Shades of grey: radiopharmaceutical chemistry in the 1990s and beyond.

Science, like the universe itself, is ever expanding. The fields of biochemistry, physiology and pharmacology are rapidly growing, and radiopharmaceutical chemistry is no exception. In a recent Editorial Dr William C. Eckelman briefly reviewed the current status of radiopharmaceutical development (Eckelman, 1991). The ballots are out and returns are mixed. Radiopharmaceutical chemists have made great strides in the development and refinement of methods for radiochemical syntheses, but have made little impact on clinical Nuclear Medicine. Dr Eckelman ends his Editorial with a plaintive plea for more validation of radiotracers. Dr Eckelman did not go far enough; radiopharmaceutical chemists need to be educated as to what are the important questions to which they should be devoting their time and talents. We do not need radiotracers to tell us where certain high affinity binding sites are in anima1 or human tissues; the fields of autoradiography and molecular biology are doing that just fine, with far greater resolution than we can ever hope for. It is not sufficient that new radiotracers are “validated” by a lack of specific binding after pharmacological pretreatments with massive amounts of cold drugs, or after severe neurochemical lesions in animal models, or in end-stage human disease. The answers we seek are not so black and white, but rather shades of grey. We need to develop sensitive in uiuo measures of biochemical parameters which impact on the prevention or therapeutic treatment of human diseases. Nothing less. In his Editorial, Bill Eckelman outlines three steps in radiotracer development. First, “develop a radiotracer that binds preferentially to a specific site”. Second, “determine the sensitivity of the radiotracer to a change in biochemistry”. Finally, “find a biochemical change as a function of a specific disease that matches that sensitivity”. Dr Eckelman may have put the cart before the horse. The first step in a radiopharmaceutical development program should be to identify the relevant disease and the needed sensitivity (the percent change in any given parameter) which will be the focus of the study. Fortunately, more and more diseases are now being defined biochemically, with definitions of both the site and extent of an abnormal biochemical process. These data should be our ticket to the fun-house and our guide through the unknown. Admittedly, in many disease states the needed sensitivity may not be known; a good example is Parkinson’s disease, where it is not known exactly what degree of change in the striatal dopaminergic system results in progression from normal function to symptomatic disease (and will this imaginary boundary be the same for all subjects?). Even in these cases, however, it might be expected that the large scale changes present at end-stage disease will not be the primary target of in vivo studies, and development of radiopharmaceuticals with a matching all-or-none sensitivity might be of lesser interest. One cannot disagree with Dr Eckelman that too little time, thought and effort is spent on the analysis of the sensitivity of a radiotracer measure, sometimes erroneously termed “validation”. But just what is “validation”? It depends upon the question being asked. If one wishes to know the rate of loss of a specific binding site during the course of a neurodegenerative disease, a sensitivity to S-10% changes in biochemistry might be required for medical utility. If one wishes, on the other hand, to differentiate between pathologies involving two optional but large scale changes in biochemistry, or simply the location of a large biochemical change, far less sensitivity might be perfectly acceptable. So what kinds of sensitivity testing should be pursued? A favorite of many is the correlation between in vivo values, such as tissue ratios, B,,,,,, binding potential, volume of distribution or other parameter and in vitro measures of the same binding site or enzyme site densities as measured in different