Don’t Rely on Biomarker Tests to Predict Risk of Premature Births, Warns AACC
After steadily declining, the rate of spontaneous preterm births in the U.S. rose to 10 percent in 2018. Preventing premature births would save the nation over $26 billion per year. However, diagnostic tests that claim they can predict premature births are unreliable and should not be used as part of routine evaluation of women with […]
After steadily declining, the rate of spontaneous preterm births in the U.S. rose to 10 percent in 2018. Preventing premature births would save the nation over $26 billion per year. However, diagnostic tests that claim they can predict premature births are unreliable and should not be used as part of routine evaluation of women with symptoms of preterm delivery. That, at least, is the conclusion of new guidance issued by the American Association of Clinical Chemistry (AACC).
The Diagnostic Challenge
“Identifying women who will deliver preterm is critical to allow selective initiation of appropriate therapy, while preventing unnecessary treatment of women who will deliver at term,” the AACC notes. But this task is “inherently challenging” because the best way to predict premature birth is by determining if the patient has a history of giving birth before term. Obviously, that solution is not available for first time pregnancies.
Another problem is that the signs that labor is beginning, things like backache and pelvic pressure, are “ubiquitous.” Heck, you do not have to be pregnant to experience a backache. (Our words, not the AACC’s.)
The New AACC Guidance
Currently available diagnostic tests using more specific biomarkers usually identify risk of premature birth by measuring the patient’s fetal fibronectin (fFN), interleukin 6 (IL-6) and placental alpha macroglobulin-1 (PAMG-1). However, the AACC says that tests for these biomarkers have low positive predictive values and there have been no studies demonstrating their “definitive clinical value.”
Thus, for example, “randomized controlled trials that assess patient outcomes and healthcare utilization by implementing fFN testing are lacking and existing studies have found conflicting results,” according to the guidance document. The AACC also cites the lack of consensus among professional societies about the utility of fFN in predicting preterm birth.
What the AACC Recommends
“At this time, AACC does not recommend measurement of fFN, PAMG-1 or IL-6 in the routine evaluation of all women with symptoms consistent with preterm delivery,” the guidance concludes. One potential solution is to “limit biomarker testing to high-risk women, thereby increasing the pre-test probability of the population being tested and improving the [positive predictive value] of currently available test methods.” In the meantime, laboratory professionals and obstetricians should discuss all available recommendations on preterm birth testing together and take a collaborative approach to developing institutional testing strategies for preterm birth.
The Silver Lining: PAMG-1 May Signal Preterm Delivery Within a Week
The AACC makes one limited exception to its non-recommendation of measuring fFN, PAMG-1 or IL-6 as part of routine evaluation of all women with symptoms consistent with preterm delivery. A positive PAMG-1 test result in patients who present with symptoms of premature labor and who are at high risk of preterm delivery due to cervical length, a positive PAMG-1 result may be helpful for identifying women who are likely to deliver within a week. In patient populations with a higher prevalence of spontaneous preterm delivery within one week, PAMG-1 would likely “provide clinical value as the majority of women with a positive test result would go on to deliver prematurely,” the report notes.
The AACC also addresses its recommendations to test makers. Among the limitations of current tests is that they measure only a single protein. Perhaps multi-marker panels may improve performance. However, the association quickly adds, the multi-marker panels that have been evaluated so far “do not demonstrate improved diagnostic performance relative to single biomarkers.” Another solution would be to come up with “novel diagnostic tools with improved PPV to predict the minority of symptomatic women who will deliver prematurely.”
This content is exclusive to Diagnostic Testing and Emerging Technologies subscribers
Start a Free Trial for immediate access to this article and our entire archive of over 20 years of DTET reports.