Periodic Colonoscopies Are Associated with Improved Survival and Prognosis of Colorectal Cancer in Ulcerative Colitis.

BACKGROUND AND STUDY AIMS: This study aimed to identify whether ulcerative colitis (UC) patients who develop colorectal cancer (CRC) present at earlier stages of CRC and have improved survival if prior to their CRC diagnosis, they underwent intermittent follow-up colonoscopies compared to those who have no follow-up colonoscopies. METHODS: Patients with UC who developed primary CRC were identified using data provided by the Institute for Clinical Evaluative Sciences. We defined low-risk CRC stage as estimated 5-year survival ≥ 80% compared to high-risk CRC as 5-year survival < 80%. RESULTS: A total of 421 patients were identified with UC and CRC. The 15-year mortality rate was significantly higher in those who did not have follow-up colonoscopy (33/74; 44.6%) compared to the follow-up group (105/347; 30.3%) (p = 0.0172). Among the 219 patients with UC with staging information available, patients who did not have follow-up colonoscopy were more likely to present with high-risk CRC (24/31; 77.4%) compared with patients who had follow-up colonoscopies (88/188; 44.4%) (p = 0.0016). Those who underwent follow-up colonoscopies at average intervals ≤ 3 years presented with high-risk CRC 41.3% of the time, which was less than the 48.6% in those with less frequent colonoscopies and 77.4% in those with no follow-up (p = 0.0048). CONCLUSIONS: Patients with UC who underwent intermittent follow-up colonoscopies had CRC detected at earlier stages and improvement in all-cause mortality, compared to those who with no follow-up colonoscopies. This may support regular surveillance colonoscopies for patients with UC.

Identifying Unexpected Deaths in Long-Term Care Homes.

OBJECTIVES: Predicting unexpected deaths among long-term care (LTC) residents can provide valuable information to clinicians and policy makers. We study multiple methods to predict unexpected death, adjusting for individual and home-level factors, and to use as a step to compare mortality differences at the facility level in the future work. DESIGN: We conducted a retrospective cohort study using Resident Assessment Instrument Minimum Data Set assessment data for all LTC residents in Ontario, Canada, from April 2017 to March 2018. SETTING AND PARTICIPANTS: All residents in Ontario long-term homes. We used data routinely collected as part of administrative reporting by health care providers to the funder: Ontario Ministry of Health and Long-Term Care. This project is a component of routine policy development to ensure safety of the LTC system residents. METHODS: Logistic regression (LR), mixed-effect LR (mixLR), and a machine learning algorithm (XGBoost) were used to predict individual mortality over 5 to 95 days after the last available RAI assessment. RESULTS: We identified 22,419 deaths in the cohort of 106,366 cases (mean age: 83.1 years; female: 67.7%; dementia: 68.8%; functional decline: 16.6%). XGBoost had superior calibration and discrimination (C-statistic 0.837) over both mixLR (0.819) and LR (0.813). The models had high correlation in predicting death (LR-mixLR: 0.979, LR-XGBoost: 0.885, mixLR-XGBoost: 0.882). The inter-rater reliability between the models LR-mixLR and LR-XGBoost was 0.56 and 0.84, respectively. Using results in which all 3 models predicted probability of actual death of a resident at <5% yielded 210 unexpected deaths or 0.9% of the observed deaths. CONCLUSIONS AND IMPLICATIONS: XGBoost outperformed other models, but the combination of 3 models provides a method to detect facilities with potentially higher rates of unexpected deaths while minimizing the possibility of false positives and could be useful for ongoing surveillance and quality assurance at the facility, regional, and national levels.

The economic burden of cancer care in Canada: a population-based cost study.

