CRISPR gene editing technology has tantalized the public with its potential to cure disease.

However, new research suggests it could be more dangerous and less precise than previously believed.

CRISPR-Cas9 was discovered in 2012 by University of California molecular biologist Jennifer Doudna and her colleagues. It allows for genetic “editing” by snipping out small bits of defective or harmful DNA and replacing it.

Gene editing has existed since the 1970s, but CRISPR-Cas9 has reinvented it as a precise, accessible technology.

The potential applications seem almost limitless.

This year, Dr. Edze Westra of the University of Exeter, told the Independent that he expects the technology to be “used to cure all inherited diseases, to cure cancers, to restore sight to people by transplanting genes.”

Read more: Scientists find gene editing with CRISPR hard to resist »

Promises and problems

Still in its infancy, CRISPR-Cas9 has yet to deliver on these promises, in humans anyway.

One of the key talking points of CRISPR-Cas9 has been its precision — its ability to accurately edit small sections of DNA without affecting nearby sections.

However, a new from Columbia University says that CRISPR-Cas9 can introduce hundreds of unexpected mutations into the genome beyond what was intended.

“We feel it’s critical that the scientific community consider the potential hazards of all off-target mutations caused by CRISPR,” said co-author Dr. Stephen Tsang, a professor at Columbia University Medical Center, in a press release.

Tsang and his team discovered the mutations while conducting research on mice, using CRISPR-Cas9 to correct a gene that caused blindness.

The technology worked effectively in curing the blindness, but when the researchers later looked at the genome of the mice, they said they found additional, unintended mutations.

Despite this, the mice appeared to be in fine health.

“We did not see any observable complications in the mice, despite having all these extra CRISPR-related mutations,” Tsang told Healthline.

Sheila Jasanoff, professor of science and technology studies at Harvard University, told Healthline that “precision” can have a slippery definition in biotechnology.

“Genetic engineering was also sold some 40 years ago as a highly precise technique. Now, CRISPR is being heralded as even more precise,” she said.

“Undoubtedly, there is some truth in that claim ... But we also know from older genetic engineering techniques that very precise interventions into one part of a genome can produce unexpected side effects — or off-target impacts — that scientists were not expecting,” Jasanoff added.

Read more: CRISPR gene editing and cancer treatment »

Risk vs. reward

Tsang frames the message of his research in two ways.

First, he hopes that his work will bring a newfound awareness to the potential side effects caused by CRISPR.

Although the mutations he and his team observed did not appear to have any malignant effects, they should be a wake-up call for researchers.

Secondly, Tsang says that no matter what kind of medicine or treatment is being used, there is the potential for side effects.

“If we apply CRISPR, it’s just like any other intervention medicine. There is always off-targeting and risks and benefits,” he says.

Jasanoff is more tempered in her assessment of the risk vs. reward of CRISPR.

“The assumption that there are untold benefits in store — long before the work has been done to establish how a new technology actually will have an impact on any disease — is a typical example of the hype that surrounds new and emerging technologies,” she said.

Tsang’s research offers no hard answers to the larger questions of efficacy, risk, and benefit of using CRISPR on humans.

“Let’s not go overboard,” said Pete Shanks, a consultant who is an expert on genetics. “Three blind mice don’t prove much.”

Tsang’s research does provide some cautionary insight into how research must be conducted in order to make the technology safer.

Currently most studies of off-target mutations depend on computer algorithms to locate and examine affected areas. Tsang and his team say that this isn’t sufficient when using live specimens.

“These predictive algorithms seem to do a good job when CRISPR is performed in cells or tissues in a dish, but whole genome sequencing has not been employed to look for all off-target effects in living animals,” Alexander Bassuk, professor of pediatrics at the University of Iowa, and co-author of the study, said in a press release.

“Researchers who aren’t using whole genome sequencing to find off-target effects may be missing potentially important mutations,” Tsang said.

Read more: Gene editing could be used to battle mosquito-borne disease »

CRISPR moving ahead rapidly

This study comes at an important time.

China has begun its first round of human testing using CRISPR-Cas9.

The United States is due to start its own tests next year.

The research field is moving quickly — perhaps too quickly.

“We hope our findings will encourage others to use whole genome sequencing as a method to determine all the off-target effects of their CRISPR techniques and study different versions for the safest, most accurate editing,” Tsang said.

Jasanoff is much blunter.

“We should put aside the notion the benefits of CRISPR are already proven, and all we need to worry about is risks,” she said.