Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Dec 1.
Published in final edited form as: Alcohol Clin Exp Res. 2021 Nov 23;45(12):2569–2578. doi: 10.1111/acer.14729

Is extended release naltrexone superior to buprenorphine-naloxone to reduce drinking among outpatients receiving treatment for opioid use disorder? A secondary analysis of the CTN X:BOT trial

John D Roache 1, Martina Pavlicova 2, Aimee Campbell 3, Tse-Hwei Choo 4, Michelle Peavy 5, Andrea S Kermack 6, Edward V Nunes Jr 7, John Rotrosen 8
PMCID: PMC8722377  NIHMSID: NIHMS1751029  PMID: 34698397

Abstract

Background:

The comparative effectiveness of extended-release naltrexone versus buprenorphine-naloxone for opioid relapse prevention (X:BOT) trial showed that following induction, treatment with the sublingual agonist (buprenorphine-naloxone, BUP-NX) or injected antagonist (extended release naltrexone, XR-NTX) produced similar reductions in opioid relapse in injection users with opioid use disorder (OUD). Because XR-NTX reduces drinking in alcohol use disorder (AUD), we conducted a secondary analysis of the X:BOT sample of patients successfully inducted onto treatment to determine whether XR-NTX (n = 204) was superior to BUP-NX (n = 270) in reducing drinking or heavy drinking in patients with OUD.

Methods:

Standard drink units consumed were measured using the Timeline Follow-back method. Mixed-models regression was used to examine the monthly frequency of any drinking and heavy drinking over 6 months of treatment. We used a proportional hazard survival analysis to examine the time to first drink.

Results:

Both treatment groups reduced drinking from baseline to posttreatment (small to medium effect), but no differences between groups were detected. However, only 29% (n = 136) of the sample had AUD and 19% (n = 26/136) of those were abstinent before treatment. Analysis of a subsample enriched for possible drinking included 136 individuals with an AUD diagnosis plus 43 who did not have AUD, but reported at least one day of heavy drinking prior to the study. However, this subsample reported only 32% of days of any drinking with a median of only 13% of days designated as “heavy.” Within this subsample, at baseline, the BUP-NX group reported more mean drinks per drinking day than the XR-NTX group (p = 0.03); however, there were no other significant group differences on drinking observed before, during, or at the end of treatment.

Conclusions:

There was an overall reduction in drinking during treatment of OUD using both agonist and antagonist medications, so that the hypothesis that XR-NTX would be superior to BUP-NX was not supported. The study is limited by low levels of comorbid AUD or heavy drinking observed in X:BOT trial participants seeking treatment for OUD.

Keywords: alcohol, alcohol use disorder, heavy drinking, treatment, naltrexone, buprenorphine, opioid, opioid use disorder

INTRODUCTION

Naltrexone was first synthesized in 1963 but was not developed as a therapeutic agent until the Food & Drug Administration (FDA) approved it as Trexan© for the treatment of opioid dependence in 1984. Despite its efficacy as an opioid-receptor antagonist, oral formulations of naltrexone were not effective for treatment of heroin dependence due to poor adherence to daily dosing of the oral formulation (Nielsen et al., 2016; Nunes et al., 2006). Based on pre-clinical data showing opioid-receptor mechanisms in alcohol-related reward pathways (Volpicelli, 1987), two independent clinical trials demonstrated the efficacy of oral naltrexone to reduce, not the frequency of “any” drinking, but rather the frequency of drinking days that were considered to be “heavy” drinking days (O’Malley et al., 1992; Volpicelli et al., 1992). Subsequently, the FDA-approved oral naltrexone for the treatment of alcohol dependence in 1995 under the brand name ReVia©. Although oral naltrexone is considered to be one of the most effective treatments for alcohol use disorders (AUDs) (Anton et al., 2006; Soyka et al., 2017), compliance with daily oral dosing also is a problem limiting its effectiveness in alcoholism treatment (Volpicelli et al., 1997). This led to the development of an intramuscular depot formulation of naltrexone where once-monthly injections showed dose-related superiority over placebo to reduce “heavy” drinking among alcohol-dependent outpatients (Garbutt et al., 2005). Thus, the FDA approved Vivitrol© for the treatment of alcohol dependence in 2006. Development of an injectable long-acting form of naltrexone also rejuvenated interest in depo injections of opioid antagonists for treatment of opioid dependence (Comer et al., 2006; Krupitsky et al., 2011), resulting in FDA approval of Vivitrol for the treatment of opioid dependence in 2010.

Independently, the μ-opioid partial agonist, buprenorphine (BUP), was developed to treat heroin dependence and has become one of the standard-of-are substitution therapies for treating opioid dependence (Kreek & Vocci, 2002; Lee et al., 2016; Mattick et al., 2014). Recently, Nunes and colleagues (Nunes et al., 2016) sought to directly compare the effectiveness of agonist and antagonist approaches to treat opioid dependence in a pragmatic open-l abel comparison of extended release naltrexone injection (XR-NTX) versus sublingual buprenorphine-naloxone (BUP-NX). Completed as part of the NIDA National Drug Abuse Treatment Clinical Trials Network (CTN), the “naltrexone versus buprenorphine for opioid treatment” (X:BOT) trial and its methods and design have been previously described (Lee et al., 2016) and the main study results reported (Lee et al., 2018). Although similar outcomes were achieved for both the antagonist and agonist treatment arms, among those successfully inducted onto treatment, drop-out rates during induction onto the antagonist (XR-NTX) were higher than with induction onto the partial agonist (BUP-NX).