BACKGROUND: Resource and cost issues are a growing concern in health care. Thus, it is important to have an accurate estimate of the economic burden of care. Previous work has estimated the economic burden of cancer care for Canada; however, there is some concern this estimate is too low. The objective of this analysis was to provide a comprehensive revised estimate of this burden. METHODS: We used a case-control prevalence-based approach to estimate direct annual cancer costs from 2005 to 2012. We used patient-level administrative health care data from Ontario to correctly attribute health care costs to cancer. We employed the net cost method (cost difference between patients with cancer and control subjects without cancer) to account for costs directly and indirectly related to cancer and its sequelae. Using average patient-level cost estimates from Ontario, we applied proportions from national health expenditures data to obtain the economic burden of cancer care for Canada. All costs were adjusted to 2015 Canadian dollars. RESULTS: Costs of cancer care rose steadily over our analysis period, from $2.9 billion in 2005 to $7.5 billion in 2012, mostly owing to the increase in costs of hospital-based care. Most expenditures for health care services increased over time, with chemotherapy and radiation therapy expenditures accounting for the largest increases over the study period. Our cost estimates were larger than those in the Economic Burden of Illness in Canada 2005-2008 report for every year except 2005 and 2006. INTERPRETATION: The economic burden of cancer care in Canada is substantial. Further research is needed to understand how the economic burden of cancer compares to that of other diseases.

Ictal Symmetric Tonic Extension Posturing and Postictal Generalized EEG Suppression Arising From Sleep in Children With Epilepsy.

BACKGROUND: The identification of a biomarker for sudden unexpected death in epilepsy (SUDEP) has the potential to save lives. Generalized convulsive seizures and postictal generalized suppression on electroencephalography (EEG) most often precede sudden unexpected death in epilepsy (SUDEP) and are potential biomarkers. We identify the EEG and seizure characteristics associated with postictal generalized EEG suppression in children with epilepsy. METHODS: Video EEGs were reviewed for seizure type, duration and semiology, and electrographic features. To identify predictors of postictal generalized EEG suppression, we identified 40 children with generalized convulsive seizures from a group of 399 patients who experienced an electroclinical seizure during video-EEG. Seventy-seven generalized convulsive seizures with and without postictal generalized EEG suppression were anayzed. RESULTS: Age of seizure onset was older in 19 children with postictal generalized EEG suppression (mean 6.8 years old, 95% CI [4.7-8.9]) than in 21 without postictal generalized EEG suppression (3.0 years old, [1.1-4.1], P = 0.041). Postictal generalized EEG suppression occurred significantly more often from sleep than awake (point of estimate 16.67; 95% CI [0.97-32.36], P < 0.038). Shorter duration of the clonic phase (-0.76; [-1.338, -0.133], P = 0.018) was significantly associated with postictal generalized EEG suppression. Ictal symmetric tonic extension posturing significantly increased the odds of postictal generalized EEG suppression (42.94; [18.77, 67.12], P = 0.001). All 15 generalized convulsive seizures with a terminal burst-suppression pattern were followed by postictal generalized EEG suppression in contrast to 19 of 62 generalized convulsive seizures without burst-suppression (15.32, P < 0.001). CONCLUSIONS: Ictal decerebrate-like symmetric tonic extension posturing with shorter clonic phase and a terminal burst-suppression pattern identify malignant generalized convulsive seizures, associated with postictal generalized EEG suppression and a potentially increased risk of sudden unexpected death in epilepsy.

Estimating the Cost of Cancer Care in British Columbia and Ontario: A Canadian Inter-Provincial Comparison.

BACKGROUND: Costing studies are useful to measure the economic burden of cancer. Comparing costs between healthcare systems can inform evaluation, development or modification of cancer care policies. OBJECTIVES: To estimate and compare cancer costs in British Columbia and Ontario from the payers' perspectives. METHODS: Using linked cancer registry and administrative data, and standardized costing methodology and analyses, we estimated costs for 21 cancer sites by phase of care to determine potential differences between provinces. RESULTS: Overall, costs were higher in Ontario. Costs were highest in the initial post-diagnosis and pre-death phases and lowest in the pre-diagnosis and continuing phases, and generally higher for brain cancer and multiple myeloma, and lower for melanoma. Hospitalization was the major cost category. Costs for physician services and diagnostic tests differed the most between provinces. CONCLUSIONS: The standardization of data and costing methodology is challenging, but it enables interprovincial and international comparative costing analyses.

Androgen deprivation therapy and the risk of parkinsonism in men with prostate cancer.

