To the naked eye, the little globs of cells are undifferentiated masses, smaller than sesame seeds. Put them under a microscope, though, and these lab-grown miniature organs show striking complexity: the tiny tubules of a kidney, the delicate folds of cerebral cortex, or a mucousy layer of intestinal lining. Now—after nearly a decade of figuring out how to make cells grow, organize, and specialize into 3D structures similar to human tissues, scientists have created a veritable zoo of “organoids,” including livers, pancreases, stomachs, hearts, kidneys, and even mammary and salivary glands. In a special issue published today in the journal Development, researchers in this young field describe what organoid research has achieved so far and report a handful of new advances. Here’s a crash course on these alluring—but imperfect—little models.

Are organoids really miniature organs?

Hardly. These clumps of cells resemble organs in many ways, but they lack certain features that allow real organs to function and grow—notably, a system of blood vessels that nourish internal tissue as the organ gets bigger. So for now, organoids don’t develop beyond tiny and simplistic models of organs. But they’re still a giant step up from 2D cultures of cells that scientists have long grown in the lab. Today’s organoids let researcher observe how organ structures emerge in early human development, and how certain genetic mutations or infections can derail an organ’s function.

How do you make an organoid?

All organoids begin as stem cells, grown in precise culture conditions that make them differentiate into multiple cell types that self-organize and cooperate. But the source of the stem cells makes a big difference. Some researchers start with pluripotent stem cells—isolated from a human embryo or reprogrammed from mature cells—which in theory can become any cell type in the body. These organoids reenact the growth of organs during the first weeks and months of life. They can help researchers spot glitches in that process, such as a genetic mutation that causes a lack of key cell types in the gut. “You can watch as a congenital defect unfolds before your eyes in the dish,” says James Wells, a developmental biologist at Cincinnati Children’s Hospital Medical Center in Ohio.

Such developmental models are especially prized in neuroscience: When researchers wanted to know how the Zika virus influences the growing brain, they turned to organoids, which revealed how the virus hijacks and then kills neural progenitor cells.

A cerebral organoid with neural stem cells (red) and neurons (green).

A separate field of organoid research relies on adult stem cells, isolated from the tissues that line our organs and help them regenerate after injury. These cells produce simpler structures than pluripotent stem cells, but they can still be powerful predictive tools. Organoids modeling the liver, stomach, intestine, and pancreas can reveal how genetic differences affect the function of an organ or how the body might respond to a drug.

Colon organoids grown from adult stem cells.

How good are predictions from organoids?

When it comes to testing how drugs work, at least one organoid is already proving itself in the clinic. A customized minigut model has shown which cystic fibrosis (CF) patients with rare forms of the disease could benefit from a new and expensive therapy developed by Vertex Pharmaceuticals. Patients provide rectal tissue samples that are then grown into a minigut model, developed by stem cell biologist Hans Clevers and colleagues at the Hubrecht Institute in Utrecht, the Netherlands. His team, along with researchers at Wilhelmina Children's Hospital in Utrecht, then test whether exposing these organoids to the drug prompts them to swell up in the same way as healthy guts—a sign that their cellular channels are moving salt and water effectively, not building up the mucus that leads to congestion and infections in the guts and lungs of CF patients. So far, seven children have received the drug—and reimbursement from insurance companies—based on its performance in their personalized gut models, Clevers says.

But as several authors point out in Development, how well an organoid mimics a true organ isn’t known. Neuroscientist Guo-Li Ming and colleagues at Johns Hopkins University School of Medicine in Baltimore, Maryland, have modeled Zika damage in cerebral organoids, and have even used them to screen promising drug candidates. But she notes that for now, these models don’t include an immune system, which likely influences how the disease develops. Researchers could simply mix immune cells in with the organoids, she notes, but “whether they function normally—we don’t know.”

And the field is facing down another big challenge: standardization. Leave stem cells to their own devices, and they’ll produce organs with various shapes and mixes of cell types. One paper published in today’s special issue describes a method of sorting intestinal cell cultures to isolate the ones that will grow into more uniform, mature organoids. But organoids remain hard to churn out in the large, consistent batches needed for drug screening and other efforts.

Can we use organoids to grow tissues for transplant?

Not yet. Many researchers hope that one day lab-grown tissues could be transplanted back into the body to regrow or repair an organ. Scientists have already shown that liver and colon organoids grafted into mouse organs can survive and grow. Wells and his collaborators at Cincinnati Children’s Hospital would like to use lab-grown tissue to heal intestinal damage after infection in premature infants. But because they can’t grow large chunks of intestine in the lab, they’re exploring how they might grow an organoid within a patient’s abdomen and then tie it into the intestinal tract. “We’re looking at a decade anyway of even thinking about trying this in a patient,” he says. Still, the progress in coaxing organ tissue to grow outside the body has been stunning. “The fact that we can ever get it to happen in a petri dish is, frankly, mind-numbing. … I can tell you, there is no way I ever saw this one coming.”