Initiation, retention, and discontinuation of extended-release buprenorphine (BUP-XR) for opioid dependence: A single-arm multi-site trial in Ukraine

Abstract

Background:

Opioid agonist therapy (OAT) is an essential tool for reducing illicit opioid use and mortality. Extended-release formulations of buprenorphine (BUP-XR) show promise in improving adherence and increasing treatment uptake, but real-world studies show mixed results. Previous studies of BUP-XR were conducted in high-income countries. This study aims to evaluate BUP-XR initiation, retention and dropout in the context of Ukraine.

Methods:

A 12-months prospective single-arm, multicenter, open-label trial was conducted among 181 participants switching from sublingual buprenorphine (BUP-SL) to monthly BUP-XR injections (CAM2038/Buvidal^®) between February-July 2023 across 10 cities of Ukraine. Correlates of BUP-XR discontinuation due to OAT dropout and switch to BUP-SL were analyzed using competing risks cluster-robust Cox proportional hazards regression models with time-varying covariates.

Results:

Despite initial interest, BUP-XR enrollment was low: 273 patients among approximately 2300 eligible agreed to switch to BUP-XR. 80% received 2 or more, 54% – 7 or more, and 44% – 13 or more BUP-XR injections. By the 13^th injection, 30% switched to BUP-SL and 26% discontinued OAT with distinct set of correlates for the two dropout reasons. 13% of participants were prescribed supplemental BUP-SL. 74% of participants starting on the lowest dose (64mg) had their BUP-XR dose subsequently increased.

Conclusion:

BUP-XR uptake and retention in Ukraine were lower than expected in the volatile wartime context. Availability of take-home dosing may reduce the perceived benefits of BUP-XR. Revising clinical guidelines on BUP-XR dosing, dose adjustment, and supplemental BUP-SL prescriptions may reduce the occurrence of opioid withdrawal and improve retention.

Introduction

Opioid use disorder (OUD) remains a critical global public health challenge, accounting for nearly 80% of drug-related deaths worldwide, with approximately one-quarter attributable to opioid overdose (World Health Organization, 2023). Pharmacological treatments for OUD, particularly opioid agonist therapies (OATs), have been shown to effectively reduce illicit opioid use and lower mortality (Shulman et al., 2019). Despite their demonstrated efficacy, the real-world implementation of OAT is hindered by limited coverage, significant barriers to accessibility (Hutchison et al., 2025) and suboptimal treatment retention (Villamil et al., 2024).

The development of extended-release injectable formulations of buprenorphine (BUP-XR) – one of the two primary OAT medications – represents a significant innovation aimed at overcoming barriers to treatment implementation (Soyka & Franke, 2021). By reducing the frequency of dosing and providing an alternative to daily clinic visits where take-home medication is not permitted, BUP-XR has the potential to enhance treatment adherence (Barnett et al., 2021). Clinical trials of BUP-XR provided favorable evidence of its efficacy compared to placebo or sublingual buprenorphine formulation (BUP-SL), which requires daily administration. Two U.S.-based randomized controlled trials assessing the efficacy of two BUP-XR formulations (RBP-6000 / Sublocade^® and CAM2038 / Brixadi^®) reported that BUP-XR was superior to placebo in promoting abstinence from illicit opioids (Haight et al., 2019) and non-inferior to BUP-SL (Lofwall et al., 2018). A pragmatic trial conducted in the U.K. established clinical superiority of BUP-XR compared to BUP-SL or methadone in terms of opioid abstinence, and favorable cost-effectiveness for patients with more severe OUD (Marsden et al., 2023). A recent systematic review of 18 studies concluded that BUP-XR showed favorable performance compared to other forms of OAT for multiple substance use-related outcomes (McMaster & Abeysundera, 2024). Although clinical trials consistently support its efficacy, evidence from real-world implementation remains mixed, with 6-month retention rates ranging from approximately 30% to 86% (Farrell et al., 2022; Frost et al., 2019; Heil et al., 2024; Iacono et al., 2024; Ivasiy et al., 2025; Morgan et al., 2021; Peckham et al., 2021; Stein et al., 2022). To the best of our knowledge, all studies of BUP-XR to date were conducted in high-income counties, primarily the United States, Canada, Australia, the UK and several nations in Western Europe.

