Researchers announced they had derived stem cells from cloned human embryos, a long-awaited research coup that Science‘s editors chose as a runner-up for Breakthrough of the Year.

After more than a decade of failed attempts, it finally worked. This year, researchers announced that they had cloned human embryos and used them as a source of embryonic stem (ES) cells—a long-cherished goal. Able to develop into any tissue while providing a perfect genetic match to the cell that was cloned, the ES cells could prove a powerful tool for research and medicine. However, concerns about destroying embryos and the emergence of a cheaper, easier rival technique might keep human cloning for stem cells from becoming standard practice.The cloning technique, called somatic cell nuclear transfer (SCNT), is the same one used to clone Dolly the sheep 17 years ago. Scientists remove the nucleus from an egg cell and then fuse the remaining cell material with a cell from the individual to be cloned. They then give the fused cell a signal to start dividing, and when things go right an embryo develops. Scientists have used SCNT to clone mice, pigs, dogs, and other animals, but human cells proved trickier to work with. Years of trying—and a high-profile fraud—yielded nothing more than a few poor-quality embryos, unable to produce ES cells.

At the Oregon National Primate Research Center in Beaverton, however, researchers finally cloned monkey embryos and derived ES cells from them in 2007. In the process, they discovered a number of tweaks that made SCNT more effective for primate cells, including human ones. The final recipe worked surprisingly well, yielding ES cells in about one in 10 tries. One key ingredient seems to be caffeine, which appears to help stabilize key molecules in the delicate human egg cells.

How important the technique will be in the long run is an open question. In the years since human cloning was first attempted, researchers found that they can make patient-specific stem cells by “reprogramming” adult cells into induced pluripotent stem (iPS) cells. That method, which scientists adapted to human cells in 2007, eliminated the need for human eggs and does not involve embryos, two aspects that make SCNT controversial and expensive. But some experiments have suggested that, at least in mice, ES cells from cloned embryos might be of better quality than iPS cells. Now, investigators will be able to compare the two types of human stem cells side by side.

The feat also raises concerns about cloned babies. But that seems unlikely for now. Despite hundreds of tries, the Oregon researchers say, none of their cloned monkey embryos have established a pregnancy in surrogate females.