(CNN) -- Potential cancer vaccine! Possible anxiety treatment! Scientific studies looking at potential therapies for physical and mental illness often sound exciting -- that is, until you read further and realize they're in mice.
The applicability to humans of studies in rodents varies widely, and some diseases are more accurately modeled in these animals than others, experts say.
Uncertainty about the applicability to humans is usually a big disclaimer in rodent studies, even in respected journals. One that's getting buzz in Nature Neuroscience suggests that a brain hormone is a "remote control" for cholesterol. The findings support previous research that the central nervous system controls processes that the body uses for energy, but must be further investigated before being translated to humans, the authors say.
Another is a recent study in Nature Medicine that found a vaccine was effective in preventing breast cancer in six mice that had been genetically engineered to develop breast cancer.
Generally, small studies like the cancer one appear to fit the adage "great news if you're a mouse," said Lois Parker, senior pharmacist at Massachusetts General Hospital. It's just too soon to draw meaningful interpretations from the available data.
"If it's just in rodents and only in a small number of rodents, personally I find it hard to get that excited about it," she said. "Maybe that's unfortunate, because maybe some of that stuff is worth getting excited about." But there's a large time gap from lab to clinical practice, she said.
Mice and rats are so often used because they can be bred easily, and they are convenient because of their small size, said Dr. Raymond Dingledine, chairman of the department of pharmacology at Emory University. Scientists also have an extensive knowledge of their physiology, so they can be compared against what is known about people.
But in most rat and mouse studies, the animals are highly inbred, and therefore represent a much more narrow sliver of genetics than seen in people, Dingledine said. Humans have a great deal more variability in their biochemical makeup than inbred mice, he said.
The U.S. Food and Drug Administration typically demands that drugs be tested in animal models before trials begin in humans. Out of every 250 compounds tested for safety in the lab or animal models, only one gets approved, according to the Pharmaceutical Research and Manufacturers of America. That's out of 5,000 to 10,000 compounds considered for testing in the first place. The entire process of development takes 10 to 15 years for those that make it to approval.
Here's the good news: There are specific conditions that rodents do model well. The effectiveness in humans of anticonvulsants have been quite predictive in mice and rat models, Dingledine said. Epilepsy turns out to be easily modeled in these animals.
But when it comes to the immune system, rodents and humans are so different that studies looking at treatments for certain diseases in mice or rats have not been successful when applied to people, he said. Increasingly, research is turning to nonhuman primates for preclinical research on immune-based therapies for a variety of conditions, according to the U.S. Department of Agriculture. These kinds of studies look at understanding the body's response to viruses, bacteria, and other invaders.
In terms of cancer, it is difficult to translate animal models to human applications because a human tumor takes several years to develop, and rodent studies often just inject human tumors into the rat or mouse, said Dr. Gabriel Lopez-Berestein, professor of medicine and cancer biology at the University of Texas M.D. Anderson Cancer Center.
In general, animal models give some hints about how to use cancer drugs in humans, he said. "But by no way should we be overexcited about it; we should be cautious about how to interpret that data" and in applying cancer studies in rodents to humans, he said.
Cancer drug researchers generally like to test potential treatments on two animal species, such as rats and dogs or mice and monkeys, and examine how the drug behaves in the fluids of the body, before proceeding to humans, Lopez-Berestein said.
"There are many genetic mutations that can result in cancer, and you can develop a compound or a drug that works great against one genetic mutation, but if that human patient doesn't have that genetic mutation, it'll be ineffective," Dingledine said.
Toxicity is also an issue for cancer research, as potential harmful side effects in humans do not always show up in rodents, Dingledine said.
We may think of our brain as an organ that is uniquely human in its intricacy and capacity for complex thought, but a great deal about us can still be learned from rodent models, researchers say.
Humans and rodents share brain circuits that are evolutionarily ancient, said Dr. Gregory Quirk of the University of Puerto Rico Medical School department of psychiatry. The amygdala, involved in the "flight or fight" response, and the prefrontal cortex, associated with decision-making, are human brain structures that can be probed in rats and mice.
Quirk's research has found that injecting a protein called brain-derived neurotrophic factor into the rat's prefrontal cortex decreases a fear response, which may be useful for the treatment of anxiety disorders such as post-traumatic stress disorder. It may be years before this can be translated into humans, but Quirk is hopeful that researchers can identify existing drugs to be used in this way.
Other research groups are also making headway on treating anxiety based on studies of mice and rats. An antibiotic called D-cycloserine was first observed in rodents to diminish fear and went into human clinical trials relatively quickly, he said.
"It can be frustrating for the public to hear about the nth rodent study," he said. "But I think fear and anxiety is one of the areas where there's low hanging fruit."
Rodent research has also been driving human brain imaging research, Quirk said. It helps scientists to form an initial hypothesis based on observed patterns of brain activity in rodents before testing humans. Of course, human brain scans may reveal activity in structures associated with higher thinking that rats do not have, he said.