Ukraine is an Eastern European country with a pre-war estimated population of people who inject drugs (PWID) of 355,000 or about 1% of adult population (Sazonova et al., 2019), about 80% of whom used opioids (Dumchev et al., 2024). In January 2023, there were 20,063 patients in the national OAT program, of whom 2,384 received BUP-SL, and the rest were on methadone (Public Health Center of the Ministry of Health of Ukraine, 2023). The COVID-19 pandemic and the war with Russia triggered substantial scale-up of take-home dosing of OAT (Meteliuk et al., 2021; Morozova et al., 2023). In January 2023, 91% of OAT patients were on take-home dosing, receiving up to 10 day-supply of medications for unsupervised consumption (Public Health Center of the Ministry of Health of Ukraine, 2023).

Following the full-scale Russian invasion of Ukraine in 2022, the national OAT program faced numerous challenges but managed to sustain operations and expand (Morozova et al., 2023). With 11% estimated coverage, the OAT program in Ukraine has become the largest in Eastern Europe and Central Asia (UNAIDS, 2025). Recognizing the potential benefits of extended-release formulations in mitigating risks of treatment disruption associated with population displacement, the Public Health Center (PHC) of the Ministry of Health of Ukraine, with facilitation and advocacy support from the WHO Regional Office for Europe, requested a humanitarian donation of BUP-XR from the manufacturer, Camurus AB (Sweden). As a result of these efforts, in January 2023 the company supplied the PHC with 500 annual courses of BUP-XR (Buvidal^®, the European trade name of CAM2038).

In advance of the shipment, to facilitate recruitment and plan the medication distribution, the PHC issued a national announcement to OAT program staff and patients, inviting physicians to compile lists of patients interested in transitioning from BUP-SL to BUP-XR. No specific eligibility criteria were established. The announcement emphasized the potential benefits of BUP-XR, including improved convenience and sustained treatment engagement. The initiative was met with strong interest from the patient community, with more than 1,200 individuals joining the initial waitlist.

The present study was conducted to evaluate the effectiveness of BUP-XR in the current context of Ukraine and to inform the decision of potential scale-up of BUP-XR as an alternative OAT formulation through the government-funded program. The specific study aims were: 1) to evaluate 12-months treatment retention; and 2) to assess the reasons for and correlates of BUP-XR discontinuation.

Methods

Study design and population

The study was designed as a 12-months prospective single-arm, multicenter, open-label trial, aiming to evaluate the patient-level outcomes after switching from BUP-SL to BUP-XR. Based on the medication availability, the maximum study sample was limited to 500 people. Ten clinics with the largest numbers of interested patients from ten geographically diverse cities of Ukraine were selected to participate (Supplementary Figure S1).

Recruitment was conducted by physicians, who screened patients using the following eligibility criteria: (i) diagnosed with OUD and receiving OAT with BUP-SL at the time of recruitment; (ii) fully completed induction period and being on a stable dose of BUP-SL; (iii) not planning to move to another city during the next 12 months; (iv) ability to comply with the study procedures. OAT patients who received methadone were excluded due to lack of guidelines for matching the daily methadone dose to the monthly BUP-XR dose at the time of the study, however patients on methadone were offered an option of switching to BUP-XR via BUP-SL induction. Patients who were pregnant, had severe liver, kidney, or respiratory disease, had severe mental health comorbidity, or were unwilling or unable to provide informed consent were excluded. Patients who initially expressed interest in switching to BUP-XR were screened first but did not fill the available 500 slots. After the trial initiation, recruitment was expanded to the entire patient population in Ukraine, and four additional clinics were added to the study (Supplementary Figure S1). Patients who met the eligibility criteria and signed informed consent were administered the first dose of BUP-XR and were included in the study. While participation in the study was not required to receive BUP-XR, every patient who received BUP-XR enrolled.

