Association of genetic ancestry with pre-eclampsia in multi-ethnic cohorts of pregnant women

Abstract

Objectives: Maternal self-reported ethnicity is recognised as a risk factor for pre-eclampsia in clinical screening tools and models. This study investigated whether ethnicity is acting as a proxy for genetic variants in this context.

Study design: A total of 436 women from multi-ethnic backgrounds recruited to two UK observational pregnancy hypertension cohort studies were genotyped. Genetically-computed individual ancestry estimates were calculated for each individual through comparison to the multi-ethnic 1000 Genomes reference panel genotypes. Regression models for pre-eclampsia using clinical risk factors including self-reported ethnicity with and without ancestry estimates were built and compared using Likelihood Ratio Tests (LRT).

Main outcome measures: Pre-eclampsia (early- and late-onset).

Results: In these multi-ethnic cohorts (mean age 34.9 years; 41.3 % White, 34.2 % Black, 13.1 % Asian ethnic backgrounds; 82.6 % chronic hypertension), discrepancies between self-reported ethnicity and genetically-computed individual ancestry estimates were present in all ethnic groups, particularly minority groups. Genetically-computed pan-African ancestry percentage was associated with early-onset (< 34 weeks) pre-eclampsia in adjusted models (aOR 100 % vs 0 % African ancestry: 3.81, 95 % CI 1.04-14.14, p-value 0.044) independently of self-reported ethnicity and established clinical risk factors. Addition of genetically-computed African ancestry to a clinical risk factor model including self-reported ethnicity, improved model fit (Likelihood ratio test p-value 0.023).

Conclusions: Self-reported maternal ethnicity is an imperfect proxy for genetically-computed individual ancestry estimates, particularly in ethnic minority groups. Genetically-computed African ancestry percentage was associated with early-onset pre-eclampsia independently of self-reported maternal ethnicity. Well-powered studies in multi-ethnic cohorts are required to delineate the genetic contribution to pre-eclampsia.

Keywords: Ancestry; Ethnicity; Pre-eclampsia; Pregnancy.

Fig. 1

Genetically-estimated individual ancestry proportions (AFR: African, AMR: Admixed American, EAS: East Asian, EUR: European, SAS: south Asian superpopulations, calculated using 1000 Genomes as reference data) in study cohort (n = 436). Each individual is denoted by a single vertical stacked bar, with individual population ancestry percentages (summing to 1) denoted by bar colouring. Individuals are grouped by self-reported ethnic group.

Fig. 2

Histograms of percentage African (A) and European (B) ancestry in whole study cohort (n = 436), with frequency bars coloured by number of ‘equivalent grandparents’. AFR = African, EUR = European, GP = grandparent(s).

Fig. 3

Sankey plots linking self-reported ethnic group (left hand panel) to genetically-estimated ancestry percentages: A) European (EUR), B) African (AFR), C) South Asian (SAS). Genetically-estimated ancestry percentages are grouped by number of ‘equivalent grandparents’ (GP).

Fig. 3

Sankey plots linking self-reported ethnic group (left hand panel) to genetically-estimated ancestry percentages: A) European (EUR), B) African (AFR), C) South Asian (SAS). Genetically-estimated ancestry percentages are grouped by number of ‘equivalent grandparents’ (GP).

Fig. 3

Sankey plots linking self-reported ethnic group (left hand panel) to genetically-estimated ancestry percentages: A) European (EUR), B) African (AFR), C) South Asian (SAS). Genetically-estimated ancestry percentages are grouped by number of ‘equivalent grandparents’ (GP).