PURPOSE: Case reports and anecdotal experiences suggest that some men develop parkinsonism after initiating androgen deprivation therapy (ADT) for the treatment of prostate cancer, possibly due to neurophysiological effects of changes in testosterone and/or estrogen. We hypothesized that ADT would increase the risk of parkinsonism. METHODS: Using linked administrative databases in Ontario, Canada, men age 40 or older with prostate cancer on continuous ADT for at least 6 months or who underwent bilateral orchiectomy (n = 38,931) were matched 1:1 with men with prostate cancer who had never received ADT. Treated and untreated groups were range-matched on age at index date and year of diagnosis, and propensity-matched on comorbidities, medications, cardiovascular risk factors, and socio-economic variables. A competing risk analysis was conducted where the primary outcome was time to a new diagnosis of parkinsonism. RESULTS: The cohort was followed for a mean of 5.76 years. Based on the results from the multivariable cause-specific hazard regression model, the adjusted relative rate of experiencing parkinsonism among ADT users compared to non-users was 0.74 (95% confidence interval (CI) 0.67-0.83, p < 0.0001). The adjusted relative rate of experiencing the competing event of death among ADT users compared to non-users was 1.33 (95% CI 1.30-1.36, p < 0.0001). The 5-year incidence of parkinsonism was 1.03% in ADT users versus 1.56% in non-users. CONCLUSION: Contrary to our hypothesis, continuous ADT use for at least 6 months in men with prostate cancer was not associated with an increased risk of parkinsonism after accounting for the substantial competing risk of death.

Phase-specific and lifetime costs of cancer care in Ontario, Canada.

BACKGROUND: Cancer is a major public health issue and represents a significant economic burden to health care systems worldwide. The objective of this analysis was to estimate phase-specific, 5-year and lifetime net costs for the 21 most prevalent cancer sites, and remaining tumour sites combined, in Ontario, Canada. METHODS: We selected all adult patients diagnosed with a primary cancer between 1997 and 2007, with valid ICD-O site and histology codes, and who survived 30 days or more after diagnosis, from the Ontario Cancer Registry (N = 394,092). Patients were linked to treatment data from Cancer Care Ontario and administrative health care databases at the Institute for Clinical and Evaluative Sciences. Net costs (i.e., cost difference between patients and matched non-cancer control subjects) were estimated by phase of care and sex, and used to estimate 5-year and lifetime costs. RESULTS: Mean net costs of care (2009 CAD) were highest in the initial (6 months post-diagnosis) and terminal (12 months pre-death) phases, and lowest in the (3 months) pre-diagnosis and continuing phases of care. Phase-specific net costs were generally lowest for melanoma and highest for brain cancer. Mean 5-year net costs varied from less than $25,000 for melanoma, thyroid and testicular cancers to more than $60,000 for multiple myeloma and leukemia. Lifetime costs ranged from less than $55,000 for lung and liver cancers to over $110,000 for leukemia, multiple myeloma, lymphoma and breast cancer. CONCLUSIONS: Costs of cancer care are substantial and vary by cancer site, phase of care and time horizon analyzed. These cost estimates are valuable to decision makers to understand the economic burden of cancer care and may be useful inputs to researchers undertaking cancer-related economic evaluations.

The Effect of Community-Based Specialist Palliative Care Teams on Place of Care.

BACKGROUND: Prior research on community-based specialist palliative care teams used outcome measures of place of death and/or dichotomous outcome measures of acute care use in the last two weeks of life. However, existing research seldom measured the diverse places of care used and their timing prior to death. OBJECTIVE: The study objective was to examine the place of care in the last 30 days of life. METHODS: In this retrospective cohort study, patients who received care from a specialist palliative care team (exposed) were matched by propensity score to patients who received usual care in the community (unexposed) in Ontario, Canada. Measured was the percentage of patients in each place of care in the last month of life as a proportion of the total cohort. RESULTS: After matching, 3109 patients were identified in each group, where 79% had cancer and 77% received end-of-life home care. At 30 days compared to 7 days before death, the exposed group's proportions rose from 33% to 41% receiving home care and 14% to 15% in hospital, whereas the unexposed group's proportions rose from 28% to 32% receiving home care and 16% to 22% in hospital. Linear trend analysis (proportion over time) showed that the exposed group used significantly more home care services and fewer hospital days (p < 0.001) than the unexposed group. On the last day of life (place of death), the exposed group had 18% die in an in-patient hospital bed compared to 29% in usual care. CONCLUSION: Examining place of care in the last month can effectively illustrate the service use trajectory over time.

Regional variation in the use of surgery in Ontario.