Medication administration procedures

The study medication was an injectable, extended-release buprenorphine solution (CAM2038/Buvidal^®), available in three once-monthly formulations: 64mg, 96mg, and 128mg. The initial BUP-XR dose was prescribed based on the previous BUP-SL dose (8–10mg, 12–16mg, and 18–24 mg, respectively; patients on 4–6mg BUP-SL received 64mg BUP-XR). BUP-XR injections were administered subcutaneously in accordance with the manufacturer’s guidelines. The first injection was administered one day after the last BUP-SL intake. The subsequent injections were scheduled 21–35 days after the previous one, in accordance with the manufacturer’s guidelines and considering individual patient factors. Before each subsequent injection, treating physicians assessed the patient and adjusted the BUP-XR dose and timing if necessary. Deviations from the recommended inter-injection intervals were permitted based on clinical assessment and individual patient circumstances. Between the injections, patients could be prescribed daily supplemental BUP-SL doses of 2–8 mg if they reported protracted opioid withdrawal symptoms.

Study procedures

Depersonalized clinical data related to BUP-XR administration were collected by clinical personnel using online forms. The dates of each BUP-XR injection and doses, as well as the last pre-transition BUP-SL dose and the date of discharge from the OAT program were recorded in the forms. Reasons for treatment discharge included the following: failure to attend the healthcare facility within 10 days of the scheduled visit and inability to contact the patient during that period; administrative rules violation, patient’s decision to terminate OAT; change of residence; adverse reactions to the medication; desire to return to BUP-SL; incarceration; and death. Patients who wished to withdraw from the study but remain in treatment were asked for their consent to use their depersonalized medical records for research purposes. Study participants who discontinued treatment before the end of the follow-up period were not followed beyond the initial treatment discharge date.

In addition, each patient completed a structured face-to-face interview at baseline (up to 7 days before the first injection), 6, and 12 months after the first injection. In case of BUP-XR discontinuation before 12 months, an exit interview was conducted. The questionnaires included the following sections: sociodemographic information, illicit drug use in the past 6 months, mental health screening instruments (depression: PHQ-9 (Kroenke et al., 2001), anxiety: GAD-7 (Spitzer et al., 2006), alcohol use: AUDIT-C (Saunders et al., 1993)), expectations of BUP-XR experience (baseline only), actual BP-XR experience (follow-up only), reasons for discontinuation (exit only). To minimize the social desirability bias, all interviews were conducted by non-clinical trained research staff. The participants were compensated an equivalent of USD 8 for each completed structured interview.

Statistical analysis

The primary outcome of interest was BUP-XR treatment retention, defined as receiving all scheduled injections. Person-time started on the date of the first injection and was censored on the date of 12th injection plus 1 month, or the date of BUP-XR discontinuation, whichever came first. BUP-XR discontinuation could occur due to one of the two main reasons: (i) complete dropout from the OAT program, or (ii) switching to BUP-SL based on the patient’s preference or due to side effects of BUP-XR. The date of BUP-XR discontinuation was defined as the documented date of discharge from the clinic, or the day before resuming BUP-SL, respectively.

We analyzed correlates of BUP-XR discontinuation using cluster-robust Cox proportional hazards regression models (Therneau & Grambsch, 2000) with time-dependent covariates (Therneau et al., 2024) for variables that could change during the observation period. The two possible reasons for discontinuation were treated as competing risks (Austin et al., 2016).

