Robert W Haile

Queensland Institute of Medical Research, Brisbane, Queensland, Australia

Are you Robert W Haile?

Claim your profile

Publications (361)

  • Source
    [Show abstract] [Hide abstract] ABSTRACT: Genome-wide association studies (GWAS) have identified many common single nucleotide polymorphisms (SNPs) associated with colorectal cancer risk. These SNPs may tag correlated variants with biological importance. Fine-mapping around GWAS loci can facilitate detection of functional candidates and additional independent risk variants. We analyzed 11,900 cases and 14,311 controls in the Genetics and Epidemiology of Colorectal Cancer Consortium and the Colon Cancer Family Registry. To fine-map genomic regions containing all known common risk variants, we imputed high-density genetic data from the 1000 Genomes Project. We tested single-variant associations with colorectal tumor risk for all variants spanning genomic regions 250-kb upstream or downstream of 31 GWAS-identified SNPs (index SNPs). We queried the University of California, Santa Cruz Genome Browser to examine evidence for biological function. Index SNPs did not show the strongest association signals with colorectal tumor risk in their respective genomic regions. Bioinformatics analysis of SNPs showing smaller P-values in each region revealed 21 functional candidates in 12 loci (5q31.1, 8q24, 11q13.4, 11q23, 12p13.32, 12q24.21, 14q22.2, 15q13, 18q21, 19q13.1, 20p12.3, and 20q13.33). We did not observe evidence of additional independent association signals in GWAS-identified regions. Our results support the utility of integrating data from comprehensive fine-mapping with expanding publicly available genomic databases to help clarify GWAS associations and identify functional candidates that warrant more onerous laboratory follow-up. Such efforts may aid the eventual discovery of disease-causing variant(s).
    Full-text Article · Jul 2016 · PLoS ONE
  • Source
    Dataset: S1 Table
    [Show abstract] [Hide abstract] ABSTRACT: Expanded list of SNPs with weak, moderate, or strong evidence of biological function based on bioinformatics (including those in Table 2). (XLSX)
    Full-text Dataset · Jul 2016
  • Source
    Dataset: S1 Text
    [Show abstract] [Hide abstract] ABSTRACT: Supplementary materials. Describes in detail the study population and case/control definition; genotyping and quality control; as well as functional annotation using bioinformatics. Also includes Supplementary Tables A-C and Supplementary Figures A-D. (DOCX)
    Full-text Dataset · Jul 2016
  • [Show abstract] [Hide abstract] ABSTRACT: Germline mutations in the DNA base excision repair gene MUTYH are known to increase a carrier's risk of colorectal cancer. However, the risks of other (extracolonic) cancers for MUTYH mutation carriers are not well defined. We identified 266 probands (91% Caucasians) with a MUTYH mutation (41 biallelic and 225 monoallelic) from the Colon Cancer Family Registry. Mutation status, sex, age, and histories of cancer from their 1,903 first- and 3,255 second-degree relatives, were analysed using modified segregation analysis conditioned on the ascertainment criteria. Compared with incidences for the general population, HRs (95% confidence intervals [CIs]) for biallelic MUTYH mutation carriers were: urinary bladder cancer, 19(3.7-97); and ovarian cancer, 17(2.4-115). The HRs (95%CI) for monoallelic MUTYH mutation carriers were: gastric cancer, 9.3(6.7-13); hepatobiliary cancer, 4.5(2.7-7.5); endometrial cancer, 2.1(1.1-3.9); and breast cancer, 1.4(1.0-2.0). There was no evidence for an increased risk of cancers at the other sites examined (brain, pancreas, kidney or prostate). Based on the USA population incidences, the estimated cumulative risks (95%CI) to age 70 years for biallelic mutation carriers were: bladder cancer, 25%(5%-77%) for males and 8%(2%-33%) for females; and ovarian cancer, 14%(2%-65%). The cumulative risks (95%CI) for monoallelic mutation carriers were: gastric cancer, 5%(4%-7%) for males and 2.3%(1.7%-3.3%) for females; hepatobiliary cancer, 3%(2%-5%) for males and 1.4%(0.8%-2.3%) for females; endometrial cancer, 3%(2%-6%); and breast cancer 11%(8%-16%). These unbiased estimates of both relative and absolute risks of extracolonic cancers for people, mostly Caucasians, with MUTYH mutations will be important for the clinical management.
