Risk of Incident Colorectal Cancer and Death After Colonoscopy: A Population-based Study in Utah.

BACKGROUND & AIMS: Colonoscopy is widely recommended for colorectal (CRC) screening in the United States, but evidence of effectiveness is limited. We examined whether exposure to colonoscopy decreases the odds of incident CRC and death from CRC in Utah. METHODS: We performed a case-control study of Utah residents, 54 to 90 years old, who received a CRC diagnosis from 2000 through 2010 (cases). Age- and sex-matched controls with no history of CRC (controls) were selected for each case. We determined receipt of colonoscopy 6 months to 10 years before the reference date for each case and control through administrative claims data. Colonoscopy exposure was compared by using conditional logistic regression. RESULTS: We identified 5128 cases and 20,512 controls; 741 cases (14%) and 5715 controls (28%) received a colonoscopy. Exposure to colonoscopy reduced the odds for a diagnosis of CRC; the odds ratios (ORs) were 0.41 for any CRC (95% confidence interval [CI], 0.38-0.44), 0.58 for proximal colon cancer (95% CI, 0.51-0.65), and 0.29 for distal colon or rectal cancer (95% CI, 0.25-0.33). This finding was consistent among sexes, age groups, and cancer stages. Similarly, in a subgroup analysis, colonoscopy was associated with decreased odds of death from CRC (OR, 0.33; 95% CI, 0.28-0.39) in both the proximal colon (OR, 0.43; 95% CI, 0.34-0.55) and distal colon or rectum (OR, 0.23; 95% CI, 0.18-0.30). CONCLUSIONS: In the population of Utah, colonoscopy is associated with a large reduction in risk of new-onset CRC and death from CRC. This reduction in risk for CRC was greatest for the distal colon and rectum, with a more modest reduction for proximal colon cancer.

Increased risk of colorectal neoplasia among family members of patients with colorectal cancer: a population-based study in Utah.

BACKGROUND & AIMS: Colorectal cancer (CRC) frequently develops in multiple members of the same families, but more data are needed to prepare effective screening guidelines. We quantified the risk of CRC and adenomas in first-degree relatives (FDRs) and second-degree relatives and first cousins of individuals with CRC, and stratified risk based on age at cancer diagnosis. METHODS: We performed a case-control study of Utah residents, 50-80 years old, who underwent colonoscopy from 1995 through 2009. Index cases (exposed to colonoscopy) were colonoscopy patients with a CRC diagnosis. Age- and sex-matched individuals, unexposed to colonoscopy (controls) were selected to form the comparison groups for determining risk in relatives. Colonoscopy results were linked to cancer and pedigree information from the Utah Population Database to investigate familial aggregation of colorectal neoplasia using Cox regression analysis. RESULTS: Of 126,936 patients who underwent a colonoscopy, 3804 were diagnosed with CRC and defined the index cases. FDRs had an increased risk of CRC (hazard rate ratio [HRR], 1.79; 95% confidence interval [CI],1.59-2.03), as did second-degree relatives (HRR, 1.32; 95% CI, 1.19-1.47) and first cousins (HRR, 1.15; 95% CI, 1.07-1.25), compared with relatives of controls. This risk was greater for FDRs when index patients developed CRC at younger than age 60 years (HRR, 2.11; 95% CI, 1.70-2.63), compared with older than age 60 years (HRR, 1.77; 95% CI, 1.58-1.99). The risk of adenomas (HRR, 1.82; 95% CI, 1.66-2.00) and adenomas with villous histology (HRR, 2.43; 95% CI, 1.96-3.01) also were increased in FDRs. Three percent of CRCs in FDRs would have been missed if the current guidelines, which stratify screening recommendations by the age of the proband, were strictly followed. CONCLUSIONS: FDRs, second-degree relatives, and first cousins of patients who undergo colonoscopy and are found to have CRC have a significant increase in the risk of colorectal neoplasia. These data should be considered when establishing CRC screening guidelines for individuals and families.