The following interview-based variables were assessed as potential correlates of retention: socio-demographic characteristics (age, sex, family status, education, employment, homelessness), illicit drug use during 6 months before the baseline (injecting and non-injecting), mental health status (depression, anxiety, alcohol use risk scores), HIV status, and having any negative expectations about BUP-XR (a positive response to any of the following statements: I expect that I will have side effects; physical health will become worse; I will experience withdrawal before the next injection; or the medication will not suit me). The clinical covariates included treatment characteristics before switching to BUP-XR (OAT duration, BUP-SL dose, BUP-SL dispensing mode) and BUP-XR dosing (initial dose as a fixed covariate, and current dose, dose increase, and supplemental prescription of BUP-SL – as time-dependent covariates).

The multivariable models included a priori selected variables that were hypothesized to be associated with treatment retention, accounting for the variability in distributions and correlation between candidate variables. No automatic variable selection methods were employed. The analysis was performed in R statistical computing environment (R Core Team, 2025), using ‘survival’ (Therneau, 2024) and ‘tidycmprsk’ (Sjoberg & Fei, 2024) packages.

Ethics statement

The study protocol was approved by the Ethics Committee of the Public Health Center of the Ministry of Health of Ukraine (protocol #IRB2022–82, dated December 21, 2022). All interviewers were trained in ethical principles governing human subject research and effective respondent interaction. All participants signed informed consent for the BUP-XR treatment and for the study participation.

Results

Participant characteristics

Between February 2023 – October 2024, a total of 1173 patients were screened at fourteen OAT clinics, and of them 273 patients started on BUP-XR. Of these 273 patients, 181 who started BUP-XR before August 2023 at ten clinics were included in the present study. Table 1 shows the baseline participant characteristics. The median age was 41 years old (IQR: 36–46), and 15 (8%) were women. Prior to switching to BUP-XR, 98 (54%) patients were receiving low (4–10 mg) doses of BUP-SL, 136 (75%) received take-home dosing, and the median duration of OAT before switching to BUP-XR was 19 months (IQR: 5–39). The most common negative expectation – the possibility of experiencing withdrawal symptoms – was expressed by 58 (32%) participants. At least one negative expectation was reported by 66 (37%) participants. Injection drug use during the 6 months before baseline was reported by 30 (17%) and non-injection illicit drug use - by 27 (15%) participants.

Table 1.

Baseline characteristics of the study participants (N=181)

Characteristic	n (%)
Age (years old)
<35	28 (15%)
35–44	95 (52%)
≥45	58 (32%)
Sex
Male	166 (92%)
Female	15 (8%)
Family Status
Partnered	115 (64%)
Single	66 (36%)
Education
Complete high school or lower	108 (60%)
College or higher	73 (40%)
Employment
Unemployed	33 (18%)
Employed	148 (81%)
Ever homeless
No	172 (95%)
Yes	9 (5%)
BUP-SL duration before switching to BUP-XR
<3 months	25 (14%)
3–23 months	78 (43%)
≥24 months	78 (43%)
BUP-SL dose before switching to BUP-XR
4–10 mg	98 (54%)
12–16 mg	65 (36%)
>16 mg	18 (10%)
BUP-SL dispensing before switching to BUP-XR
On-site	45 (25%)
Take-home dosing	136 (75%)
HIV status
Negative	140 (77%)
Positive	41 (23%)
Injected illicit drugs in the past 6 months
No	151 (83%)
Yes	30 (17%)
Used non-injection illicit drugs in the past 6 months
No	154 (85%)
Yes	27 (15%)
Depression screening (PHQ-9)
Minimal (total score ≤ 4)	115 (64%)
Present (total score ≥ 5)	66 (36%)
Anxiety screening (GAD-7)
Minimal (total score ≤ 4)	128 (71%)
Present (total score ≥ 5)	53 (29%)
Alcohol use disorder screening (AUDIT-C)
Low risk	169 (94%)
High risk	11 (6%)
Any negative expectations about BUP-XRa
No	115 (64%)
Yes	66 (36%)
Initial dose of BUP-XR (CAM2038/Buvidal®)
64 mg	77 (43%)
96 mg	82 (45%)
128 mg	22 (12%)

^aAny negative expectations of switching to BUP-XR included a positive response to any of the following statements: I expect that I will have side effects; physical health will become worse; I will experience withdrawal before the next injection; or the medication will not suit me.

