Should Intracerebroventricular Nerve Growth Factor Be Used to Treat Alzheimer's Disease?

SHOULD INTRACEREBROVENTRICULAR NERVE GROWTH FACTOR BE USED TO TREAT ALZHEIMER'S DISEASE? BRUCE NATHAN SAFFRAN* Introduction Alzheimer's disease (AD), first described in a clinicopathological case report in 1907 [1], is a neurological disorder identified clinically by progressive memory loss and dementia. AD is now recognized as the most common form of adult-onset dementia [2—5] and it is currently estimated that AD affects approximately 10 percent of the population over 65. AD primarily affects the elderly and is characterized by a slow deterioration of cognitive function such that patients late in the course of the disease require virtually constant care. Furthermore, at the end stages of AD, patients often recognize neither friends nor family: a situation that itself can cause much suffering and frustration on the part of the patient's closest associates. Perhaps as a consequence of care-giver frustration , AD patients currently account for the majority of U.S. nursing home admissions [5]. Because this disease places such an enormous emotional and financial burden on both patients and their families, the pressure to find a cure for AD has never been more acute. Toward this end, many laboratories have investigated the histopathology of AD and, through the study of lesion-induced biochemical correlates in rats, have proposed a specific strategy for medical intervention in the early stages of the disease. Over the past eight years, several authors have suggested that the administration of nerve growth factor (NGF) into the cerebroventricular system could potentially slow the clinical manifestations of AD by rescuing neurons destined to die during the course of the disease [6, 7]. *Department of Internal Medicine, St. Elizabeth's Hospital, and the Tufts University School of Medicine, 736 Cambridge Street, Boston, Massachusetts 02135.© 1992 by The University of Chicago. All rights reserved. 003 1 -5982/92/3504-0797$0 1 .00 Perspectives in Biology and Medicine, 35, 4 ¦ Summer 1992 471 Since the initial formulation of the "NGF hypothesis," NGF therapy has been the focus of considerable excitement within both the basic science and the clinical communities, and clinical trials have been proposed to begin within the next year. However, among those working in the NGF field, considerable controversy still exists over whether NGF treatment will benefit Alzheimer's patients. In fact, some authors have suggested that intracerebroventricular NGF administration may do more harm than good [8, 9]. What is NGF, and what is the theoretical basis for its use in the treatment of AD? In this essay, I will examine the evidence supporting the use of NGF in the treatment of AD and will also present recent work suggesting that intracerebroventricular NGF injections may impair cerebral function or otherwise exacerbate the pathological phenomena associated with AD. However, in order to place in perspective the arguments both for and against the use of NGF to treat AD, I will first review the major pathological findings associated with AD and then will discuss how one of these findings, the loss of cholinergic neurons in the basal forebrain, is thought to be related to a deficiency of NGF produced in the cerebral cortex. Histopathological Features ofAlzheimer's Brains Histologically, the brains of patients afflicted with AD exhibit numerous ß amyloid—containing plaques within the cerebral cortex. The disease is also characterized by large numbers of cortical neurons filled with paired helical intermediate filaments, structures referred to as "neurofibrillary tangles." One notable aspect of AD histopathology is the marked degeneration of several populations of subcortical neurons whose axons project to cortical targets [4, 5, 8]. The most well-characterized population of neurons undergoing such degeneration is the magnocellular cholinergic neurons in the basal forebrain [10]. Early neurochemical studies of AD brains obtained at autopsy revealed a substantial reduction of presynaptic cholinergic markers in the cerebral cortex and hippocampus in patients with AD [11, 12]. In 1981, Price and coworkers succeeded in localizing the source of these enzymes to large cholinergic neurons in the substantia innominata, specifically to neurons in the nucleus basalis of Meynert (nbM), the medial septum, and the diagonal band of Broca [10]. These basal forebrain cholinergic neurons, routinely identified by their large size (when stained for Nissl substance) and their positive immunoreactivity for the enzyme choline...