While the question of whether to stimulate μ-opioid receptors with an agonist (i.e., with buprenorphine) or block the receptor with an antagonist (i.e., with naltrexone) has direct pharmacological relevance for the treatment of opioid use disorder (OUD) (Lee et al., 2016; Nunes et al., 2016), the same question becomes important when considering how best to address alcohol use comorbidity among opioid users. Unfortunately, AUD comorbidity among individuals with OUD is not well studied (Soyka, 2015). AUD has been reported to occur in as many as one-third of patients in opioid maintenance treatment where it is associated with clinically worse treatment outcomes (see Soyka, 2015 and Nolan et al., 2016). A longitudinal trajectory analysis over 5 years of methadone or buprenorphine substitution therapy reported that among the 50% of the sample with significant levels of alcohol consumption, about a third increased, a third decreased, and a third remained unchanged in the frequency of alcohol intake (Eastwood et al., 2019). Although improved drinking outcomes would be a desirable goal for individuals in the treatment for OUD, controlled clinical trials evaluating treatment outcomes for dually comorbid opioid and AUDs are generally lacking. An early study (Liebson et al., 1973, 1978) demonstrated reduced alcohol use when methadone dose was administered only contingent upon disulfiram administration compared to the control condition where methadone dosing did not depend upon taking disulfiram. One pilot trial evaluated the feasibility and safety of giving naltrexone to HIV patients with opioid or AUDs where only 5 of 12 patients with comorbid alcohol and OUD were inducted onto naltrexone and safely treated (Korthuis et al., 2017). A case report of naltrexone treatment for a patient with alcohol and OUD observed increased depressive symptoms in response to naltrexone that were relieved when the patient was switched to buprenorphine (Schurks et al., 2005). Despite the paucity of data addressing the question, a recent review of pharmacotherapy for comorbid alcohol and opioid use (Hood et al., 2020) concluded that naltrexone may be the preferred option to address AUD comorbidity. However, that recommendation was based only upon known psychopharmacological mechanisms and no randomized clinical trials were cited to have examined the effects of naltrexone on alcohol use among patients with OUD.

The current study seeks to address the research gap in our knowledge of how to treat AUD comorbidity in OUD. It is a secondary analysis of alcohol use outcomes in the per-protocol sample of outpatients with OUD who participated in the randomized X:BOT effectiveness trial of injectable XR-NTX versus sublingual BUP-NX. We sought to determine whether treatment with naltrexone differentially affects concurrent alcohol use for patients receiving treatment for OUD. Specifically, we hypothesized that as an FDA-approved medication to reduce heavy drinking, XR-NTX should be superior to BUP-NX to reduce heavy drinking.

METHODS

Original study design

The original study was conducted by the NIDA CTN (protocol CTN-0051) at eight community treatment sites as a 2-arm randomized trial comparing treatment with XR-NTX (Vivitrol injection, 380 mg of naltrexone base) versus BUP-NX (Suboxone, flexibly dosed in the range of 2 to 24 mg of BUP per day) delivered open label in a pragmatic trial design (Nunes et al., 2016). The patients were first detoxified sufficiently to start one of the two medications within a hospital or residential facility. Outpatient treatment was then continued for 6 months with weekly study visits completed to collect urine samples, monitor safety, and assess daily alcohol and drug use using calendar-based timeline follow-back (Sobell & Sobell, 1995). Other details of methodology and trial design were published in a design manuscript (Lee et al., 2016) and/or in the main trial result (Lee et al., 2018).

Sample for analysis

The current study utilizes data taken from the main outcome per-protocol sample (Lee et al., 2018) which was restricted to the 474 patients (270 BUP-NX, 204 XR-NTX) successfully inducted onto treatment. Briefly, study inclusion required patients with Diagnostic & Statistical Manual—Version 5 (DSM-5) OUD who had used non-prescribed opioids within the past 30 days and did not have clinically serious medical or mental illness including other substance use disorder, excepting tobacco use disorder or mild-to-moderate AUD. There were no inclusion or exclusion criteria related to alcohol other than the exclusion of patients with liver transaminase levels >5×normal. Table 1 describes the patient sample.

TABLE 1.

Patient sample description

Total (N = 474) n (%) XR-NTX (N = 204) n (%) BUP-NX (N = 270) n (%) p-value
Demographics
 Age (years) [mean (SD)] 33.7 (9.6) 33.7 (9.3) 33.7 (9.8) 0.9912
 Number (%) male 331 (70%) 138 (68%) 193 (71%) 0.3678
 Number (%) white 358 (76%) 157 (77%) 201 (74%) 0.5281
 Number (%) never married 314 (66%) 134 (66%) 180 (67%) 0.9597
 Number (%) currently unemployed 297 (63%) 125 (61%) 172 (64%) 0.3560
Opioid use measures
 % using intravenous opioids 301 (64%) 131 (64%) 170 (63%) 0.8539
 Duration of opioid use (years) [mean (SD)] 12.5 (9.1) 12.9 (9.1) 12.3 (9.1) 0.4608
% using substance past 30 days
 Number using stimulants 254 (54%) 100 (49%) 154 (57%) 0.1010
 Number using sedatives 139 (29%) 53 (26%) 86 (32%) 0.1976
 Number using cannabis 216 (46%) 86 (42%) 130 (48%) 0.2287
 Number reporting “Any” alcohol drinking 259 (55%) 110 (54%) 149 (55%) 0.7506
 Number reporting any “Heavy” alcohol drinking 214 (45%) 94 (46%) 120 (45%) 0.9396
Drinking measures (in past 30 days)
 % days of any drinking [PDD mean (SD)] 13.3 (27.2) 12.9 (25.9) 13.7 (28.1) 0.7503
 # drinks per drinking day [DDD median (IQR)] 5.0 (2.0 to 9.4) 4.6 (2.0 to 8.0) 5.0 (2.0 to 10.7) 0.4825
 % days of “heavy” (among drinkers) [PHDD median (IQR)] 3.3 (0.0 to 40.0) 3.3 (0.0 to 36.7) 3.3 (0.0 to 40.0) 0.5985
Open in a new tab

Note: Patient sample characteristics (at baseline) of those (N = 474) included in analyses. Patients were randomized to receive extended release naltrexone (XR-NTX) and buprenorphine-naloxone (BUP-NX). Data are number of patients or percentages (%), unless specified as years (years), means, medians, standard deviations (SD), or interquartile range (IQR). Drinking is described as % days drinking (PDD), drinks per drinking day (DDD), and % heavy drinking days (PHDD).

