Polygenic Risk Scores for Prediction of Breast Cancer and Breast Cancer Subtypes

NBCS Collaborators, ABCTB Investigators, kConFab/AOCS Investigators

Research output: Contribution to journalArticle

Abstract

Stratification of women according to their risk of breast cancer based on polygenic risk scores (PRSs) could improve screening and prevention strategies. Our aim was to develop PRSs, optimized for prediction of estrogen receptor (ER)-specific disease, from the largest available genome-wide association dataset and to empirically validate the PRSs in prospective studies. The development dataset comprised 94,075 case subjects and 75,017 control subjects of European ancestry from 69 studies, divided into training and validation sets. Samples were genotyped using genome-wide arrays, and single-nucleotide polymorphisms (SNPs) were selected by stepwise regression or lasso penalized regression. The best performing PRSs were validated in an independent test set comprising 11,428 case subjects and 18,323 control subjects from 10 prospective studies and 190,040 women from UK Biobank (3,215 incident breast cancers). For the best PRSs (313 SNPs), the odds ratio for overall disease per 1 standard deviation in ten prospective studies was 1.61 (95%CI: 1.57–1.65) with area under receiver-operator curve (AUC) = 0.630 (95%CI: 0.628–0.651). The lifetime risk of overall breast cancer in the top centile of the PRSs was 32.6%. Compared with women in the middle quintile, those in the highest 1% of risk had 4.37- and 2.78-fold risks, and those in the lowest 1% of risk had 0.16- and 0.27-fold risks, of developing ER-positive and ER-negative disease, respectively. Goodness-of-fit tests indicated that this PRS was well calibrated and predicts disease risk accurately in the tails of the distribution. This PRS is a powerful and reliable predictor of breast cancer risk that may improve breast cancer prevention programs.

Original languageEnglish (US)
Pages (from-to)21-34
Number of pages14
JournalAmerican Journal of Human Genetics
Volume104
Issue number1
DOIs
StatePublished - Jan 3 2019

Fingerprint

Breast Neoplasms
Estrogen Receptors
Prospective Studies
Single Nucleotide Polymorphism
Genome
Area Under Curve
Odds Ratio

Keywords

  • breast
  • cancer
  • epidemiology
  • genetic
  • polygenic
  • prediction
  • risk
  • score
  • screening
  • stratification

ASJC Scopus subject areas

  • Genetics
  • Genetics(clinical)

Cite this

Polygenic Risk Scores for Prediction of Breast Cancer and Breast Cancer Subtypes. / NBCS Collaborators; ABCTB Investigators; kConFab/AOCS Investigators.

In: American Journal of Human Genetics, Vol. 104, No. 1, 03.01.2019, p. 21-34.

