A Bupropion Smoking Cessation Clinical Trial For Cancer Patients.pdf

Cancer Causes Control (2010) 21:811–820 DOI 10.1007/s10552-010-9507-8

ORIGINAL PAPER

A bupropion smoking cessation clinical trial for cancer patients Robert A. Schnoll • Elisa Martinez • Kristina L. Tatum • Dorothy M. Weber Natalie Kuzla • Marcella Glass • John A. Ridge • Corey Langer • Curtis Miyamoto • E. Paul Wileyto • Frank Leone

•

Received: 19 June 2009 / Accepted: 6 January 2010 / Published online: 20 January 2010 Ó Springer Science+Business Media B.V. 2010

Abstract Objective Many cancer patients continue to smoke post diagnosis, yet there have been few smoking cessation trials for this population. Depression, which is prevalent among cancer patients, may be a barrier to cessation. Methods This double-blind placebo-controlled trial randomized 246 cancer patients to 9 weeks of placebo or bupropion, stratifying by pre-treatment depression symptoms.

All work within this manuscript is original. Preliminary study results were presented in April 2005 at the Society for Research on Nicotine and Tobacco annual conference and in July 2008 at the Annual International Conference on Head and Neck Cancer. R. A. Schnoll
E. P. Wileyto Department of Psychiatry, University of Pennsylvania, 3535 Market Street, 4th Floor, Philadelphia, PA 19104, USA E. Martinez
K. L. Tatum
D. M. Weber
N. Kuzla
M. Glass
J. A. Ridge Fox Chase Cancer Center, 333 Cottman Avenue, Philadelphia, PA 19111, USA C. Langer Abramson Cancer Center, University of Pennsylvania, 3400 Spruce Street, 12 Penn Tower, Philadelphia, PA 19104, USA C. Miyamoto Department of Radiation Oncology, Temple University Hospital, 3401 N. Broad Street, Philadelphia, PA 19140, USA F. Leone Department of Pulmonary Medicine, Presbyterian Medical Center - 1 PHI, University of Pennsylvania, 51 N. 39th Street, Philadelphia, PA 19104, USA R. A. Schnoll (&) Department of Psychiatry, University of Pennsylvania, 3535 Market Street, 4th Floor, Philadelphia, PA 19104, USA e-mail: [email protected]

In addition, all patients received transdermal nicotine and behavioral counseling. Primary outcomes were 7-day pointprevalence abstinence, biochemically confirmed, at the end of treatment (Week 12), and at 6 months post quit day (Week 27). Additional outcomes included: withdrawal, affect, quality of life, compliance, and side effects. Results There was no main effect of bupropion vs. placebo on abstinence (Odds Ratio [OR] = 1.36, 95% CI: 0.38–4.81, p = .64). Patients with depression symptoms reported significantly lower abstinence rates vs. patients without depression symptoms (OR = .14, 95% CI: 0.02–0.80, p = .03). Bupropion increased abstinence rates, vs. placebo, more for participants with depression vs. those without depression symptoms. For patients with depression symptoms, bupropion reduced withdrawal symptoms and improved quality of life vs. placebo. Conclusions For patients with depression symptoms, bupropion increases abstinence rates, lowers withdrawal, and increases quality of life. However, abstinence rates among patients with depression symptoms were low vs. patients without depression symptoms, who exhibited similar abstinence rates when treated with bupropion or transdermal nicotine and counseling alone. These results can guide future smoking cessation intervention studies with cancer patients. Keywords Cancer patients
Smoking
Bupropion
Depression

Introduction Continued tobacco use has been documented in cancer patients. In one study, 40% of lung cancer patients continue to use tobacco post diagnosis [1], and in other studies with

123

812

cancer patients who quit, relapse is common [2, 3]. Continued smoking by cancer patients can decrease survival time [4], increase risk for a second primary cancer [5], reduce medical treatment effectiveness [6–11], and diminish quality of life (QOL) [12]. Consequently, there has been growing recognition of the need to integrate smoking cessation treatment into the oncologic context [13]. Unfortunately, there have been remarkably few smoking cessation clinical trials for cancer patients, and many of the past trials have used very small samples and/or lacked biochemical verification of cessation [14]. Previous trials have found that physician-based (vs. no treatment control) and tailored behavioral interventions (vs. general health education) do not significantly increase abstinence rates among cancer patients [15–17]. This may be due to the use of relatively weak interventions which do not address barriers to cessation among cancer patients. Depression symptoms impede cessation among cancer patients [18] and the general population [19–21] and are prevalent among cancer patients vs. the general population [22, 23]. Bupropion, an antidepressant originally marketed as Wellbutrin is an FDA-approved medication for nicotine dependence and is thought to promote cessation by, among other mechanisms, reducing negative affect (e.g., depression symptoms) [24]. This study, for the first time, examined the efficacy of bupropion for treating tobacco dependence among cancer patients. A primary objective was to examine the effects of bupropion among patients with significant depression symptoms. In contrast to past trials, which performed post hoc analysis of depression symptoms and treatment response [25, 26], randomization to placebo or bupropion in this trial was stratified by baseline depression symptoms a priori. This ensured an equal number of patients with significant depressive symptoms allocated to each arm and reduced the risk for confounding of treatment arm effects. Bupropion-depression symptoms interaction effects on positive and negative affect, withdrawal, and QOL were assessed. The main effect of depression symptoms on abstinence rates was examined as was adherence to, and side effects from, treatment.