BACKGROUND: Regional variation in the use of surgery implies that there is uncertainty regarding appropriate use. The objectives of this study were to identify which surgical procedures are most commonly performed in the province of Ontario and measure the extent of variation in the use of surgical procedures across Ontario counties. METHODS: We used the Canadian Institute for Health Information Discharge Abstract Database, Same Day Surgery Database and National Ambulatory Care Reporting System to retrieve information on all inpatient and day surgery visits in Ontario between Apr. 1, 2002, and Mar. 31, 2011. We identified the 84 most common procedures according to Canadian Classification of Interventions codes. We calculated rates of use for each procedure throughout the 49 Ontario counties and then calculated measures of variation (quartile ratio and systematic component of variation) in use between the counties. RESULTS: Colonoscopy was the most commonly performed procedure during the study period, with an average adjusted rate of 2012 per 100 000 population. The procedure with the highest measure of variation was iridectomy, with a quartile ratio of 6.7, followed by colposcopy (5.2), cervical biopsy (4.2) and femoral arteriography (4.1). These procedures were less commonly performed. Common procedures such as colonoscopy, cataract extraction and vaginal delivery had lower quartile ratios. Analysis using the systematic component of variation as the measure of variation gave similar results. INTERPRETATION: Colonoscopy was the most commonly performed procedure in Ontario, and cataract extraction was the most common surgical procedure. Procedures with the highest measures of variation between counties tended to be those that occurred less commonly in Ontario, and common procedures were associated with less regional variation.

Impact of aggressive management and palliative care on cancer costs in the final month of life.

BACKGROUND: A significant share of the cost of cancer care is concentrated in the end-of-life period. Although quality measures of aggressive treatment may guide optimal care during this timeframe, little is known about whether these metrics affect costs of care. METHODS: This study used population data to identify a cohort of patients who died of cancer in Ontario, Canada (2005-2009). Individuals were categorized as having received or having not received aggressive end-of-life care according to quality measures related to acute institutional care or chemotherapy administration in the end-of-life period. Costs (2009 Canadian dollars) were collected over the last month of life through the linkage of health system administrative databases. Multivariate quantile regression was used to identify predictors of increased costs. RESULTS: Among 107,253 patients, the mean per-patient cost over the final month was $18,131 for patients receiving aggressive care and $12,678 for patients receiving nonaggressive care (P < .0001). Patients who received chemotherapy in the last 2 weeks of life also sustained higher costs than those who did not (P < .0001). For individuals receiving end-of-life care in the highest cost quintile, early and repeated palliative care consultation was associated with reduced mean per-patient costs. In a multivariate analysis, chemotherapy in the 2 weeks of life remained predictive of increased costs (median increase, $536; P < .0001), whereas access to palliation remained predictive for lower costs (median decrease, $418; P < .0001). CONCLUSIONS: Cancer patients who receive aggressive end-of-life care incur 43% higher costs than those managed nonaggressively. Palliative consultation may partially offset these costs and offer resultant savings.

Impact of community based, specialist palliative care teams on hospitalisations and emergency department visits late in life and hospital deaths: a pooled analysis.

OBJECTIVE: To determine the pooled effect of exposure to one of 11 specialist palliative care teams providing services in patients' homes. DESIGN: Pooled analysis of a retrospective cohort study. SETTING: Ontario, Canada. PARTICIPANTS: 3109 patients who received care from specialist palliative care teams in 2009-11 (exposed) matched by propensity score to 3109 patients who received usual care (unexposed). INTERVENTION: The palliative care teams studied served different geographies and varied in team composition and size but had the same core team members and role: a core group of palliative care physicians, nurses, and family physicians who provide integrated palliative care to patients in their homes. The teams' role was to manage symptoms, provide education and care, coordinate services, and be available without interruption regardless of time or day. MAIN OUTCOME MEASURES: Patients (a) being in hospital in the last two weeks of life; (b) having an emergency department visit in the last two weeks of life; or (c) dying in hospital. RESULTS: In both exposed and unexposed groups, about 80% had cancer and 78% received end of life homecare services for the same average duration. Across all palliative care teams, 970 (31.2%) of the exposed group were in hospital and 896 (28.9%) had an emergency department visit in the last two weeks of life respectively, compared with 1219 (39.3%) and 1070 (34.5%) of the unexposed group (P<0.001). The pooled relative risks of being in hospital and having an emergency department visit in late life comparing exposed versus unexposed were 0.68 (95% confidence interval 0.61 to 0.76) and 0.77 (0.69 to 0.86) respectively. Fewer exposed than unexposed patients died in hospital (503 (16.2%) v 887 (28.6%), P<0.001), and the pooled relative risk of dying in hospital was 0.46 (0.40 to 0.52). CONCLUSIONS: Community based specialist palliative care teams, despite variation in team composition and geographies, were effective at reducing acute care use and hospital deaths at the end of life.

