The central concept in toxicology is the dose-response relationship. It is used to determine acceptable doses of medicines, additives to foods, exposure standards for environmental and occupational hazards, radiation exposures from TVs, cell phones, microwaves, X-rays as well as numerous medical devices. It is hard to imagine that this quiet concept can play such an insidiously dominant role in our lives. But it does. Since the dose response concept is so significant to our welfare, it is essential that those who derive the "acceptable" exposure values get the "right" answer. That is the problem. For the past 25 years the field of toxicology has taught (i.e., indoctrinated) students throughout the country (and here at UMass) that the dose response is a threshold response for non-carcinogens while being linear at low doses for carcinogens. These two concepts have determined what the levels of lead, cadmium, mercury as well as benzene, trichloroethylene, and dozens of other potentially toxic substances can be in your food, drinking water, air and soil.

Ongoing studies by Linda Baldwin, departmental assistant in the Environmental Health Sciences Department, and myself, at UMass are challenging these concepts. We have amassed substantial information that the most fundamental shape of the dose response is neither threshold, nor linear (Figure 1), but rather U-shaped (or inverted U) (Figure 2). This type of dose response which is characterized by a low-dose stimulation, high-dose inhibition is referred to as hormesis, a term that originated in 1942, but the concept has a much longer history going back to the late 1880's. This information has been the recent object of numerous papers in the toxicological literature and the subject of several major conferences. The hormesis concept strikes at the very heart of how studies are designed, how risk is estimated, what exposure standards are and how new drugs may be designed and medical treatments employed.

What are the implications of a switch in perspective from threshold/linear models, to a U-shaped hormetic model?

1. The most important environmental implication is that carcinogen risk would no longer be considered linear at low dose. It would, in fact, display both a threshold and an actual decrease in risk at lower doses. While this conclusionary interpretation would appear environmentally sacrilegious, so be it. Our goal is to follow the data, not the prevailing and well-entrenched paradigm. Recognition of the hormesis phenomenon would directly change how risk is determined (e.g., radon levels in homes) and how clean-up standards are derived.

2. Recognition of U-shaped dose responses can be used to enhance adaptive responses either in prevention or in treatment of disease. For example, it is now well known that a moderate consumption of ethanol may reduce heart disease, moderate stress from exercise may enhance numerous molecular adaptive functions, and that a low whole body dose of X-rays can enhance immune responsiveness and reduce risks of certain tumor types.

3. That U-shaped dose responses may not always be a positive thing for people. That is, while high doses of antibiotics kill bacteria, very low doses have often been shown to enhance their growth. This is also the case for anti-viral drugs. This concept also applies to anti-tumor drugs regardless of mechanism. This ignored phenomenon is an important area which could enhance the success of many medical treatments.

A major question is how could the field of toxicology have missed this concept for so many years.

Toxicology has typically focused on high doses; emphasis on high doses and extrapolation procedures that assumed low-dose linearity for carcinogens created false impressions and poor guidance. In fact, this is still in practice within all federal and state environmental and public health agencies in the U.S.

Proving that U-shaped dose responses exist can be difficult and requires much more testing than "high"-dose toxicity.

Low-dose stimulatory phenomena have often been incorrectly linked to the medical practice of homeopathy. Since traditional medicine has had long standing antipathy toward homeopathy it led to hostility toward the hormesis concept in pharmacology and toxicology which had their origins in medical schools.

Low doses are now driving the future of toxicology. This suggests that the future of toxicology and risk assessment will be filled with U-shaped dose responses. The key will be using this information to assist environmental leaders in making data driven conclusions on environmental standards and clean-up activities and to assist pharmaceutical companies in deriving drugs that do not fall victim to a concentration-based response flip-flop in which there are therapeutic effects at a high doses but harmful effects at much lower doses. While the dose-response relationship is more complex than previously believed, such new developments will lead to better decisions that should profit our health and the economy.

Dr. Edward Calabrese is a professor in the Environmental Health Sciences Department.