Materials and methods

Cancer Causes Control (2010) 21:811–820

smoke at least 2 cigarettes/day on average, and have a cancer diagnosis. Patients were excluded if they reported no current smoking; had stage IV lung cancer or had brain metastases; reported current drug or alcohol dependence; reported current Axis I psychiatric conditions (major depression; eating disorders; generalized anxiety disorder; bipolar disorder; psychotic disorders) assessed by the Mini International Neuropsychiatric Interview [27]; were pregnant/lactating; reported a seizure disorder, cardiac, renal, pulmonary, endocrine, or neurological disorders; were currently using an MAO inhibiter or a pharmacotherapy for nicotine dependence; and reported the recent discontinuation of benzodiazepines. Over 7,500 patients were phonescreened, and 291 attended in-person screening; 246 patients were randomized to treatment arm. There were few differences between eligible enrollees and decliners [28]. See Fig. 1 for accrual and retention. Study design and procedures This was a double-blind, placebo-controlled trial. All patients received 5 counseling sessions and 8 weeks of nicotine replacement therapy (NRT; transdermal patch). In addition, patients were randomized to bupropion or placebo for 9 weeks. Thus, patients received either NRT, counseling, and placebo or NRT, counseling, and bupropion. The primary outcome was 7-day point-prevalence abstinence, confirmed with CO, at Week 12, the end of treatment, and at Week 27, 6 months post quit day. Patients were screened for eligibility initially by phone and subsequently in person. If eligible, patients completed informed consent and HIPAA forms and an initial assessment of smoking, affect, withdrawal, and QOL. The Center for Epidemiologic Studies Depression (CES-D) scale stratified patients when randomized. A score of C16 denoted the presence of significant depression symptoms [29]. At Week 1, patients received a counseling session to prepare for a Week 3 quit day and began placebo or bupropion. At Week 3, patients began NRT and received a second counseling session. Patients received 3 additional counseling sessions (Weeks 5, 7, 9) and continued medication for 9 weeks and NRT for 8 weeks. Mid-treatment measures were ascertained at Week 5. Abstinence rate was assessed at Weeks 12 (end of treatment) and 27 (6 months post target quit date). All procedures were standardized across sites.

Patients Measures After approval from site Institutional Review Boards, research assistants proactively screened cancer patients from physician schedules for eligibility from October 1, 2002 to April 1, 2008. To be eligible, patients had to be C18 years of age, speak English, possess a telephone,

123

Demographics The age, gender, education, marital status, family income, race, and ethnicity were recorded for patients.

Cancer Causes Control (2010) 21:811–820 Fig. 1 Participant accrual. Note: W12 = Week 12; W27 = Week 27; ITT = Intent-to-treat

813

7503 Phone Screens Enrollment Ineligible = 28 Refused = 17

291 History & Physical

Excluded Ineligible = 5826 Declined = 678 Missed = 40 Not Reached = 668

246 Stratified and Randomized (ITT) Allocation

27 Depression Symptoms/Placebo

105 Non-Depression Symptoms/Placebo

28 Depression Symptoms/Bupropion

86 Non-Depression Symptoms/Bupropion

Follow-up Lost to Follow-up 7 Withdrew 3 Unreachable

Lost to Follow-up 8 Withdrew 7 Unreachable

Lost to Follow-up 4 Withdrew 7 Unreachable

105 Analyzed at W12

28 Analyzed at W12

Lost to Follow-up 11 Withdrew 8 Unreachable

Analysis 27 Analyzed at W12

86 Analyzed at W12

Follow-up Lost to Follow-up 7 Withdrew 3 Unreachable

Lost to Follow-up 10 Withdrew 17 Unreachable

Lost to Follow-up 4 Withdrew 5 Unreachable

Lost to Follow-up 13 Withdrew 18 Unreachable

Analysis 27 Analyzed at W27

105 Analyzed at W27

28 Analyzed at W27

86 Analyzed at W27

Disease characteristics

Affect

Tumor type and stage was ascertained from medical records. Course of treatment was characterized as: awaiting treatment; undergoing treatment, recovering from treatment, or in remission/completed treatment.

The positive and negative affect scale assessed positive (e.g., enthusiasm) and negative (e.g., distressed) affect changes during treatment [32]. The 20 items yield 2 subscales that are reliable and valid [32]. The Cronbach’s alphas in this study were .89 for negative affect and .84 for positive affect.

Smoking characteristics Patients indicated the age at which they started smoking, years smoked, current smoking rate, and nicotine dependence level assessed by the 6-item Fagerstro¨m Test for Nicotine Dependence (FTND) [30]. Depression symptoms The CES-D [29], a 20-item measure of depression symptoms used previously with cancer patients [31], stratified patients into treatment arms. This scale is reliable and is a valid predictor of clinical depression [29]. A score of C16 indicates clinically relevant symptoms of depression [29]. The Cronbach’s alpha in this study was .89.

