Preterm birth (PTB) is the second most common cause of infant death in the United States and a major cause of costly—and sometimes lifelong—health and social problems. As a result, clinicians and laboratorians have a keen interest in detecting women at risk. A new AACC guideline does not recommend routinely measuring interleukin 6 (IL-6), placental alpha microglobulin 1 (PAMG-1), or fetal fibronectin (fFN) in pregnant women with symptoms consistent with preterm delivery. These biomarkers have low positive predictive values (PPVs) and provide limited utility in diagnostic algorithms most commonly used in the U.S., according to the guidance.
Despite the guideline’s negative stance on the markers’ routine use, its authors say labs should not issue unilateral decrees about these tests. “Decisions should be made with the obstetricians in a collaborative effort,” said senior author Robert D. Nerenz, PhD, DABCC, assistant director, clinical chemistry and assistant professor of pathology and laboratory medicine at the Geisel School of Medicine and Dartmouth-Hitchcock Medical Center in Lebanon, New Hampshire.
Preterm Birth: A National Problem
The precise mechanisms involved in PTB are not fully known, but proposed etiologies leading to PTB include fetal stress, inflammation, decidual hemorrhage, and pathological uterine distension, according to the guideline. Each of the tests measures a single protein associated with a specific PTB etiology: IL-6 for infection and inflammation, fFN for degradation of the extracellular matrix, and PAMG-1 for presence of amniotic fluid.
PTB can lead to low birth weight, respiratory distress, underdeveloped organs, neurodevelopmental disabilities, cognitive impairment, visual and hearing impairments, developmental coordination disorders, and behavioral and emotional difficulties. After a period of declining PTBs, they accounted for 10% of all 2018 U.S. births and disproportionately affected racial and ethnic minorities, especially African Americans.
Some treatments may improve outcomes for babies of patients who present with signs and symptoms of premature labor. “However, 90% of these patients won’t deliver preterm. It’s useful to have tools to see who’s genuinely at risk,” said Phillip R. Bennett, MD, PhD, a PTB biomarker researcher and professor of obstetrics and gynecology at Queen Charlotte and Hammersmith Hospitals in London. He added, however, that some therapies may be harmful when symptoms are not related to PTB.
Recommendations on Preterm Birth Biomarkers
The guidelines say biomarkers are unlikely to provide much clinical value in populations with a pretest probability of less than 5%. “In a population with a low pretest probability, these markers don’t have much clinical utility,” Nerenz said. “While the three markers do have high negative predictive values, a negative result doesn’t substantially change the diagnostic uncertainty, as most women in the population being evaluated will not deliver prematurely. To distinguish the small proportion of women who will give birth prematurely from a larger pool with similar signs and symptoms who will not, a test with a high PPV is needed.”
Limiting biomarker testing to only high-risk women identified on the basis of cervical length or other characteristics will increase the pretest probability in the tested population and improve PPV, Nerenz added.
PAMG-1 may be useful in some cases. In populations with a pretest probability of 5%, a positive PAMG-1 result increases the post-test probability to about 20%–30% while a negative result leaves the post-test probability essentially unchanged, the guideline notes. However, PAMG-1 does not provide sufficient assurance to either rule in or rule out PTB within 7 days. In populations with a pretest probability of 20%–25%, however, a positive PAMG-1 result substantially increases the post-test probability to about 75%. In this patient population, a positive PAMG-1 result may identify women likely to deliver within 7 days.
fFN research in populations with a pretest probability of 3% shows a positive fFN result modestly increases the post-test probability to about 15%, while a negative result does not substantially reduce the post-test probability. Studies of fFN testing show it does not consistently improve clinical outcomes, as measured by reduced hospitalization rates, rates of PTB, or healthcare costs. fFN should not be measured in asymptomatic women due to its low specificity and PPV, according to the guideline.