    Article · May 2016 · International Journal of Cancer
  • [Show abstract] [Hide abstract] ABSTRACT: ndividuals diagnosed with colorectal cancer (CRC) are at risk of developing a metachronous colorectal cancer. We examined the associations between personal, tumour-related and lifestyle risk factors, and risk of metachronous CRC. A total of 7,863 participants with incident colon or rectal cancer who were recruited in the USA, Canada and Australia to the Colon Cancer Family Registry during 1997-2012, except those identified as high-risk e.g. Lynch syndrome, were followed up approximately every 5 years. We estimated the risk of metachronous CRC, defined as the first new primary CRC following an interval of at least one year after the initial colorectal cancer diagnosis. Observation time started at the age at diagnosis of the initial CRC and ended at the age at diagnosis of the metachronous CRC, last contact or death whichever occurred earliest, or were censored at the age at diagnosis of any metachronous colorectal adenoma. Cox regression was used to derive hazard ratios (HRs) and 95% confidence intervals (CIs). During a mean follow-up of 6.6 years, 142 (1.81%) metachronous colorectal cancers were diagnosed (mean age at diagnosis 59.8; incidence 2.7/1000 person-years). An increased risk of metachronous CRC was associated with the presence of a synchronous CRC (HR=2.73; 95% CI: 1.30-5.72) and the location of cancer in the proximal colon at initial diagnosis (compared with distal colon or rectum, HR=4.16; 95% CI: 2.80-6.18). The presence of a synchronous CRC and the location of the initial CRC might be useful for deciding the intensity of surveillance colonoscopy for individuals diagnosed with CRC.
    Article · Apr 2016 · International Journal of Cancer
  • [Show abstract] [Hide abstract] ABSTRACT: Background: People with a DNA mismatch repair (MMR) gene mutation have a substantially elevated risk of colorectal cancer (CRC) but the modifiers of this risk are not well established. We investigated the association between dietary supplement intake and CRC risk for carriers. Methods: This study included 1966 (56% female) carriers of an MMR gene mutation (719MLH1, 931MSH2, 211MSH6and 105PMS2) who were recruited from the USA, Canada, Australia and New Zealand into the Colon Cancer Family Registry between 1997 and 2012. Information on lifestyle factors including supplement intake was collected at the time of recruitment. Using Cox proportional hazards regression weighted to correct for ascertainment bias, we estimated hazard ratios (HRs) and 95% confidence intervals (CIs) for associations between self-reported multivitamin, calcium and folic acid supplement intake and CRC risk. Results: Of 744 carriers with CRC, 18%, 6% and 5% reported intake of multivitamin, calcium and folic acid supplements for at least 1 month, respectively, compared with 27%, 11% and 10% of 1222 carriers without CRC. After adjusting for identified confounding variables, a decreased CRC risk was associated with multivitam inintake for at least 3 years (HR 0.47, 95% CI 0.32-0.69) and calcium intake for at least 3 years(HR 0.42, 95% CI 0.23-0.74), compared with never users. There was no evidence of an association between folic acid supplement intake and CRC risk (P= 0.82). Conclusion: Intake of multivitamin and calcium supplements might be associated with a decreased risk of CRC for MMR gene mutation carriers.
    Article · Apr 2016 · International Journal of Epidemiology
  • Source
    [Show abstract] [Hide abstract] ABSTRACT: Objectives: Immunohistochemistry for DNA mismatch repair proteins is used to screen for Lynch syndrome in individuals with colorectal carcinoma (CRC). Although solitary loss of PMS2 expression is indicative of carrying a germline mutation in PMS2, previous studies reported MLH1 mutation in some cases. We determined the prevalence of MLH1 germline mutations in a large cohort of individuals with a CRC demonstrating solitary loss of PMS2 expression. Design: This cohort study included 88 individuals affected with a PMS2-deficient CRC from the Colon Cancer Family Registry Cohort. Germline PMS2 mutation analysis (long-range PCR and multiplex ligation-dependent probe amplification) was followed by MLH1 mutation testing (Sanger sequencing and multiplex ligation-dependent probe amplification). Results: Of the 66 individuals with complete mutation screening, we identified a pathogenic PMS2 mutation in 49 (74%), a pathogenic MLH1 mutation in 8 (12%) and a MLH1 variant of uncertain clinical significance predicted to be damaging by in silico analysis in 3 (4%); 6 (9%) carried variants likely to have no clinical significance. Missense point mutations accounted for most alterations (83%; 9/11) in MLH1. The MLH1 c.113A> G p.Asn38Ser mutation was found in 2 related individuals. One individual who carried the MLH1 intronic mutation c.677+3A>G p.Gln197Argfs*8 leading to the skipping of exon 8, developed 2 tumours, both of which retained MLH1 expression. Conclusions: A substantial proportion of CRCs with solitary loss of PMS2 expression are associated with a deleterious MLH1 germline mutation supporting the screening for MLH1 in individuals with tumours of this immunophenotype, when no PMS2 mutation has been identified.