Withdrawal The Minnesota nicotine withdrawal scale evaluated tobacco withdrawal [33]. This is an 8-item scale widely used in tobacco control research [34]. The Cronbach’s alpha in this study was .75. Smoking behavior The primary outcome was 7-day point-prevalence abstinence at Week 12 and 27, confirmed with breath CO (abstinence = B 10 ppm). Daily smoking was also assessed using a validated time-line follow-back method [35].

123

814

QOL The short-form health survey (SF-12) assessed QOL [36]. This measure yields physical and mental subscales, has been used to assess cancer patient QOL [37], and has demonstrated validity and reliability [38]. The Cronbach’s alphas in this study were .77 for physical QOL and .67 for mental QOL. Side effects A 32-item side-effects checklist from bupropion and transdermal nicotine was completed (e.g., nausea, insomnia). The severity of each side effect was graded (0–3), and a total score was computed by summing Week 5 responses. The checklist was developed from past bupropion or transdermal nicotine trials [17, 40–42]. Adherence Self-reported daily number of pills and patches used and counseling sessions completed was assessed during treatment. Treatments Bupropion or placebo Patients received 9 weeks of placebo or sustained release bupropion. Patients began medication 2 weeks before a quit date as recommended [39]. The recommended bupropion dose of 300 mg/day (two 150 mg tablets) was used and treatment began with one 150 mg pill taken each day for 7 days and then two 150 mg pills/day (separated by 8 h) for days 8–63 [40, 41]. A 9-week treatment period was selected as in previous bupropion trials [42].

Cancer Causes Control (2010) 21:811–820

avoiding relapse based on social cognitive theory. The session content conformed to treatment guidelines [39] and our past smoking cessation trial with cancer patients [17]. Data analysis Chi-square and ANOVA compared treatment arms, stratified by depression symptoms, on demographic, medical, and smoking data, and completion rate of Week 12 and 27 surveys to identify covariates. We used longitudinal logistic regression (with random intercepts) to evaluate the main effect of treatment and depression symptoms on Week 12 and 27 abstinence rates, controlling for covariates (gender, education, time); the interaction between depression symptoms and treatment arm was described and not included in the regression model since the small number of depressed participants led to unstable parameters in the model. We tested for interactions of predictors with time using the Wald v2. Analysis was based on intent-to-treat, with those with missing data at Week 12 and 27 data presumed to be smokers. ANOVA evaluated main and interaction effects on smoking rate among non-abstainers, controlling for covariates. The main effect of treatments and interaction effects between treatment and depression symptoms on changes in withdrawal and positive and negative affect were assessed using ANOVA and data from baseline to Week 5. The main effects of treatment and interaction effects between treatment and depression symptoms on changes in mental and physical QOL from baseline to Week 12 were assessed using ANOVA. Using ANOVA, we examined differences in side effects at Week 5 across treatment arms, and adherence to bupropion vs. placebo, NRT, and counseling across the treatment arms.

Results

NRT

Sample characteristics

All patients received 8 weeks of transdermal nicotine (Nicoderm CQ) based on guidelines [39]. Patients smoking \10 cigarettes/day were given 4 weeks of 14 mg patches, followed by 4 weeks of 7 mg patches; patients smoking C10 cigarettes/day were given 2 weeks of 21 mg patches followed by 2 weeks of 14 mg patches and 4 weeks of 7 mg patches.

Table 1 shows the sample characteristics. Fifty-five patients (22% of the sample) had CES-D scores of C16. Almost half of the sample were women and had a college degree or higher. Close to one-third of the sample were ethnic/racial minorities, including 29% African American. The average patient age was 54.8 years. The average FTND score was 3.2 (SD = 1.2; range = 1–5), and the average cigarettes/ day was 17.5 (SD = 9.3). Close to one-third of the sample had stage 3 or 4 cancer. Close to one-third of the sample had tobacco-related tumors (head and neck or lung), 21% had breast cancer, 15% had prostate cancer, 9% had lymphoma, 5% had colorectal cancer; 4% had kidney, pancreas, or liver tumors; 3% had genitourinary cancer; 3% had esophageal cancer; and 5% had other tumor types or multiple primaries;

Counseling All patients received 5 smoking cessation counseling sessions (3 in-person, 2 phone) from a trained smoking cessation counselor. Sessions were manual-based and focused on behavioral strategies for preparing to quit smoking and

123

Cancer Causes Control (2010) 21:811–820

815

Table 1 Participant characteristics by treatment arm and depression symptoms (ITT, N = 246) Characteristics

Depression symptom placebo (N = 27)

Non-depression symptom placebo (N = 105)

Depression symptom bupropion (N = 28)

Non-depression symptom bupropion (N = 86)

Overall p

Sex (% female)

70.4

47.6

57.1

38.4

48

.02

Age (mean)

50.7

54.1

52.6

54.0

54.8

.98

Education (% Bcollege) Race (% European ancestry)