Assessment of alcohol and substance use and disorder

Diagnoses of AUD involved nonstructured clinical assessments using DSM-5 criteria for those patients who reported regular patterns of alcohol use. All patients were assessed for daily opioid, alcohol, and other drug use using calendar-based timeline follow-back techniques (Sobell & Sobell, 1995). At baseline, the assessment tracked drug and alcohol use over the previous 30 days; then, throughout the trial, weekly assessments collected the same information for the past week. Following the standards of the National Institute on Alcohol Abuse and Alcoholism (NIAAA) assessing drink type and container size, alcohol drinking each day was calculated as the number of standard drinks (1 drink = 12 oz beer = 5 oz wine = 1.5 oz liquor). As in previous studies (O’Malley et al., 1992; Volpicelli et al., 1992), the current analysis defined “heavy” drinking as ≥5 standard drinks in a day for males or ≥4 drinks in a day for females. AUD diagnostic severity followed DSM-5 recommendations with mild (2 to 3 symptoms), moderate (4 to 5 symptoms), and severe (≥ 6 symptoms) designations.

Data analysis

As commonly done for alcohol treatment studies, daily drinking reports obtained by timeline follow-back were reduced to “monthly” (4-week period) aggregate values of three drinking parameters, namely: (i) % days “Any” drinking (PDD); (ii) number of drinks per drinking day (DDD); and (iii) % “Heavy” drinking days (PHDD). Following examination of the drinking outcomes, the PDD and PHDD variables were dichotomized for further analysis into “Any” versus “No” days of any drinking or heavy drinking, respectively. A mixed-effects linear regression model was fit with DDD as the outcome, whereas mixed-effects logistic regression models were fit for any drinking days and any heavy drinking days outcomes. Each of these models included random effects for subject and site and an autoregressive correlation structure (AR(1)) for within-subject observations. Predictors for each model included treatment (XR-NTX vs. BUP-NX), time (months 1 to 6, as categorical) postrandomization, and the 2-way interaction between treatment and time. Baseline values of alcohol use, obtained from the 30 days prior to randomization, were also included as covariates. From these models, contrasts were made to estimate the treatment effects at 6 months.

Additionally, after testing the proportional-hazards assumption, Cox proportional-hazards models were fit to test for treatment differences in the time (from baseline) to first day of alcohol use. This model was adjusted for proportion of days with alcohol use in the 30 days prior to randomization. In a subsample analysis of 179 participants with AUD or some amount of heavy drinking at baseline, across-group reductions in PDD, PHDD, and at DDD during the full treatment period were assessed using paired t-tests. Cohen’s d effect sizes were calculated as mean differences from baseline to 6 months, divided by the standard deviation at baseline. All analyses were performed using SAS version 9.4. All relevant hypothesis tests were tested as 2-sided on level of significance 5%.

RESULTS

The baseline (pretreatment) demographics and substance use characteristics of the analysis sample are shown in (Table 1). Patients were primarily white, male, never-married, opioid injection users (primarily heroin) who were currently unemployed. There were no significant differences between groups on these characteristics (all p > 0.35). Beyond opioids, the use of other drugs including alcohol was fairly common; however, the sample as a whole did not drink frequently (PDD = 27% days), and the frequency of heavy drinking was rare (PHDD = 3.3%).

Our first analysis looked simply at the per-protocol sample of 474 patients. There were no significant differences related to treatment for any of the drinking parameters and no significant 2-way interactions between treatment and time (all p > 0.17). The failure to detect any effect of treatment on drinking in the full sample led us to look more closely at the levels of drinking observed in the 30 days prior to the study (see Table 2). Only 29% of the sample were identified as having at least mild AUD, although only 21% were of at least moderate severity. Also seen in Table 2 is that only 10% patients drank frequently (i.e., n = 48 drank on more than half of the days) and few (only n = 129 or 27%) reported any heavy drinking at all. These observations and percentages were comparable for each randomized group broken out for the n = 204 XR-NTX and n = 270 BUP-NX participants as well. There were no significant differences in the odds of the XR-NTX group versus BUP-NX in having an AUD diagnosis (OR = 0.86; 95% CI = 0.57, 1.29; p = 0.4689) or the proportion having “Any” drinking (OR = 1.06; 95% CI = 0.74, 1.53; p = 0.7506) or the proportion having any “Heavy” drinking (OR = 1.02; 95% CI = 0.68, 1.53; p = 0.9396) during the 30 days prior to treatment.

TABLE 2.

Patient sample drinking characteristics in the 30 days prior to study

AUD diagnostic severity codes
 Total %
None Mild Moderate Severe Total #
Total sample
 Total # 338 34 26 76 474
 Total % 71 7 5 16 100
# “Any” drinking days
 Unknown 1 0 0 0 1 0.2
 None 233 9 6 11 259 55
 1 to 15 days 103 23 13 27 166 35
 >15 days 1 2 7 38 48 10
# “Heavy” drinking days
 Unknown 1 0 0 0 1 0.2
 None 294 22 11 17 344 73
 1 to 8 days 40 6 6 16 68 14
 9 to 14 days 2 4 2 7 15 3
 ≥15 days 1 2 7 36 46 10
XR-NTX
 Total # 149 16 11 28 204
 Total % 73 8 5 14 100
# “Any” drinking days
 Unknown 0 0 0 0 0 0
 None 103 3 1 3 110 55
 1 to 15 days 45 11 6 12 74 35
 >15 days 1 2 4 13 20 10
# “Heavy” drinking days
 Unknown 0 0 0 0 0 0
 None 126 12 3 7 148 73
 1 to 8 days 22 1 3 4 30 15
 9 to 14 days 0 1 1 6 8 4
 ≥15 days 1 2 4 11 18 9
BUP-NX
 Total # 189 18 15 48 270
 Total % 70 7 6 18 100
# “Any” drinking days
 Unknown 1 0 0 0 1 0.4
 None 130 6 5 8 149 55
 1 to 15 days 58 12 7 15 92 34
 >15 days 0 0 3 25 28 10
# “Heavy” drinking days
 Unknown 1 0 0 0 1 0.4
 None 168 10 8 10 196 73
 1 to 8 days 18 5 3 12 38 14
 9 to 14 days 2 3 1 1 7 3
 ≥15 days 0 0 3 25 28 10
Open in a new tab

Note: Shown are the numbers of patients (or %) of N-474 patients who reported the designated number of “Any” drinking days or the number of “Heavy” drinking days, in the 30 days before baseline, displayed as a function of their Diagnostic & Statistical Manual—Version 5 (DSM-5) severity codes for alcohol use disorder (AUD).