BUP-SL, sublingual buprenorphine; BUP-XR, extended-release buprenorphine; HIV, human immunodeficiency virus; PHQ-9, Patient Health Questionnaire 9; GAD-7, Generalized Anxiety Disorder – 7 item; AUDIT-C, Alcohol Use Disorders Identification Test – Consumption.

The initial dose of BUP-XR matched the pre-transition BUP-SL dose according to the clinical guidelines among 150 (83%) participants. The main source of mismatch were participants on low BUP-SL dose (4–10 mg), who were prescribed BUP-XR dose of 96 mg (n=24). Supplementary Table S1 provides details of the BUP-SL/BUP-XR dose correspondence.

BUP-XR retention, dropout and inter-dose intervals

BUP-XR dosing and retention trajectories for each study participant are visualized in Figure 1. Table 2 shows the summary of participant outcomes. The second injection was received by 80% (144/181), and 13 or more injections were received by 44% (80/181) of the participants. Among participants who dropped out, 53% (54/101) switched back to BUP-SL and 47% (47/101) discontinued OAT. Supplementary Table S2 shows detailed distributions of dropout reasons at 2^nd, 7^th and 13^th monthly dose checkpoints. Figure 2 shows the temporal dynamics of the cumulative incidence of BUP-XR discontinuation since the first injection (time zero) by reason during the study follow-up. Supplementary Figure S2 shows the distributions of the study participants by status at each monthly injection checkpoint.

Figure 1.

Treatment trajectories of the study participants (N=181)

BUP-XR, extended-release buprenorphine (CAM2038/Buvidal^®); BUP-SL, sublingual buprenorphine; OAT, opioid agonist therapy.

Table 2.

Summary of the study participant outcomes (N=181)

Outcome	n (%)
12-months retention on BUP-XR (N=181)
Received 2 injections or more	144 (80%)
Received 7 injections or more	98 (54%)
Received 13 injections or more	80 (44%)
Type of drop-out (N=101)
Switch to BUP-SL	54 (53%)
Discontinuation of OAT	47 (47%)
Reasons for switch to BUP-SL (N=54)
Personal preference	34 (63%)
Side effects	20 (37%)
Reasons for OAT discontinuation (N=47)
Personal decision	18 (38%)
Other reasons	29 (62%)
Any BUP-XR dose change during follow-up (N=144)
No	77 (53%)
Yes	67 (47%)
Any supplemental BUP-SL (N=181)
No	157 (87%)
Yes	24 (13%)
Common issues reported by participants who switched to BUP-SL (N=50)a
Sweating	33 (66%)
Anxiety	37 (74%)
Withdrawal symptoms	38 (76%)
Depression	34 (68%)
Sleeplessness	39 (78%)
Reduced contact with clinicians	33 (66%)
Common issues reported by participants who discontinued OAT (N=26)a
Sweating	3 (12%)
Anxiety	4 (15%)
Withdrawal symptoms	3 (12%)
Depression	5 (19%)
Sleeplessness	4 (15%)
Reduced contact with clinicians	4 (15%)
Proportion of patients who would recommend BUP-XR
Among patient who switched to BUP-SL (N=50)a	13 (26%)
Among patient who discontinued OAT (N=35)a	19 (54%)

^aThese statistics are based on exit interviews, which were completed by a subset of study participants who discontinued treatment; for some participants answers to some exit interview questions were missing.

BUP-SL, sublingual buprenorphine; BUP-XR, extended-release buprenorphine; OAT, opioid agonist therapy.

