Preterm birth is one of the leading causes of infant mortality. Approximately 10% of babies born in the U.S. in 2018 were born before 37 weeks gestation, according to the U.S. Centers for Disease Control and Prevention (CDC). The CDC also reports that the risk of preterm birth in 2016 was 14% for African American women versus 9% for white women. Right now, physicians are stymied by the fact that few reliable indicators of preterm birth exist.
But a new study suggests that the vaginal microbiome may allow doctors to identify women who are at high risk for preterm birth. “We compared the microbiome signatures in vaginal swab samples taken in pregnancy and found a signature early in pregnancy that distinguished women who would go on to deliver spontaneously preterm from case-matched controls who delivered at term,” said the study’s lead author, Jennifer Fettweis, Ph.D., an assistant professor of microbiology and immunology at the Virginia Commonwealth University.
The study, which was published in Nature Medicine, was funded by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (NICHD).
Currently, doctors often don’t know a woman is going to deliver a baby prematurely until she is already in labor. “Our knowledge of predicting preterm birth is very rudimentary, and there are no great predictors of preterm birth, which is a major public health issue,” said Andrew Bremer, M.D., Ph.D., acting chief of the NICHD pregnancy and perinatology branch that funded the study. According to Bremer, the current study’s findings offer hope that the fight against preterm labor can go from being reactive to proactive. “If a woman were to get a screen based on her vaginal microbiome, then other interventions could be implemented to potentially carry that pregnancy to term,” he told Clinical OMICs.
Multi-omics approach
Fettweis and colleagues analyzed omics data for 597 women who were part of the Multi-Omic Microbiome Study-Pregnancy Initiative (MOMS-PI), which included 1572 pregnancies. They generated omics data for approximately 12,000 samples as part of the integrative Human Microbiome Project, a program of the National Institutes of Health Common Fund. To study the potential association of microbiome composition in preterm birth, they obtained vaginal swab samples from 45 pregnant women who ultimately delivered preterm and compared them to samples from 90 pregnant women who delivered at term. Eighty percent of the women in the preterm cohort identified as African American. “We case-matched women based on ethnicity, ancestry, age and annual household income to women who delivered 39 weeks or later,” Fettweis explained.
The researchers wanted to know if there was a difference in microbiomes between the two groups—a signature that could retrospectively predict preterm birth. They used 16S ribosomal RNA data to determine microbial profiles, metagenomics to identify microbial species, metatranscriptomics to determine which genes were turned on and what these microbial species were doing biologically, and cytokine profiles to determine the women’s immune response. The researchers found that African American women had significantly lower levels of Lactobacillus crispatus—the main component of the vaginal microbiome—and higher levels of 15 other taxa. They also found that preterm birth associated taxa were correlated with inflammatory cytokines. “We can’t claim causation, at this point, but there are signatures when we compare these two groups.”
The study also offers new ways of approaching complex biological problems using Big Data and high-performance computing. The study required a large multi-disciplinary team of researchers, key people who could work across disciplines to solve novel computational challenges. “The datasets we have generated will stimulate work in those areas,” Fettweis said.
Microbiome-based diagnostics
Fettweis’ findings are a starting point for researchers who want to decrease the frequency and disparities in preterm birth. “We’re hopeful we developed a proof of principal model that suggests that, using just the microbiome, we can do just about as well as using some of the best clinical measures,” Fettweis said. Fettweis also said she can envision a day when a microbiome-based diagnostic test that uses a vaginal swab done during a regular prenatal exam is used to screen women in early pregnancy for preterm birth risk.
The U.S. Food and Drug Administration has yet to approve a microbiome-based diagnostic test. That day is coming, said Greg Kuehn, vice president and chief operating officer of Prescient Metabiomics, a joint venture between Metabiomics and Prescient Medicine. Prescient Metabiomics is in the preclinical stages of development, preparing to conduct larger and more complex clinical trials as it moves toward FDA clearance of microbiome-based diagnostic tests for colorectal adenoma, colon cancer, and inflammatory bowel disease.
Kuehn agreed that Fettweis’ work has the potential to be the basis for such a test for preterm pregnancy risk in the future. “For a whole range of diseases there could be tests and therapeutics developed from microbiome diagnostics. It’s a new paradigm in human health,” he said. Kuehn also pointed to the irony that researchers are basically assessing human health based on microbial populations of microenvironments within the human body, such as the vagina, gut, and colon. These environments depend on the microbiome to maintain biofilms, maintain the integrity of mucosa, and the suppression of pathogeneic bacteria. “It’s sort of a regulated environment where there is interaction between the microbiome and the host system.”
In addition to replicating this study’s results, Fettweis said that researchers in the field need to standardize the way swabs are collected, the primer sets used in analysis, and the number of weeks gestation used to define preterm birth. For example, some studies include women whose labor was induced. Fettweis and her colleagues only included those who gave birth early spontaneously. “We need to harmonize studies moving forward so that we can compare across studies and across populations.”