66.7 59.3

50 70.5

82.1 57.1

52.3 65.1

56.3 65.9

.01 .48

FTND (mean) Cigarettes per day (mean)

3.3

3.2

3.2

3.2

3.2

.98

19.1

16.9

16.0

18.3

17.5

.46

Marital status (% married)

37

58

57

57

55

.25

Disease stage (% 3 or 4)

22.2

29.5

37

26.8

28.9

.70

Disease type (% tobacco)

40.7

30.5

39.5

29.3

32.2

.58

Disease course (% completed treatment)

66.7

61.4

46.4

60

59.8

.44

Note: FTND Fagerstro¨m Test for Nicotine Dependence, ITT Intent-to-treat; There were also no significant differences between treatment arms when collapsing across depression symptom and non-depression symptom groups (all ps [ .05)

7% of the sample were awaiting treatment, 23% were undergoing treatment; 10% were recovering from treatment, and 60% had completed treatment. While there were no significant differences between treatment arms, women and those with less than a college education were more likely to be in the depression symptoms group; these variables were included as covariates in cessation models. There was no significant difference across treatment arms in the rate of completion of Week 12 (bupropion = 74%; placebo = 81%; p = .17) or Week 27 (bupropion = 65%; placebo = 72%; p = .23) assessments. Treatment arm effects on smoking Treatment arm did not influence abstinence rates at Week 12 or Week 27 (OR = 1.36, 95% CI: 0.38–4.81). At Week 12, the abstinence rates were 24.2% for placebo and 27.2% for bupropion; at Week 27, abstinence rates were 17.4% for placebo and 18.4% for bupropion. Figure 2 shows the effects of treatment on Week 12 and Week 27 abstinence rates depending on depression symptoms. For non-depression symptom patients, quit rates at week 12 were 28.6% for placebo and 31.4% for bupropion and quit rates at week 27 were 20% for placebo and 19.8% for bupropion. For patients with depression symptoms, quit rates at week 12 were 7.4% for placebo and 14.3% for bupropion and at week 27 were 7.4% for placebo and 14.3% for bupropion. While the number of participants within certain cells is small and the results are not statistically significant, patients without depression symptoms quit equally well on bupropion vs. placebo but, among patients with depression symptoms, those provided with bupropion had higher abstinence rates vs. those provided with placebo.

Fig. 2 Count of abstinence, ITT, CO-Confirmed by Treatment Arm and Depression Group. Note: The bars are exact binomial 95% CIs. The numerator is the number of participants quit and the denominator is the number of participants in the respective category

There was no significant difference between placebo (M = 5.2; SE = .95) and bupropion patients (M = 5.3; SE = 1.0) in average daily smoking to Week 12 (p = .97), controlling for gender and education. There was no significant difference between placebo (M = 5.4; SE = .88) and bupropion patients (M = 6.1; SE = .90) in average daily smoking to Week 27 (p = .64), controlling for gender and education. The treatment 9 depression symptoms interaction on average daily smoking to Week 12, controlling for gender and education, approached significance (F[1,142] = 2.99, p = .09). In the non-depression symptoms group, patients on bupropion reported smoking more cigarettes/day vs. placebo patients (Fig. 3), although this was not statistically different. In the depression symptom group, patients on bupropion reported smoking fewer cigarettes/day vs. those on placebo (Fig. 3), although this was not statistically different. The interaction effect for treatment arm and depression symptoms on average daily

123

816

Cancer Causes Control (2010) 21:811–820 9 8

13

Placebo 7.4

Bupropion

7

6.2

6

5.5

11

5.7

5.1

10.3

10.1

10

5

4.5

9 4

Mean

Mean

12.3

12 6.5

Depression Symptoms-Placebo Depression Symptoms-Bupropion Non-Depression Symptoms-Placebo Non-Depression Symptoms-Bupropion

3

3

8 7

7.3 6.7

2

6 1

5.3

5

6 5.5

0 Depression Symptoms

Non-depression Symptoms

Week 12

Depression Symptoms

4

Non-depression Symptoms

3 Baseline

Week 27

Fig. 3 Average daily cigarette smoking, baseline to Week 12 and Week 27 by Treatment Arm and Depression Group

Fig. 4 Change in withdrawal symptoms from baseline to Week 5

25

smoking to Week 27, controlling for gender and education, was not significant (p = .34).