    Full-text Article · Feb 2016 · BMJ Open
  • [Show abstract] [Hide abstract] ABSTRACT: Background: Menopausal hormone therapy (MHT) use has been consistently associated with a decreased risk of colorectal cancer (CRC) in women. Our aim was to use a genome-wide gene-environment interaction analysis to identify genetic modifiers of CRC risk associated with use of MHT. Methods: We included 10 835 postmenopausal women (5419 cases and 5416 controls) from 10 studies. We evaluated use of any MHT, oestrogen-only (E-only) and combined oestrogen-progestogen (E+P) hormone preparations. To test for multiplicative interactions, we applied the empirical Bayes (EB) test as well as the Wald test in conventional case-control logistic regression as primary tests. The Cocktail test was used as secondary test. Results: The EB test identified a significant interaction between rs964293 at 20q13.2/CYP24A1 and E+P (interaction OR (95% CIs)=0.61 (0.52-0.72), P=4.8 × 10(-9)). The secondary analysis also identified this interaction (Cocktail test OR=0.64 (0.52-0.78), P=1.2 × 10(-5) (alpha threshold=3.1 × 10(-4)). The ORs for association between E+P and CRC risk by rs964293 genotype were as follows: C/C, 0.96 (0.61-1.50); A/C, 0.61 (0.39-0.95) and A/A, 0.40 (0.22-0.73), respectively. Conclusions: Our results indicate that rs964293 modifies the association between E+P and CRC risk. The variant is located near CYP24A1, which encodes an enzyme involved in vitamin D metabolism. This novel finding offers additional insight into downstream pathways of CRC etiopathogenesis.British Journal of Cancer advance online publication, 14 January 2016; doi:10.1038/bjc.2015.443 www.bjcancer.com.
    Article · Jan 2016 · British Journal of Cancer
  • [Show abstract] [Hide abstract] ABSTRACT: This study was aimed to characterize the distribution of colorectal cancer risk using family history of cancers by data mining. Family histories for 10,066 colorectal cancer cases recruited to population cancer registries of the Colon Cancer Family Registry were analyzed using a data mining framework. A novel index was developed to quantify familial cancer aggregation. Artificial neural network was used to identify distinct categories of familial risk. Standardized incidence ratios (SIRs) and corresponding 95 % confidence intervals (CIs) of colorectal cancer were calculated for each category. We identified five major, and 66 minor categories of familial risk for developing colorectal cancer. The distribution the major risk categories were: (1) 7 % of families (SIR = 7.11; 95 % CI 6.65-7.59) had a strong family history of colorectal cancer; (2) 13 % of families (SIR = 2.94; 95 % CI 2.78-3.10) had a moderate family history of colorectal cancer; (3) 11 % of families (SIR = 1.23; 95 % CI 1.12-1.36) had a strong family history of breast cancer and a weak family history of colorectal cancer; (4) 9 % of families (SIR = 1.06; 95 % CI 0.96-1.18) had strong family history of prostate cancer and weak family history of colorectal cancer; and (5) 60 % of families (SIR = 0.61; 95 % CI 0.57-0.65) had a weak family history of all cancers. There is a wide variation of colorectal cancer risk that can be categorized by family history of cancer, with a strong gradient of colorectal cancer risk between the highest and lowest risk categories. The risk of colorectal cancer for people with the highest risk category of family history (7 % of the population) was 12-times that for people in the lowest risk category (60 %) of the population. Data mining was proven an effective approach for gaining insight into the underlying cancer aggregation patterns and for categorizing familial risk of colorectal cancer.