Identification of genes expressed by immune cells of the colon that are regulated by colorectal cancer-associated variants.

A locus on human chromosome 11q23 tagged by marker rs3802842 was associated with colorectal cancer (CRC) in a genome-wide association study; this finding has been replicated in case-control studies worldwide. In order to identify biologic factors at this locus that are related to the etiopathology of CRC, we used microarray-based target selection methods, coupled to next-generation sequencing, to study 103 kb at the 11q23 locus. We genotyped 369 putative variants from 1,030 patients with CRC (cases) and 1,061 individuals without CRC (controls) from the Ontario Familial Colorectal Cancer Registry. Two previously uncharacterized genes, COLCA1 and COLCA2, were found to be co-regulated genes that are transcribed from opposite strands. Expression levels of COLCA1 and COLCA2 transcripts correlate with rs3802842 genotypes. In colon tissues, COLCA1 co-localizes with crystalloid granules of eosinophils and granular organelles of mast cells, neutrophils, macrophages, dendritic cells and differentiated myeloid-derived cell lines. COLCA2 is present in the cytoplasm of normal epithelial, immune and other cell lineages, as well as tumor cells. Tissue microarray analysis demonstrates the association of rs3802842 with lymphocyte density in the lamina propria (p = 0.014) and levels of COLCA1 in the lamina propria (p = 0.00016) and COLCA2 (tumor cells, p = 0.0041 and lamina propria, p = 6 × 10(-5)). In conclusion, genetic, expression and immunohistochemical data implicate COLCA1 and COLCA2 in the pathogenesis of colon cancer. Histologic analyses indicate the involvement of immune pathways.

Complications of daytime elective laparoscopic cholecystectomies performed by surgeons who operated the night before.

IMPORTANCE: The effect of surgeons' disrupted sleep on patient outcomes is not clearly defined. OBJECTIVE: To assess if surgeons operating the night before have more complications of elective surgery performed the next day. DESIGN, SETTING, AND PARTICIPANTS: Population-based, matched, retrospective cohort study using administrative health care databases in Ontario, Canada (2012 population, 13,505,900). Participants were 2078 patients who underwent elective laparoscopic cholecystectomies performed by surgeons who operated the night before, matched with 4 other elective laparoscopic cholecystectomy recipients (n = 8312). EXPOSURE: In total, 94,183 eligible elective laparoscopic cholecystectomies were performed between 2004 and 2011. Of these surgeries, there were 2078 procedures in which 331 different surgeons across 102 community hospitals had operated between midnight and 7 am the night before. Each "at-risk" surgery was randomly matched with 4 other elective laparoscopic cholecystectomies (n = 8312) performed by the same surgeon, who had no evidence of having operated the night before. MAIN OUTCOMES AND MEASURES: The primary outcome was conversion from a laparoscopic cholecystectomy to open cholecystectomy. Secondary outcomes included evidence of iatrogenic injuries or death. Risks were quantified using generalized estimating equations. RESULTS: No significant association was found in conversion rates to open operations between surgeons when they operated the night before compared with when they did not operate the previous night (46/2031 [2.2%] vs 157/8124 [1.9%]; adjusted odds ratio [OR], 1.18; 95% CI, 0.85-1.64). There was no association between operating the night before vs not operating the night before, and risk of iatrogenic injuries (14/2031 [0.7%] vs 72/8124 [0.9%]; adjusted OR, 0.77; 95% CI, 0.43-1.37) or death (≤5/2031 [≤0.2%] vs 7/8124 [0.1%]). CONCLUSIONS AND RELEVANCE: No significant association was found between operating the night before and not operating the previous night for conversion to open cholecystectomy, risk of iatrogenic complications, or death for elective daytime cholecystectomy. These findings do not support safety concerns related to surgeons operating the night before performing elective surgery.