123

24.1

24 23

Depression Symptoms-Placebo Depression Symptoms-Bupropion Non-Depression Symptoms-Placebo Non-Depression Symptoms-Bupropion

23.7

22

Treatment arm effects on withdrawal, affect, and QOL

21

Mean

There were no main effects of bupropion on changes in withdrawal (p = .39), negative (p = .41) and positive (p = .77) affect, and physical (p = .31) and mental (p = .42) QOL. A significant treatment 9 depression symptom effect was found for change in withdrawal (F[1,184] = 7.78, p = .006; Fig. 4). There was a significant reduction in withdrawal for depression symptom patients on bupropion vs. depression symptom patients on placebo (F[1,33] = 11.05, p = .002). There was no significant difference in change in withdrawal for non-depression symptom patients on bupropion vs. placebo (F[1,151] = 2.06, p = .15). For negative affect, the analyses indicated no significant treatment 9 depression symptoms interaction (p = .15). Although not significant, there was a decrease in negative affect over time among depression symptom patients on bupropion vs. placebo (F[1,33] = 2.7, p = .11; Fig. 5). There was little difference in the change in negative affect over time between the nondepression symptom patients on bupropion vs. placebo (p = .55). There was no significant interaction effect on change in positive affect (p = .84). There was a significant interaction of treatment 9 depression symptoms for change in physical (F[1,169) = 7.08, p = .009) and mental (F[1,177) = 7.93, p = .005) QOL. For physical QOL, depression symptom patients on bupropion had a significant increase in QOL vs. depression symptom subjects on placebo (F[1,29) = 16.8, p = .001; Fig. 6); non-depression symptom patients on bupropion also showed a significant increase in physical QOL vs. nondepression symptom patients on placebo, but the effect was considerably smaller (F[1,140) = 7.89, p = .006; Fig. 6). For mental QOL, depression symptom patients on

Week 5

20.9

20 19 18

16.8

17 16.5

16

16

15

14.6

14.9

14 13 Baseline

Week 5

Fig. 5 Change in negative affect from baseline to Week 5

bupropion reported a significant increase in QOL vs. depression symptom patients on placebo (F[1,28) = 7.5, p = .01; Fig. 5); there was no significant difference in change in mental QOL for non-depression symptom patients on bupropion vs. placebo (p = .31; Fig. 6). Treatment arm effects on side effects and adherence There was no significant difference between treatment arms in total (p = .80) or individual (e.g., headache, nausea, seizures) side effects. There was no significant difference in the number of pills taken between placebo (M = 91.3; SD = 36.0) and bupropion (M = 84.7; SD = 41.8; p = .21); 64% of placebo patients took C75% of their pills vs. 60% for bupropion. There was no significant difference in the number of counseling sessions completed by placebo (M = 4.2; SD = 1.2) vs. bupropion (M = 3.9; SD = 1.4) patients (p = .14); 78% of placebo patients completed C4 of 5 sessions vs. 69% for bupropion. A greater proportion of placebo patients used NRT (94%) vs. bupropion (82%; p = .004). However, when the mean

Cancer Causes Control (2010) 21:811–820

22

20.8

21

Mean

20.4

Discussion

19.8

19.7

19 18

abstinent, vs. 19.9% of the non-depression symptom patients (p = .11).

Depression Symptoms-Placebo Depression Symptoms-Bupropion Non-Depression Symptoms-Placebo Non-Depression Symptoms-Bupropion

23

20

817

19.4 18.2

17 16

16.6

15

15.5

14 13 Baseline

Week 12 Depression Symptoms-Placebo Depression Symptoms-Bupropion Non-Depression Symptoms-Placebo Non-Depression Symptoms-Bupropion

20 19

18.2 18.2

18.6

18

18.2

Mean

17 16.9

16

15.3

15 14

14.5 14.5

13 12 Baseline

Week 12

Fig. 6 Change in physical and mental quality of life from baseline to Week 12

number of days that the patch was used was compared across treatment arms, there was no significant difference between placebo (M = 44.6; SE = 13.7) and bupropion (M = 44.6; SD = 14.3; p = .99). Depression symptoms effects on abstinence rates The regression model (Table 2) also indicated a significant main effect for depression symptoms on abstinence rates (OR = 0.14, 95% CI: 0.02–0.80, p = .03). At Week 12, 10.9% of depression symptom patients were abstinent vs. 29.8% of the non-depression symptom patients (p = .002). At Week 27, 10.9% of depression symptom patients were

Table 2 Random intercept linear mixed effects regression models of Week 12 and Week 27 abstinence Variable

OR

95% CI

p

Sex

3.16

0.87–11.5

.08

Education

2.26

0.63–8.07

.21

Time

0.32

0.16–0.68

.003

Treatment arm

1.36

0.38–4.81

.64

Depression symptom group

0.14

0.02–0.80

.03

Note: OR odds ratio, CI confidence interval; the effect of time indicated that abstinence rates diminished over time

