The Stanford Medicine study, published online Oct. 21 in Science Advances, reports that a combination of two widely available drugs—low-dose aspirin and lansoprazole—can delay labor significantly more than aspirin alone. The research, led by Brice Gaudilliere, used a novel artificial intelligence technique to create “digital twins” of individual pregnant women’s immune systems, simulating drug effects before any patient receives treatment.
The Immune Clock and the Search for Protection
For decades, scientists have tracked how the maternal immune system changes throughout pregnancy. Early on, the mother’s body adjusts its defenses to tolerate the growing fetus. Later in gestation, a shift toward pro-inflammatory signals helps ready the body for labor. When that immune clock runs too quickly—pro-inflammatory signals activated too soon—preterm labor can result. Prematurity, defined as birth three or more weeks before the due date, remains the leading cause of death for children under age five worldwide.
The Stanford team built a comprehensive atlas of the immune system, mapping it cell by cell during pregnancy. From this atlas, they created digital twins—computational models that continuously sync with real patient data—to virtually test how different drugs might shift the immune clock. The goal was to find a combination that could slow the process without harming the mother or fetus.
From Atlas to Digital Twin
The researchers began by collecting blood samples from pregnant women. Immune cells from these samples were exposed to many different drugs, and the team measured the cellular responses, generating data from a quarter of a billion cells. This massive dataset was then fed into the new system, dubbed SIMCO, which uses artificial intelligence and machine-learning techniques to integrate how all the body’s immune cells behave during pregnancy.
In the virtual simulations, the drug pair of aspirin and lansoprazole emerged as the most effective. In the digital twin system, the combination delayed labor 15 times higher than aspirin alone. Lansoprazole is sometimes prescribed in pregnancy to treat acid reflux, and low-dose aspirin has previously been shown to have a modest ability to prevent premature birth. The study notes that no new pharmacological intervention for preventing preterm birth has emerged in decades.
Hope in a Clinical Trial
Building on the simulation results, Stanford Medicine is launching a clinical trial to test the aspirin and lansoprazole combination in real patients. Senior study author Brice Gaudilliere said the work puts the field “on the brink of having a real impact for the treatment of preterm birth, a global health problem that affects 10% of pregnancies worldwide.” For families facing the uncertainty of early labor, the research offers a potential new path forward—one grounded in the complex, shifting biology of the pregnant immune system.
The Science Behind SIMCO’s Predictive Power
The Stanford team’s digital twin platform, SIMCO, represents a significant advancement in computational immunology. Unlike standard models, SIMCO continuously synchronizes with real patient data to emulate the dynamic behavior of an individual’s immune system. The system was trained on a massive dataset generated from immune cells exposed to various drugs, allowing it to predict how specific interventions would alter the “immune clock”—the biological signals that govern the timing of labor. By integrating cell-by-cell data from a quarter of a billion cells, SIMCO can simulate the complex interplay of pro-inflammatory and tolerance-related signals that precede childbirth.
Crucially, the model’s ability to identify the aspirin and lansoprazole combination stemmed from its capacity to test numerous pharmacological agents in a risk-free virtual environment. In these simulations, the drug pair demonstrated a dramatic effect, delaying labor 15 times more effectively than aspirin administered alone. This outcome highlights the platform’s utility in uncovering synergistic effects between existing medications. While low-dose aspirin has long been associated with a modest reduction in preterm birth risk, and lansoprazole is commonly used to manage gestational acid reflux, the study provides the first strong evidence that their combined use can significantly shift the immune pathways responsible for triggering labor.
The Role of Inflammation in Preterm Birth
Stanford Medicine research over the past two decades has established that the maternal immune system undergoes a critical shift late in gestation. Early in pregnancy, the mother’s body adjusts its defenses to tolerate the immunologically distinct fetus. However, later in gestation, a series of immune changes occurs, including a switch toward pro-inflammatory signals that help ready the body for labor. When this “immune clock” runs too quickly—with pro-inflammatory signals activated prematurely—preterm labor can result. Prematurity, defined as birth three or more weeks before the due date, remains the leading cause of death for children under age five worldwide, underscoring the urgent need for interventions that can safely modulate this biological process.
The creation of the SIMCO digital twin platform was driven by the need to map these complex immune dynamics. To build the comprehensive atlas required for the model, researchers collected blood samples from pregnant women and exposed the immune cells to a wide array of drugs. They measured cellular responses, generating data from a quarter of a billion cells. This massive dataset allowed the AI-powered system to integrate how all the body’s immune cells behave during pregnancy, creating a virtual model that continuously syncs with real patient data to emulate individual immune behavior and predict the effects of pharmacological interventions.
The Path to Clinical Testing
Following the promising results from the SIMCO simulations, Stanford Medicine has initiated a clinical trial to evaluate the aspirin and lansoprazole combination in human patients. Senior study author Brice Gaudilliere emphasized that the research places the field “on the brink of having a real impact for the treatment of preterm birth, a global health problem that affects 10% of pregnancies worldwide.” The trial builds directly on the virtual screening of a quarter of a billion immune cells, which identified the drug pair as capable of delaying labor 15 times more effectively than aspirin monotherapy. While low-dose aspirin has previously demonstrated a modest capacity to reduce preterm birth risk, and lansoprazole is commonly used to manage gestational acid reflux, this study provides the first robust evidence that their synergistic use can meaningfully shift the immune pathways governing labor timing. The transition from computational model to bedside represents a critical step toward addressing a condition that has seen few pharmacological advances in decades.
Participants in the upcoming trial will be monitored for both efficacy and safety as the combination therapy is administered under clinical supervision. The research team notes that because both drugs are already widely available and have established safety profiles in pregnancy, the path to potential approval may be accelerated compared to entirely new compounds. Gaudilliere stated that the work marks a significant shift in how preterm birth might be managed, moving from reactive treatment to targeted intervention based on individual immune profiling. For the millions of families affected by early labor each year, this trial offers a scientifically grounded avenue for hope, rooted in the detailed mapping of the maternal immune system’s transition from tolerance to parturition readiness.
Sources
- ‘Digital twins’ of immune system in pregnancy identify drugs to prevent early labor, Stanford Medicine-led study finds
- Digital twin
- Developing an AI Factory Digital Twin with an Agent-Assisted Workflow | NVIDIA Technical Blog
Photo: LPS.1 (CC0)
