Extended Duration Varenicline for Smoking Among Cancer Patients: A Clinical Trial

BACKGROUND Prevalence of Smoking among Cancer Patients The rate of smoking among individuals with cancer who are age 40 or under are substantially higher (38-40%) than rates of smoking in the comparable age group in the general population (\~26%; Bellizzi et al., 2005; Coups \& Ostroff, 2005). Studies with patients that have traditional tobacco-related cancers show extremely high rates of smoking; upwards of 50% of head and neck (Duffy et al., 2008) and lung (Cooley et al., 2009) cancer patients report current smoking. However, high rates of smoking are not unique to such traditional tobacco-related disease sites. Significant rates of current smoking have been reported among testicular (19%; Shinn et al., 2007), prostate (16-17%; Gong et al., 2008; Pantarotto et al., 2007), cervical (21%; Beesley et al., 2008), breast (19%; Li et al., 2009), bladder (18%; Blanchard et al., 2008), esophageal (39%; Sundelof et al., 2008), colorectal (22%; Vincenzi et al., 2009), and lymphoma (19%; Geyer et al., 2010) cancer patients. Overall, about one-third to one-half of cancer patients who were smokers prior to their diagnosis continue to smoke following diagnosis (Gritz et al., 2006). Adverse Health Consequences of Smoking among Individuals with Cancer Continued smoking by cancer patients has been associated with diminished QOL, reduced survival probability and duration, and increased risk for disease recurrence and a second primary tumor (Gritz et al., 2006; 2007). Continued smoking by cancer patients is associated with greater treatment side effects or diminished QOL among head and neck (Duffy et al., 2007; Zevallos et al., 2009), lung (Daniel et al., 2009), prostate (Ku et al., 2009), and a heterogeneous group of (Schnoll et al., 2010a) cancer patients. A recent meta-analysis of studies with lung cancer patients found that continued smoking was associated with an increased risk of death, recurrence, and a second primary tumor (Parsons et al., 2010). Likewise, studies with head and neck cancer patients have reported that patients who continue to smoke following their diagnosis have a lower survival rate and an increased risk for a recurrence and a second primary tumor (Browman et al., 2002; Hilgert et al., 2009; Fortin et al., 2009; Leon et al., 2009). Continued smoking has also been associated with reduced survival among breast (Aksoy et al., 2007), lymphoma (Geyer et al., 2010), esophageal (Sundelof et al., 2008), prostate (Gong et al., 2008), cervical (Coker et al., 2009), and bladder (Aveyard et al., 2002) cancer patients and with an increased risk of recurrence or a second primary tumor among bladder (Fleshner et al., 1999), breast (Li et al., 2009), lymphoma (Moser et al., 2006), and colorectal (Jacobson et al., 1994) cancer patients. Continued smoking may worsen prognosis by reducing the effectiveness of chemotherapy (Duarte et al., 2008; van der Bol et al., 2007; Vincenzi et al., 2009; Hotta et al., 2008) and radiotherapy (Browman et al., 1993). Nicotine Dependence Treatments for Those with Cancer Very few smoking cessation trials have been conducted with this population (Gritz et al., 2006; 2007) and many of these past trials have used small samples and relied on self-report for smoking abstinence outcomes (de Moor et al., 2008). After nearly two decades of research in this area, not a single smoking cessation randomized clinical trial has yielded significant treatment effects (excluding Emmons et al., 2009, which studied adult survivors of childhood cancer; de Moor et al., 2008). Nurse-led (Griebel et al., 1998; Stanislaw \& Wewers, 1994; Wewers et al., 1994), physician-led (Browning et al., 2000; Gritz et al., 1993; Schnoll et al., 2003b), and behavioral (Schnoll et al., 2005; Wakefield et al., 2004) smoking cessation trials have failed to yield treatment effects for cancer patients. Our recent clinical trial with bupropion (Schnoll et al., 2010a) found no main effect for the medication, but bupropion increased abstinence rates, reduced withdrawal symptoms, and improved QOL more for patients with depression, vs. those without depression. Lastly, a very recent varenicline study reported end-of-treatment quit rates of 34%, vs. 14% for the comparison group (OR = 3.14), and a side effect profile that mirrored the general population (Park et al., 2011). While these data are encouraging, as were reported feasibility data, the study was under-powered (n = 49) and did not use a randomized design. Thus, there is currently no empirically-based treatment model for addressing nicotine dependence in the oncologic context. As such, a recent NCI meeting, with representatives from NCI cancer centers, concluded that the evaluation of novel smoking cessation interventions for cancer patients is a critical priority (Morgan et al., 2010). A Novel Treatment for Cancer Patients who Smoke: Extended Duration Varenicline Thus, nicotine dependence treatments for cancer patients may show greater efficacy if they adequately address the patient's relatively high level of nicotine dependence, risk for psychological distress and cognitive impairment, and delayed relapse process. We hypothesize that extended duration varenicline (24-weeks) will address these barriers to cessation and significantly increase quit rates, vs. standard varenicline treatment (12 weeks). Our rationale for selecting varenicline is as follows. First, the high rate of nicotine dependence among cancer patients underscores the need to include a pharmacotherapy as part of treatment. Varenicline is currently the most efficacious FDA-approved medication for nicotine dependence, yielding quit rates that significantly exceed those produced by bupropion (Gonzales et al., 2006; Jorenby et al., 2006) and nicotine patch (Aubin et al., 2008; Biazzo et al., 2010; Stapleton et al., 2008). Second, varenicline mitigates adverse psychological effects and cognitive impairment associated with quitting smoking (Patterson et al., 2009; Smith et al., 2009; Philip et al., 2009; Rollema et al., 2009; Sofuoglu et al., 2009). The anti-depressant-like (Rollema et al., 2009) and cognitive enhancing (Loughead et al., 2010) effects of varenicline is consistent with what we know about how varenicline works. As a nicotinic acetylcholine receptor (nAChRs) partial agonist, varenicline binds to nAChRs and blocks the entry of nicotine (from smoking) into the receptor and stimulates a moderate release of dopamine. This reduces the rewarding effects of smoking and reduces withdrawal symptoms (Rollema et al., 2009). Preclinical studies also indicate that α4β2 nAChRs subtypes are critical for cognition (Levin et al., 2006) and stimulation of these receptors by varenicline yields improved cognitive function (Loughead et al., 2010). Likewise, animal studies indicate that the simultaneous activation and desensitization of nAChRs receptors produced by nicotinic partial agonists like varenicline can yield antidepressant-like effects (Mineur \& Picciotto, 2010), which underlies current evaluations of varenicline as a treatment for major depression. Third, varenicline is efficacious and safe for treating nicotine dependence among various clinical populations, including: cardiovascular disease patients (Rigotti et al., 2010), COPD patients (Tashkin et al., 2010), smokers with comorbid alcohol (Hays et al., 2010) and cocaine (Poling et al., 2010) dependence, and smokers with comorbid affective or psychotic disorders (McClure et al., 2010; Smith et al., 2009; Philip et al., 2009). It has been shown to be safe when taken over 52 weeks (Williams et al., 2007). Although there have been reports of adverse psychiatric events following varenicline use, leading the FDA to mandate a boxed warning for varenicline, pooled data from controlled efficacy trials (Cahill et al., 2009; Tonstad et al., 2010), effectiveness trials (McClure et al., 2010), and large cohort studies (Gunnell et al., 2009; Kasliwal et al., 2009) demonstrate that varenicline is safe for treating nicotine dependence, even among smokers with psychiatric comorbidity, including depression (Stapleton et al., 2008; McClure et al., 2010; Steinberg et al., 2010). Our rationale for selecting an extended duration treatment is as follows. First, we have shown, in a placebo-controlled randomized trial with general population smokers, that 24-weeks of transdermal nicotine, vs. the standard 8-weeks, increases 6-month quit rates by an OR of 1.81 (32% vs. 20%; Schnoll et al., 2010b). Second, and importantly, extending treatment with the nicotine patch to 24-weeks significantly helps smokers with high levels of nicotine dependence and cognitive impairment, in particular, to overcome their liability to relapse (see below). While extended therapy with transdermal nicotine did not offset the effect of depression symptoms on relapse rates, the quit rate among depressed smokers in extended treatment was almost 2-times higher than it was for depressed smokers in standard treatment (see below). Third, we showed in our placebo-controlled randomized clinical trial with general population smokers (Schnoll et al., 2010b) that extending treatment with transdermal nicotine to 24-weeks (vs. 8-weeks) significantly reduced the probability that smokers would experience a lapse and, importantly, increased the likelihood that smokers would recover to abstinence following a lapse. Extended duration treatment offered smokers who lapsed the opportunity to re-start their quit attempt and eventually achieve abstinence. Thus, extended duration varenicline may be particularly efficacious at addressing the barriers to cessation that are evident among cancer patients.