Research output: Contribution to journalArticle

NBCS Collaborators, ABCTB Investigators & kConFab/AOCS Investigators 2019, 'Polygenic Risk Scores for Prediction of Breast Cancer and Breast Cancer Subtypes', American Journal of Human Genetics, vol. 104, no. 1, pp. 21-34. https://doi.org/10.1016/j.ajhg.2018.11.002
NBCS Collaborators ; ABCTB Investigators ; kConFab/AOCS Investigators. / Polygenic Risk Scores for Prediction of Breast Cancer and Breast Cancer Subtypes. In: American Journal of Human Genetics. 2019 ; Vol. 104, No. 1. pp. 21-34.
@article{582b522d82f94ee785c6fa4f7504fa8f,
title = "Polygenic Risk Scores for Prediction of Breast Cancer and Breast Cancer Subtypes",
abstract = "Stratification of women according to their risk of breast cancer based on polygenic risk scores (PRSs) could improve screening and prevention strategies. Our aim was to develop PRSs, optimized for prediction of estrogen receptor (ER)-specific disease, from the largest available genome-wide association dataset and to empirically validate the PRSs in prospective studies. The development dataset comprised 94,075 case subjects and 75,017 control subjects of European ancestry from 69 studies, divided into training and validation sets. Samples were genotyped using genome-wide arrays, and single-nucleotide polymorphisms (SNPs) were selected by stepwise regression or lasso penalized regression. The best performing PRSs were validated in an independent test set comprising 11,428 case subjects and 18,323 control subjects from 10 prospective studies and 190,040 women from UK Biobank (3,215 incident breast cancers). For the best PRSs (313 SNPs), the odds ratio for overall disease per 1 standard deviation in ten prospective studies was 1.61 (95{\%}CI: 1.57–1.65) with area under receiver-operator curve (AUC) = 0.630 (95{\%}CI: 0.628–0.651). The lifetime risk of overall breast cancer in the top centile of the PRSs was 32.6{\%}. Compared with women in the middle quintile, those in the highest 1{\%} of risk had 4.37- and 2.78-fold risks, and those in the lowest 1{\%} of risk had 0.16- and 0.27-fold risks, of developing ER-positive and ER-negative disease, respectively. Goodness-of-fit tests indicated that this PRS was well calibrated and predicts disease risk accurately in the tails of the distribution. This PRS is a powerful and reliable predictor of breast cancer risk that may improve breast cancer prevention programs.",
keywords = "breast, cancer, epidemiology, genetic, polygenic, prediction, risk, score, screening, stratification",
author = "{NBCS Collaborators} and {ABCTB Investigators} and {kConFab/AOCS Investigators} and Nasim Mavaddat and Kyriaki Michailidou and Joe Dennis and Michael Lush and Laura Fachal and Andrew Lee and Tyrer, {Jonathan P.} and Chen, {Ting Huei} and Qin Wang and Bolla, {Manjeet K.} and Xin Yang and Adank, {Muriel A.} and Thomas Ahearn and Kristiina Aittom{\"a}ki and Jamie Allen and Andrulis, {Irene L.} and Hoda Anton-Culver and Antonenkova, {Natalia N.} and Volker Arndt and Aronson, {Kristan J.} and Auer, {Paul L.} and P{\"a}ivi Auvinen and Myrto Barrdahl and {Beane Freeman}, {Laura E.} and Beckmann, {Matthias W.} and Sabine Behrens and Javier Benitez and Marina Bermisheva and Leslie Bernstein and Carl Blomqvist and Bogdanova, {Natalia V.} and Bojesen, {Stig E.} and Bernardo Bonanni and B{\o}rresen-Dale, {Anne Lise} and Hiltrud Brauch and Michael Bremer and Hermann Brenner and Adam Brentnall and Brock, {Ian W.} and Angela Brooks-Wilson and Brucker, {Sara Y.} and Thomas Br{\"u}ning and Barbara Burwinkel and Daniele Campa and Carter, {Brian D.} and Castelao, {Jose E.} and Chanock, {Stephen J.} and Rowan Chlebowski and Hans Christiansen and Nilanjan Chatterjee",
year = "2019",
month = "1",
day = "3",
doi = "10.1016/j.ajhg.2018.11.002",
language = "English (US)",
volume = "104",
pages = "21--34",
journal = "American Journal of Human Genetics",
issn = "0002-9297",
publisher = "Cell Press",
number = "1",

}

TY - JOUR

T1 - Polygenic Risk Scores for Prediction of Breast Cancer and Breast Cancer Subtypes

AU - NBCS Collaborators

AU - ABCTB Investigators

AU - kConFab/AOCS Investigators

AU - Mavaddat, Nasim

AU - Michailidou, Kyriaki

AU - Dennis, Joe

AU - Lush, Michael

AU - Fachal, Laura

AU - Lee, Andrew

AU - Tyrer, Jonathan P.

AU - Chen, Ting Huei

AU - Wang, Qin

AU - Bolla, Manjeet K.

AU - Yang, Xin

AU - Adank, Muriel A.

AU - Ahearn, Thomas

AU - Aittomäki, Kristiina

AU - Allen, Jamie

AU - Andrulis, Irene L.

AU - Anton-Culver, Hoda

AU - Antonenkova, Natalia N.

AU - Arndt, Volker

AU - Aronson, Kristan J.

AU - Auer, Paul L.

AU - Auvinen, Päivi

AU - Barrdahl, Myrto

AU - Beane Freeman, Laura E.

AU - Beckmann, Matthias W.

AU - Behrens, Sabine

AU - Benitez, Javier

AU - Bermisheva, Marina

AU - Bernstein, Leslie

AU - Blomqvist, Carl

AU - Bogdanova, Natalia V.

AU - Bojesen, Stig E.

