Role of Naltrexone in Initial Smoking Cessation: Preliminary Findings

E FOR AN association between nicotine and the endogenous opioid system comes from several lines of research. Studies have shown that administration of nicotine and/or stimulation of nicotinic receptors activates -endorphin and enkephalin release and biosynthesis (Eiden et al., 1984; Gilbert et al., 1992; Houdi et al., 1991; Pomerleau et al., 1983). In addition to having higher rates of smoking than in the general population, heroin addicts have been shown to smoke more during heroin selfinjection (Mello et al., 1980) and smoke less during methadone dose tapering (Bigelow et al., 1981). Furthermore, opioid antagonists block certain effects of nicotine, including antinociception, operant responding, and prolactin release (Aceto et al., 1993; Corrigall et al., 1988; Flores et al., 1989). The opioid antagonist naloxone has also been shown to induce withdrawal-like states in nicotine-dependent rats and humans (Krishnan-Sarin et al., 1999; Malin et al., 1993, 1996). Conversely, nicotine has also been shown to suppress opioid antagonist effects such as decreasing naloxone-induced jumping in morphine-dependent mice (Brase et al., 1974; Zarrindast and Farzin, 1996). Although there seems to be evidence of an opioid–nicotine interaction from in vivo and in vitro studies, human preclinical studies have largely shown mixed results. Specifically, there are varied results from laboratory studies on the effects of the opioid antagonists naltrexone (or intravenous naloxone) on aspects of smoking response. Some studies have shown that naltrexone or naloxone significantly reduce subjects’ self-reported pleasure or satisfaction from smoking; perceived difficulty in abstaining, craving, and urge to smoke; and the number of cigarettes smoked or puffs taken (Gorelick et al., 1989; Houtsmuller et al., 1997; Karras and Kane, 1980; King and Meyer, 2000; Palmer and Berens, 1983; Sutherland et al., 1995; Wewers et al., 1998). In conjunction with the nicotine patch, naltrexone has also been shown to significantly decrease negative affect and craving after presentation of a smoking cue (Hutchison et al., 1999). However, these findings are not ubiquitous, as other studies have failed to show significant effects of opioid antagonists on altering ad libitum smoking, number of cigarettes or puffs taken, or desire to smoke (Gorelick et al., 1989; Houtsmuller et al., 1997; NemethCoslett and Griffiths, 1986; Sutherland et al., 1995). Also, naltrexone or naloxone has been shown to increase smokers’ negative mood and withdrawal-like symptoms (Brauer et al., 1999; Krishnan-Sarin et al., 1999). Collectively, there seems to be a lack of consistency in findings from preclinical human laboratory studies on opioid antagonist effects on smoking behaviors. Although there are no simple explanations, several methodological issues may be contributing to this disparity, such as the use of small sample sizes, lack of statistical power, a range of naltrexone or naloxone doses or routes of administration, and fundamental differences in paradigms (timing issues, degree of naturalism, baseline abstinence, etc.). A few studies have examined naltrexone in smoking cessation, and no studies have examined concurrent smoking and alcohol behaviors in study patients. As in laboratory studies, the results from clinical trials using adjunct treatment with naltrexone for nicotine dependence have shown mixed results, with negative findings in one study (Wong et al., 1999) and positive or mixed findings in two other studies (Covey et al., 1999; O’Malley et al., 1997). Although it is well established that alcohol may increase the risk for smoking relapse (Baer and Lichtenstein, 1988; Brandon et al., 1990; Shiffman 1982, 1986), smoking cessation trials have not systematically targeted pharmacological treatment strategies that may affect both behaviors. It is possible that naltrexone, a medication used for alcohol dependence (O’Malley et al., 1992; Volpicelli et al., 1992) and posited to block alcohol’s pleasurable effects (King et al., 1997; O’Malley et al., 1996; Swift et al., 1994; Volpicelli et al., 1995), might theoretically be an effective smoking cessation treatment agent, especially in smokers who drink. This may be due in part to direct nicotine–opioid system interactions (as stated earlier), as well as blockade of alcohol reward and concomitant drinking episodes, which bode poorly for smoking outcome. Finally, the timing of naltrexone may be an important variable. Data from our previous laboratory study (King and Meyer, 2000) indicate that naltrexone attenuated smoking urge and pleasure during and after cigarette exposure; therefore, medication randomization From the Department of Psychiatry, University of Chicago Pritzker School of Medicine, Chicago, Illinois. Received for publication August 2, 2002; accepted September 5, 2002. Supported by NIH K08-AA00276, P30 CA14599-28, and M01-RR00055. Reprint requests: Reprint requests: Andrea King, PhD, University of Chicago, Department of Psychiatry, 5841 S. Maryland Avenue, Chicago, IL 60637; Fax: 773-702-6454; E-mail: aking@yoda.bsd.uchicago.edu Copyright © 2002 by the Research Society on Alcoholism.