A Genome-Wide Association Study Meta-Analysis of Alpha Angle Suggests Cam-Type Morphology May Be a Specific Feature of Hip Osteoarthritis in Older Adults

Benjamin G Faber; Monika Frysz; April E Hartley; Raja Ebsim; Cindy G Boer; Fiona R Saunders; Jennifer S Gregory; Richard M Aspden; Nicholas C Harvey; Lorraine Southam; William Giles; Christine L Le Maitre; J Mark Wilkinson; Joyce B J van Meurs; Eleftheria Zeggini; Timothy Cootes; Claudia Lindner; John P Kemp; George Davey Smith; Jonathan H Tobias

doi:10.1002/art.42451

A Genome-Wide Association Study Meta-Analysis of Alpha Angle Suggests Cam-Type Morphology May Be a Specific Feature of Hip Osteoarthritis in Older Adults

Benjamin G Faber, Monika Frysz, April E Hartley, Raja Ebsim, Cindy G Boer, Fiona R Saunders, Jennifer S Gregory, Richard M Aspden, Nicholas C Harvey, Lorraine Southam, William Giles, Christine L Le Maitre, J Mark Wilkinson, Joyce B J van Meurs, Eleftheria Zeggini, Timothy Cootes, Claudia Lindner, John P Kemp, George Davey Smith, Jonathan H Tobias

Division of Informatics, Imaging & Data Sciences (L5)

Research output: Contribution to journal › Article › peer-review

Abstract

OBJECTIVES: To examine the genetic architecture of cam morphology, using alpha angle (AA) as a proxy measure, we conducted an AA genome wide association study (GWAS), followed by Mendelian randomisation (MR) to evaluate its causal relationship with hip osteoarthritis (HOA).

METHODS: Observational analyses examined associations between AA derived from hip DXA scans in UK Biobank (UKB), and radiographic HOA (rHOA) and subsequent total hip replacement (THR). Afterwards, an AA GWAS meta-analysis was performed (n=44,214), using AA previously derived in the Rotterdam Study (RS). Linkage disequilibrium score regression assessed the genetic correlation between AA and HOA. Genetic associations with P<5x10 -8 instrumented AA for two-sample MR.

RESULTS: DXA-derived AA showed expected associations between AA and rHOA (OR 1.63 [95% CI 1.58-1.67]) and THR (HR 1.45 [1.33-1.59]) in UKB. The heritability of AA was 10% and AA had a moderate genetic correlation with HOA (r g =0.26 [0.10-0.43]). Eight independent genetic signals were associated with AA. Two-sample MR provided weak evidence of causal effects of AA on HOA risk (inverse variance weighted (IVW): OR=1.84 [1.14-2.96], P 0.01). In contrast, genetic predisposition for HOA had stronger evidence of a causal effect on increased AA (IVW: β=0.09 [0.04-0.13], P 4.58 x 10 -05 ).

CONCLUSIONS: Expected observational associations between AA and related clinical outcomes provided face-validity for the DXA-derived AA measures. Evidence of bidirectional associations between AA and HOA, particularly in the reverse direction, suggests that hip shape modelling secondary to a genetic predisposition to HOA contributes to the well-established relationship between HOA and cam morphology in older adults.

Original language	English
Pages (from-to)	900-909
Journal	Arthritis & rheumatology (Hoboken, N.J.)
Volume	75
Issue number	6
Early online date	20 Jan 2023
DOIs	https://doi.org/10.1002/art.42451
Publication status	Published - 1 Jun 2023

Access to Document

10.1002/art.42451Licence: CC BY

Cite this

Faber, B. G., Frysz, M., Hartley, A. E., Ebsim, R., Boer, C. G., Saunders, F. R., Gregory, J. S., Aspden, R. M., Harvey, N. C., Southam, L., Giles, W., Le Maitre, C. L., Wilkinson, J. M., van Meurs, J. B. J., Zeggini, E., Cootes, T., Lindner, C., Kemp, J. P., Davey Smith, G., & Tobias, J. H. (2023). A Genome-Wide Association Study Meta-Analysis of Alpha Angle Suggests Cam-Type Morphology May Be a Specific Feature of Hip Osteoarthritis in Older Adults. Arthritis & rheumatology (Hoboken, N.J.), 75(6), 900-909. Advance online publication. https://doi.org/10.1002/art.42451

