NWNaomi Wilcox

Naomi Wilcox

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Naomi Wilcox has a total of 83 co-authors. Here are the top co-authors by number of shared publications:

  • Joe Dennis

    Centre for Cancer Genetic Epidemiology, Department of Public Health and Primary Care, Strangeways Research Laboratory, University of Cambridge, Cambridge, UK

    6 shared publications
  • Douglas F. Easton

    Centre for Cancer Genetic Epidemiology, Department of Public Health and Primary Care, University of Cambridge, Cambridge, UK; Centre for Cancer Genetic Epidemiology, Department of Oncology, University of Cambridge, Cambridge, UK

    4 shared publications
  • Douglas F. Easton

    Centre for Cancer Genetic Epidemiology, Department of Public Health and Primary Care, University of Cambridge , Cambridge CB1 8RN

    4 shared publications
  • Eugene J. Gardner

    MRC Epidemiology Unit, Wellcome-MRC Institute of Metabolic Science, University of Cambridge, Cambridge, UK

    4 shared publications
  • Jonathan P. Tyrer

    Centre for Cancer Genetic Epidemiology, Department of Public Health and Primary Care, University of Cambridge, Cambridge, UK

    4 shared publications
  • Xin Yang

    Centre for Cancer Genetic Epidemiology, Department of Public Health and Primary Care, University of Cambridge, Cambridge, United Kingdom

    4 shared publications

Latest Publications

  • Open Access

    The contribution of coding variants to the heritability of multiple cancer types using UK Biobank whole-exome sequencing data

    by Naomi Wilcox, Jonathan P. Tyrer, Joe Dennis, Xin Yang, John R.B. Perry, Eugene J. Gardner, Douglas F. Easton


  • Open Access

    Using Family History Data to Improve the Power of Association Studies: Application to Cancer in UK Biobank

    by Naomi Wilcox, Jonathan P. Tyrer, Joe Dennis, Xin Yang, John R. B. Perry, Eugene J. Gardner, Douglas F. Easton

    In large cohort studies the number of unaffected individuals outnumbers the number of affected individuals, and the power can be low to detect associations for outcomes with low prevalence. We consider how including recorded family history in regression models increases the power to detect associations between genetic variants and disease risk. We show theoretically and using Monte‐Carlo simulations that including a family history of the disease, with a weighting of 0.5 compared with true cases, increases the power to detect associations. This is a powerful approach for detecting variants with moderate effects, but for larger effect sizes a weighting of > 0.5 can be more powerful. We illustrate this both for common variants and for exome sequencing data for over 400,000 individuals in UK Biobank to evaluate the association between the burden of protein‐truncating variants in genes and risk for four cancer types.


  • Open Access

    Evaluating the performance of the Breast and Ovarian Analysis of Disease Incidence Algorithm model in predicting 10-year breast cancer risks in UK Biobank

    by Carmen Petitjean, Naomi Wilcox, Lorenzo Ficorella, Joe Dennis, Jonathan Tyrer, Michael Lush, Jacques Simard, Douglas Easton, Antonis C Antoniou, Xin Yang

