BALTIMORE – Following compelling clinical results, pregnancy health startup Mirvie is poised to develop noninvasive, RNA-based blood tests that can predict pregnancy complications ahead of their symptoms.
Earlier this month, the South San Francisco, California-based company published a study in the American Journal of Obstetrics and Gynecology (AJOG) demonstrating that cell-free RNA (cfRNA) profiles in maternal blood could help predict spontaneous preterm birth. The study echoes Mirvie's previous study, published in Nature in January, which showed that cfRNA in blood can predict preeclampsia, a pregnancy complication characterized by the onset of hypertension, months before symptoms occur.
The company's near-term goals involve conducting more clinical trials to further validate its cfRNA profiles and commercialize tests that can predict conditions such as preeclampsia, preterm birth, and gestational diabetes. By uncovering the underlying biology of these conditions, it also hopes, in the long run, to discover new therapeutics to remedy these complications.
Mirvie was founded in 2018 to address the "absolutely staggering" unmet need in the field of pregnancy health, according to Cofounder and CEO Maneesh Jain, who noted that approximately 1 in 5 pregnancies is impacted by complications. Since its inception, the company has raised more than $30 million in early-stage financing, with backing from Khosla Ventures and Mayfield Fund.
Despite the high prevalence of pregnancy complications, there are limited ways for clinicians to gauge and prevent these conditions right now, said Mirvie Chief Medical Adviser Michal Elovitz, a maternal fetal medicine specialist at the Hospital of the University of Pennsylvania. For instance, when it comes to spontaneous preterm birth — early deliveries that happen for no apparent reason and account for about 75 percent of all preterm birth cases — there is almost no effective way for a physician to foresee whether the patient is at risk, especially for first-time mothers, Elovitz said.
Although previous efforts have tried to detect pregnancy complications with biomarkers, Jain said these tended to only involve a handful of protein markers, which were inadequate to pick up the often complex and heterogeneous pregnancy conditions.
In contrast, Mirvie is "bringing to bear the full toolkit" for capturing blood cfRNA signatures potentially associated with adverse pregnancy outcomes and combs through "large numbers of diverse samples" using genome-wide analysis and machine learning technologies, Jain said. The company's technology builds on a cfRNA liquid biopsy method developed by Stephen Quake's group at Stanford University, he added, and Mirvie has since secured "a pretty broad patent portfolio."
Compared with DNA biomarkers, one advantage of cfRNA signals is that they reflect what's currently happening in the tissues, said Elovitz. About 25 percent of the maternal blood flows to the placenta every time the mother's heart pumps, she said, and a maternal blood sample, which is a confluence of molecular signatures from the mother, baby, and placenta, can offer a noninvasive window into the maternal-fetal dyad.
Another strength of cfRNA markers is that they appear to relate to the underlying biology of the complications rather than general risk factors, Elovitz pointed out. "The goal is that the biology itself will reveal the risk," she said, adding that the company's Nature paper demonstrated that, in the case of preeclampsia, the current clinical risk factors — such as maternal age, body mass index, and race — can be "pretty biased" and do not offer a powerful predictive value. In contrast, cfRNA profiles, which are independent of those risk factors, have shown they can indicate the onset of the condition months in advance.
That said, social demographic factors can also predispose some groups to have higher complication risks than others, Jain said, but these influences will be reflected through gene expression patterns of the patient and are thus detectable via the cfRNA markers.
In brief, Mirvie's diagnostic workflow starts with a maternal blood draw in the second trimester. Next, cfRNA is extracted from the plasma and subsequently sequenced using an Illumina platform. Afterward, cfRNA profiles are analyzed using machine learning algorithms to identify any patterns that are indicative of healthy or complicated pregnancies, he said.
Cognizant of the mistakes made by prior studies in this field, Elovitz emphasized that Mirvie's approach did not merely focus on any single analyte, but rather cast a wide net on all detectable cfRNA signatures for an answer.
Importantly, she said that some previous preterm birth studies did not generate enough metadata to separate sub-phenotypes of the condition, thus failing to identify appropriate biomarkers for specific disease subtypes.
Mirvie, on the other hand, recruited large and diverse sample cohorts and amassed granular clinical data on each participant, she said, in order to tease apart the nuances within the disorder. So far, the company has analyzed approximately 3,000 samples, including the cohorts included in the Nature and AJOG studies, according to Jain. While most samples are collected from collaborating medical centers in the US, the company has also enrolled patients from a few UK clinical sites and one in Africa, he added.
Although interventions are limited, clinicians are currently able to offer steroid treatments, such as corticosteroid therapy, during pregnancy to help improve the outcomes of preterm birth, Elovitz said. While the steroid shot cannot alter the course of preterm birth, she said, the medication can cross the placental barrier and help the baby's lungs mature before delivery.
Still, Elovitz said such remedies are not being fully utilized since clinicians cannot effectively pinpoint who is at risk for preterm birth at this point. By deploying predictive tests such as Mirvie's, she added, clinicians can take better advantage of the existing interventions to benefit mothers at risk.
Furthermore, she believes that by understanding the biology behind these conditions, Mirvie can also help develop new therapeutics and interventions. Jain said that although he considers Mirvie "clearly a diagnostic company," he definitely envisions the company to partner with biotech and pharmaceutical companies for drug development down the road.
For now, Jain said the company's mission is "working as fast as possible" to commercialize the technology into a noninvasive blood test that can be ordered by a physician. However, he did not disclose a specific commercialization timeline and said it is too early to discuss pricing, adding that Mirvie is still exploring the exact format for the test.
In terms of competition, Jain said, the biggest rivalries for the company are the lack of awareness and the reactive mindset of the field. Due to the lack of predictive tools, people have been taking a reactive approach when it comes to pregnancy health, he said, taking action only when symptoms arrive. The company's task is therefore to convince doctors to proactively prevent adverse pregnancy outcomes.
"I think there is an overall undervaluation of women's health and pregnancy health in general," Elovitz concurred, adding that there is a widespread lack of funding for pregnancy health research from NIH and industry.
Meanwhile, other research groups and companies are also working on identifying diagnostic markers for preterm birth, preeclampsia, and other pregnancy complications. For example, in an early exploratory study published in Scientific Reports earlier this year, Washington State University researchers found DNA methylation profiles in buccal cells may indicate a mother's inherent risk of a pregnancy with preterm birth. Similarly, researchers from Canada's Princess Margaret Cancer Centre and the University of Toronto are exploring a cell-free methylated DNA profiling method for prenatal testing applications.
Moving forward, Jain said Mirvie is on track to further validate its tests in samples from thousands of pregnant women through prospective studies. The company, currently with over 30 employees and a dozen or so contractors, is also "absolutely growing and hiring" to prepare for its commercial launch, he added.
"Over time, we do expect to develop tests for as many [conditions] as we can," he said.