    Article · Dec 2015 · Familial Cancer
  • Dataset · Dec 2015
  • Dataset · Dec 2015
  • Source
    [Show abstract] [Hide abstract] ABSTRACT: High-risk mutations in several genes predispose to both colorectal cancer (CRC) and endometrial cancer (EC). We therefore hypothesised that some lower-risk genetic variants might also predispose to both CRC and EC. Using CRC and EC genome-wide association series, totalling 13,265 cancer cases and 40,245 controls, we found that the protective allele [G] at one previously-identified CRC polymorphism, rs2736100 near TERT, was associated with EC risk (odds ratio (OR) = 1.08, P = 0.000167); this polymorphism influences the risk of several other cancers. A further CRC polymorphism near TERC also showed evidence of association with EC (OR = 0.92; P = 0.03). Overall, however, there was no good evidence that the set of CRC polymorphisms was associated with EC risk, and neither of two previously-reported EC polymorphisms was associated with CRC risk. A combined analysis revealed one genome-wide significant polymorphism, rs3184504, on chromosome 12q24 (OR = 1.10, P = 7.23 × 10−9) with shared effects on CRC and EC risk. This polymorphism, a missense variant in the gene SH2B3, is also associated with haematological and autoimmune disorders, suggesting that it influences cancer risk through the immune response. Another polymorphism, rs12970291 near gene TSHZ1, was associated with both CRC and EC (OR = 1.26, P = 4.82 × 10−8), with the alleles showing opposite effects on the risks of the two cancers.
    Full-text Article · Dec 2015 · Scientific Reports
  • [Show abstract] [Hide abstract] ABSTRACT: Background: The CpG island methylator phenotype (CIMP) is a major molecular pathway in colorectal cancer. Approximately 25% to 60% of CIMP tumors are microsatellite unstable (MSI-H) due to DNA hypermethylation of the MLH1 gene promoter. Our aim was to determine if the distributions of clinicopathologic factors in CIMP-positive tumors with MLH1 DNA methylation differed from those in CIMP-positive tumors without DNA methylation of MLH1. Methods: We assessed the associations between age, sex, tumor-site, MSI status BRAF and KRAS mutations, and family colorectal cancer history with MLH1 methylation status in a large population-based sample of CIMP-positive colorectal cancers defined by a 5-marker panel using unconditional logistic regression to assess the odds of MLH1 methylation by study variables. Results: Subjects with CIMP-positive tumors without MLH1 methylation were significantly younger, more likely to be male, and more likely to have distal colon or rectal primaries and the MSI-L phenotype. CIMP-positive MLH1-unmethylated tumors were significantly less likely than CIMP-positive MLH1-methylated tumors to harbor a BRAF V600E mutation and significantly more likely to harbor a KRAS mutation. MLH1 methylation was associated with significantly better overall survival (HR, 0.50; 95% confidence interval, 0.31-0.82). Conclusions: These data suggest that MLH1 methylation in CIMP-positive tumors is not a completely random event and implies that there are environmental or genetic determinants that modify the probability that MLH1 will become methylated during CIMP pathogenesis. Impact: MLH1 DNA methylation status should be taken into account in etiologic studies.
    Article · Oct 2015 · Cancer Epidemiology Biomarkers & Prevention
  • Article · Oct 2015 · Nature Communications
  • [Show abstract] [Hide abstract] ABSTRACT: Over 50 loci associated with colorectal cancer (CRC) have been uncovered by genome-wide association studies (GWAS). Identifying additional loci has the potential to help elucidate aspects of the underlying biological processes leading to better understanding of the pathogenesis of the disease. We re-evaluated a GWAS by excluding controls that have family history of CRC or personal history of colorectal polyps, as we hypothesized that their inclusion reduces power to detect associations. This is supported empirically and through simulations. Two-phase GWAS analysis was performed in a total of 16,517 cases and 14,487 controls. We identified rs17094983, a SNP associated with risk of CRC [p = 2.5 × 10(-10); odds ratio estimated by re-including all controls (OR) = 0.87, 95 % confidence interval (CI) 0.83-0.91; minor allele frequency (MAF) = 13 %]. Results were replicated in samples of African descent (1894 cases and 4703 controls; p = 0.01; OR = 0.86, 95 % CI 0.77-0.97; MAF = 16 %). Gene expression data in 195 colon adenocarcinomas and 59 normal colon tissues from two different studies revealed that this locus has genotypes that are associated with RTN1 (Reticulon 1) expression (p = 0.001), a protein-coding gene involved in survival and proliferation of cancer cells which is highly expressed in normal colon tissues but has significantly reduced expression in tumor cells (p = 1.3 × 10(-8)).