    Background: The Breast and Ovarian Analysis of Disease Incidence Algorithm (BOADICEA) model predicts breast cancer risk using cancer family history, epidemiological, and genetic data. We evaluated its validity in a large prospective cohort. Methods: We assessed model calibration, discrimination and risk classification ability in 217 885 women (6838 incident breast cancers) aged 40-70 years of self-reported White ethnicity with no previous cancer from the UK Biobank. Age-specific risk classification was assessed using relative risk thresholds equivalent to the absolute lifetime risk categories of less than 17%, 17%-30%, and 30% or more, recommended by the National Institute for Health and Care Excellence guidelines. We predicted 10-year risks using BOADICEA v.6 considering cancer family history, questionnaire-based risk factors, a 313–single nucleotide polymorphisms polygenic score, and pathogenic variants. Mammographic density data were not available. Results: The polygenic risk score was the most discriminative risk factor (area under the curve [AUC] = 0.65). Discrimination was highest when considering all risk factors (AUC = 0.66). The model was well calibrated overall (expected-to-observed ratio = 0.99, 95% confidence interval [CI] = 0.97 to 1.02; calibration slope = 0.99, 95% CI = 0.99 to 1.00), and in deciles of predicted risks. Discrimination was similar in women aged younger and older than 50 years. There was some underprediction in women aged younger than 50 years (expected-to-observed ratio = 0.89, 95% CI = 0.84 to 0.94; calibration slope = 0.96, 95% CI = 0.94 to 0.97), which was explained by the higher breast cancer incidence in UK Biobank than the UK population incidence in this age group. The model classified 87.2%, 11.4%, and 1.4% of women in relative risk categories less than 1.6, 1.6-3.1, and at least 3.1, identifying 25.6% of incident breast cancer patients in category relative risk of at least 1.6. Conclusion: BOADICEA, implemented in CanRisk (www.canrisk.org), provides valid 10-year breast cancer risk, which can facilitate risk-stratified screening and personalized breast cancer risk management.


  • Open Access

    The contribution of coding variants to the heritability of multiple cancer types using UK Biobank whole-exome sequencing data

    by Naomi Alice Wilcox, Jonathan P. Tyrer, Joe Dennis, Xin Yang, John R. B. Perry, Eugene J. Gardner, Douglas F. Easton

    Genome-wide association studies have been highly successful at identifying common variants associated with cancer, however, they do not explain all the inherited risk of cancer. Family-based studies, targeted sequencing and, more recently, exome-wide association studies have identified rare coding variants in some genes associated with cancer risk, but the overall contribution of these variants to the heritability of cancer is less clear. Here we describe a method to estimate the genome-wide contribution of rare coding variants to heritability that fits models to the burden effect sizes using an empirical Bayesian approach. We apply this method to the burden of protein-truncating variants in over 15,000 genes for 11 cancers in UK Biobank, using whole-exome sequencing data on over 400,000 individuals. We extend the method to consider the overlap of genes contributing to pairs of cancers. We found ovarian cancer to have the greatest proportion of heritability attributable to protein-truncating variants in genes (46%). The joint cancer models highlight significant clustering of cancer types, including a near complete overlap in susceptibility genes for breast, ovarian, prostate and pancreatic cancer. Our results provide insights into the contribution of rare coding variants to the heritability of cancer and identify additional genes with strong evidence of susceptibility to multiple cancer types.

  • Open Access

    Protein-truncating and rare missense variants in ATM and CHEK2 and associations with cancer in UK Biobank whole-exome sequence data

    by Toqir K Mukhtar, Naomi Wilcox, Joe Dennis, Xin Yang, Marc Naven, Nasim Mavaddat, John R B Perry, Eugene Gardner, Douglas F Easton

    Background: Deleterious germline variants in ATM and CHEK2 have been associated with a moderately increased risk of breast cancer. Risks for other cancers remain unclear. Methods: Cancer associations for coding variants in ATM and CHEK2 were evaluated using whole-exome sequence data from UK Biobank linked to cancer registration data (348 488 participants), and analysed both as a retrospective case-control and a prospective cohort study. Odds ratios, hazard ratios, and combined relative risks (RRs) were estimated by cancer type and gene. Separate analyses were performed for protein-truncating variants (PTVs) and rare missense variants (rMSVs; allele frequency ATM were associated with increased risks of nine cancers at pCHEK2 were associated with three cancers at pATM and CHEK2 are associated with a wide range of cancers, with the highest RR for pancreatic cancer in ATM PTV carriers. These findings can inform genetic counselling of carriers.


  • Open Access

    Analysis of variants in untranslated and promoter regions and breast cancer risk using whole genome sequencing data

    by Naomi Wilcox, Jonathan P. Tyrer, Leila Dorling, Joe Dennis, Marc Naven, Mustapha Abubakar, Thomas U. Ahearn, Irene L. Andrulis, Antonis C. Antoniou, Natalia V. Bogdanova, et al.