This is the first bupropion smoking cessation trial with cancer patients and the first study to prospectively examine the efficacy of bupropion for nicotine dependence based on depression symptoms. While there was no main effect of bupropion vs. placebo on abstinence rates, there was some indication that bupropion can be more effective for smokers with depression symptoms vs. those without depression symptoms. Among depression symptom smokers, those on bupropion reported higher quit rates, fewer cigarettes per day, lower nicotine withdrawal, lower negative affect, and higher QOL vs. depression symptom smokers on placebo. In contrast, non-depression symptom smokers were effectively treated with counseling and NRT. It should be noted, however, that the abstinence rates among depression symptom smokers, vs. non-depression symptom smokers, were very low, indicating that even with bupropion, depression symptoms undermine the ability to quit smoking. This strong main effect for depression symptoms further supports previous findings showing that current depression symptoms impairs the ability to quit smoking [18–21]. Previous trials of bupropion for smoking cessation have not found bupropion to be more effective for smokers with depression symptoms vs. smokers without depression symptoms [25, 43]. Compared to past trials, the present trial stratified randomization to treatment arm a priori based on current depression symptoms, whereas previous studies examined the relationship between depression symptoms and bupropion response post hoc. The implementation of this trial within current cancer care, which can increase depression symptoms, provided the opportunity to stratify randomization by depression symptoms, and this may have led to somewhat divergent results. Further, in contrast to previous bupropion trials [44], the present study found no main effect for bupropion vs. placebo. However, in contrast to the present trial, most previous trials excluded patients with current depression symptoms and were conducted with the general population of smokers, not clinical populations (e.g., cancer patients) [44]. A bupropion trial with smokers with cardiovascular disease found no effect vs. placebo [45], and a review of cessation trials for hospitalized smokers found no support for any pharmacotherapy vs. behavioral treatment [46]. Factors related to smoking cessation treatment response (e.g., depression symptoms) may mask pharmacotherapy effects in clinical populations. Thus, a priori stratification of depression symptoms to examine bupropion treatment response may

123

818

be important when treating smokers in the context of medical illness. Finally, as seen with cancer patients [18] and smokers in the general population [19–21], depression symptoms adversely affected the likelihood of abstinence, underscoring the need for smoking cessation interventions to address this barrier to smoking cessation. There were no main effects of bupropion vs. placebo on withdrawal and affect, which is consistent with previous studies [24, 47–49]. The present findings, however, suggest that bupropion’s effects on withdrawal may be larger for smokers with depression symptoms. Likewise, the results for negative affect lend further support to previous studies [24, 47]. However, contrary to past studies [25, 43], the present findings suggest that bupropion’s effects on negative affect may be more salient for smokers with depression symptoms, although these analyses were not statistically significant. Previous studies that found no differential effect of bupropion on negative affect between smokers with and without depression symptoms did not stratify randomization by depression symptoms and assessed past vs. current depression. The present study suggests that bupropion is safe for cancer patients who smoke (in terms of bupropion-related side effects) and may increase QOL for patients with depression symptoms. Bupropion may help smokers with depression symptoms achieve cessation and this improves QOL; or bupropion may directly improve patient QOL, and this effect is more noticeable among patients with depression symptoms because of ceiling effects among patients with depression symptoms. Regardless, this result supports the use of bupropion with cancer patients, especially with cancer patients who exhibit depression symptoms. The current study has limitations. First, the sample was small vs. general population bupropion trials, and only 22% of the sample had significant depression symptoms, which limited statistical power and options for the analysis of interaction effects. The rates of accrual and completion of outcome assessments were lower than desired. The challenges of conducting smoking cessation trials with cancer patients include a high rate of contraindicated medication use, contraindicated medical conditions, spontaneous cessation, and disinclination to consider cessation [28, 50]. Relative to smoking cessation trials with cancer patients, this study is one of the largest trials and the first bupropion trial. In addition, the accrual rate should be conceptualized in the context of our proactive recruiting approach, and our outcome assessment completion rate compares to large smoking cessation clinical trials [51]. Second, the effect of bupropion vs. placebo for patients with depression symptoms was relatively small given the number of patients with depression symptoms in this trial. Nevertheless, the effects were consistent across smoking rate, and changes in withdrawal, affect, and QOL. Third,

123

Cancer Causes Control (2010) 21:811–820

we used a self-report measure of depression symptoms vs. a diagnostic interview. Use of a clinical interview prerandomization was not feasible and may have reduced the proportion of patients with depression. Fourth, the present sample had an average FTND score that was lower than is seen in the general population. Two FTND items which may have lower values among cancer patients include number of cigarettes smoked per day and smoking despite illness. Fifth, although we controlled for the potential effects of education and gender in the analyses, these variables may still influence the study findings. Sixth, we relied on self-report measures of adherence, which may have over-estimated the proportion of patients who adhered to prescribed treatments. Lastly, we excluded those with major depression. While it may have been worthwhile to allow those with major depression to enroll, we did not want patients to discontinue antidepressants to enroll in this study. Thus, results are generalizable to smokers with possible sub-threshold depression only. Despite these limitations, this study suggests that cancer patients with depression symptoms who smoke may benefit from bupropion, while cancer patients without depression symptoms who smoke may be effectively treated with counseling and NRT. This effect, however, should be conceptualized in the context of a strong main effect of depression symptoms on quit rates. A priority for future smoking cessation research with this group of smokers is to evaluate behavioral counseling methods tailored to address depression symptoms, and several treatment approaches have recently been recommended [52]. The abstinence rates among depression symptom patients were low, vs. non-depression symptom patients, and bupropion did not fully offset this relapse liability. As such, assessment of current depression symptoms among cancer patients who smoke should be integrated into smoking cessation programs to enhance treatments for these patients, including the potential use of extended treatment with FDA-approved medications for nicotine dependence since abstinence rates diminished after the treatment phase ended. Acknowledgments The authors would like to thank Ms. Jeanne Pomenti for assisting with the preparation of this manuscript. This study was funded by grant R01 CA95678 from the National Cancer Institute. Transdermal nicotine patches were provided free-of-charge by GlaxoSmithKline. All sites provided Institutional Review Board approval. All patients provided informed consent for this study.