Abbreviations: BUP-NX, buprenorphine-naloxone; XR-NTX, extended release naltrexone.

In order to evaluate treatment effects in a subsample of patients enriched for possible drinking, we selected n = 136 who had an AUD diagnosis at baseline, plus n = 43 who did not have AUD, but reported at least one day of heavy drinking during the 30-day baseline period before randomization. This gave us a subset (i.e., only 38% of analysis sample) of n = 101 patients receiving XR-NTX and n = 78 receiving BUP-NX. The baseline drinking behavior of the two groups is shown in Table 3. Among this sample, there was a significant difference indicating that the BUP-NX group drank significantly more DDD (p = 0.0353) than the XR-NTX group at baseline; but the two groups were not different on PHDD or PDD.

TABLE 3.

Analysis sample drinking in past 30 days at baseline

Total (n = 179) XR-NTX (n = 101) BUP-NX (n = 78) p
Drinking measures
 % days of any drinking [PDD mean (SD)] 32.0 (36.5) 30.4 (34.6) 33.2 (38.0) 0.6125
 # drinks per drinking day [DDD median (IQR)] 8.0 (4.3 to 12.0) 7.0 (3.8 to 10.2) 8.2 (5.0 to 13.0) 0.0353
 % days of “heavy” drinking [PHDD median (IQR)] 13.3 (3.3 to 63.3) 10.0 (3.3 to 53.3) 18.3 (3.3 to 70.0) 0.0966
Open in a new tab

Note: Drinking levels reported for a subsample (n = 179) who either had an alcohol use disorder (AUD) diagnosis or who reported >1 day of “heavy” drinking in the 30 days prior to the baseline assessment. Other details are same as Table 1.

Abbreviations: BUP-NX, buprenorphine-naloxone; DDD, drinks per drinking day; IQR, interquartile range; PHDD, % heavy drinking days; PDD, % days drinking; SD, standard deviations; XR-NTX, extended release naltrexone.

This expanded subset of patients with some level of heavy drinking and/or AUD exhibited a median of 8.0 DDD at baseline. However, the frequency of any drinking at all was still low and occurred on only 32% of days with the median frequency of heavy drinking being only 13% of days. Figure 1 displays the unadjusted 6-month drinking outcomes for this subset of patients on the three main drinking measures.

FIGURE 1.

FIGURE 1

Open in a new tab

Drinking outcomes for each of 6 months of treatment with extended release naltrexone (XR-NTX) or buprenorphine-naloxone (BUP-NX). Data are observed means for the subsample of n = 179 patients—unadjusted for any statistical model parameters

Comparisons of drinking reported for the 30 days prior to treatment with the observations of drinking over 6 months postrandomization suggested there were general reductions in drinking within the first month of entering treatment. Among the subsample of n = 176 participants with AUD and/or some amount of heavy drinking, this was seen as a significant reduction from 32% to 7% in frequency of any drinking (PDD Cohen’s d = 0.67, p < 0.0001), from 8.0 to 6.2 in DDD (DDD Cohen’s d = 0.30, p = 0.0073), and from 28% to 4% in the frequency of heavy drinking (PHDD Cohen’s d = 0.65, p < 0.0001). Thereafter, drinking levels in this study remained stable over the 6 months of treatment. There were no significant differences attributable to treatment group for any of the three drinking measures (p > 0.17).

In order to be certain that there were no differential effects of treatment on drinking in this study, we also conducted a survival analysis of “relapse” defined as time to first day of “any” drinking (Figure 2A) or time to first day of “heavy” drinking day (Figure 2B) from the point of randomized induction during inpatient treatment. The percentage of patients without any drinking at all decreased to 69% by the end of the first month and reached a final value of 37% never drinking by 6 months. For the heavy drinking measure, those numbers were 83% at one month, and 55% by 6 months. For both of these analyses, there were no significant differences between the two randomized treatment groups.

FIGURE 2.

FIGURE 2

Open in a new tab

Survival curves for the times to first event of any drinking or heavy drinking. Data are survival from the point of randomization to XR-NTX (n = 78) or BUP-NX (n = 101) of the subsample of n = 179 patients

DISCUSSION

The original pragmatic trial comparing XR-NTX injection with sublingual BUP-NX found that the two treatments were similarly effective at preventing opioid relapse among those patients successfully inducted onto treatment (Lee et al., 2018). Contrary to our hypothesis, the current analysis of that treatment sample did not find evidence that naltrexone was superior to buprenorphine to reduce heavy alcohol drinking. It is likely that the hypothesis test of naltrexone superiority over buprenorphine suffered from floor-effects brought about by the low frequencies of heavy drinking observed in this per-protocol treatment sample. With such low levels of heavy drinking observed at baseline, and with further reductions observed during treatment, there was little room to show between group differences in drinking reduction. Notably, only 21% of our study sample met DSM-5 criteria for moderate severity or greater AUD and 18% of those were alcohol abstinent in the 30 days before treatment. Even among a subsample enriched for increased likelihood of drinking (38% of the total sample), any drinking at all was still relatively infrequent (i.e., a mean of 33% of days involved alcohol consumption) and heavy drinking occurred on only 13% of days in the 30 days before beginning treatment. Furthermore, drinking significantly declined within the first month for both treatment groups and persisted throughout 6 months of treatment. There were no significant differences between treatment groups in drinking reduction.

