For the first time, laboratory-grown models of human brain tissue have matured over several years in a way that mirrors natural development, according to research funded by the US National Institutes of Health. A team led by Paola Arlotta at Harvard University kept cerebral organoids alive for nearly six years and recorded them passing through the same developmental milestones seen in a growing child’s brain.
Organoids are three-dimensional clumps of cells grown from stem cells that stand in for real tissue. Until now they have been short-lived, usually mimicking only the earliest months of brain development and varying widely from batch to batch. The Harvard team had already solved the consistency problem in 2019; in the new work they tested how long their organoids could survive and keep developing.
Over the nearly six-year period, the researchers tracked gene activity, the chemical tags that mark biological age, and cell structure at single-cell resolution. They watched neurons and their supporting glial cells emerge and mature, with hallmarks of postnatal development appearing in the same order and timing as they would in a human brain. Molecular “age clocks” trained on real-world data showed the organoids’ biological and chronological ages lining up, meaning they aged at a slow, realistic pace despite growing outside the body.
The neurons also behaved like living tissue. They formed connections with one another and fired electrical signals that the organoids sustained for at least two years. To confirm the pattern was real, the scientists moved older neurons into younger organoids; the older cells kept advancing as if still in their original environment, suggesting the cells carry an internal record of time in culture. The findings were published in Nature on 19 August 2026 and are described in an NIH news release.
The work could open the door to longer studies of conditions such as autism that unfold over many years, a process that is difficult to study directly in people and only partly captured by animal models. The study also points to new ways to model how the brain builds itself after birth.