Spring 1995 Volume 4, Number 1

---

BY HUNTINGTON POTTER, PH.D.

lzheimer's disease is the most feared of the afflictions of old age. While heart attack and cancer have yielded encouragingly to molecular medicine, Alzheimer's still has no cure, and patients face an inexorable, slow decline in memory and cognitive ability. Nonetheless, we can be optimistic that recent advances in our understanding of the basic science of Alzheimer's will ultimately produce new approaches to therapy.

In clinical medicine, however, treatment must be coupled with early diagnosis. This is especially true of neurodegenerative disorders such as Alzheimer's, for once neurons in the brain die, they are not replaced. To be effective, therefore, any new therapy for Alzheimer's will need to begin as early as possible.

Unfortunately, the neuropsychological tests currently used to diagnose Alzheimer's are somewhat imprecise, and they often identify patients rather late in the disease process. Thus, a great deal of research is focused on finding a diagnostic test that will accurately identify Alzheimer's patients before they show symptoms of cognitive decline. One such effort - a collaboration between basic scientists at the Harvard Medical School Department of Neurobiology and clinical scientists at the Harvard-affiliated Beth Israel and Brigham and Women's hospitals - has recently led to a promising new approach to diagnostic testing, based on a neurobiological analysis of the eye.

If this test - which seems simplicity itself - is confirmed by further studies, it will be of major value in the fight against Alzheimer's disease. But that simplicity will have people asking, "How did they come up with the idea, and why didn't they think of it sooner?" It's a good question, and the answer says a lot about how science is done.

Except for pure serendipity (a frequent friend), all scientific results start with a plain question: "What if...?" For our potential Alzheimer's test, that question started taking shape one evening deep in the New Hampshire woods, at a scientific retreat for Harvard Medical School faculty and students. In that relaxed atmosphere, my thoughts wandered to a medical mystery that has intrigued many Alzheimer's researchers: Virtually every individual with Down Syndrome who lives beyond the age of 30 or 40 develops the same brain lesions and dementia that characterize Alzheimer's patients. Although the mechanistic relationship between Down Syndrome and Alzheimer's disease was unknown, it seemed to me logical that these two illnesses might have a similar etiology and, therefore, might share other clinical characteristics besides dementia and lesions in the brain.

Once back in Boston, I spent many hours in the library searching the voluminous literature on Down Syndrome for any feature that might also be present in Alzheimer's disease and have the potential to serve as a diagnostic test. Several turned up, and my longtime colleague, Dr. David Dressler, and I spent still more hours discussing them; we even tested some in the laboratory.

Reports in three obscure papers dating back to 1958 intrigued us most: that in Down Syndrome patients, the neuronal controls of the muscles of the heart and the iris of the eye were hypersensitive to a certain class of drug that inhibits acetylcholine-mediated neurotransmission. This hypersensitivity had been detected as an increase in heart rate upon injection of the drug, atropine, or by an increase in pupil diameter after a drop of atropine (or its synthetic analog, tropicamide) was placed in the eye. Neither Down's effect had ever been explored in patients with Alzheimer's disease, but they struck a special chord with us: First, Alzheimer patients had been shown to have a deficit of acetylcholine neurotransmission in the brain, and, second, we had just finished writing a book (Discovering Enzymes, Scientific American Library) in which the heart response to acetylcholine figures prominently.

Upon further consideration, the heart effect seemed too risky to use as a diagnostic test, because Alzheimer patients are often elderly and might suffer adverse reactions to a heart drug. In contrast, the tropicamide eye test would be simple and non-invasive, and only contraindicated for a person with the eye disease, glaucoma. ("Atropine plus glaucoma equals no diploma," goes the old medical school maxim.) Moreover, millions of people already receive tropicamide every year during routine eye exams. The only difference is that, normally, the ophthalmologist uses a 1 percent solution of tropicamide to dilate the pupil to allow examination of the retina behind, while a similar pupil dilation in the hypersensitive Down Syndrome patients is readily induced by only 0.01 percent - one one-hundredth of the standard dose.

We took this theory to Dr. Marsel Mesulam, an expert on cholinergic dysfunction in Alzheimer's, who was then at the Beth Israel Hospital. In one of those bits of luck that come along every once in a while, Mesulam had within his unit a colleague, Dr. Leonard Scinto, with the expertise and special equipment that could be used to measure pupil dilation. With the addition of a neurologist, Dr. Kirk Daffner, a collaboration was born.

Our team would carefully measure the rate and extent of pupil dilation in both diagnosed Alzheimer's disease patients and other, control, individuals without the disease, after placing a drop of tropicamide in one eye (the other eye serving as a standard for comparison). In testing an initial cohort of 19 Alzheimer patients and 32 individuals without the disease, all but one of the Alzheimer patients were indeed found to be hypersensitive to tropicamide, while 30 of the 32 normal individuals were relatively insensitive to the low concentration of drug used - statistically, a highly significant result.

Normal Eye                           Normal Eye, pupil dilated with tropicamide

This project, however, was not without its unique setback. The fourth "normal" individual we tested ("SG") was hypersensitive to tropicamide on the eye test, but showed no indications of Alzheimer's. If such individuals were common, the test would be useless. Nonetheless, we persevered, and over time it became clear that this first false-positive was in fact "the exception that proved the rule."

When we reexamined "SG" nine months later, he was still hypersensitive to tropicamide in the pupil dilation assay, but now showed severe cognitive deficits and had to be reclassified as an Alzheimer's patient. This was most important -the first indication that the eye test might be able to identify Alzheimer individuals before the onset of symptoms.

The news media reported our findings widely when they were published in the journal Science last November. Such a press reaction alarms scientists as much as it pleases them, because every good laboratory result faces years of confirmation and refinement before it enters clinical practice. Indeed, we and our colleagues around the world will conduct further trials with many more patients, to rule out the possibility that the results reflect some unknown phenomenon other than susceptibility to Alzheimer's disease, and to confirm the specificity of the pupil analysis for Alzheimer-type dementia. Only then can the Alzheimer neurological eye test move out of the laboratory and become a standard diagnostic procedure available from your doctor.

If the pupil test lives up to its potential as an early diagnostic, it will be particularly useful, because the Alzheimer treatments now being sought will only be able to halt the progress of the disease, not restore the neurons already lost. Thus, identifying patients in the earliest possible stages of the disease will allow the most successful therapy. *

Dr. Potter is Associate Professor of Neurobiology at Harvard Medical School.