Researchers have identified two common drugs that may, when given together during pregnancy, prevent premature births.
The finding comes from a new, Stanford Medicine-led, artificial intelligence-powered technique for screening drugs to prevent early labour. It uses a comprehensive atlas of the immune system, accurate at cell-by-cell level, to create ‘digital twins’ of the immune systems of individual pregnant women for virtual drug testing.
The study, published online 21 October in Science Advances, builds on two decades of Stanford Medicine research on maternal immune shifts during pregnancy. When the immune clock of pregnancy runs too quickly, preterm labour can result.
After screening several possible treatments with the digital twin system, the researchers identified one drug combination that promises to markedly lower the risk of early labour. It shows sufficient potential that researchers are starting a clinical trial.
“We’re on the brink of having a real impact for the treatment of preterm birth, a global health problem that affects 10% of pregnancies worldwide,” said senior study author Dr Brice Gaudilliere. It has been decades since a new pharmacological intervention for preventing preterm birth.
The winning combination was aspirin and lansoprazole. Low-dose aspirin alone has a modest ability to prevent premature birth; lansoprazole is sometimes used in pregnancy for acid reflux.
“In the digital twin system, these two drugs in combination had a large effect on the delay of labour, 15 times higher than aspirin alone,” Gaudilliere said.
Modelling how the immune clock ticks
Although triggers of labour remain incompletely explained, Stanford Medicine research has shown immune signals play a key role. Early in pregnancy the maternal immune system adjusts to tolerate the fetus. Late in pregnancy, immune changes – including a switch toward pro-inflammatory signals – prepare the body for labour. When pro-inflammatory signals activate too soon, early labour can result.
The new system, SIMCO (simulated immune modelling of clinical outcomes), uses AI and machine-learning to integrate extensive data on immune-cell behaviour during pregnancy.
Researchers collected blood samples from pregnant women and exposed immune cells to multiple drugs. “We measured responses from a quarter of a billion cells, training the algorithm on a massive single-cell dataset,” Gaudilliere said.
“A digital twin has the prerequisite that you understand a biological system very well,” said co-lead author Dr Jakob Einhaus, postdoctoral scholar at Stanford Medicine. “The digital twin represents the complexity and connections in the immune system.”
Einhaus shares lead authorship with graduate students Peter Neidlinger and Olivier Fondeur, and with Dr Masaki Sato (now at the National Center for Child Health and Development, Tokyo).
Longitudinal data from 63 pregnancies were integrated to link cell-behaviour patterns to clinical outcomes and capture inter-patient variability. The algorithm predicts individual immune-cell responses to drugs and, from limited patient data (e.g., a small blood sample), the likely clinical effect on the immune clock of pregnancy.
“You’re making the prediction for one individual based on knowledge of millions of single cells from dozens of patients,” Einhaus said.
Common medications may delay delivery
After validating that SIMCO forecasts time to labour, the team evaluated nine pregnancy-safe drug regimens identified via a drug-repurposing pipeline developed by a group led by Dr Marina Sirota (University of California, San Francisco). Multiple digital-twin copies compared treated versus untreated scenarios for each individual.
The model strongly predicted that the aspirin–lansoprazole combination would delay labour. “Our findings suggest that salicylic acid [aspirin] and lansoprazole act synergistically … this synergy may enhance their immunomodulatory effect,” the investigators write.
Folic acid also delayed immunological progression toward labour. “We’re giving folic acid for spina bifida prevention, but maybe, unknowingly, we’re also helping the maternal immune system adapt to pregnancy,” Einhaus said.
The model indicated specific immune-signal changes: for the aspirin–lansoprazole combination, processes supporting pregnancy continuation are enhanced while those triggering labour are inhibited.
Because the model captures patient variation, it can identify likely responders and non-responders, potentially allowing smaller, more efficient trials restricted to predicted responders.
The combination delayed labour onset in an animal model of inflammatory pregnancy; FDA approval was obtained for a human trial.
Source: Stanford Medicine