    Article · Sep 2015 · Human Genetics
  • A. Win · J. Reece · J. Dowty · [...] · M. Jenkins
    Conference Paper · Sep 2015
  • [Show abstract] [Hide abstract] ABSTRACT: Individuals diagnosed with colorectal cancer (CRC) are at risk of developing a metachronous colorectal cancer. We examined the associations between personal, tumour-related and lifestyle risk factors, and risk of metachronous CRC. A total of 7,863 participants with incident colon or rectal cancer who were recruited in the USA, Canada and Australia to the Colon Cancer Family Registry during 1997-2012, except those identified as high-risk e.g. Lynch syndrome, were followed up approximately every 5 years. We estimated the risk of metachronous CRC, defined as the first new primary CRC following an interval of at least one year after the initial colorectal cancer diagnosis. Observation time started at the age at diagnosis of the initial CRC and ended at the age at diagnosis of the metachronous CRC, last contact or death whichever occurred earliest, or were censored at the age at diagnosis of any metachronous colorectal adenoma. Cox regression was used to derive hazard ratios (HRs) and 95% confidence intervals (CIs). During a mean follow-up of 6.6 years, 142 (1.81%) metachronous colorectal cancers were diagnosed (mean age at diagnosis 59.8; incidence 2.7/1000 person-years). An increased risk of metachronous CRC was associated with the presence of a synchronous CRC (HR=2.73; 95% CI: 1.30-5.72) and the location of cancer in the proximal colon at initial diagnosis (compared with distal colon or rectum, HR=4.16; 95% CI: 2.80-6.18). The presence of a synchronous CRC and the location of the initial CRC might be useful for deciding the intensity of surveillance colonoscopy for individuals diagnosed with CRC. This article is protected by copyright. All rights reserved.
    Conference Paper · Sep 2015
  • [Show abstract] [Hide abstract] ABSTRACT: Inheritance of a germline mutation in one of the DNA mismatch repair (MMR) genes MLH1, MSH2, MSH6, and PMS2 causes a high risk of colorectal and other cancers (Lynch Syndrome). Use of aspirin has been shown to be associated with a reduced risk of colorectal cancer for the general population as well as for MMR gene mutation carriers. The aim of this study was to determine whether use of aspirin and ibuprofen in a nontrial setting is associated with the risk of colorectal cancer risk for MMR gene mutation carriers. We included 1858 participants in the Colon Cancer Family Registry who had been found to have a pathogenic germline mutation in a MMR gene (carriers). We used weighted Cox proportional hazards regression to estimate hazard ratios (HRs) and 95% confidence intervals (CIs). All statistical tests were two-sided. A total of 714 carriers (38%) were diagnosed with colorectal cancer at a mean age of 42.4 (standard deviation 10.6) years. A reduced risk of colorectal cancer was associated with aspirin use (for 1 month to 4.9 years: HR = 0.49, 95% CI = 0.27 to 0.90, P = .02; for ≥5 years: HR = 0.25, 95% CI = 0.10 to 0.62, P = .003) and ibuprofen use (for 1 month to 4.9 years: HR = 0.38, 95% CI = 0.18 to 0.79, P = .009; for ≥5 years: HR = 0.26, 95% CI = 0.10 to 0.69, P = .007), compared with less than one month of use. Our results provide additional evidence that, for MMR gene mutation carriers, use of aspirin and ibuprofen might be effective in reducing their high risk of colorectal cancer. © The Author 2015. Published by Oxford University Press. All rights reserved. For Permissions, please e-mail: journals.permissions@oup.com.
