to regional lymph nodes), lymphovascular invasion and perineural invasion. Sections were stored for future research at each CCFRC site. Two sites (Ontario and University of Southern California consortium) have made tumour microarrays (TMAs) from colorectal cancers (n 1278). CCFRC has created a digitized library of pathology slides (electronic representations of traditional glass slides). A total of 4510 H&E stained slides of histological sections of colorectal tumours from the probands were scanned using either the NanoZoomer Digital Pathology scanner (Hamamatsu Corp.) or the Aperio ScanScope digital slide scanner. Each image is stored as a series of 752 x 480 pixel jpeg image tiles that are reconstructed with relevant software. Typical size of these images is between 200mb and 1.5gb per slide. All images were archived on five image servers: one for short-term storage and four for long-term storage. Clinical treatment and outcome records were requested from 3830 case-probands and for 111 relatives with an incident colorectal cancer diagnosed since baseline, and have been abstracted into standardized items for analysis. Probands’ colorectal cancers were characterized for DNA mismatch repair (MMR) deficiency by polymerase chain reaction (PCR)-based microsatellite instability (MSI) tests and/or by immunohistochemistry for the four DNA Colorectal cancer tumour DNA was for the and in and of were also for of the gene used to that deficiency is more to have been by a in than by a in of the was also by across five gene and of participants (probands and relatives) at the Colon Cancer Family Registry Cohort, as of June 2017 high probands relatives probands relatives high probands relatives probands relatives were as if more than of five percentage of or of participants (probands and relatives) at the Colon Cancer Family Registry Cohort, as of June 2017 high probands relatives probands relatives high probands relatives probands relatives were as if more than of five percentage of or for in and was for all population-based probands who had a colorectal tumour an or a of of one or more of the by and for the colorectal cancer case participant from each clinic-based family, of or All case-probands were for identified and (detail in et blood samples from the relatives of probands with a were for the gene or identified in the Of the CCFRC were identified as a in one of the in in in in and in and were identified as either a (n or (n in baseline, these have a total of 18 person-years and person-years. In addition, was conducted of known or colorectal cancer (including the for cases cases with familial colorectal cancer type 50 years at colorectal cancer or Lynch syndrome. CCFRC has single data for participants cases and controls) by (including all to the We known of and The CCFRC resource has been used for more than a that the of this cohort to genetic and risk for colorectal cancer. Lynch by in DNA and or has been a research of the These studies have included of the of Lynch and risk of cancer of extent of on the risk of metachronous of Lynch in colorectal cancer and in the general pathology of Lynch and of genetic and of for Lynch A example is the cohort of gene to cancer of Lynch A total of gene with an average age of 40 years and who had no cancer diagnosis were followed for years study of Lynch in which participants completed a risk factor questionnaire. The and of cancer were with the risk for colorectal cancer risk population cancer risk population and ovarian cancer risk population A of this study was that these also to have an risk of cancer. A total of of their relatives who were not were also for cancer were to be at the same risk as the general population and can be as at average risk, a family history of cancer. The CCFRC also to the for new genetic risk factors, to large sample size and of family-based This for the of DNA samples from case-probands as well as their relatives, and the verification of reports of cancer by relatives than on This research includes the of and of of these studies from the research within the CCFRC on of familial colorectal cancer, the to the of colorectal cancer type the of colorectal cancer cases who the for which has an of familial colorectal The of the CCFRC is with familial and Participants have been for familial risk, and the CCFRC from the cancer research cohort in that not This a and research that genetic and and research relevant to people at familial Participants can be on their familial risk profile based on their family history and risk factor which the of risk to be for of genetic for studies of The of data studies of the of genetic and factors, and the of by genetic The CCFRC can and has used for a of gene research linkage studies and and a large proportion of CCFRC participants were diagnosed with colorectal cancer before recruitment and have risk factor data as well as blood studies of can be The CCFRC also and research for a family studies can be of the of familial as well as the additional of that to be to within families (for to that information is not to other family We have that these can be through study protocols and We that the of a family cohort and that more this research on risk across the risk the CCFRC has under the that is a resource for research on the aetiology, risk and of colorectal cancer for all those not with To this CCFRC collaborative to access and both data medical records, family history and blood, and tumour Of the total to CCFRC have from The CCFRC provides and and access to this investigators have established funded For information on to and access data for the cohort data described in a The Colon Cancer Family Registry Cohort was established for the of research on the genetic and aetiology of colorectal cancer. The 42 489 study participants from 15 049 families were recruited between 1998 and 2012 in the and recently diagnosed colorectal cancer cases from population-based cancer controls from population-based from family cancer with a family history of colorectal cancer; and their relatives. 4–5 years after baseline, all population-based case-families and clinic-based families were followed and The total follow-up of 37 436 participants 000 person-years follow-up 9.1 baseline, 824 (2.2%) participants were diagnosed with a colorectal cancer and 3582 (9.5%) were diagnosed with a non-colorectal cancer. At baseline, all participants completed the same risk factor questionnaire for a detailed personal and family history of cancer, and a of risk factors. At each follow-up, participants were asked for updates on their personal and family history of cancer, screening, surgery, death and some risk factors. samples and tumour specimens have been and used for genetic and Lynch syndrome. CCFRC are available for collaborative research The all study participants of the CCFRC and for their to this The also the to the CCFRC since in made by the following and This was by grant from the National Cancer Institute and through with the following CCFRC sites: Colorectal Cancer Family Registry and Mayo Clinic Family Registry for Colon Cancer Ontario Colorectal Cancer Registry Colorectal Cancer Family Registry University of Hawaii Colorectal Cancer Family Registry and Consortium Colorectal Cancer Family Registry The minority recruitment was by grant The studies were by and The and was by support for case was provided from the Epidemiology and of the National Cancer Institute to the Fred Hutchinson Cancer Research Center and and the of Health and and and the California of Health awarded to the University of Southern California and awarded to the Cancer Prevention Institute of the following cancer AZ, MN, and and by the Cancer Registry, and the Ontario Cancer Registry, is an is an Research is a Research is a University of Melbourne Research at Melbourne Research and The of this not the or of the National Cancer Institute or any of the sites in the of or by the or the had for the of the study, the of the the and of the the to the for and the of the of The have no of to with to this

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.