The new genome-editing tool called CRISPR is sparking enormous excitement about the possibility of repairing DNA defects and curing disease. A biologist explains its promise while pointing out technical perils and troubling ethical questions.

Scientists have been manipulating DNA for decades — adding to the mouse genome, for example, to create living models with Alzheimer’s and other diseases, with the aim of exploring their causes and potential treatments. And, just as we can add, we can delete genes, in order to gather clues about the role they play in development and physiology.

But early genetic-manipulation techniques were slow, expensive and geared to individual species. Enter CRISPR* — a fast, cheap and flexible way to make precise changes in any cell’s DNA.

The CRISPR system was first identified as a sort of immune system in bacteria, enabling cells to recognize and destroy the DNA of invading viruses. Scientists harnessed this system to devise a genome-editing tool, which rapidly entered the research mainstream. It’s tremendously exciting, because it’s simple and it can be used in practically any living thing. My own group of undergraduates at Kenyon is using CRISPR to genetically modify frogs as part of our research on the toxic effects of dioxin pollution.

Even before the emergence of CRISPR, scientists were experimenting with gene-editing technologies to treat medical conditions in humans. Some studies, for instance, have aimed at genetically altering immune cells so that they’ll destroy leukemias or inhibit the proliferation of HIV. But CRISPR’s enormous promise has both scientists and investors aggressively seeking new therapeutic applications. One current example: Editas Medicine — a genome-editing company founded in 2013 — hopes to use CRISPR to alter a gene within retinal cells so as to restore sight to patients with Leber congenital amaurosis, a rare cause of heritable blindness. Initial clinical trials are slated for 2017.

It’s easy to envision CRISPRs that repair, replace or block production of defective proteins. Red blood cell progenitors could be engineered and transplanted into bone marrow to fix the gene that causes sickle cell anemia, or the production of blood clotting factors could be restored in the livers of hemophiliacs.

These scenarios share an important limitation, however. They affect only one patient, the recipient of the genome edits.

A more ambitious goal would be to modify the “germ line,” the cells that give rise to sperm and egg, thereby creating genetic alterations that could be passed on to a patient’s children. The hope would be to eliminate inherited diseases like cystic fibrosis, muscular dystrophy or Huntington’s disease.

Edits in the human germ line must be made on in-vitro-fertilized human embryos. And that’s where serious concerns arise.

The most immediate problem is technical. In the only publicized attempt to perform CRISPR edits in an early human embryo, Chinese scientists detected numerous “off-target” effects, unanticipated changes at multiple sites in the genome. If a CRISPR-modified embryo were actually implanted and brought to term, unpredictable off-target mutations might cause birth defects or diseases.

Moreover, it’s impossible to ensure that every cell in the developing embryo is altered. The procedure could yield a “mosaic” child, in whom only a random fraction of cells contain the therapeutic edit, which then might not provide the medical benefit. Clearly, CRISPR technology is not yet safe enough for medical use in human embryos.

Then, of course, there are profound ethical questions. While some medical scientists argue that it’s morally wrong to withhold the cure to a genetic disease, the alteration of the human genome raises the specter of controlled human breeding. “Eugenics” is clearly unacceptable when it involves forced sterilization or murder. But what if we used technology to select specific traits in offspring — height, for example, or skin color or intelligence? Would we be blurring the definition of what a “normal” person is? Would some reasonably common traits — traits that now fall within the wide spectrum of human variability — come to be seen as “defects”? How would widespread genome editing affect the population genetics of our species in the future?

Additionally, although CRISPR is touted for its low cost and ease of use, it remains a sophisticated experimental technique. If available only to the rich, it could exacerbate and entrench economic inequality at the biological level.

Issues such as these cry out for international consensus. Until then, scientists in all nations would be wise to follow the laws of many western European countries and the policy of the U.S. National Institutes of Health, which for now establish a strict moratorium on genetic manipulation of the human germ line.

* CRISPR stands for “Clustered Regularly-Interspaced Short Palindromic Repeats,” a reference to the bacterial system on which the genome-editing technology is based.