A Double-Blind Crossover, Placebo-Controlled Study of the Adenosine A 2A Antagonist Theophylline in Parkinson’s Disease *Jaime Kulisevsky, †Manel Barbanoj, *Alexandre Gironell, †Rosa Antonijoan, †Miquel Casas, and *Berta Pascual–Sedano *Movement Disorders Unit, Department of Neurology; and †Clinical Pharmacology Service, Pharmacological Research Area, Sant Pau Hospital, Autonomous University of Barcelona, Spain Summary: Blockade of the adenosine A 2A receptor potentiates the effects of levodopa in experimental animals and may offer a novel nondopaminergic target for drug therapy in Parkinson’s disease (PD). Open-label trials suggest that the nonspecific adenosine antagonist theophylline improves parkinsonian symptoms and increases ON time in ad- vanced patients with PD. In a double-blind, crossover, placebo-controlled trial, the au- thors investigated the ability of stable plasma levels of theophylline (between 10–20 μg/mL after 15 days of treatment) to modulate the long-duration response and the short- duration response of levodopa in 10 patients with PD. Although theophylline induced a longer duration of the effect of levodopa in all Unified Parkinson’s Disease Rating Scale variables considered, including dyskinesias, maximal levodopa-induced improvement and the duration of the effect of levodopa did not differ significantly from placebo. Only the secondary variable “akinesia” showed a statistical tendency to a more prolonged beneficial response with theophylline during an acute levodopa test (short-duration re- sponse), and tremor worsened with theophylline during levodopa withdrawal (long- duration response). No differences were observed during the subacute course of study medication added to levodopa. During this exploratory study, the effects of theophylline were not strong enough to potentiate clearly the antiparkinsonian action of levodopa or to increase ON time in patients with advanced PD. Key Words: Theophylline— Parkinson’s disease—Adenosine antagonists Adenosine is an endogenous purine released from the nerve terminal that acts as a major modulator of neuronal activity and neurotransmitter release in the central nervous system. Its biologic effects are medi- ated through interaction with one of four G protein- coupled, membrane-bound receptor subtypes known as A 1 ,A 2A ,A 2B , and A 3 (1,2). Only the A 1 and A 2A receptor subtypes, which are blocked by methylxanthines (3), are reported to be en- riched in the mammalian central nervous system, with A 1 receptors inhibiting transmitter release and A 2 re- ceptors generally enhancing release of several trans- mitter systems relevant to the control of movement (4). The adenosine A 2A receptor subtype, which is of special interest in Parkinson’s disease (PD) (5,6), has restricted distribution in the mammalian brain and is abundant in the striatum, where it modulates the neu- ronal activity of striatal -aminobutyric-releasing out- put neurons (7,8). An enrichment of A 2A striatal recep- tor binding sites is found within -aminobutyric and enkephalin target sites (namely the GPe), but not within the GPi and SNr—targets of -aminobutyric and sub- stance P direct neurons (9). In situ hybridization studies of the adult monkey and human striatum demonstrate that there is a high expression of adenosine A 2A recep- tor messenger ribonucleic acid in the striatum that is colocalized extensively with the dopamine D2 recep- tor and preproenkephalin A messenger ribonucleic acids (10). Blockade of the A 2A receptor results in marked stimulation of locomotor activity in dopamine- denervated experimental animals (11–13), whereas stimulation of A 2A receptors causes hypomobility (14– 16)—an effect mediated presumably through a direct Address correspondence and reprint requests to Jaime Kulisevsky, Movement Disorders Unit, Department of Neurology, Sant Pau Hospi- tal, Sant Antoni M. Claret 167, 08025 Barcelona, Spain. Clinical Neuropharmacology Vol. 25, No. 1, pp. 25–31 © 2002 Lippincott Williams & Wilkins, Inc., Philadelphia 25