Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2023 Oct 2;14(1):6147.
doi: 10.1038/s41467-023-41819-0.

Combining Asian and European genome-wide association studies of colorectal cancer improves risk prediction across racial and ethnic populations

Minta Thomas  1 Yu-Ru Su  1   2 Elisabeth A Rosenthal  3 Lori C Sakoda  1   4 Stephanie L Schmit  5   6 Maria N Timofeeva  7   8 Zhishan Chen  9 Ceres Fernandez-Rozadilla  10   11 Philip J Law  12 Neil Murphy  13 Robert Carreras-Torres  14 Virginia Diez-Obrero  15   16   17 Franzel J B van Duijnhoven  18 Shangqing Jiang  1 Aesun Shin  19 Alicja Wolk  20 Amanda I Phipps  1   21 Andrea Burnett-Hartman  22 Andrea Gsur  23 Andrew T Chan  24   25   26   27   28   29 Ann G Zauber  30 Anna H Wu  31 Annika Lindblom  32   33 Caroline Y Um  34 Catherine M Tangen  35 Chris Gignoux  36 Christina Newton  34 Christopher A Haiman  37 Conghui Qu  1 D Timothy Bishop  38 Daniel D Buchanan  39   40   41 David R Crosslin  42 David V Conti  37 Dong-Hyun Kim  43 Elizabeth Hauser  44   45 Emily White  1   46 Erin Siegel  47 Fredrick R Schumacher  48 Gad Rennert  49   50 Graham G Giles  51 Heather Hampel  52 Hermann Brenner  53   54 Isao Oze  55 Jae Hwan Oh  56 Jeffrey K Lee  57   58 Jennifer L Schneider  4 Jenny Chang-Claude  59   60 Jeongseon Kim  61 Jeroen R Huyghe  1 Jiayin Zheng  1 Jochen Hampe  62 Joel Greenson  58 John L Hopper  63   64 Julie R Palmer  65 Kala Visvanathan  66 Keitaro Matsuo  67 Koichi Matsuda  68 Keum Ji Jung  69 Li Li  70 Loic Le Marchand  71 Ludmila Vodickova  72   73   74 Luis Bujanda  75 Marc J Gunter  13 Marco Matejcic  76 Mark A Jenkins  40   63 Martha L Slattery  77 Mauro D'Amato  78   79 Meilin Wang  80 Michael Hoffmeister  53 Michael O Woods  81 Michelle Kim  1 Mingyang Song  24   26   82 Motoki Iwasaki  83   84 Mulong Du  85   86 Natalia Udaltsova  4 Norie Sawada  84 Pavel Vodicka  72   73 Peter T Campbell  87 Polly A Newcomb  1 Qiuyin Cai  9 Rachel Pearlman  52 Rish K Pai  88 Robert E Schoen  89 Robert S Steinfelder  1 Robert W Haile  90 Rosita Vandenputtelaar  91 Ross L Prentice  1 Sébastien Küry  92 Sergi Castellví-Bel  93 Shoichiro Tsugane  84 Sonja I Berndt  94 Soo Chin Lee  95 Stefanie Brezina  23 Stephanie J Weinstein  94 Stephen J Chanock  94 Sun Ha Jee  96 Sun-Seog Kweon  97   98 Susan Vadaparampil  82 Tabitha A Harrison  1 Taiki Yamaji  83 Temitope O Keku  99 Veronika Vymetalkova  72   73 Volker Arndt  53 Wei-Hua Jia  100 Xiao-Ou Shu  101 Yi Lin  1 Yoon-Ok Ahn  19 Zsofia K Stadler  102 Bethany Van Guelpen  103   104 Cornelia M Ulrich  105 Elizabeth A Platz  66 John D Potter  1 Christopher I Li  1 Reinier Meester  91 Victor Moreno  106   107   108   109 Jane C Figueiredo  37   110 Graham Casey  111 Iris Lansdorp Vogelaar  91 Malcolm G Dunlop  8 Stephen B Gruber  112 Richard B Hayes  113 Paul D P Pharoah  114 Richard S Houlston  12 Gail P Jarvik  3 Ian P Tomlinson  11 Wei Zheng  9 Douglas A Corley  4   115 Ulrike Peters  116   117 Li Hsu  118   119
Affiliations

Combining Asian and European genome-wide association studies of colorectal cancer improves risk prediction across racial and ethnic populations

Minta Thomas et al. Nat Commun. .

Abstract

Polygenic risk scores (PRS) have great potential to guide precision colorectal cancer (CRC) prevention by identifying those at higher risk to undertake targeted screening. However, current PRS using European ancestry data have sub-optimal performance in non-European ancestry populations, limiting their utility among these populations. Towards addressing this deficiency, we expand PRS development for CRC by incorporating Asian ancestry data (21,731 cases; 47,444 controls) into European ancestry training datasets (78,473 cases; 107,143 controls). The AUC estimates (95% CI) of PRS are 0.63(0.62-0.64), 0.59(0.57-0.61), 0.62(0.60-0.63), and 0.65(0.63-0.66) in independent datasets including 1681-3651 cases and 8696-115,105 controls of Asian, Black/African American, Latinx/Hispanic, and non-Hispanic White, respectively. They are significantly better than the European-centric PRS in all four major US racial and ethnic groups (p-values < 0.05). Further inclusion of non-European ancestry populations, especially Black/African American and Latinx/Hispanic, is needed to improve the risk prediction and enhance equity in applying PRS in clinical practice.