    Recent exome-wide association studies have explored the role of coding variants in breast cancer risk, highlighting the role of rare variants in multiple genes including BRCA1, BRCA2, CHEK2, ATM and PALB2 , as well as new susceptibility genes e.g., MAP3K1 . These genes, however, explain a small proportion of the missing heritability of the disease. Much of the missing heritability likely lies in the non-coding genome. We evaluated the role of rare variants in the 5’ and 3’ untranslated regions (UTRs) of 18,676 genes, and 35,201 putative promoter regions, using whole-genome sequencing data from UK Biobank on 8,001 women with breast cancer and 92,534 women without breast cancer. Burden tests and SKAT-O tests were performed in UTR and promoter regions. For UTR regions of 35 putative breast cancer susceptibility genes, we additionally performed a meta-analysis with a large breast cancer case-control dataset. Associations for 8 regions at PCDK5R1 (P=8.5×10 −7 ). These results highlight the potential role of non-coding regulatory regions in breast cancer susceptibility.

  • Open Access

    Using Family History Data to Improve the Power of Association Studies: Application to Cancer in UK Biobank

    by Naomi Wilcox, Jonathan P. Tyrer, Joe Dennis, Xin Yang, John R. B. Perry, Eugene J. Gardner, Douglas F. Easton

    In large cohort studies the number of unaffected individuals outnumbers the number of affected individuals, and the power can be low to detect associations for outcomes with low prevalence. We consider how including recorded family history in regression models increases the power to detect associations between genetic variants and disease risk. We show theoretically and using Monte-Carlo simulations that including family history of the disease, with a weighting of 0.5 compared to true cases, increases the power to detect associations. This is a powerful approach for detecting variants with moderate effects, but for larger effect sizes a weighting of >0.5 can be more powerful. We illustrate this both for common variants and for exome sequencing data for over 400,000 individuals in UK Biobank to evaluate the association between the burden of protein-truncating variants in genes and risk for 4 cancer types.


  • Open Access

    CanRisk-Prostate: A Comprehensive, Externally Validated Risk Model for the Prediction of Future Prostate Cancer

    by Tommy Nyberg, Mark N. Brook, Lorenzo Ficorella, Andrew Lee, Joe Dennis, Xin Yang, Naomi Wilcox, Tokhir Dadaev, Koveela Govindasami, Michael Lush, et al.

    PURPOSE: Prostate cancer (PCa) is highly heritable. No validated PCa risk model currently exists. We therefore sought to develop a genetic risk model that can provide personalized predicted PCa risks on the basis of known moderate- to high-risk pathogenic variants, low-risk common genetic variants, and explicit cancer family history, and to externally validate the model in an independent prospective cohort.MATERIALS AND METHODS: We developed a risk model using a kin-cohort comprising individuals from 16,633 PCa families ascertained in the United Kingdom from 1993 to 2017 from the UK Genetic Prostate Cancer Study, and complex segregation analysis adjusting for ascertainment. The model was externally validated in 170,850 unaffected men (7,624 incident PCas) recruited from 2006 to 2010 to the independent UK Biobank prospective cohort study.RESULTS: The most parsimonious model included the effects of pathogenic variants in BRCA2, HOXB13, and BRCA1, and a polygenic score on the basis of 268 common low-risk variants. Residual familial risk was modeled by a hypothetical recessively inherited variant and a polygenic component whose standard deviation decreased log-linearly with age. The model predicted familial risks that were consistent with those reported in previous observational studies. In the validation cohort, the model discriminated well between unaffected men and men with incident PCas within 5 years (C-index, 0.790; 95% CI, 0.783 to 0.797) and 10 years (C-index, 0.772; 95% CI, 0.768 to 0.777). The 50% of men with highest predicted risks captured 86.3% of PCa cases within 10 years.CONCLUSION: To our knowledge, this is the first validated risk model offering personalized PCa risks. The model will assist in counseling men concerned about their risk and can facilitate future risk-stratified population screening approaches.