Researchers have identified cell-free RNA transcripts obtained from a noninvasive blood test during pregnancy that can predict risk of preterm birth in addition to predicting gestational age with an accuracy similar to ultrasound, which may soon pave the way for a low-cost alternative to ultrasound for prenatal care in developing areas, according to recent results from two pilot studies.
“Our results are thus generally comparable to ultrasound measurements, can be performed throughout pregnancy, and do not require a priori physiological knowledge such as the woman’s last menstrual period,” Stephen Quake, PhD, of Stanford (Calif.) University, and his colleagues wrote in Science.
Dr. Quake and his colleagues recruited 31 women from Denmark who provided weekly blood samples (521 samples) during pregnancy up until they delivered full-term. After analyzing the cell-free RNA (cfRNA) genes, researchers found cfRNA placenta, fetal, and immune genes were highly correlated with one another. They created a random forest model based on nine cfRNA genes (CGA, CAPN6, CGB, ALPP, CSHL1, PLAC4, PSG7, PAPPA, and LGALS14) that corresponded with the placenta. They estimated that those nine genes would predict gestational age and tested the model using 306 samples from 21 women in a training cohort and validated the test using 215 samples from 10 women in a validation cohort. The blood test predicted gestational age within 14 days of delivery in 32% of cases at the second trimester (T2), 23% at the third trimester (3T), and 45% at T2 and T3, compared with a 48% with ultrasound.
In a second pilot study, Dr. Quake and his colleagues created a polymerase chain reaction panel for 38 genes identified from sequencing RNA from patients in Pennsylvania, Alabama, and Denmark, with full-term and preterm deliveries up to 2 months before labor to determine “cfRNA transcripts that might be able to discriminate a spontaneous preterm delivery from a full-term delivery.” The top seven cfRNA transcripts (CLCN3, DAPP1, PPBP, MAP3K7CL, MOB1B, RAB27B, and RGS18), when grouped in “unique combinations” of three genes, predicted 75% of preterm samples and misclassified 1 of 26 samples (4%) from Denmark and Pennsylvania; in a validated cohort of Alabama patients, the test predicted 4 of 5 preterm samples (80%) and misclassified 3 of 18 full-term samples (17%).
“These cfRNA [polymerase chain reaction]–based tests have two advantages over alternatives: broader applicability and lower cost,” Dr. Quake and his colleagues wrote. “They can be applied across the globe as a complement to or substitute for ultrasound, which can be expensive and inaccurate during the second and third trimesters.”
The authors noted a larger sample size and blinded testing on a broader patient population is needed before clinics can apply this blood test in a diagnostic or screening tool for widespread use.
Dr. Quake and three other authors have a patent application submitted by the Chan Zuckerberg Biohub relating to “noninvasive estimates of gestational age, delivery, and preterm birth.” The other authors have no relevant financial disclosures.
SOURCE: Ngo TTM et al. Science. 2018 Jun 7. doi: 10.1126/science.aar3819.