AU - Bonanni, Bernardo

AU - Børresen-Dale, Anne Lise

AU - Brauch, Hiltrud

AU - Bremer, Michael

AU - Brenner, Hermann

AU - Brentnall, Adam

AU - Brock, Ian W.

AU - Brooks-Wilson, Angela

AU - Brucker, Sara Y.

AU - Brüning, Thomas

AU - Burwinkel, Barbara

AU - Campa, Daniele

AU - Carter, Brian D.

AU - Castelao, Jose E.

AU - Chanock, Stephen J.

AU - Chlebowski, Rowan

AU - Christiansen, Hans

AU - Chatterjee, Nilanjan

PY - 2019/1/3

Y1 - 2019/1/3

N2 - Stratification of women according to their risk of breast cancer based on polygenic risk scores (PRSs) could improve screening and prevention strategies. Our aim was to develop PRSs, optimized for prediction of estrogen receptor (ER)-specific disease, from the largest available genome-wide association dataset and to empirically validate the PRSs in prospective studies. The development dataset comprised 94,075 case subjects and 75,017 control subjects of European ancestry from 69 studies, divided into training and validation sets. Samples were genotyped using genome-wide arrays, and single-nucleotide polymorphisms (SNPs) were selected by stepwise regression or lasso penalized regression. The best performing PRSs were validated in an independent test set comprising 11,428 case subjects and 18,323 control subjects from 10 prospective studies and 190,040 women from UK Biobank (3,215 incident breast cancers). For the best PRSs (313 SNPs), the odds ratio for overall disease per 1 standard deviation in ten prospective studies was 1.61 (95%CI: 1.57–1.65) with area under receiver-operator curve (AUC) = 0.630 (95%CI: 0.628–0.651). The lifetime risk of overall breast cancer in the top centile of the PRSs was 32.6%. Compared with women in the middle quintile, those in the highest 1% of risk had 4.37- and 2.78-fold risks, and those in the lowest 1% of risk had 0.16- and 0.27-fold risks, of developing ER-positive and ER-negative disease, respectively. Goodness-of-fit tests indicated that this PRS was well calibrated and predicts disease risk accurately in the tails of the distribution. This PRS is a powerful and reliable predictor of breast cancer risk that may improve breast cancer prevention programs.

AB - Stratification of women according to their risk of breast cancer based on polygenic risk scores (PRSs) could improve screening and prevention strategies. Our aim was to develop PRSs, optimized for prediction of estrogen receptor (ER)-specific disease, from the largest available genome-wide association dataset and to empirically validate the PRSs in prospective studies. The development dataset comprised 94,075 case subjects and 75,017 control subjects of European ancestry from 69 studies, divided into training and validation sets. Samples were genotyped using genome-wide arrays, and single-nucleotide polymorphisms (SNPs) were selected by stepwise regression or lasso penalized regression. The best performing PRSs were validated in an independent test set comprising 11,428 case subjects and 18,323 control subjects from 10 prospective studies and 190,040 women from UK Biobank (3,215 incident breast cancers). For the best PRSs (313 SNPs), the odds ratio for overall disease per 1 standard deviation in ten prospective studies was 1.61 (95%CI: 1.57–1.65) with area under receiver-operator curve (AUC) = 0.630 (95%CI: 0.628–0.651). The lifetime risk of overall breast cancer in the top centile of the PRSs was 32.6%. Compared with women in the middle quintile, those in the highest 1% of risk had 4.37- and 2.78-fold risks, and those in the lowest 1% of risk had 0.16- and 0.27-fold risks, of developing ER-positive and ER-negative disease, respectively. Goodness-of-fit tests indicated that this PRS was well calibrated and predicts disease risk accurately in the tails of the distribution. This PRS is a powerful and reliable predictor of breast cancer risk that may improve breast cancer prevention programs.

KW - breast

KW - cancer

KW - epidemiology

KW - genetic

KW - polygenic

KW - prediction

KW - risk

KW - score

KW - screening

KW - stratification

UR - http://www.scopus.com/inward/record.url?scp=85059498503&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85059498503&partnerID=8YFLogxK

U2 - 10.1016/j.ajhg.2018.11.002

DO - 10.1016/j.ajhg.2018.11.002

M3 - Article

VL - 104

SP - 21

EP - 34

JO - American Journal of Human Genetics

JF - American Journal of Human Genetics

SN - 0002-9297

IS - 1

ER -