References 1. Videtic GM, Stitt LW, Dar AR et al (2003) Continued cigarette smoking by patients receiving concurrent chemoradiotherapy for limited-stage small-cell lung cancer is associated with decreased survival. J Clin Oncl 21:1544–1549

Cancer Causes Control (2010) 21:811–820 2. Schnoll RA, James C, Malstrom M et al (2003) Longitudinal predictors of continued tobacco use among patients diagnosed with cancer. Ann Behav Med 25:214–222 3. Walker MS, Vidrine DJ, Gritz ER et al (2006) Smoking relapse during the first year after treatment for early-stage non-small-cell lung cancer. Cancer Epidemiol Biomarkers Prev 15:2370–2377 4. Zhou W, Heist RS, Liu G et al (2006) Smoking cessation before diagnosis and survival in early stage non-small cell lung cancer patients. Lung Cancer 53:375–380 5. Garces YI, Schroeder DR, Nirelli LM et al (2007) Second primary tumors following tobacco dependence treatments among head and neck cancer patients. Am J Clin Oncol 30:531–539 6. Browman GP, Mohide EA, Willan A et al (2002) Association between smoking during radiotherapy and prognosis in head and neck cancer: a follow-up study. Head Neck 24:1031–1037 7. Carter CL, Key J, Marsh L et al (2001) Contemporary perspectives in tobacco cessation: what oncologists need to know. Oncologist 6:496–505 8. Tsao As, Liu D, Lee JJ, Sptiz M, Hong WK (2006) Smoking affects treatment outcome in patients with advanced nonsmall cell lung cancer. Cancer 106:2428–2436 9. Rades D, Setter C, Schild SE, Dunst J (2008) Effect of smoking during radiotherapy, respiratory insufficiency, and hemoglobin levels on outcome in patients irradiated for non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 71:1134–1142 10. Vander Ark W, DiNardo LJ, Oliver DS (1997) Factors affecting smoking cessation in patients with head and neck cancer. Laryngoscope 107:888–892 11. Duarte RL, Luiz RR, Paschoal ME (2008) The cigarette burden (measured by the number of pack years smoked) negatively impacts the response rate to platinum-based chemotherapy in lung cancer patients. Lung Cancer 61:244–254 12. Baser S, Shannon VR, Eapen GA et al (2006) Smoking cessation after diagnosis of lung cancer is associated with a beneficial effect on performance status. Chest 130:1784–1790 13. Gritz ER, Fingeret MC, Vidrine DJ et al (2006) Successes and failures of the teachable moment: smoking cessation in cancer patients. Cancer 106:17–27 14. Gritz ER, Vidrine DJ, Fingeret MC (2007) Smoking cessation a critical component of medical management in chronic disease populations. Am J Prev Med 33:S414–S422 15. Gritz ER, Carr CR, Rapkin D et al (1993) Predictors of long-term smoking cessation in head and neck cancer patients. Cancer Epidemiol Biomarkers Prev 2:261–270 16. Schnoll RA, Zhang B, Rue M et al (2003) Brief physician-initiated quit-smoking strategies for clinical oncology settings: a trial coordinated by the Eastern Cooperative Oncology Group. J Clin Oncol 21:355–365 17. Schnoll RA, Rothman RL, Wielt DB et al (2005) A randomized pilot study of cognitive-behavioral therapy versus basic health education for smoking cessation among cancer patients. Ann Behav Med 30:1–11 18. Schnoll RA, Malstrom M, James C et al (2002) Correlates of tobacco use among smokers and recent quitters diagnosed with cancer. Patient Educ Couns 46:137–145 19. Covey LS, Glassman AH, Stetner F (1990) Depression and depressive symptoms in smoking cessation. Compr Psychiatry 31:350–354 20. Hayford KE, Patten CA, Rummans TA et al (1999) Efficacy of bupropion for smoking cessation in smokers with a former history of major depression or alcoholism. Br J Psychiatry 174:173–178 21. Hughes JR (1992) Tobacco withdrawal in self-quitters. J Consult Clin Psychol 60:689–697 22. Pirl WF (2004) Evidence report on the occurrence, assessment, and treatment of depression in cancer patients. J Natl Cancer Inst Monogr 32:32–39