Epidemiology and familial risk of synchronous and metachronous colorectal cancer: a population-based study in Utah.

BACKGROUND & AIMS: Patients diagnosed with colorectal cancer (CRC) are at risk for synchronous and metachronous lesions at the time of diagnosis or during follow-up evaluation. We performed a population-based study to evaluate the rate, predictors, and familial risk for synchronous and metachronous CRC in Utah. METHODS: All newly diagnosed cases of CRC between 1980 and 2010 were obtained from the Utah Cancer Registry and linked to pedigrees from the Utah Population Database. RESULTS: Of the 18,782 patients diagnosed with CRC, 134 were diagnosed with synchronous CRC (0.71%) and 300 were diagnosed with metachronous CRC (1.60%). The risk for synchronous CRC was significantly higher in men (odds ratio [OR], 1.45; 95% confidence interval [CI], 1.02-2.06) and in patients aged 65 years or older (OR, 1.50; 95% CI, 1.02-2.21). Synchronous CRCs were located more often in the proximal colon (OR, 1.70; 95% CI, 1.20-2.41). First-degree relatives of cases with synchronous (OR, 1.86; 95% CI, 1.37-2.53), metachronous (OR, 2.34; 95% CI, 1.62-3.36), or solitary CRC (OR, 1.75; 95% CI, 1.63-1.88) were at increased risk for developing CRC, compared with relatives of CRC-free individuals. Four percent of first-degree relatives of patients with synchronous or metachronous cancer developed CRC at younger ages than the age recommended for initiating CRC screening (based on familial risk), and therefore would not have been screened. CONCLUSIONS: Of patients diagnosed with CRC, 2.3% are found to have synchronous lesions or develop metachronous CRC during follow-up evaluation. Relatives of these patients have a greater risk of CRC than those without a family history of CRC. These results highlight the importance of obtaining a thorough family history and adhering strictly to surveillance guidelines during management of high-risk patients.

Characteristics of missed or interval colorectal cancer and patient survival: a population-based study.

BACKGROUND & AIMS: Colorectal cancers (CRCs) diagnosed within a few years after an index colonoscopy can arise from missed lesions or the development of a new tumor. We investigated the proportion, characteristics, and factors that predict interval CRCs that develop within 6-60 months of colonoscopy. METHODS: We performed a population-based cohort study of Utah residents who underwent colonoscopy examinations from 1995 through 2009 at Intermountain Healthcare or the University of Utah Health System, which provide care to more than 85% of state residents. Colonoscopy results were linked with cancer histories from the Utah Population Database to identify patients who underwent colonoscopy 6-60 months before a diagnosis of CRC (interval cancer). Logistic regression was performed to identify risk factors associated with interval cancers. RESULTS: Of 126,851 patients who underwent colonoscopies, 2659 were diagnosed with CRC; 6% of these CRCs (159 of 2659) developed within 6 to 60 months of a colonoscopy. Sex and age were not associated with interval CRCs. A higher percentage of patients with interval CRC were found to have adenomas at their index colonoscopy (57.2%), compared with patients found to have CRC detected at colonoscopy (36%) or patients who did not develop cancer (26%) (P < .001). Interval CRCs tended to be earlier-stage tumors than those detected at index colonoscopy, and to be proximally located (odds ratio, 2.24; P < .001). Patients with interval CRC were more likely to have a family history of CRC (odds ratio, 2.27; P = .008) and had a lower risk of death than patients found to have CRC at their index colonoscopy (hazard ratio, 0.63; P < .001). CONCLUSIONS: In a population-based study in Utah, 6% of all patients with CRC had interval cancers (cancer that developed within 6 to 60 months of a colonoscopy). Interval CRCs were associated with the proximal colon, earlier-stage cancer, lower risk of death, higher rate of adenoma, and family history of CRC. These findings indicate that interval colorectal tumors may arise as the result of distinct biologic features and/or suboptimal management of polyps at colonoscopy.