AUD has been reported to occur in one-third (or more) of patients in opioid substitution therapy where it is often associated with clinically worse treatment outcomes (see Nolan et al., 2016; Soyka, 2015). A previous analysis of the X:BOT study sample suggested that preexisting AUD diagnosis did not predict differential opioid-treatment outcomes (Nunes, personal communication). However, such a finding differs from previous reports suggesting that AUD comorbidity is associated with clinically or prognostically worse outcomes. For the current study analysis, only 28% of 474 patients met criteria for AUD, and 55% were alcohol abstinent in the 30 days before treatment. Additionally, both the frequency and amount of alcohol use decreased even further with the initiation of treatment for a subsample of 179 patients with AUD or heavy drinking patterns at baseline. This suggests that for some reason, this particular study sample of largely injection opioid users enrolled in outpatient treatment for OUD generally did not experience major consequences from alcohol use-related comorbidity. The low level of alcohol use observed during treatment coupled with a lack of difference between treatment arms may suggest that effective treatment of OUD with either an agonist or antagonist approaches may simultaneously reduce other substance use (including alcohol) for patients with primary OUD. However, there are equivocal findings about whether changes in alcohol use should be expected, once individuals initiate and successfully undergo treatment for OUD.

A longitudinal trajectory analysis of 7,717 patients receiving methadone or buprenorphine substitution therapy found that among the 50% of the sample with significant levels of alcohol consumption, about a third increased, a third decreased, and a third remained unchanged in the frequency of alcohol intake over a 5-year period (Eastwood et al., 2019). Other studies have shown decreases (Caputo et al., 2002; Nava et al., 2008) or no change in alcohol use (Klimas et al., 2016, 2018) with methadone or buprenorphine treatment, although an older literature had reported treatment-related increases in alcohol use (Anglin et al., 1989; Gelb et al., 1978; Simpson & Lloyd, 1978). For the current study, we can certainly conclude that treatment for OUD was not associated with compensatory increases in alcohol consumption. However, there may be more nuanced explanations for discrepant findings about comorbid alcohol consumption during OUD treatment. The issues of comorbid alcohol or other substance use are complex and relate to whether or not alcohol or other substance use complements with, substitutes for, or is independent of opioid use (Caputo et al., 2002; Klimas et al., 2018; Staiger et al., 2012). Another factor may involve behavioral or personality characteristics of individuals with OUD. In the longitudinal trajectory analysis of substitution therapy (Eastwood et al., 2019), those categorized into a “continued high-level heroin use” trajectory category also were more likely to use alcohol and crack than those in other trajectory categories. Such results suggest that alcohol use may be one of the multiple drugs of misuse in a subset of patients receiving opioid treatment. Another study of patients in a methadone program (Moses & Greenwald, 2019) concluded that it was not the alcohol use per se that increased problems, but rather the fact that alcohol use was associated with an increased likelihood of sedative use, and that sedative use led to poor outcomes. Indeed, among opioid-treatment patients with alcohol use comorbidity, opioid overdose deaths are more likely (Joseph & Appel, 1985; Zador & Sunjic, 2000), benzodiazepine use is increased (Simon et al., 2019), as is the use of other drugs (Hunt et al., 1986; Preston et al., 2016). This patient group also struggles with other predictors of negative treatment outcomes such as difficulties with engagement and retention in treatment (Joseph & Appel, 1985; Rowan-Szal et al., 2000), lower psychosocial functioning (Hartzler et al., 2010), and poor quality of life (Senbanjo et al., 2007). Consistent with this observation, a post hoc mediational analysis of the large Project COMBINE treatment trial for AUD reported that comorbid opioid use was associated with worse drinking outcomes (Witkiewitz et al., 2018).

Several studies have reported outcomes of using naltrexone within a medication-assisted treatment for alcohol and/or OUD (Crits-Christoph et al., 2016; Hartung et al., 2014; Korthuis et al., 2017), but these studies did not break out specifically the outcomes for individuals with opioid and comorbid AUD. One case report (Schurks et al., 2005) observed a naltrexone-related depression in a patient with comorbid alcohol and OUD. The only studies we could find that specifically targeted treatment of comorbid alcohol and opioid use were older studies utilizing disulfiram to reduce alcohol use among methadone maintenance patients (Bickel et al., 1987, 1988; Liebson et al., 1973, 1978), and an integrated psychosocial treatment simultaneously targeting alcohol and OUDs (Watkins et al., 2017). Interestingly, a positive suggestion of a possible benefit for naltrexone was reported by a recent blood toxicology study of fatalities involving alcohol and other drugs (Kelty et al., 2020) where alcohol was less frequently detected among fatalities with opioid-positive toxicology findings if they also had a naltrexone implant. Given the co-occurrence of alcohol and other drug use in opioid-treatment seeking populations, we were surprised to note that so few studies have specifically been designed to target the treatment of comorbid substance use disorders. Given a recent review of pharmacotherapy options for treatment of alcohol and opioid use comorbidity (Hood et al., 2020) was forced to make recommendations based upon putative mechanisms rather than clinical trial evidence, it seems even more important for future research to specifically answer the question of naltrexone efficacy as a treatment to benefit patients who are dually afflicted with both AUD and OUD.

