TL;DR
Stanford researchers built a computer copy of the immune system, trained on blood from just 19 people, and asked it which existing drugs might keep a pregnancy going longer. Its top pick was a cheap pairing, aspirin plus the antacid lansoprazole, which it predicted would delay delivery a median 17.3 days in simulated patients counted as responders, about 15 times the roughly 1.1 days it predicted for aspirin alone. Every one of those numbers comes from simulation, and a human trial of the pair is now registered. If the trial bears the model out, babies at risk of arriving early could get more time in the womb from two familiar drugs, and the same method could sort through old medicines for other conditions. Computer copies of patients are a story we are following, because they could speed up nearly every medical timeline we track. To see our odds on when that trial reports and whether the pair works, plus the timeline of breakthroughs that leads there, subscribe.
Below the line, for paid subscribers: The full story: where it stands, how we got here and what's dated next; our labeled speculation on what it adds up to; and every source.