Risk of colorectal cancer and adenomas in the families of patients with adenomas: a population-based study in Utah.

BACKGROUND: Guidelines recommend that individuals with a first-degree relative (FDR) diagnosed with colorectal cancer (CRC) or advanced adenoma before age 60 years should undergo colonoscopy starting at age 40 years. The authors quantified the risk of adenomas and CRC in FDRs, second-degree relatives (SDRs), and third-degree relatives (TDRs) of patients diagnosed with adenomas and advanced adenomas. METHODS: A population-based, retrospective, case-control study was performed of residents of the state of Utah aged 50 years to 80 years who underwent colonoscopy between 1995 and 2009 at Intermountain Healthcare or the University of Utah. Controls were selected from the population of colonoscopy patients who were free of adenomas or CRC and matched to each case based on sex and birth year. Colonoscopy results were linked with cancer and pedigree information from the Utah Population Database to investigate the familial aggregation of adenomas and CRC using Cox regression analysis. The unit of analysis was the relatives of cases and controls. RESULTS: Of 126,936 patients who underwent colonoscopy, 43,189 had adenomas and 5563 had advanced adenomas and defined the case population. An elevated risk of CRC was found in FDRs (relative risk [RR], 1.35; 95% confidence interval [95% CI], 1.25-1.46), SDRs (RR, 1.15; 95% CI, 1.07-1.23) of adenoma cases, and in FDRs of advanced adenoma cases (RR, 1.68; 95% CI, 1.29-2.18) compared with controls. Approximately 10% of CRCs diagnosed in relatives would have been missed if the current screening guidelines were strictly adhered to. CONCLUSIONS: Relatives of colonoscopy patients with adenomas and advanced adenomas appear to have a significantly elevated risk of developing colorectal neoplasia. These data should be considered when establishing CRC screening guidelines for individuals and their families.

Familial aggregation of childhood and adult cancer in the Utah genealogy.

A small proportion of childhood cancer is attributable to known hereditary syndromes, but whether there is any familial component to the remainder remains uncertain. We explored familial aggregation of cancer in a population-based case-control study using genealogical record linkage and designed to overcome limitations of previous studies. Subjects were selected from the Utah Population Database. We compared risk of cancer in adult first-degree relatives of children who were diagnosed with cancer with the risk in relatives of children who had not had a cancer diagnosed. We identified 1,894 childhood cancer cases and 3,788 controls; 7,467 relatives of cases and 14,498 relatives of controls were included in the analysis. Relatives of children with cancer had a higher risk of cancer in adulthood than relatives of children without cancer [odds ratio (OR) 1.31, 95% confidence interval (CI) 1.11-1.56]; this was restricted to mothers and siblings and was not evident in fathers. Familial aggregation appeared stronger among relatives of cases diagnosed before 5 years of age (OR 1.48, 95% CI 1.13-1.95) than among relatives of cases who were older when diagnosed (OR 1.22, 95% CI 0.98-1.51). These findings provide evidence of a generalized excess of cancer in the mothers and siblings of children with cancer. The tendency for risk to be higher in the relatives of children who were younger at cancer diagnosis should be investigated in other large data sets. The excesses of thyroid cancer in parents of children with cancer and of any cancer in relatives of children with leukemia merit further investigation.

Identifying clinical/translational research cohorts: ascertainment via querying an integrated multi-source database.

BACKGROUND: Ascertainment of potential subjects has been a longstanding problem in clinical research. Various methods have been proposed, including using data in electronic health records. However, these methods typically suffer from scaling effects-some methods work well for large cohorts; others work for small cohorts only. OBJECTIVE: We propose a method that provides a simple identification of pre-research cohorts and relies on data available in most states in the USA: merged public health data sources. MATERIALS AND METHODS: The Utah Population Database Limited query tool allows users to build complex queries that may span several types of health records, such as cancer registries, inpatient hospital discharges, and death certificates; in addition, these can be combined with family history information. The architectural approach incorporates several coding systems for medical information. It provides a front-end graphical user interface and enables researchers to build and run queries and view aggregate results. Multiple strategies have been incorporated to maintain confidentiality. RESULTS: This tool was rapidly adopted; since its release, 241 users representing a wide range of disciplines from 17 institutions have signed the user agreement and used the query tool. Three examples are discussed: pregnancy complications co-occurring with cardiovascular disease; spondyloarthritis; and breast cancer. DISCUSSION AND CONCLUSIONS: This query tool was designed to provide results as pre-research so that institutional review board approval would not be required. This architecture uses well-described technologies that should be within the reach of most institutions.

