Some babies seem to have a genetic predisposition to a higher risk of being born too soon, according to a paper being presented Thursday at the Society for Maternal-Fetal Medicine's annual meeting in San Diego.

The study Birth found that variants in the fetus's DNA - not the mother's - may be what triggers some early births. 

More than 450,000 babies are born too soon each year in the U.S. Preterm birth (birth before 37 weeks of pregnancy) is the leading cause of newborn death, and babies who survive an early birth often face an increased risk of a lifetime of health challenges, such as breathing problems, cerebral palsy, intellectual disabilities and others. Even babies born just a few weeks early have higher rates of hospitalization and illness than full-term infants. It is a serious health problem that costs the United States more than $26 billion annually.

The preterm birth rate in the United States dropped more than 10 percent between 2013 and 2006, with most of the improvement focused in late preterm births (those between 34 and 37 weeks of pregnancy). The new findings focused on early preterm births, births before 34 weeks of pregnancy, in which there has been little improvement in recent years. 

The study was premised on the idea that copy number variants (CNVs) contribute to human genetic diversity and are associated with complex disorders. Given the complex nature of spontaneous preterm birth (SPTB), the authors sought to examine the association between maternal or neonatal CNVs and SPTB at <34 wks. 

Case-control study of singleton SPTB <34 wks compared to spontaneous labor at 39-40 wks. Maternal and neonatal DNA obtained from either blood or saliva was analyzed on Affymetrix Human SNP Array 6.0. CNVs were identified using PennCNV software; those containing > 5 consecutive SNPs and >10kb were retained. Results were filtered based on genotyping quality controls. CNVs with copy number >2 were defined as duplications and <2 as deletions. CNV distribution, type, and size were compared in a global burden analysis. Gene- and region-based CNV association strategies were utilized in logistic regression models as well as in PLINK software. Bonferroni correction adjusted for multiple comparisons.

Results: 901 maternal cases and 900 controls were analyzed. Mean CNV size and numbers were similar, but the proportion of CNV that were deletions was higher in cases (63.4 vs 62.2%, p=0.04). Neither gene- nor region-based association studies correlated with SPTB. 787 neonatal cases and 785 controls were analyzed.

CNV size and number did not differ, but the proportion of CNV that were deletions was higher in cases (73.7 vs 71.3%, p=1.6x10-11). SPTB associated deletions and duplications were noted across the genome including chromosomes 2, 4, 5, 6, 8, 10, 13, and 21.(Table) In gene-based analysis, a duplication of 3 consecutive genes on 11q11-12 (OR4P4, OR4C6 and OR4S2) and one on 11p15 involving RASSF7 were associated with SPTB. In gene region-based PLINK analysis, deletions involving 7 genes and duplications involving 3 were significantly associated with SPTB. Conclusion: Neonatal, but not maternal, CNVs are associated with early SPTB.

The effects of the involved genes require further investigation and these CNVs should be examined in additional cohorts.

Presentation: Neonatal, not Maternal, Copy Number Variants are Associated with Spontaneous Preterm Birth,  Joseph R. Biggio, Feifei Xiao, Don Baldwin, Radek Bukowski, Samuel Parry, M. S. Esplin, William W. Andrews, George Saade, Michael Varner, Yoel Sadovsky, Uma Reddy, John Ilekis, Heping Zhang