In lab, brain organoids mirror human development for nearly six years

Two junior scientists arrived in Paola Arlotta’s Harvard office with images of what they called the “old” brain organoids. The tissue in their dishes had stayed alive for more than five years—far beyond the months usually expected of these 3D models—and the researchers wanted to know what it had become.

The answer, from a study led by Arlotta’s team and published in Nature, is that the organoids continued to mature. They developed neurons and supporting glial cells, with several milestones—including signatures of postnatal development—appearing at the same cadence as in the human brain. Biological-age clocks trained on human data also matched the organoids’ molecular and chronological ages.

That longevity addresses a practical gap in neuroscience. Animal models capture some features of human brain development, but the process unfolds over nearly two decades, while direct cellular study in people is generally not feasible. Earlier organoids mainly modeled early development over months and were also vulnerable to variation between batches. Arlotta’s group had previously addressed that consistency problem; in the new work, the researchers used a fluid that supported neuronal activity, which they suspect helped the neurons last.

The cells also appeared to carry their own developmental history. When researchers moved older neurons into younger organoids, the cells continued along their previous trajectory instead of simply following signals from their new surroundings. Arlotta described this as a “warping of developmental time,” suggesting an internal cellular clock rather than development driven only by the surrounding tissue.

So what changes in practice? Researchers could have a longer-lived bench model for studying how disorders such as autism emerge and unfold later in life, and for testing experimental interventions at stages that short-lived organoids cannot reach. The limits remain clear: the organoids are simplified models, the brain normally interacts with many other systems, and the team says future work must add anatomical complexity and better simulate developmental disorders.