Genetic association study of Preterm birth and Gestational age in a population-based case-control study in Peru: Genetics of PTB and GA in pregnant women in Peru

Abstract

Preterm birth (PTB) is an adverse pregnancy outcome affecting ~15 million pregnancies worldwide. Genetic studies have identified several candidate loci for PTB, but results remain inconclusive and limited to European populations. Thus, we conducted a genome-wide association study (GWAS) of PTB and gestational age at delivery (GA) among 2,212 Peruvian women. PTB cases delivered ≥ 20 weeks' but < 37 weeks' gestation, while controls delivered at term (≥ 37 weeks but < 42 weeks). After imputation (TOPMED) and quality control, we assessed the association of ~6 million SNPs with PTB and GA using multivariable regression models adjusted for maternal age and the first two genetic principal components. In silico functional analysis (FUMA-GWAS) was conducted among top signals detected with an arbitrary P < 1.0×10^-5 in each GWAS. We sought to replicate genetic associations with PTB and GA identified in Europeans, and we developed a genetic risk score for GA based on European markers. Mean GA was 30 ± 4 weeks in PTB cases (N=933) and 39 ± 1 in the controls (N=1,279). PTB cases were slightly older and had higher C-sections and vaginal bleeding than controls. No association was identified at genome-wide level. Top suggestive (P < 1.0×10^-5) signals were seen at rs13151645 (LINC01182) for PTB, and at rs72824565 (CTNNA2) for GA. Top PTB variants were enriched for biological pathways associated with polyketide, progesterone, steroid hormones, and glycosyl metabolism. Top GA variants were enriched in intronic regions and cancer pathways, and these genes were upregulated in the brain and subcutaneous adipose tissue. In combination with non-genetic risk factors, top SNPs explained 14% and 15% of the phenotypic variance of PTB and GA in our sample, but these results need to be interpreted with caution. Variants in WNT4 associated with GA in Europeans were replicated in our study. The genetic risk score based in European markers, was associated with a 2-day longer GA (R²=0.003, P=0.002) per standard deviation increase in the score in our sample. This genetic association study identified various signals suggestively associated with PTB and GA in a non-European population; they were linked to relevant biological pathways related to the metabolism of progesterone, prostanoid, and steroid hormones, and genes associated with GA were significantly upregulated in relevant tissues for the pathophysiology of PTB based on the in-silico functional analysis. None of these top variants overlapped with signals previously identified for PTB or GA in Europeans.

Keywords: GWAS; genetic risk score; gestational age at delivery; maternal genotype; pregnancy; preterm birth.

Fig 1.. Genetic ancestry analysis showing the clustering of PAGE samples relative to global populations from different ancestries included in the 1000 Genomes project.

More genetically related populations clustered more closely with each other than genetically distant populations. Most of the PAGE participants (black and grey dots) aligned close to the 1000 Genomes PEL samples (Peruvians from Lima, Peru), and few others were dispersed among samples from other populations in the Americas (CLM Colombians, MXL Mexican Americans). PC1-PC3: first three principal components. Subpopulations: ACB African Caribbean in Barbados, ASW African Ancestry in Southwest US, BEB Bengali in Bangladesh, CDX Chinese Dai in Xishuangbanna, CEU Utah residents with Northern and Western European ancestry, CHB Han Chinese in Beijing, CHS Southern Han Chinese, CLM Colombian in Medellin, ESN Esan in Nigeria, FIN Finnish in Finland, GBR British in England and Scotland, GIH Gujarati Indian in Houston, GWD Gambian in Western Division, IBS Iberian populations in Spain, ITU Indian Telugu in the UK, JPT Japanese in Tokyo, KHV Kinh in Ho Chi Minh City, LWK Luhya in Webuye, MSL Mende in Sierra Leone, MXL Mexican Ancestry in Los Angeles, PEL Peruvian in Lima, PJL Punjabi in Lahore, PUR Puerto Rican in Puerto Rico, STU Sri Lankan Tamil in the UK, TSI Toscani in Italy, YRI Yoruba in Ibadan.

Fig 2.. Quantile and Manhattan plots summarizing results of the GWAS of Preterm Birth (A, B) and Gestational age at delivery (C, D) conducted among women in the PAGE study (N=2,212).

In the Manhattan plots (B, D), the horizontal blue line corresponds to the suggestive genome-wide association threshold at P < 1.0×10⁻⁵; the horizontal red line is the Bonferroni significant threshold at P < 5.0×10⁻⁸. Associations were adjusted for maternal age and the first two genetic PCs. Top SNPs were annotated with their rsID using the Haplotype Reference Consortium panel (GRCh37/hg19), or the SNP coordinates reported in TOPMED (GRCh38/hg38).

Fig 3.. Regional plots showing the genetic context of top signals detected with suggestive association with Preterm Birth (rs13151645 in chr4) and Gestational age at delivery (rs72824565 in chr2).

The colored legend represents the correlation (LD) of nearby SNPs with our sentinel SNP shown by the purple diamond. SNPs more correlated with the SNP of interest are shown as red dots. Pairwise SNP correlations were calculated using the 1,000 genomes (phase 3) and the admixed American population as reference. At the bottom of the plot, showing representative genes in the region and their transcriptional orientation.