@article{6adb90acf6f047eba115ddbae1f27ed4,

title = "A Genome-Wide Association Study Meta-Analysis of Alpha Angle Suggests Cam-Type Morphology May Be a Specific Feature of Hip Osteoarthritis in Older Adults",

abstract = "OBJECTIVES: To examine the genetic architecture of cam morphology, using alpha angle (AA) as a proxy measure, we conducted an AA genome wide association study (GWAS), followed by Mendelian randomisation (MR) to evaluate its causal relationship with hip osteoarthritis (HOA).METHODS: Observational analyses examined associations between AA derived from hip DXA scans in UK Biobank (UKB), and radiographic HOA (rHOA) and subsequent total hip replacement (THR). Afterwards, an AA GWAS meta-analysis was performed (n=44,214), using AA previously derived in the Rotterdam Study (RS). Linkage disequilibrium score regression assessed the genetic correlation between AA and HOA. Genetic associations with P<5x10 -8 instrumented AA for two-sample MR. RESULTS: DXA-derived AA showed expected associations between AA and rHOA (OR 1.63 [95% CI 1.58-1.67]) and THR (HR 1.45 [1.33-1.59]) in UKB. The heritability of AA was 10% and AA had a moderate genetic correlation with HOA (r g =0.26 [0.10-0.43]). Eight independent genetic signals were associated with AA. Two-sample MR provided weak evidence of causal effects of AA on HOA risk (inverse variance weighted (IVW): OR=1.84 [1.14-2.96], P 0.01). In contrast, genetic predisposition for HOA had stronger evidence of a causal effect on increased AA (IVW: β=0.09 [0.04-0.13], P 4.58 x 10 -05 ). CONCLUSIONS: Expected observational associations between AA and related clinical outcomes provided face-validity for the DXA-derived AA measures. Evidence of bidirectional associations between AA and HOA, particularly in the reverse direction, suggests that hip shape modelling secondary to a genetic predisposition to HOA contributes to the well-established relationship between HOA and cam morphology in older adults.",

author = "Faber, {Benjamin G} and Monika Frysz and Hartley, {April E} and Raja Ebsim and Boer, {Cindy G} and Saunders, {Fiona R} and Gregory, {Jennifer S} and Aspden, {Richard M} and Harvey, {Nicholas C} and Lorraine Southam and William Giles and {Le Maitre}, {Christine L} and Wilkinson, {J Mark} and {van Meurs}, {Joyce B J} and Eleftheria Zeggini and Timothy Cootes and Claudia Lindner and Kemp, {John P} and {Davey Smith}, George and Tobias, {Jonathan H}",

year = "2023",

month = jun,

day = "1",

doi = "10.1002/art.42451",

language = "English",

volume = "75",

pages = "900--909",

journal = "Arthritis & rheumatology (Hoboken, N.J.)",

issn = "2326-5191",

publisher = "John Wiley & Sons Ltd",

number = "6",

}

Faber, BG, Frysz, M, Hartley, AE, Ebsim, R, Boer, CG, Saunders, FR, Gregory, JS, Aspden, RM, Harvey, NC, Southam, L, Giles, W, Le Maitre, CL, Wilkinson, JM, van Meurs, JBJ, Zeggini, E, Cootes, T , Lindner, C, Kemp, JP, Davey Smith, G & Tobias, JH 2023, 'A Genome-Wide Association Study Meta-Analysis of Alpha Angle Suggests Cam-Type Morphology May Be a Specific Feature of Hip Osteoarthritis in Older Adults', Arthritis & rheumatology (Hoboken, N.J.), vol. 75, no. 6, pp. 900-909. https://doi.org/10.1002/art.42451

TY - JOUR

T1 - A Genome-Wide Association Study Meta-Analysis of Alpha Angle Suggests Cam-Type Morphology May Be a Specific Feature of Hip Osteoarthritis in Older Adults