Figure 2.

Cumulative incidence of BUP-XR discontinuation by reason (N=181)

BUP-XR, extended-release buprenorphine; BUP-SL, sublingual buprenorphine; OAT, opioid agonist therapy

Transitions between BUP-XR doses among those who received at least 2 doses are summarized in Supplementary Table S3. Most patients who started on 64mg subsequently switched to a higher dose: 96mg (50%, 31/62) and 128mg (24%, 15/62).

The median BUP-XR inter-dose interval was 28 days (IQR: 25–29). Among 144 participants who received at least two BUP-XR doses, 92% of injections were given within the recommended interval of 21–35 days; 4% were given less than 21 days and 4% - more than 35 days after the previous injection. Supplementary Figure S3 shows the distribution of inter-injection intervals.

Twenty-four participants (13%) were prescribed supplemental BUP-SL, eight of them had two prescriptions, others had one. In most cases (21/32 episodes), supplemental BUP-SL doses of 2–6mg were prescribed during the first month of BUP-XR (Figure 1). Median duration of supplemental prescription was 8.5 days (IQR: 5–21 days). Half of the prescriptions (16/32) were issued in one clinic.

In the exit interviews, the most common issues reported by the patients transitioning to BUP-SL were sweating (66%), anxiety (74%), opioid withdrawal symptoms (76%), depression (68%), difficulty sleeping (78%), and reduced contact with clinicians (66%). Among patients who discontinued OAT, these negative experiences were less prevalent, with depression being the most frequent (19%) (Table 2). Notably, no adverse drug reactions, whether expected or unexpected, to BUP-XR were reported to the national pharmacovigilance system. The proportion of patients who said that they would recommend BUP-XR to another person was 26% among those who switched to BUP-SL, and 54% among those who discontinued OAT (Table 2).

Correlates of BUP-XR dropout

In the multivariable model (Table 3), transition to BUP-SL was associated with the baseline negative expectations about BUP-XR (aHR=1.9; 95% CI: 1.3–3.0). The initial high dose of BUP-XR (128mg) was associated with a lower hazard of transition to BUP-SL (the hazard ratio could not be calculated due to lack of BUP-SL transition events in the 128mg group; Supplementary Figure S4).

Table 3.

Multivariable Cox proportional hazards regression for the characteristics associated with 12-month BUP-XR dropout by the main reason (N =181).

