American Association for Clinical Chemistry
Better health through laboratory medicine
October 2007 Clinical Laboratory News: Session Addresses the Need for Predicting Preterm Labor

October 2007: Volume 33, Number 10

Session Addresses the Need for Predicting Preterm Labor

IL-6 Assay Could Reduce Cost of Testing

By Deborah Levenson

An experimental assay that identifies pregnant women who are at increased risk of giving birth prematurely may someday offer a less costly alternative to the fetal fibronectin (fFN) test. The assay for interleukin 6 (IL-6), which was the focus of a presentation during the symposium “Emerging Biomarkers for Assessing Pregnancy” and a recent paper (Clin Chem 2007; 53: 1534-1540), could not only reduce testing costs but also the incidence of preterm birth, which has jumped 20% since 1980, according to presenter and co-author Alison Woodworth, PhD, Assistant Professor of Pathology and Director of Esoteric Chemistry at Vanderbilt University Medical Center in Nashville, Tenn.

Currently, more than 12.7% of U.S. births are preterm, in part because preterm labor is difficult to diagnose. Initial symptoms are often mild and indistinguishable from those of normal pregnancy, while later symptoms often occur too late for effective intervention. The only test commonly used to identify women at risk for preterm birth, fFN, is sold as Full Term, The Fetal Fibronectin Test by Cytyc in Sunnyvale, Calif., and costs more than $100 per test for the necessary reagents. fFN is one of two FDA-approved tests for preterm labor, along with salivary estriol, which the American Academy of Obstetricians and Gynecologists no longer recommends. “fFN is essentially the only marker being used in clinical practice for the prediction of preterm birth,” Woodworth noted. In addition to its lower cost at about $7 per test, IL-6 may offer the additional advantage of an automated platform. “Currently fFN testing is not available on an automated platform,” added Woodworth.

Similar Negative Predictive Values

IL-6 and fFN have similar performance characteristics, according to the paper by Woodworth and colleagues at Washington University in St. Louis, Mo., and University of North Carolina at Chapel Hill. They measured fFn and IL-6 concentrations in 660 cervicovaginal fluid samples and determined IL-6’s clinical utility for predicting delivery within 14 days of collection. Samples came from women at 24 and 36 weeks of gestation, the usual time for fFN testing, but as Woodworth noted, researchers don’t know the optimal time for IL-6 testing. The results showed that both assays are more useful for ruling out preterm birth than predicting it. Negative predictive values were 97% for IL-6 and 98% for fFN, with a preterm delivery prevalence of 4.7%. The sensitivity, specificity, positive and negative likelihood ratios for delivery within 14 days were 65%, 87%, 4.8, and 0.4, respectively for fFN and 35%, 91%, 3.8, and 0.7 for IL-6, with a 250 ng/L cutoff. Normalizing IL-6 to albumin didn’t improve performance characteristics.

“IL-6and fFN perform quite similarly, but neither is an ideal test for predicting preterm delivery. Their strengths lie in their negative predictive values,” said Woodworth.

New markers for preterm labor may help researchers understand the different etiologies of preterm birth and improve the diagnostic accuracy of both markers. “It may be that one marker isn’t the answer,” Woodworth remarked, referring to research on other markers including hCG, MMP-8, PIGFBPI, profillin 1, and annexin V. “Preterm labor has many different etiologies, therefore a panel of markers representing the different pathologic causes of preterm birth may better predict patients who will deliver preterm.”

While IL-6 assays are commercially available for research use, they are not FDA approved for determining preterm labor risk, coauthor Ann Gronowski, Associate Professor at Washington University School of Medicine, Department of Pathology & Immunology in St. Louis, Mo. pointed out during an interview. “The challenge would be getting a company that makes the assay interested enough to put the resources into a study to get 510(k) approval.”