Effects of BRCA1 and BRCA2 mutations on female fertility.

Women with BRCA1/2 mutations have a significantly higher lifetime risk of developing breast or ovarian cancer. We suggest that female mutation carriers may have improved fitness owing to enhanced fertility relative to non-carriers. Here we show that women who are carriers of BRCA1/2 mutations living in natural fertility conditions have excess fertility as well as excess post-reproductive mortality in relation to controls. Individuals who tested positive for BRCA1/2 mutations who linked into multi-generational pedigrees within the Utah Population Database were used to identify putative obligate carriers. We find that women born before 1930 who are mutation carriers have significantly more children than controls and have excess post-reproductive mortality risks. They also have shorter birth intervals and end child-bearing later than controls. For contemporary women tested directly for BRCA1/2 mutations, an era when modern contraceptives are available, differences in fertility and mortality persist but are attenuated. Our findings suggest the need to re-examine the wider role played by BRCA1/2 mutations. Elevated fertility of female mutation carriers indicates that they are more fecund despite their elevated post-reproductive mortality risks.

Evaluation of record linkage between a large healthcare provider and the Utah Population Database.

OBJECTIVE: Electronically linked datasets have become an important part of clinical research. Information from multiple sources can be used to identify comorbid conditions and patient outcomes, measure use of healthcare services, and enrich demographic and clinical variables of interest. Innovative approaches for creating research infrastructure beyond a traditional data system are necessary. MATERIALS AND METHODS: Records from a large healthcare system's enterprise data warehouse (EDW) were linked to a statewide population database, and a master subject index was created. The authors evaluate the linkage, along with the impact of missing information in EDW records and the coverage of the population database. The makeup of the EDW and population database provides a subset of cancer records that exist in both resources, which allows a cancer-specific evaluation of the linkage. RESULTS: About 3.4 million records (60.8%) in the EDW were linked to the population database with a minimum accuracy of 96.3%. It was estimated that approximately 24.8% of target records were absent from the population database, which enabled the effect of the amount and type of information missing from a record on the linkage to be estimated. However, 99% of the records from the oncology data mart linked; they had fewer missing fields and this correlated positively with the number of patient visits. DISCUSSION AND CONCLUSION: A general-purpose research infrastructure was created which allows disease-specific cohorts to be identified. The usefulness of creating an index between institutions is that it allows each institution to maintain control and confidentiality of their own information.

Pregnancy outcomes in female childhood and adolescent cancer survivors: a linked cancer-birth registry analysis.