Characteristic	Category	Transition to BUP-SL				Dropout of OAT
		Coefficient	p-value	aHR	95% CI	Coefficient	p-value	aHR	95% CI
Age	35–44	Ref				Ref
	<35	−0.41	0.070	0.7	(0.4; 1.0)	−0.08	0.879	0.9	(0.3; 2.6)
	≥45	−0.12	0.761	0.9	(0.4; 1.9)	−0.15	0.839	0.9	(0.2; 3.6)
Sex	Male	Ref				Ref
	Female	−0.93	0.260	0.4	(0.1; 2.0)	−1.31	0.025	0.3	(0.1; 0.9)
Family status	Partnered	Ref				Ref
	Single	0.37	0.203	1.4	(0.8; 2.5)	0.36	0.347	1.4	(0.7; 3.0)
Education	High school or lower	Ref				Ref
	College or higher	0.37	0.258	1.4	(0.8; 2.7)	−0.03	0.920	1.0	(0.6; 1.6)
Employment	Employed	Ref				Ref
	Unemployed	0.36	0.373	1.4	(0.7; 3.2)	0.33	0.343	1.4	(0.7; 2.7)
BUP-SL duration before switching to BUP-XR	≥24 months	Ref				Ref
	3–23 months	−0.19	0.470	0.8	(0.5; 1.4)	0.20	0.430	1.2	(0.7; 2.0)
	<3 months	0.09	0.762	1.1	(0.6;1.9)	−0.48	0.189	0.6	(0.3; 1.3)
BUP-SL dispensing before switching to BUP-XR	On-site	Ref				Ref
	Take-home dosing	0.10	0.871	1.1	(0.3; 3.6)	−1.01	0.014	0.4	(0.2; 0.8)
Injected illicit drugs in the past 6 months	No	Ref				Ref
	Yes	−0.76	0.177	0.5	(0.2; 1.4)	0.92	<0.001	2.5	(1.6; 3.8)
HIV status	Negative	Ref				Ref
	Positive	−0.18	0.524	0.8	(0.5; 1.5)	0.27	0.482	1.3	(0.6; 2.8)
Depression screening (PHQ-9)	Minimal (total score ≤ 4)	Ref				Ref
	Present (total score ≥ 5)	−0.29	0.445	0.8	(0.4; 1.6)	0.41	0.390	1.5	(0.6; 3.8)
Anxiety screening (GAD-7)	Minimal (total score ≤ 4)	Ref				Ref
	Present (total score ≥ 5)	0.74	0.159	2.1	(0.8; 5.8)	0.60	0.002	1.8	(1.2; 2.7)
Any negative expectations about BUP-XRa	No	Ref				Ref
	Yes	0.66	0.002	1.9	(1.3; 3.0)	−0.25	0.215	0.8	(0.5; 1.2)
Initial dose of BUP-XR (CAM2038/Buvidal®)	64mg	Ref				Ref
	96mg	−0.15	0.818	0.9	(0.2; 3.2)	0.63	0.261	1.9	(0.6; 5.6)
	128mg	−18.59	<0.001	<0.1	(<0.1)	0.94	0.414	2.6	(0.3; 24.3)
BUP-XR dose (time-dependent)	64mg	Ref				Ref
	96mg	0.47	0.521	1.6	(0.4; 6.6)	0.21	0.807	1.2	(0.2; 6.4)
	128mg	0.83	0.304	2.3	(0.5; 11.3)	−0.14	0.918	0.9	(0.1; 11.5)
BUP-XR dose increase from baseline (time-dependent)	No	Ref				Ref
	Yes	−0.50	0.437	0.6	(0.2; 2.1)	−0.47	0.630	0.6	(0.1; 4.2)
Supplemental BUP-SL (time-dependent)	No	Ref				Ref
	Yes	−0.94	0.387	0.4	(0.1; 3.3)	−18.69	<0.001	<0.1	(<0.1)

Bolded values represent associations that are significant at 0.05 level.

^aAny negative expectations of switching to BUP-XR included a positive response to any of the following statements: I expect that I will have side effects; physical health will become worse; I will experience withdrawal before the next injection; or the medication will not suit me.

BUP-SL, sublingual buprenorphine; BUP-XR, extended-release buprenorphine; OAT, opioid agonist therapy; HIV, human immunodeficiency virus; PHQ-9, Patient Health Questionnaire 9; GAD-7, Generalized Anxiety Disorder – 7 item

Higher hazard of BUP-XR discontinuation due to OAT dropout was associated with recent injecting drug use (aHR=2.5; 95% CI: 1.6–3.8) and higher level of anxiety (aHR=1.8; 95% CI: 1.2–2.7). The correlates of lower hazard of OAT discontinuation included female sex (aHR=0.3; 95% CI: 0.1–0.9), receiving BUP-SL for take-home dosing before starting BUP-XR (aHR=0.4; 95% CI: 0.2–0.8), and receiving supplemental BUP-SL (the hazard ratio could not be calculated due to lack of OAT dropout events in this group). We did not find significant associations between the time-dependent BUP-XR dose and the BUP-XR drop-out for either reason. Supplementary Table S4 shows the regression results for the combined dropout outcome.

