NEW YORK – Sera Prognostics has published new data demonstrating the effectiveness of its PreTRM test for predicting risk of preterm birth as it works to build out the clinical evidence supporting the assay.
Detailed last month in a paper published in the American Journal of Obstetrics and Gynecology, the results found that the PreTRM test was useful for identifying women at high risk of delivering prior to 32 weeks.
Sera had previously published data showing the test's performance at predicting risk of birth at or before 37 and 35 weeks of gestation, but had not provided data on its performance at 32 weeks, which Gregory Critchfield, the company's chairman and CEO, noted was a particularly important time point given the sometimes dramatic differences in outcomes for children born either before or after the 32 week mark.
"We always wanted to be able to better characterize what happened early on in preterm birth because the earlier the delivery takes place the greater the negative impact on that child," he said.
Indeed, one of Sera's competitors, Phoenix, Arizona-based Nixxi, recently highlighted its preterm birth test's focus on the 32-week time point as a key differentiator of the product, with the company's CEO Avi Patil saying that feedback from clinicians and payors indicated that they were most interested in identifying patients at risk of delivering prior to that time, with the 35 and 37 week cut-offs being lesser priorities.
Critchfield likewise said that in Sera's conversations with insurers they were especially interested in the 32-week mark but added that the 37-week mark, which has traditionally been the point at which births are considered full term, remained relevant.
"Large insurance companies have models that show that screening even for risk of delivery before 37 weeks is cost effective," he said. "They know that the greatest impact will be from the earlier [births] but that there is benefit across all pregnancies that are identified of being at higher risk. They don't want to limit it to only having a prediction of less than 32 [weeks]."
Sera presented its initial validation study, the Proteomic Assessment of Preterm Risk (PAPR) study, in 2016. While that study looked at 5,501 patients, it did not contain enough women who delivered prior to 32 weeks to collect good statistics on its test's performance in that group.
The recent study, titled TREETOP (The Multicenter Assessment of a Spontaneous Preterm Birth Risk Predictor), enrolled 5,011 women at 18 sites across the country, with 847 of these women enrolled in a subanalysis looking specifically at the 32-week time point.
Of those 847 women, nine gave birth prior to 32 weeks, and the PreTRM test identified those women with an area under the curve, or AUC, of .71. When body mass index was included with the PreTRM measurement, the AUC was .76.
In the PAPR study, Sera found the test was able to identify women who gave birth at or before 35 weeks gestation with an area under the curve (AUC) of .93 and that it identified women who gave birth at or before 37 weeks with an AUC of .75.
The PreTRM test measures the ratio of two proteins, insulin-like growth factor-binding protein 4/sex hormone-binding globulin (IBP4/SHBG), that Critchfield said are markers of placental dysfunction. He said data from the PAPR study showed that patients who, according to the test, had the highest probability of delivering before 37 weeks were also the likeliest to deliver at earlier time points, which he noted suggested to the company that the test signature might be useful for predicting delivery prior to 32 weeks.
Sera is also exploring whether the test could be useful for predicting which premature infants would have more severe outcomes or longer hospital stays. In the TREETOP study they found that among infants born prior to 32 weeks it could identify those infants who would die following delivery with an AUC of .67. It also correlated with length of hospital stay.
Sera signed a licensing deal with Laboratory Corporation of America for the PreTRM test in 2017, but Critchfield said this week that the company is still laying the groundwork for commercialization of the assay. The company plans to offer the test as a laboratory-developed test run out of its CLIA laboratory.
Having published validation data in the PAPR and TREETOP studies, the company is now in the process of completing outcome studies where it aims to demonstrate the test's impact on clinical care and outcomes and is in the process of reporting out such data, Critchfield said.
He said that Sera was currently using some of this data in conversations with insurers and that it expected to publish the data this year. He said lining up coverage from payors was the next step in Sera's commercialization plan, noting that before it begins offering the test broadly it wants "the question of insurance coverage to be largely answered."
"We're not going out broadly to physician practices yet because we want to make sure the data are published from the outcome studies before we do that, we want to make sure that the data is well understood by the clinical community before we do that, and you want to make sure you put as much [insurance] coverage in place prior to doing that as you can," Critchfield said.
Sera runs the assay using an LC-MS platform, which is still a relative rarity for multiplexed protein tests. And while throughput has long been a concern for clinical mass spec protein testing, Critchfield said the company has the ability to scale its mass spec operations to be able to serve all of the intended-use patients for its test, which he said was potentially in the range of 3.5 million to 3.8 million women per year.
"We have the ability to scale it up and do it very cost effectively with automation," he said, noting that serving the entire intended-use population would require on the order of tens of mass spec instruments.