    Article · Sep 2015 · Journal of the National Cancer Institute
  • [Show abstract] [Hide abstract] ABSTRACT: BACKGROUND & AIMS: African Americans (AAs) have the highest incidence and mortality of colorectal cancer (CRC) in the United States (US). Few data are available on genetic and non-genetic risk factors for CRC among AAs. Little is known about cancer risks and mutations in mismatch repair (MMR) genes in AAs with the most common inherited CRC syndrome, Lynch syndrome. We aimed to characterize phenotype, mutation spectrum, and risk of CRC in AAs with Lynch Syndrome. METHODS: We performed a retrospective study of AAs with mutations in MMR genes (MLH1, MSH2, MSH6, and PMS2) using databases from 13 US referral centers. We analyzed data on personal and family histories of cancer. Modified segregation analysis conditioned on ascertainment criteria was used to estimate age- and sex-specific CRC cumulative risk studying members of the mutation-carrying families. RESULTS: We identified 51 AA families with deleterious mutations that disrupt function of the MMR gene product: 31 in MLH1 (61%), 11 in MSH2 (21%), 3 in MSH6 (6%), and 6 in PMS2 (12%); 8 mutations were detected in more than 1 individual and 11 have not been previously reported. In the 920 members of the 51 families with deleterious mutations, the cumulative risks of CRC at an age of 80 y were estimated to be 36.2% (95% confidence interval [CI], 10.5%-83.9%) for men and 29.7% (95% CI, 8.31%-76.1%) for women. CRC risk was significantly higher among individuals with mutations in MLH1 or MSH2 (hazard ratio, 13.9; 95% CI, 3.44-56.5). CONCLUSIONS: We estimate the cumulative risk for CRC in AAs with MMR gene mutations to be similar to that of individuals of European descent with Lynch syndrome. Two-thirds of mutations were found in MLH1-some were found in multiple individuals and some have not been previously reported. Differences in the mutation spectrum are likely to reflect the genetic diversity of this population.
    Article · Aug 2015 · Gastroenterology
  • Source
    [Show abstract] [Hide abstract] ABSTRACT: The base excision repair protein, MUTYH, functionally interacts with the DNA mismatch repair (MMR) system. As genetic testing moves from testing one gene at a time, to gene panel and whole exome next generation sequencing approaches, understandin g the risk associated with co-existence of germline mutations in these genes will be important for clinical interpretation and management. From the Colon Cancer Family Registry, we identified 10 carriers who had both a MUTYH mutation (6 with c.1187G>A p.(Gly396Asp), 3 with c.821G>A p.(Arg274Gln), and 1 with c.536A>G p.(Tyr179Cys)) and a MMR gene mutation (3 in MLH1, 6 in MSH2, and 1 in PMS2), 375 carriers of a single (monoallelic) MUTYH mutation alone, and 469 carriers of a MMR gene mutation alone. Of the 10 carriers of both gene mutations, 8 were diagnosed with colorectal cancer. Using a weighted cohort analysis, we estimated that risk of colorectal cancer for carriers of both a MUTYH and a MMR gene mutation was substantially higher than that for carriers of a MUTYH mutation alone [hazard ratio (HR) 21.5, 95 % confidence interval (CI) 9.19-50.1; p < 0.001], but not different from that for carriers of a MMR gene mutation alone (HR 1.94, 95 % CI 0.63-5.99; p = 0.25). Within the limited power of this study, there was no evidence that a monoallelic MUTYH gene mutation confers additional risk of colorectal cancer for carriers of a MMR gene mutation alone. Our finding suggests MUTYH mutation testing in MMR gene mutation carriers is not clinically informative.
    Full-text Article · Jul 2015 · Familial Cancer

Publication Stats

19k Citations

Institutions

  • 2011
    • Queensland Institute of Medical Research
      • Cancer and Population Studies Group
      Brisbane, Queensland, Australia
  • 2009
    • Samuel Lunenfeld Research Institute
      Toronto, Ontario, Canada
  • 2008
    • Memorial Sloan-Kettering Cancer Center
      • Epidemiology & Biostatistics Group
      New York City, NY, United States
  • 2007
    • Geisel School of Medicine at Dartmouth
      • Department of Community and Family Medicine
      Hanover, New Hampshire, United States
  • 2002-2006
    • Columbia University
      • Department of Epidemiology
      New York, New York, United States
    • University of Southern California
      • Department of Preventive Medicine
      Los Ángeles, California, United States
    • University of Melbourne
      Melbourne, Victoria, Australia
  • 1998-2002
    • Harbor-UCLA Medical Center
      • Department of Pediatrics
      Torrance, California, United States
    • University of North Carolina at Chapel Hill
      North Carolina, United States
  • 1996
    • University of Toronto
      Toronto, Ontario, Canada
  • 1992-1994
    • University of California, Los Angeles
      • Department of Epidemiology
      Los Angeles, CA, United States