Psychiatry has spent a century trying to treat what it could not see. Patient-derived neurons are about to change that.
Depression touches a staggering portion of the U.S. population. According to Gallup’s 2026 polling,1 19.1% of American adults, roughly 51 million people, are currently being treated for depression, and 29.5% report having been diagnosed in their lifetime. Among adults under the age of 30, the rate has more than doubled in less than a decade, climbing from 13% in 2017 to 28% in 2026. The disease accounts for billions of dollars in lost productivity and is consistently ranked as one of the most commonly diagnosed disabilities globally.
Yet, for how common it is, and despite a large market of FDA-approved antidepressants, psychology still hasn’t found a reliable way to treat patients quickly. Two-thirds of patients don’t respond to their first antidepressant2 and even after a few different medication switches, a third of patients still only experience partial relief. Each failed attempt to find the right medication prolongs patients’ suffering. A typical antidepressant takes six to twelve weeks to confirm whether it is working, and a patient may cycle through four to six different medications before something fits. With such a long journey towards effective treatment, many patients give up when the extended trial and error and constant side effects don’t seem to be letting up. This leads to a substantial share of patients never finding an effective treatment.
Another issue that we encounter more frequently is treatment resistant depression (TRD), defined as an insufficient response to two or more antidepressant medications within a depressive episode or period of sustained depression symptoms. The framing, however, is misleading. These patients are not necessarily resistant to all medications. In many cases, their underlying biology was simply never matched to the right medication type to begin with.
