In a Barcelona laboratory and research institutions across the globe, scientists are learning to build tiny copies of human embryos from scratch—not from fertilized eggs, but from stem cells. These laboratory-grown models, published for the first time in an international white paper coordinated by ICREA researcher Alfonso Martínez-Arias of Pompeu Fabra University, could finally unlock why human reproduction remains so puzzlingly fragile, even as we've mastered the technology to help millions have children.

The stakes are enormous. One in six people of reproductive age experience fertility problems at some point in their lives, affecting 48 million couples worldwide. Yet despite assisted reproduction techniques having enabled the birth of more than 10 million babies over the past 30 years, a stubborn biological mystery persists: only one in three fertilized eggs progress beyond the third week of development. No one truly understands why.

The barrier is both scientific and legal. International law prohibits researchers from studying human embryos beyond 14 days of development—a boundary established for ethical reasons that has become a knowledge ceiling. "We are very knowledgeable about what happens during the first seven days of embryonic development; we know much less about what happens during the following week, in which the embryo becomes implanted in the uterus," Martínez-Arias explains. The third week is especially crucial because this is when gastrulation—the process that transforms a simple disc of cells into a complex, organized structure—occurs. During gastrulation, many cardiovascular disorders, metabolic conditions, and even limb malformations that appear after birth originate.

Stem cell-based embryo models offer a path around this impasse. Rather than using scarce surplus embryos from fertility clinics—which come with strict ethical limitations—researchers can now generate embryo-like structures in the laboratory from pluripotent stem cells, which have the capacity to become any cell type in the body. These models are reproducible, more abundant, and crucially, not subject to the same 14-day restrictions as actual human embryos. Scientists have already used them in monkeys to study gastrulation itself, demonstrating their potential.

The white paper, published in Human Reproduction following a November 2024 meeting of international experts at the Institut Pasteur in Paris, sets out the first comprehensive framework for how these models should be developed, standardized, and regulated. The experts emphasize that embryos and models must not be treated under identical regulations. Protecting embryos with the 14-day rule remains ethically sound, they argue, but stem cell models deserve recognition as "an extraordinary resource" to advance human understanding.

The challenge now is standardization. Current embryo models use disparate methodologies and cell lines, making results difficult to compare to actual human development. But Martínez-Arias is optimistic. "If we want embryonic models based on stem cells to be useful for research and the treatment of infertility, we must have reliable, reproducible models with a legal framework that allows us to use them," he says. The experts hope to break through this barrier in the near term—creating standardized models robust enough for clinical practice and the development of devices that could improve fertility treatments themselves.

For millions struggling with infertility, the white paper signals a shift: from watching the frontier of human development from behind a legal wall to finally being able to study what happens in those crucial, mysterious weeks after conception.