Validity of physician billing claims to identify deceased organ donors in large healthcare databases.

OBJECTIVE: We evaluated the validity of physician billing claims to identify deceased organ donors in large provincial healthcare databases. METHODS: We conducted a population-based retrospective validation study of all deceased donors in Ontario, Canada from 2006 to 2011 (n = 988). We included all registered deaths during the same period (n = 458,074). Our main outcome measures included sensitivity, specificity, positive predictive value, and negative predictive value of various algorithms consisting of physician billing claims to identify deceased organ donors and organ-specific donors compared to a reference standard of medical chart abstraction. RESULTS: The best performing algorithm consisted of any one of 10 different physician billing claims. This algorithm had a sensitivity of 75.4% (95% CI: 72.6% to 78.0%) and a positive predictive value of 77.4% (95% CI: 74.7% to 80.0%) for the identification of deceased organ donors. As expected, specificity and negative predictive value were near 100%. The number of organ donors identified by the algorithm each year was similar to the expected value, and this included the pre-validation period (1991 to 2005). Algorithms to identify organ-specific donors performed poorly (e.g. sensitivity ranged from 0% for small intestine to 67% for heart; positive predictive values ranged from 0% for small intestine to 37% for heart). INTERPRETATION: Primary data abstraction to identify deceased organ donors should be used whenever possible, particularly for the detection of organ-specific donations. The limitations of physician billing claims should be considered whenever they are used.

Characterization of gene-environment interactions for colorectal cancer susceptibility loci.

Genome-wide association studies (GWAS) have identified more than a dozen loci associated with colorectal cancer (CRC) risk. Here, we examined potential effect-modification between single-nucleotide polymorphisms (SNP) at 10 of these loci and probable or established environmental risk factors for CRC in 7,016 CRC cases and 9,723 controls from nine cohort and case-control studies. We used meta-analysis of an efficient empirical-Bayes estimator to detect potential multiplicative interactions between each of the SNPs [rs16892766 at 8q23.3 (EIF3H/UTP23), rs6983267 at 8q24 (MYC), rs10795668 at 10p14 (FLJ3802842), rs3802842 at 11q23 (LOC120376), rs4444235 at 14q22.2 (BMP4), rs4779584 at 15q13 (GREM1), rs9929218 at 16q22.1 (CDH1), rs4939827 at 18q21 (SMAD7), rs10411210 at 19q13.1 (RHPN2), and rs961253 at 20p12.3 (BMP2)] and select major CRC risk factors (sex, body mass index, height, smoking status, aspirin/nonsteroidal anti-inflammatory drug use, alcohol use, and dietary intake of calcium, folate, red meat, processed meat, vegetables, fruit, and fiber). The strongest statistical evidence for a gene-environment interaction across studies was for vegetable consumption and rs16892766, located on chromosome 8q23.3, near the EIF3H and UTP23 genes (nominal P(interaction) = 1.3 × 10(-4); adjusted P = 0.02). The magnitude of the main effect of the SNP increased with increasing levels of vegetable consumption. No other interactions were statistically significant after adjusting for multiple comparisons. Overall, the association of most CRC susceptibility loci identified in initial GWAS seems to be invariant to the other risk factors considered; however, our results suggest potential modification of the rs16892766 effect by vegetable consumption.

Meta-analysis of new genome-wide association studies of colorectal cancer risk.