The limitations of our study include the fact that this was a secondary analysis of data from a study focused on the treatment of OUD without regard to levels of alcohol consumption which were decidedly low. Specifically, AUD comorbidity was not an inclusion criterion for the study, randomization did not control for levels of drinking or the frequency of any drinking or heavy drinking in the 30 days leading up to treatment induction. Because of these low levels of drinking, we attempted to enrich the likelihood of drinking by analyzing only a subset of the sample with AUD or heavy drinking patterns. However, this action further limited the analysis to a reduced sample size with unknown bias in sample selection. Although naltrexone has been reported to increase abstinence in patients seeking treatment for AUD (Anton et al., 2006; Soyka et al., 2017), its primary effect has been to reduce drinking at “heavy” levels. Unfortunately for this subsample, only 13% of the patients reported heavy drinking on 9 or more days in the past 30 days. We did not attempt to calculate effect sizes for XR-NTX versus BUP-NX differences because all comparisons observed only very small effects that would not be clinically significant if indeed differences existed at all. It is certain that the current study sample size did not have the statistical power to detect these effects. We also did not engage a formal power calculation because a study sufficiently powered to detect small clinically insignificant effects would not be justified. We do not know whether larger effects may have been observed if the baseline levels of heavy drinking had been higher such as might have been observed if the study had specifically targeted a population having both OUD and concurrent AUD. There also could have been some bias in the XR-NTX effects observed, because in the original study, fewer patients were successfully inducted onto XR-NTX (73%) than onto BUP-NX (94%) (Lee et al., 2018). We do not know whether the patients who were not successfully inducted onto treatment may have included those with a poorer prognosis due in part, to increased amounts of alcohol use.

CONCLUSIONS

A secondary analysis of alcohol drinking outcomes from a randomized clinical trial comparing XR-NTX and BUP-NX treatments for OUD did not find differences in the frequency or amount of any drinking or heavy drinking during 6 months of outpatient treatment. However, the study was limited by higher than expected rates of alcohol abstinence at baseline, a low overall prevalence of AUD, and a low frequency of heavy drinking among the study sample. As we were able to analyze only the per-protocol sample who had been successfully inducted onto treatment in the original trial, we do not know if induction failures may have had higher rates of AUD. For the minority of study participants with AUD or some level of heavy drinking in the 30 days before treatment, there was evidence that drinking may have been further reduced by treatment, but the pre-post reduction did not differ between the two treatment groups. Even though we failed to detect any difference between XR-NTX and BUP-NX on drinking outcomes, future research should determine whether or not there is any evidence-based reason to suggest that naltrexone may be a preferred treatment option for those individuals with OUD who have a clearly defined AUD comorbidity and elevated frequencies of heavy drinking.

ACKNOWLEDGMENTS

This research was supported by grants from the NIDA National Drug Abuse Treatment Clinical Trials Network (U10DA013046, UG1/U10DA013035, UG1/U10DA013034, U10DA013045, UG1/U10DA013720, UG1/U10DA013732, UG1/U10DA013714, UG1/U10DA015831, U10DA015833, HHSN271201200017C, and HHSN271201500065C) and K24DA022412 (EVN Jr).

Funding information

NIDA, Grant/Award Number: HHSN271201200017C, HHSN271201500065C, K24DA022412, U10DA013045, U10DA013046, U10DA015833, UG1/U10DA013035, UG1/U10DA013034, UG1/U10DA013714, UG1/U10DA013720, UG1/U10DA013732 and UG1/U10DA015831

Footnotes

CONFLICT OF INTEREST

None of the authors have any conflicts of interest to disclose.