PubMed Disclaimer

Conflict of interest statement

D.A.C. receives funds from NCI. K.V. receives related Research support from Cepheid and non-financial collaboration with Optra Health. L.B. is a consultant or has received research funding from Ikan Biotech. R.E.S. got research support from Immunovia, Freenome, Exact Sciences. Z.K.S.’s immediate family member serves as a consultant in Ophthalmology for Adverum, Genentech, Gyroscope Therapeutics Limited, Neurogene, Optos Plc, Outlook Therapeutics, RegenexBio, and Regeneron (outside the submitted work). The remaining authors declare no competing interests.

Figures

Fig. 1
Fig. 1. Distribution of PRS.
A PRS distributions varied across racial/ethnic groups, B PRS distribution after ancestry adjustment, C Additional mean adjustment for the Asian MG (Minor GWAS Japanese Study) study that has a different imputation panel, and D forest plot by racial and ethnic group for OR estimates +/−1.96 standard error of PRS per SD using N = 120,25; 9756; 10,377 and 80,542 of unrelated samples of Asian, Black or African American (AA), Hispanic and non-Hispanic White, respectively. The p-values in the table are two-sided. PRS is based on single cross-ancestry Asian-European PRS.
Fig. 2
Fig. 2. Relative Risk Estimation.
The relative risk of individuals at different percentiles of the single cross-ancestral Asian-European PRS compared to a population average odds ratio, stratified by race and ethnicity.
Fig. 3
Fig. 3. Relative Risk Calibration of PRS.
The relative risk calibration of PRS, stratified by race and ethnicity, using N = 120,25; 9756; 10,377 and 80,542 of unrelated samples of Asian, Black or AA, Hispanic and non-Hispanic White, respectively. The x-axis is the log-transformed predicted RR values and the y-axis is the log-transformed observed RR +/− 1.96 standard error with the middle bin (40–60) as the reference group.
Fig. 4
Fig. 4. Standardized net benefit analysis.
a Standardized net benefit for none, all, family history (FamHx) model, and FamHx+PRS model. For the FamHx and Famhx+PRS models b true- and false-positive rates, c number of high-risk, and d number of high risk participants developed CRC at different risk thresholds, in 22,628 participants aged 40–49 from the GERA cohort.
Fig. 5
Fig. 5. Approaches for deriving polygenic risk scores (PRS) for colorectal cancer.
Known Loci PRS and the details of the two different approaches for deriving PRS (1) PRS-CSx PRS and (2) LDpred PRS.

Update of

  • Combining Asian-European Genome-Wide Association Studies of Colorectal Cancer Improves Risk Prediction Across Race and Ethnicity.
    Thomas M, Su YR, Rosenthal EA, Sakoda LC, Schmit SL, Timofeeva MN, Chen Z, Fernandez-Rozadilla C, Law PJ, Murphy N, Carreras-Torres R, Diez-Obrero V, van Duijnhoven FJ, Jiang S, Shin A, Wolk A, Phipps AI, Burnett-Hartman A, Gsur A, Chan AT, Zauber AG, Wu AH, Lindblom A, Um CY, Tangen CM, Gignoux C, Newton C, Haiman CA, Qu C, Bishop DT, Buchanan DD, Crosslin DR, Conti DV, Kim DH, Hauser E, White E, Siegel E, Schumacher FR, Rennert G, Giles GG, Hampel H, Brenner H, Oze I, Oh JH, Lee JK, Schneider JL, Chang-Claude J, Kim J, Huyghe JR, Zheng J, Hampe J, Greenson J, Hopper JL, Palmer JR, Visvanathan K, Matsuo K, Matsuda K, Jung KJ, Li L, Marchand LL, Vodickova L, Bujanda L, Gunter MJ, Matejcic M, Jenkins MA, Slattery ML, D'Amato M, Wang M, Hoffmeister M, Woods MO, Kim M, Song M, Iwasaki M, Du M, Udaltsova N, Sawada N, Vodicka P, Campbell PT, Newcomb PA, Cai Q, Pearlman R, Pai RK, Schoen RE, Steinfelder RS, Haile RW, Vandenputtelaar R, Prentice RL, Küry S, Castellví-Bel S, Tsugane S, Berndt SI, Lee SC, Brezina S, Weinstein SJ, Chanock SJ, Jee SH, Kweon SS, Vadaparampil S, Harrison TA, Yamaji T, Keku TO, Vymetalkova V, Arndt V, Jia WH, Shu XO, Lin Y, Ahn YO, Stadler ZK, Van Guelpen B, Ul… See abstract for full author list ➔ Thomas M, et al. medRxiv [Preprint]. 2023 Jan 19:2023.01.19.23284737. doi: 10.1101/2023.01.19.23284737. medRxiv. 2023. Update in: Nat Commun. 2023 Oct 2;14(1):6147. doi: 10.1038/s41467-023-41819-0. PMID: 36789420 Free PMC article. Updated. Preprint.

References

    1. Murphy CC, et al. Decrease in incidence of colorectal cancer among individuals 50 years or older after recommendations for population-based screening. Clin. Gastroenterol. Hepatol. 2017;15:903–909.e6. - PMC - PubMed
    1. Hikino K, et al. Genome-wide association study of colorectal polyps identified highly overlapping polygenic architecture with colorectal cancer. J. Hum. Genet. 2022;67:149–156. - PubMed
    1. Thomas M, et al. Genome-wide modeling of polygenic risk score in colorectal cancer risk. Am. J. Hum. Genet. 2020;107:432–444. - PMC - PubMed
    1. Peterson RE, et al. Genome-wide association studies in ancestrally diverse populations: opportunities, methods, pitfalls, and recommendations. Cell. 2019;179:589–603. - PMC - PubMed
    1. Vassos E, et al. An examination of polygenic score risk prediction in individuals with first-episode psychosis. Biol. Psychiatry. 2017;81:470–477. - PubMed

Publication types