OBJECTIVE: To compare birth outcomes among female survivors of childhood and adolescent cancer who subsequently bear children, relative to those of women without a history of cancer. DESIGN: Retrospective cohort study. SETTING: Four US regions. PARTICIPANTS: Cancer registries identified girls younger than 20 years who were diagnosed as having cancer from 1973 through 2000. Linked birth records identified the first live births after diagnosis (n = 1898). Comparison subjects were selected from birth records (n = 14 278). Survivors of genital tract carcinomas underwent separate analysis. MAIN EXPOSURE: Cancer diagnosis at younger than 20 years. MAIN OUTCOME MEASURES: Infant low birth weight, preterm delivery, sex ratio, malformations, mortality, and delivery method, and maternal diabetes, anemia, and preeclampsia. RESULTS: Infants born to childhood cancer survivors were more likely to be preterm (relative risk [RR], 1.54; 95% confidence interval [CI], 1.30-1.83) and to weigh less than 2500 g (1.31; 1.10-1.57). For the offspring of genital tract carcinoma survivors, RRs were 1.33 (95% CI, 1.13-1.56) and 1.29 (1.10-1.53), respectively. There were no increased risks of malformations, infant death, or altered sex ratio, suggesting no increased germ cell mutagenicity. In exploratory analysis, bone cancer survivors had an increased risk of diabetes (RR, 4.92; 95% CI, 1.60-15.13), and anemia was more common among brain tumor survivors (3.05; 1.16-7.98) and childhood cancer survivors whose initial treatment was chemotherapy only (2.45; 1.16-5.17). CONCLUSIONS: Infants born to female survivors of childhood and adolescent cancer were not at increased risk of malformations or death. Increased occurrence of preterm delivery and low birth weight suggest that close monitoring is warranted. Increased diabetes and anemia among subgroups have not been reported, suggesting areas for study.

Reproductive outcomes in male childhood cancer survivors: a linked cancer-birth registry analysis.

OBJECTIVE: To compare the risk of reproductive and infant outcomes between male childhood cancer survivors and a population-based comparison group. DESIGN: Retrospective cohort study. SETTING: Four US regions. PARTICIPANTS: Cancer registries identified males younger than 20 years diagnosed with cancer from 1973 to 2000. Linked birth certificates identified first subsequent live offspring (N = 470). Comparison subjects were identified from remaining birth certificates, frequency-matched on year and age at fatherhood, and race/ethnicity (N = 4150). MAIN EXPOSURE: Cancer diagnosis before age 20 years. OUTCOME MEASURES: Pregnancy and infant outcomes identified from birth certificates. RESULTS: Compared with infants born to unaffected males, offspring of cancer survivors had a borderline risk of having a birth weight less than 2500 g (relative risk, 1.43 [95% confidence interval, 0.99-2.05]) that was associated most strongly with younger age at cancer diagnosis and exposure to any chemotherapy (1.96 [1.22-3.17]) or radiotherapy (1.95 [1.14-3.35]). However, they were not at risk of being born prematurely, being small for gestational age, having malformations, or having an altered male to female ratio. Overall, female partners of male survivors were not more likely to have maternal complications recorded on birth records vs the comparison group. However, preeclampsia was associated with some cancers, especially central nervous system tumors (relative risk, 3.36 [95% confidence interval, 1.63-6.90]). CONCLUSIONS: Most pregnancies resulting in live births among partners of male childhood cancer survivors were not at significantly greater risk of complications vs comparison subjects. However, there remains the possibility that prior cancer therapy may affect male germ cells with some effects on progeny and on female partners.

Common familial colorectal cancer linked to chromosome 7q31: a genome-wide analysis.

Present investigations suggest that approximately 30% of colorectal cancer cases arise on the basis of inherited factors. We hypothesize that the majority of inherited factors are moderately penetrant genes, common in the population. We use an affected sibling pair approach to identify genetic regions that are coinherited by siblings with colorectal cancer. Individuals from families with at least two siblings diagnosed with colorectal adenocarcinoma or high-grade dysplasia were enrolled. Known familial colorectal cancer syndromes were excluded. A genome-wide scan on 151 DNA samples from 70 kindreds was completed using deCODE 1100 short tandem repeat marker set at an average 4-cM density. Fine mapping on a total of 184 DNAs from 83 kindreds was done in regions suggesting linkage. Linkage analysis was accomplished with Merlin analysis package. Nonparametric linkage analysis revealed three genetic regions with logarithm of the odds (LOD) scores >or=2.0: Ch. 3q29, LOD 2.61 (P = 0.0003); Ch. 4q31.3, LOD 2.13 (P = 0.0009); and Ch. 7q31.31, LOD 3.08 (P = 0.00008). Affected siblings with increased sharing at the 7q31 locus have a 3.8-year (+/- 3.5) earlier age of colorectal cancer onset although this is not statistically significant (P = 0.11). No significant linkage was found near genes causing known syndromes or regions previously reported (8q24, 9q22, and 11q23). The chromosome 3q21-q24 region reported to be linked in colorectal cancer relative pairs is supported by our study, albeit a minor peak (LOD 0.9; P = 0.02). No known familial cancer genes reside in the 7q31 locus, and thus the identified region may contain a novel susceptibility gene responsible for common familial colorectal cancer.

