The release of a study Friday linking cancer in rats to the type of radiation emitted by cell phones presents some of the strongest implications in more than two decades of research that higher doses of such signals could be linked to tumors in laboratory animals—unsettling news for billions of mobile phone users worldwide. Still missing, however, is a clear understanding of exactly how radiofrequency (RF) radiation used by mobile phones might create cellular-level changes that could lead to cancer.

The study by the U.S. National Toxicology Program (NTP) found that as the thousands of rats studied were exposed to greater intensities of RF radiation, more of them developed rare forms of brain and heart cancer that could not be easily explained away, exhibiting a direct dose-response relationship. NTP acknowledges that the research is not definitive and that more research needs to be done.

This is familiar territory for Jerry Phillips, a biochemist and director of the Excel Science Center at the University of Colorado at Colorado Springs. Phillips conducted Motorola-funded research into the potential health impacts of cell phones during the 1990s while he was with the U.S. Department of Veterans Affairs' Pettis VA Medical Center in Loma Linda, Calif. Phillips and his colleagues looked at the effects of different RF signals on rats, and on cells in a dish. “The most troublesome finding to Motorola at the time is that these radiofrequency signals could interact with living tissues, which is what we saw in the rats,” he says.

Scientific American spoke with Phillips about the NTP’s announcement, as well as his own experiences trying to understand how RF signals could be causing the DNA damage seen in his lab’s rats.

[An edited transcript of the interview follows.]

How is cell phone radiation different from other forms of radiation?

Cell phone radiation is non-ionizing radiation. X-rays, for example, are ionizing radiation and contain sufficient energy to break chemical bonds. Non-ionizing radiation associated with radiofrequency fields is very, very low-energy, so there’s insufficient energy to break chemical bonds. It was always assumed that because the power being created by the handsets was low enough, there would be insufficient energy for heat production—and without heat production there would be no biological effects [on users] whatsoever.

What happens to living cells when they are exposed to RF radiation?

The signal couples with those cells, although nobody really knows what the nature of that coupling is. Some effects of that reaction can be things like movement of calcium across membranes, the production of free radicals or a change in the expression of genes in the cell. Suddenly important proteins are being expressed at times and places and in amounts that they shouldn’t be, and that has a dramatic effect on the function of the cells. And some of these changes are consistent with what’s seen when cells undergo conversion from normal to malignant. These effects vary depending on the nature of the signal, the length of the exposure and the specifics of the signal itself.

How does the use of rats impact the validity of a study designed to determine whether cell phones are safe for people?

We try to find the best model system available based on physiology, genetics and what we know about biochemistry. Rats are really a pretty good model for humans. Of course, the question you’ve asked is now what the [wireless device] industry is going to hit on. Their primary rebuttal is that these are rats and not people.

NTP studied both Code Division Multiple Access (CDMA) and Global System for Mobile (GSM) modulations, which dictate how signals carry information. Why test more than one modulation in a study like this?

You test those two modulations because both are in wide use today. I don’t know exactly what [the NTP’s] rationale was, but the rationale we used for our study in the 1990s was to find out if signal modulation had an effect on what we were looking at. Part of the problem studying radiofrequency radiation is that we have not a clue what constitutes a dose. If you have a chemical, you can weigh it out and you know what the dose is. But with radiofrequency radiation there are too many parameters—power intensity, carrier frequency, length of exposure, signal intermittency or some combination—and nobody knows what’s most important.

What has been the prevailing argument against non-ionizing radiation causing cancer?

It’s a complicated issue. If you look as something as simple as smoking—for so long people had no clue what was in cigarette smoke that caused cancer. You could see when a smoker died that the lungs were different from those of a non-smoker, but at first it was hard to identify the mechanism causing the change in the lungs. It’s been the same sort of argument here—there’s no plausible explanation that something with such low energy could cause significant biological effects that are adverse to human health and development. Those of us working in the area of gene expression saw those effects, but there had been no way to explain them.

What should people take away from the NTP’s latest study results?

All this really does is provide a couple of answers but raise even more questions. My guess is that the needle won’t move much at all in this country. If you look at all of the research being done on this, it’s all from outside this country. People want to believe their technology is safe. I do. I would love to believe it, but I know better.

How do you reconcile your own cell phone use with the potential health hazards?

I’ll connect the phone to Bluetooth in my car. Or I’ll text. Or I if I have to make a phone call I put it on speaker. But you have to realize that this issue opens up a much bigger can of worms than cell phones. If this radiation, this form of energy can interact with biological tissue then it’s going to reopen a lot of what were supposedly settled issues regarding the safety of wireless communications. If we’re going to be bathed in a whole new electromagnetic environment, how safe is it?