Bennett noted a movement toward quantifying fFN values to give better predictive statistics and guide practice. The guideline says this tactic is promising but calls for studies to further demonstrate improved clinical outcomes.
In study populations with a pretest probability of 2%, a positive IL-6 result modestly increases the post-test probability to about 15%, while a negative result does not substantially reduce the post-test probability. IL-6 and fFN have similar diagnostic performance characteristics. IL-6 does not provide sufficient assurance to either rule in or rule out PTB within 7 days, the guideline says.
Some researchers have proposed using multimarker panels to more accurately predict women who will deliver prematurely. But multimarker panels evaluated to date do not improve diagnostic performance relative to single biomarkers, the guideline says.
Talking to Clinicians
Some U.S. clinicians may find it challenging to stop fFN testing because some guidelines recommend it. Additionally, malpractice is a consideration in obstetrics and gynecology practice, said guideline co-author Joely Straseski, PhD, MT(ASCP), DABCC, medical director of endocrinology and automatic core laboratories at ARUP and associate professor at the University of Utah School of Medicine in Salt Lake City.
AACC’s new PTB testing guideline exists alongside others from the American College of Obstetricians and Gynecologists (ACOG), Society for Maternal-Fetal Medicine (SMFM), and the United Kingdom’s National Institute for Health and Care Excellence (NICE). ACOG does not recommend routine use of fFN to stratify risk for preterm delivery, while SMFM recommends fFN for women presenting with symptoms of preterm labor prior to 34 weeks and transvaginal ultrasound demonstrating cervical length between 20–29 mm. NICE recommends testing fFN if a cervical length measure is not available or not acceptable.
Straseski sees the guideline as a conversation-starter for labs and obstetricians. The guideline can help clinicians understand that the three biomarkers “are not helpful for all women with symptoms of premature birth,” plus the specific situations where they may be useful, she said.
Added Nerenz, “The guidance’s goal is to make clinicians think critically about why they order these tests. Do they help make better clinical decisions? Are patient outcomes better?” He noted that his institution has dropped fFN.
Penn State Hershey Medical Center also stopped offering fFN after a test utilization review revealed that volume dropped to the point where the lab ran just two fFN tests in 2020, said Yusheng Zhu, PhD, DABCC, professor of pathology, laboratory medicine, and pharmacology and medical director of clinical chemistry and the automated testing laboratory at Penn State Hershey, where he also chairs the laboratory utilization committee.
In a discussion with his hospital’s departments of obstetrics and gynecology and family and community medicine, Zhu shared a literature review that revealed a growing consensus about fFN’s limited utility. He suggests that other clinical laboratorians focus similar discussions with clinicians on prevalence of PTB in an institution’s patient population and clinician practice. Some laboratorians and clinicians may need to do their own research to determine the utility of preterm labor biomarkers in their institutions, he added.
Looking to the Future
“We do need better biomarkers with higher PPVs,” Nerenz noted. The search continues for other potential biomarkers to determine risk at various stages of pregnancy.
For example, Bennett studies first trimester circulating microRNA to predict PTB and cervical shortening in women at risk of preterm delivery. The hope is that early identification of PTB risk allows time to deliver outcome-modifying interventions.
Zhu pointed to other research examining multiplex biomarkers based on proteomics, metabolomics, and exosomics. Preliminary data on one such panel of novel protein markers in cervicovaginal fluid samples collected from asymptomatic women at gestational weeks 16–24 had sensitivity and specificity of 91% and 78%, respectively (Am J Obstet Gynecol MFM 2020;2:100084). Another study showed that cervicovaginal fluid microbiota metabolic profiling identified eight vaginal microbiota metabolites to predict preterm labor (Metabolites 2020;10:349). Other research focuses on proteins in plasma exosome.
Time will tell which of these and other potential biomarkers will be useful. “The biomarkers with the best potential have the better PPVs. That’s the key, across the board,” Straseski said.
Deborah Levenson is a freelance writer in College Park, Maryland. +Email: [email protected]