AU - Faber, Benjamin G

AU - Frysz, Monika

AU - Hartley, April E

AU - Ebsim, Raja

AU - Boer, Cindy G

AU - Saunders, Fiona R

AU - Gregory, Jennifer S

AU - Aspden, Richard M

AU - Harvey, Nicholas C

AU - Southam, Lorraine

AU - Giles, William

AU - Le Maitre, Christine L

AU - Wilkinson, J Mark

AU - van Meurs, Joyce B J

AU - Zeggini, Eleftheria

AU - Cootes, Timothy

AU - Lindner, Claudia

AU - Kemp, John P

AU - Davey Smith, George

AU - Tobias, Jonathan H

PY - 2023/6/1

Y1 - 2023/6/1

N2 - OBJECTIVES: To examine the genetic architecture of cam morphology, using alpha angle (AA) as a proxy measure, we conducted an AA genome wide association study (GWAS), followed by Mendelian randomisation (MR) to evaluate its causal relationship with hip osteoarthritis (HOA).METHODS: Observational analyses examined associations between AA derived from hip DXA scans in UK Biobank (UKB), and radiographic HOA (rHOA) and subsequent total hip replacement (THR). Afterwards, an AA GWAS meta-analysis was performed (n=44,214), using AA previously derived in the Rotterdam Study (RS). Linkage disequilibrium score regression assessed the genetic correlation between AA and HOA. Genetic associations with P<5x10 -8 instrumented AA for two-sample MR. RESULTS: DXA-derived AA showed expected associations between AA and rHOA (OR 1.63 [95% CI 1.58-1.67]) and THR (HR 1.45 [1.33-1.59]) in UKB. The heritability of AA was 10% and AA had a moderate genetic correlation with HOA (r g =0.26 [0.10-0.43]). Eight independent genetic signals were associated with AA. Two-sample MR provided weak evidence of causal effects of AA on HOA risk (inverse variance weighted (IVW): OR=1.84 [1.14-2.96], P 0.01). In contrast, genetic predisposition for HOA had stronger evidence of a causal effect on increased AA (IVW: β=0.09 [0.04-0.13], P 4.58 x 10 -05 ). CONCLUSIONS: Expected observational associations between AA and related clinical outcomes provided face-validity for the DXA-derived AA measures. Evidence of bidirectional associations between AA and HOA, particularly in the reverse direction, suggests that hip shape modelling secondary to a genetic predisposition to HOA contributes to the well-established relationship between HOA and cam morphology in older adults.

AB - OBJECTIVES: To examine the genetic architecture of cam morphology, using alpha angle (AA) as a proxy measure, we conducted an AA genome wide association study (GWAS), followed by Mendelian randomisation (MR) to evaluate its causal relationship with hip osteoarthritis (HOA).METHODS: Observational analyses examined associations between AA derived from hip DXA scans in UK Biobank (UKB), and radiographic HOA (rHOA) and subsequent total hip replacement (THR). Afterwards, an AA GWAS meta-analysis was performed (n=44,214), using AA previously derived in the Rotterdam Study (RS). Linkage disequilibrium score regression assessed the genetic correlation between AA and HOA. Genetic associations with P<5x10 -8 instrumented AA for two-sample MR. RESULTS: DXA-derived AA showed expected associations between AA and rHOA (OR 1.63 [95% CI 1.58-1.67]) and THR (HR 1.45 [1.33-1.59]) in UKB. The heritability of AA was 10% and AA had a moderate genetic correlation with HOA (r g =0.26 [0.10-0.43]). Eight independent genetic signals were associated with AA. Two-sample MR provided weak evidence of causal effects of AA on HOA risk (inverse variance weighted (IVW): OR=1.84 [1.14-2.96], P 0.01). In contrast, genetic predisposition for HOA had stronger evidence of a causal effect on increased AA (IVW: β=0.09 [0.04-0.13], P 4.58 x 10 -05 ). CONCLUSIONS: Expected observational associations between AA and related clinical outcomes provided face-validity for the DXA-derived AA measures. Evidence of bidirectional associations between AA and HOA, particularly in the reverse direction, suggests that hip shape modelling secondary to a genetic predisposition to HOA contributes to the well-established relationship between HOA and cam morphology in older adults.

U2 - 10.1002/art.42451

DO - 10.1002/art.42451

M3 - Article

C2 - 36662418

SN - 2326-5191

VL - 75

SP - 900

EP - 909

JO - Arthritis & rheumatology (Hoboken, N.J.)

JF - Arthritis & rheumatology (Hoboken, N.J.)

IS - 6

ER -

A Genome-Wide Association Study Meta-Analysis of Alpha Angle Suggests Cam-Type Morphology May Be a Specific Feature of Hip Osteoarthritis in Older Adults

Abstract

Access to Document

Fingerprint

Cite this