REFERENCES

  1. Anglin MD, Almog IJ, Fisher DG & Peters KR (1989) Alcohol use by heroin addicts: evidence for an inverse relationship. A study of methadone maintenance and drug-free treatment samples. The American Journal of Drug and Alcohol Abuse, 15(2), 191–207. 10.3109/00952998909092720 [DOI] [PubMed] [Google Scholar]
  2. Anton RF, O’Malley SS, Ciraulo DA, Cisler RA, Couper D, Donovan DM et al. (2006). Combined pharmacotherapies and behavioral interventions for alcohol dependence. The COMBINE study: a randomized controlled trial. JAMA, 295(17), 2003–2017. 10.1001/jama.295.14.2003 [DOI] [PubMed] [Google Scholar]
  3. Bickel WK, Marion I. & Lowinson JH (1987) The Treatment of alcoholic methadone patients: a review. Journal of Substance Abuse, 4, 15–19. [DOI] [PubMed] [Google Scholar]
  4. Bickel WK, Rizzuto P, Zielony RD, Klobas J, Pangiosonlis P, Mernit R. et al. (1988) Combined behavioral and pharmacological treatment of alcoholic methadone patients. Journal of Substance Abuse, 11, 161–171. [DOI] [PubMed] [Google Scholar]
  5. Caputo F, Addolorato G, Domenicali M, Mosti A, Viaggi M, Trevisani F. et al. (2002) Short-term methadone administration reduces alcohol consumption in non-alcoholic heroin addicts. Alcohol and Alcoholism, 37(2), 164–168. 10.1093/alcalc/37.2.164 [DOI] [PubMed] [Google Scholar]
  6. Comer SD, Sullivan MA, Yu E, Rothenberg JL, Kleber HD, Kampman K. et al. (2006) Injectable, sustained-release naltrexone for the treatment of opioid dependence: a randomized, placebo-controlled trial. Archives of General Psychiatry, 63(2), 210–218. 10.1001/srchpsyc.63.2.210 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Crits-Christoph P, Markell HM, Connolly-Gibbons MB, Gallop R, Lundy C, Stringer M. et al. (2016) A naturalistic evaluation of extended-release naltrexone in clinical practice in Missouri. Journal of Substance Abuse Treatment, 70, 50–57. 10.1016/j.jsat.2016.07.014 [DOI] [PubMed] [Google Scholar]
  8. Eastwood B, Strang J. & Marsden J. (2019) Change in alcohol and other drug use during five years of continuous opioid substitution treatment. Drug and Alcohol Dependence, 194, 438–446. 10.1016/j.drugalcdep.2018.11.008 [DOI] [PubMed] [Google Scholar]
  9. Garbutt JC, Kranzler HR, O’Malley SS, Gastfriend DR, Pettinati HM, Silverman BL et al. (2005) Efficacy and tolerability of long-acting injectable naltrexone for alcohol dependence: a randomized controlled trial. JAMA, 293(13), 1617–1625. 10.1001/jama.293.13.1617 [DOI] [PubMed] [Google Scholar]
  10. Gelb AM, Richman BL & Anand OM (1978) Quantitative and temporal relationships of alcohol use in narcotic addicts and methadone maintenance patients undergoing alcohol detoxification. The American Journal of Drug and Alcohol Abuse, 5(2), 191–198. [DOI] [PubMed] [Google Scholar]
  11. Hartung DM, McCarty D, Fu R, Wiest K, Chalk M. & Gastfriend DR (2014) Extended-release naltrexone for alcohol and opioid dependence: a meta-analysis of healthcare utilization studies. Journal Substance Abuse Treatment, 47(2), 113–121. 10.1016/j.jsat.2014.03.007 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hartzler B, Donovan DM & Huang Z. (2010) Comparison of opiate-primary treatment seekers with and without alcohol use disorder. Journal of Substance Abuse Treatment, 39(2), 114–123. 10.1016/j.jsat.2010.05.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Hood LE, Leyrer-Jackson JM & Olive MF (2020) Pharmacological management of co-morbid alcohol and opioid use. Expert Opinion on Pharmacotherapy, 21(7), 823–839. 10.1080/14656566.2020.1732349 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hunt DE, Strug DL, Goldsmith DS, Lipton DS, Robertson K. & Truitt L. (1986) Alcohol use and abuse: heavy drinking among methadone clients. The American Journal of Drug and Alcohol Abuse, 12(1–2), 147–164. 10.3109/00952998609083749 [DOI] [PubMed] [Google Scholar]
  15. Joseph H. & Appel P. (1985) Alcoholism and methadone treatment: consequences for the patient and program. The American Journal of Drug and Alcohol Abuse, 11(1–2), 37–53. 10.3109/00952998509016848 [DOI] [PubMed] [Google Scholar]
  16. Kelty E, Hulse G. & Joyce D. (2020) A comparison of blood toxicology in fatalities involving alcohol and other drugs in patients with an opioid use disorder treated with methadone, buprenorphine, and implant naltrexone. American Journal of Drug and Alcohol Abuse, 46(2), 241–250. 10.1080/00952990.2019.1698587 [DOI] [PubMed] [Google Scholar]
  17. Klimas J, Wood E, Nguyen P, Dong H, Milloy MJ, Kerr T. et al. (2016) The impact of enrolment in methadone maintenance therapy on initiation of heavy drinking among people who use heroin. European Addiction Research, 22(4), 210–214. 10.1159/000444513 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Klimas J, Wood E, Nosova E, Milloy MJ, Kerr T. & Hayashi K. (2018) Prevalence of heavy alcohol use among people receiving methadone following change to methadose. Substance Use & Misuse, 53(2), 270–275. 10.1080/10826084.2017.1302960 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Korthuis PT, Lum PJ, Vergara-Rodriguez P, Ahamad K, Wood E, Kunkel LE et al. (2017) Feasibility and safety of extended-release naltrexone treatment of opioid and alcohol use disorder in HIV clinics: a pilot/feasibility randomized trial. Addiction, 112(6), 1036–1044. 10.1111/add.13753 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kreek MJ & Vocci FJ (2002) History and current status of opioid maintenance treatments: Blending conference session. Journal of Substance Abuse Treatment, 23(2), 93–105. 10.1016/s0740-5472(02)00259-3 [DOI] [PubMed] [Google Scholar]
  21. Krupitsky E, Nunes EV, Ling W, Illeperuma A, Gastfriend DR & Silverman BL (2011) Injectable extended-release naltrexone for opioid dependence: a double-blind, placebo-controlled, multicenter randomised trial. The Lancet, 377(9776), 1506–1513. 10.1016/s0140-6736(11)60358-9 [DOI] [PubMed] [Google Scholar]
  22. Lee JD, Nunes EV, Novo P, Bailey GL, Brigham GS, Cohen AJ et al. (2016) NIDA Clinical Trials Network CTN-0051, extended-release naltrexone vs. buprenorphine for opioid treatment (X:BOT): study design and rationale. Contemporary Clinical Trials, 50, 253–264. 10.1016/j.cct.2016.08.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lee JD, Nunes EV, Novo P, Bachrach K, Bailey GL, Bhatt S. et al. (2018) Comparative effectiveness of extended-release naltrexone versus buprenorphine-naloxone for opioid relapse prevention (X:BOT): a multicentre, open-label, randomised controlled trial. The Lancet, 391(10118), 309–318. 10.1016/s0140-6736(17)32812-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Liebson I, Bigelow G. & Flamer R. (1973) Alcoholism among methadone patients: a specific treatment method. American Journal of Psychiatry, 130(4), 483–485. 10.1176/ajp.130.4.483 [DOI] [PubMed] [Google Scholar]