Colonic adenoma risk in familial colorectal cancer--a study of six extended kindreds.

OBJECTIVES: Most colorectal cancers (CRCs) arise from adenomatous polyps, but the effects of CRC family history on adenoma risk are not well known. This issue is clinically relevant since several medical societies currently recommend earlier and more rigorous colorectal screening for individuals with a strong family history of CRC. METHODS: Colonoscopies were performed in 236 first-, second-, and third-degree relatives of 40 index CRC cases from six large kindreds selected from a large population database. The kindreds were selected for significantly greater risk of CRCs compared with the overall population. Known hereditary colon cancer syndromes were clinically and genetically excluded. RESULTS: Thirty-seven percent of relatives were found to have adenomas on colonoscopy. The average age of diagnosis for colon cancer was 63 yr and advanced adenomas 56 yr. Independent predictors of adenomatous polyps in the relatives were advancing age (P < 0.0001), male gender (P < 0.001), and greater degree of relation to CRC cases (P < 0.01). There was no significant predilection of colorectal tumors for the right or left colon. A higher degree of relationship to CRC cases was a significant predictor of having simple and advanced adenomas, but not hyperplastic polyps after adjustment for age and gender. CONCLUSIONS: These data support the current recommendations for colonoscopy starting before the age of 50 yr in individuals with a strong family history of CRC.

Preeclampsia and subsequent risk of cancer in Utah.

OBJECTIVE: The purpose of this study was to determine if preeclampsia is associated with a reduced risk of cancer later in life. STUDY DESIGN: We performed a cohort study where women with preeclampsia over the interval 1947 to 1999 were identified from the Utah Population Database. Preeclamptics (n = 17,432) were matched 1:3 with nonpreeclamptics (n = 52,296) on maternal age and birth year. Pregnancy, demographic, and cancer information was extracted from subjects and their offspring in linked datasets. Relative risk and hazard ratios were calculated. RESULTS: In a matched analysis using univariable random-effects Poisson regression, preeclampsia was protective against the development of cancer later in life (RR 0.91, 95% CI 0.84-0.99 with P = .027). In a multivariable clustered Cox regression model with the end point of cancer later in life, preeclampsia was associated with a lower risk of cancer (HR 0.92, 95% CI 0.85-0.99 with P = .039). These findings were supported by stratified and competing risk analyses. CONCLUSION: Women whose pregnancies were affected by preeclampsia have a decreased risk of developing cancer.

Twinning and the incidence of breast and gynecological cancers (United States).

Increasing epidemiological and experimental evidence indicates that the carcinogenic pathway in the breast and female reproductive organs is driven, at least in part, by factors associated with reproduction. We conducted a retrospective cohort study, comparing the risk of ovarian, breast, endometrial, and cervical cancers among women who had records of at least one twin pregnancy, compared with women who had given birth to only single children. Subjects were selected from the Utah Population Database, which consists of multiple linked datasets including genealogy, births and deaths and cancer registries. We used Poisson regression to calculate relative risks, adjusted for the number of pregnancies and the age of the mother at the birth of first and last children, with singleton mothers as the reference group in each case. The risks of breast and ovarian cancers did not differ between mothers of twins and mothers of single children. The risk of endometrial cancer was slightly lower in mothers of twins than in mothers of singleton children (RR = 0.90, 95% CI 0.67-1.21). Conversely the risk of cervical cancer was higher among twin mothers (RR = 1.78, 95% CI 0.88-3.52). This latter finding supports previous data suggesting that reproductive hormones act as cofactors in the etiology of cervical cancer.