819 23. van Wilgen CP, Dijkstra PU, Stewart RE et al (2006) Measuring somatic symptoms with the CES-D to assess depression in cancer patients after treatment: comparison among patients with oral/ oropharyngeal, gynecological, colorectal, and breast cancer. Psychosomatics 47:465–470 24. Lerman C, Roth D, Kaufmann V et al (2002) Mediating mechanisms for the impact of bupropion in smoking cessation treatment. Drug Alcohol Depend 67:219–223 25. Catley D, Harris KJ, Okuyemi KS et al (2005) The influence of depressive symptoms on smoking cessation among African Americans in a randomized trial of bupropion. Nicotine Tob Res 7:859–870 26. Hitsman B, Pingitore R, Spring B et al (1999) Antidepressant pharmacotherapy helps some cigarette smokers more than others. J Consult Clin Psychol 67:547–554 27. Sheehan DV, Lecrubier Y, Sheehan KH et al (1998) The MiniInternational Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 59(Suppl 20):22–33 28. Martinez E, Tatum KL, Weber DM, et al (2009) Issues related to implementing a smoking cessation clinical trial for cancer patients. Cancer Causes Control 20:97–104 29. Radloff L (1977) The CES-D scale: a new self-report depression scale for research in the general population. Appl Psychol Meas 1:385–401 30. Heatherton TF, Kozlowski LT, Frecker RC et al (1991) The Fagerstrom test for nicotine dependence: a revision of the Fagerstrom Tolerance Questionnaire. Br J Addict 86:1119–1127 31. de Graeff A, de Leeuw JR, Ros WJ et al (2000) Pretreatment factors predicting quality of life after treatment for head and neck cancer. Head Neck 22:398–407 32. Watson D, Clark LA, Tellegen A (1988) Development and validation of brief measures of positive and negative affect: the PANAS scales. J Pers Soc Psychol 54:1063–1070 33. Hughes JR, Hatsukami D (1986) Signs and symptoms of tobacco withdrawal. Arch Gen Psychiatry 43:289–294 34. Heil SH, Higgins ST, Mongeon JA et al (2006) Characterizing nicotine withdrawal in pregnant cigarette smokers. Exp Clin Psychopharmacol 14:165–170 35. Brown R, Burgess E, Sales S et al (1998) Reliability and validity of a smoking timeline follow-back interview. Psychol Addict Behav 12:101–112 36. Ware J Jr, Kosinski M, Keller SD (1996) A 12-item short-form health survey: construction of scales and preliminary tests of reliability and validity. Med Care 34:220–233 37. McCorkle R, Dowd M, Ercolano E, et al (2009) Effects of a nursing intervention on quality of life outcomes in postsurgical women with gynecological cancers. Psychooncology 18:62–70 38. Gandek B, Ware JE, Aaronson NK et al (1998) Cross-validation of item selection and scoring for the SF-12 Health Survey in nine countries: results from the IQOLA Project. International Quality of Life Assessment. J Clin Epidemiol 51:1171–1178 39. Fiore MC, Jae´n CR, Baker TB, et al (2008) Treating tobacco use and dependence: 2008 update. Clinical Practice Guideline. Rockville, MD: U.S. Department of Health and Human Services, Public Health Service, May 2008 40. Glaxo_Wellcome (1999) Bupropion hydrochloride sustainedrelease tablets prescribing information. Research Triangle Park, North Carolina, USA 41. Holm KJ, Spencer CM (2000) Bupropion: a review of its use in the management of smoking cessation. Drugs 59:1007–1024 42. Jorenby DE, Leischow SJ, Nides MA et al (1999) A controlled trial of sustained-release bupropion, a nicotine patch, or both for smoking cessation. N Engl J Med 340:685–691

123

820 43. Cox LS, Patten CA, Niaura RS et al (2004) Efficacy of bupropion for relapse prevention in smokers with and without a past history of major depression. J Gen Intern Med 19:828–834 44. Hughes JR, Stead LF, Lancaster T (2007) Antidepressants for smoking cessation. Cochrane Database Syst Rev: CD000031 45. Rigotti NA, Munafo MR, Stead LF (2007) Interventions for smoking cessation in hospitalised patients. Cochrane Database Syst Rev: CD001837 46. Rigotti NA, Munafo MR, Stead LF (2008) Smoking cessation interventions for hospitalized smokers: a systematic review. Arch Intern Med 168:1950–1960 47. Shiffman S, Johnston JA, Khayrallah M et al (2000) The effect of bupropion on nicotine craving and withdrawal. Psychopharmacology (Berl) 148:33–40 48. Piper ME, Federmen EB, McCarthy DE et al (2008) Using mediational models to explore the nature of tobacco motivation and tobacco treatment effects. J Abnorm Psychol 117:94–105

123

Cancer Causes Control (2010) 21:811–820 49. Teneggi V, Tiffany ST, Squassante L et al (2005) Effect of sustained-release (SR) bupropion on craving and withdrawal in smokers deprived of cigarettes for 72 h. Psychopharmacology 183:1–12 50. Schnoll RA, Rothman RL, Lerman C et al (2004) Comparing cancer patients who enroll in a smoking cessation program at a Comprehensive Cancer Center with those who decline. Head Neck 26:278–286 51. Gonzales D, Rennard SI, Nides M et al (2006) Varenicline, an alpha4beta2 nicotinic acetylcholine receptor partial agonist, vs. sustained-release bupropion and placebo for smoking cessation: a randomized controlled trial. JAMA 296:47–55 52. Dy SM, Lorenz KA, Naeim A et al (2008) Evidence-based recommendations for cancer fatigue, anorexia, depression, and dyspnea. J Clin Oncol 26:2886–2895