Colorectal cancer is the second leading cause of cancer death in developed countries. Genome-wide association studies (GWAS) have successfully identified novel susceptibility loci for colorectal cancer. To follow up on these findings, and try to identify novel colorectal cancer susceptibility loci, we present results for GWAS of colorectal cancer (2,906 cases, 3,416 controls) that have not previously published main associations. Specifically, we calculated odds ratios and 95% confidence intervals using log-additive models for each study. In order to improve our power to detect novel colorectal cancer susceptibility loci, we performed a meta-analysis combining the results across studies. We selected the most statistically significant single nucleotide polymorphisms (SNPs) for replication using ten independent studies (8,161 cases and 9,101 controls). We again used a meta-analysis to summarize results for the replication studies alone, and for a combined analysis of GWAS and replication studies. We measured ten SNPs previously identified in colorectal cancer susceptibility loci and found eight to be associated with colorectal cancer (p value range 0.02 to 1.8 × 10(-8)). When we excluded studies that have previously published on these SNPs, five SNPs remained significant at p < 0.05 in the combined analysis. No novel susceptibility loci were significant in the replication study after adjustment for multiple testing, and none reached genome-wide significance from a combined analysis of GWAS and replication. We observed marginally significant evidence for a second independent SNP in the BMP2 region at chromosomal location 20p12 (rs4813802; replication p value 0.03; combined p value 7.3 × 10(-5)). In a region on 5p33.15, which includes the coding regions of the TERT-CLPTM1L genes and has been identified in GWAS to be associated with susceptibility to at least seven other cancers, we observed a marginally significant association with rs2853668 (replication p value 0.03; combined p value 1.9 × 10(-4)). Our study suggests a complex nature of the contribution of common genetic variants to risk for colorectal cancer.

Genotype-environment interactions in microsatellite stable/microsatellite instability-low colorectal cancer: results from a genome-wide association study.

BACKGROUND: Genome-wide association studies (GWAS) have led to the identification of a number of common susceptibility loci for colorectal cancer (CRC); however, none of these GWAS have considered gene-environment (G × E) interactions. Therefore, it is unclear whether current hits are modified by environmental exposures or whether there are additional hits whose effects are dependent on environmental exposures. METHODS: We conducted a systematic search for G × E interactions using genome wide data from the Colon Cancer Family Registry that included 1,191 cases of microsatellite stable (MSS) or microsatellite instability-low (MSI-L) CRC and 999 controls genotyped using either the Illumina Human1M or Human1M-Duo BeadChip. We tested for interactions between genotypes and 14 environmental factors using 3 methods: a traditional case-control test, a case-only test, and the recently proposed 2-step method by Murcray and colleagues. All potentially significant findings were replicated in the ARCTIC Study. RESULTS: No G × E interactions were identified that reached genome-wide significance by any of the 3 methods. When analyzing previously reported susceptibility loci, 7 significant G × E interactions were found at a 5% significance level. We investigated these 7 interactions in an independent sample and none of the interactions were replicated. CONCLUSIONS: Identifying G × E interactions will present challenges in a GWAS setting. Our power calculations illustrate the need for larger sample sizes; however, as CRC is a heterogeneous disease, a tradeoff between increasing sample size and heterogeneity needs to be considered. IMPACT: The results from this first genome-wide analysis of G × E in CRC identify several challenges, which may be addressed by large consortium efforts.

Efficient p-value estimation in massively parallel testing problems.

We present a new method to efficiently estimate very large numbers of p-values using empirically constructed null distributions of a test statistic. The need to evaluate a very large number of p-values is increasingly common with modern genomic data, and when interaction effects are of interest, the number of tests can easily run into billions. When the asymptotic distribution is not easily available, permutations are typically used to obtain p-values but these can be computationally infeasible in large problems. Our method constructs a prediction model to obtain a first approximation to the p-values and uses Bayesian methods to choose a fraction of these to be refined by permutations. We apply and evaluate our method on the study of association between 2-way interactions of genetic markers and colorectal cancer using the data from the first phase of a large, genome-wide case-control study. The results show enormous computational savings as compared to evaluating a full set of permutations, with little decrease in accuracy.

Summary of Genetic Analysis Workshop 15: Group 9 linkage analysis of the CEPH expression data.

Group 9 participants carried out linkage analysis of the Centre d'Etude de Polymorphism Humain (CEPH) expression data, using strategies that ranged from focused investigation of a small number of traits to full genome scans of all available traits. Results from five key areas encompass the most important results within and across the 17 participating groups. First, both extensive genetic heterogeneity and poor predictability of mapping results based on heritability have key implications for study design. Second, choice of the map used for linkage analysis is influential, with the implication that meiotic maps are preferable to physical maps. Third, performance of different analytic methods was in general fairly consistent, with the exception of one variance-component method that uses marker allele sharing as the dependent rather than independent variable. Fourth, multivariate analysis approaches did not generally appear to provide advantages over univariate approaches for linkage detection. Finally, there were computational and analytic challenges in working with a large public data set, along with need for more data documentation.