  25. Liebson IA, Tommasello A. & Bigelow GE (1978) A Behavioral treatment of alcoholic methadone patients. Annals of Internal Medicine, 89(3), 342–344. 10.7326/0003-4819-89-3-342342 [DOI] [PubMed] [Google Scholar]
  26. Mattick RP, Breen C, Kimber J. & Davoli M. (2014) Buprenorphine maintenance versus placebo or methadone maintenance for opioid dependence. Cochrane Database Systematic Reviews, 10.1002/14651858.cd002207.pub4 [DOI] [PubMed]
  27. Moses TE & Greenwald MK (2019) History of regular nonmedical sedative and/or alcohol use differentiates substance-use pattern and consequences among chronic heroin users. Addictive Behaviors, 97, 14–19. 10.1016/j.addbeh.2019.05.017 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Nava F, Manzato E, Leonardi C. & Lucchini A. (2008) Opioid maintenance therapy suppresses alcohol intake in heroin addicts with alcohol dependence: preliminary results of an open randomized study. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 32(8), 1867–1872. [DOI] [PubMed] [Google Scholar]
  29. Nielsen S, Larance B, Degenhardt L, Gowing L, Kehler C. & Lintzeris N. (2016) Opioid agonist treatment for pharmaceutical opioid dependent people. Cochrane Database of Systematic Reviews, 10.1002/14651858.cd011117.pub2 [DOI] [PubMed]
  30. Nolan S, Klimas J. & Wood E. (2016) Alcohol use in opioid agonist treatment. Addiction Science & Clinical Practice, 11(1), 1–7. 10.1186/s13722-016-0065-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Nunes EV, Lee JD, Sisti D, Segal A, Caplan A, Fishman M. et al. (2016) Ethical and clinical safety considerations in the design of an effectiveness trial: a comparison of buprenorphine versus naltrexone treatment for opioid dependence. Contemporary Clinical Trials, 51, 34–43. 10.1016/j.cct.2016.09.006 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Nunes EV, Rothenberg JL, Sullivan MA, Carpenter KM & Kleber H. (2006) Behavioral therapy to augment oral naltrexone for opioid dependence: a ceiling on effectiveness? American Journal of Drug and Alcohol Abuse, 32(4), 503–517. [DOI] [PubMed] [Google Scholar]
  33. O’Malley SS, Jaffe AJ, Chang G, Schottenfeld RS, Meyer RE & Rounsaville B. (1992) Naltrexone and coping skills therapy for alcohol dependence: a controlled study. Archives of General Psychiatry, 49(11), 881–887. 10.1001/archpsyc.1992.01820110045007 [DOI] [PubMed] [Google Scholar]
  34. Preston KL, Jobes ML, Phillips KA & Epstein DH (2016) Real time assessment of alcohol drinking and drug use in opioid-dependent polydrug users. Behavioral Pharmacology, 27, 579–584. 10.1097/FBP.000000000000000250 [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Rowan-Szal GA, Chatham LR & Simpson DD (2000) Importance of identifying cocaine and alcohol dependent methadone clients. American Journal on Addictions, 9(1), 38–50. 10.1080/1055049005014900501 [DOI] [PubMed] [Google Scholar]
  36. Schurks M, Overlack M. & Bonnet U. (2005) Naltrexone treatment of combined alcohol and opioid dependence: deterioration of co-morbid major depression. Pharmacopsychiatry, 38(2), 100–102. 10.1055/s-2005-837812837812 [DOI] [PubMed] [Google Scholar]
  37. Senbanjo R, Wolff K. & Marshall J. (2007) Excessive alcohol consumption is associated with reduced quality of life among methadone patients. Addiction, 102(2), 257–263. 10.1111/j.1360-0443.2006.01683.x [DOI] [PubMed] [Google Scholar]
  38. Simon J, Gehret J, Stolzenberg D, Beredijiklian PK, Teng J, Paskey T. et al. (2019) Concomitant use of opioids and benzodiazepines in the outpatient setting. Pain Medicine & Rehabilitation PM&R, 11(4), 337–343. 10.1016/j.pmrj.2018.09.026 [DOI] [PubMed] [Google Scholar]
  39. Simpson DD & Lloyd MR (1978) Alcohol and illicit drug use: follow-up study of treatment admissions to DARP during 1969–1971. The American Journal of Drug and Alcohol Abuse, 5(1), 1–22. 10.3109/00952997809029257 [DOI] [PubMed] [Google Scholar]
  40. Sobell LC & Sobell MB (1995) Alcohol timeline followback users’ manual. Toronto, ON: Addiction Research Foundation. [Google Scholar]
  41. Soyka M. (2015) Alcohol use disorder in opioid maintenance therapy: prevalence, clinical correlates and treatment. European Addiction Research, 21(2), 78–87. 10.1159/0003632322 [DOI] [PubMed] [Google Scholar]
  42. Soyka M, Kranzler HR, Hesselbrock V, Kasper S, Mutschler J, Moller H. et al. (2017) Guidelines for biological treatment of substance use and related disorders, part 1: alcoholism, first revision. The World Journal of Biological Psychiatry, 18(2), 86–119. 10.1080/15622975.2016.1246752 [DOI] [PubMed] [Google Scholar]
  43. Staiger PK, Richardson B, Long CM, Carr V. & Marlatt GA (2012) Overlooked and underestimated? Problematic alcohol use in clients recovering from drug dependence. Addiction, 108, 1188–1193. 10.1111/j.1360-0443.2012.04075. [DOI] [PubMed] [Google Scholar]
  44. Volpicelli JR (1987) Uncontrollable events and alcohol drinking. British Journal of Addiction, 82(4), 381–392. 10.1111/j.1360-0443.1987.tb01494.x [DOI] [PubMed] [Google Scholar]
  45. Volpicelli JR, Alterman AI, Hayashida M. & O’Brien CP (1992) Naltrexone in the treatment of alcohol dependence. Archives of General Psychiatry, 49(11), 876–880. 10.1001/archpsyc.1992.01820110040006 [DOI] [PubMed] [Google Scholar]
  46. Volpicelli JR, Rhines KC, Rhines JS, Volpicelli LA, Alterman AI & O’Brien CP (1997) Naltrexone and alcohol dependence: Role of subject compliance. Archives of General Psychiatry, 54(8), 737–742. 10.1001/archpsyc.1997.01830200071010 [DOI] [PubMed] [Google Scholar]
  47. Watkins KE, Ober AlJ, Lamp K, Lind M, Setodigj C, Osilla KC et al. (2017) Collaborative Care for opioid and alcohol use disorders in primary care. The SUMMIT Randomized clinical trial. JAMA Internal Medicine, 177(10), 1480–1488. 10.1001/jamainternmed.2017.3947 [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Witkiewitz K, Votaw VR, Vowles KE & Kranzler HR (2018) Opioid misuse as a predictor of alcohol treatment outcomes in the COMBINE study: mediation by medication adherence. Alcohol Clinical Experimental Research, 42, 1249–1259. 10.1111/acer.13772 [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Zador D. & Sunjic S. (2000) Deaths in methadone maintenance treatment in New South Wales, Australia 1990–1995. Addiction, 95(1), 77–84. 10.1046/j.1360-0443.2000.951778.x [DOI] [PubMed] [Google Scholar]

RESOURCES