Discussion

In this study, we assessed the initiation and retention on CAM2038/Buvidal^® – one of the two available BUP-XR medications – after switching from BUP-SL, and correlates of BUP-XR discontinuation due to transition to BUP-SL or complete dropout of OAT among patients with OUD in Ukraine in a prospective single-arm, multicenter, open-label trial. To the best of our knowledge, this is one of the first real-world studies of BUP-XR in a low/middle-income country. Most studies investigating BUP-XR efficacy, implementation, and retention were conducted in the US, Canada, Australia, and Western Europe (Frost et al., 2019; Iacono et al., 2024; Marsden et al., 2023; Martin et al., 2022; Raza et al., 2023) and few explicitly differentiated between the types of dropout (complete dropout vs transition to BUP-SL) (Ivasiy et al., 2025; Peckham et al., 2021; Stein et al., 2022). Several important findings emerged from our study.

Despite the initial interest and the substantial pool of potentially eligible patients — averaging approximately 2300 on BUP-SL in Ukraine during the study period, with 50–70 new patients starting monthly (Public Health Center of the Ministry of Health of Ukraine, 2023, 2024) — the BUP-XR enrollment was unexpectedly slow. The PHC undertook extensive efforts to facilitate recruitment, including active engagement of the OAT patients’ community during all planning and implementation stages, regular nationwide webinars, targeted education for providers, and online meetings with BUP-XR patients from Australia. Nevertheless, only 273 patients agreed to switch to BUP-XR over 21 months, utilizing less than half of the available 500 annual courses (when accounting for discontinuations). This suggests that most patients with OUD in Ukraine, similar to other countries, prefer short-acting oral formulations (Epstein et al., 2021; Muthulingam et al., 2023; Saunders et al., 2020, 2024). These preferences appeared to outweigh the potential advantages of extended-release formulations, such as a greater sense of stability in the volatile wartime context with the risk of medication supply disruption (Morozova et al., 2023).

In our study, retention on BUP-XR was 80% after the first, 54% after the 6th, and 44% after the 12th injection. Other real-world studies reported varying 12-month retention levels, ranging from 30% in a Canadian population-based cohort (Iacono et al., 2024) to 75% in a multi-site study from 7 countries (Frost et al., 2019) and a community-based Australian sample (Farrell et al., 2022). The latter study followed participants through week 96 and reported retention of 47% (Farrell et al., 2024). Of note, some of these studies investigated another BUP-XR medication – RBP-6000/Sublocade^® – which has a standard dosing schedule involving two monthly doses of 300mg followed by 100mg (Greenwald et al., 2023). In contrast, CAM2038/Buvidal^®, which has four monthly formulations (64 mg, 96 mg, 128 mg, and 160 mg), is initially prescribed based on preceding BUP-SL dose and does not include a recommendation for dose tapering. A study directly comparing RBP-6000/Sublocade^® and CAM2038/Buvidal^® did not find a significant difference in retention (Daglish et al., 2024), suggesting that the retention level observed in our study is likely representative for the available BUP-XR medications in the context of Ukraine.

Importantly, we found similar cumulative probability of 12-months BUP-XR discontinuation for each of the two major reasons – complete OAT dropout and transition to BUP-SL.

Direct comparison of the OAT dropout rate with other studies is challenging due to differences in definitions and study designs. In our study, person-time started at transition to BUP-XR, on average 19 months after buprenorphine treatment initiation, whereas most studies measure person-time from the beginning of a treatment episode. Since the risk of OAT dropout is considerably higher during the first several months of treatment, the retention rate of 44% after 12 months in a stable patient cohort was lower than expected given the previous buprenorphine studies in Ukraine, which reported up to 65–85% retention in the first year (Dumchev et al., 2017; Farnum et al., 2021; Schaub et al., 2009). However, these proportions cannot be compared directly because we did not follow patients after the switch to BUP-SL. Given that the participants in this study represented a highly motivated group of patients who chose to transition to BUP-XR, the level of BUP-XR retention observed in our study likely represents the best-case scenario that can be achieved in Ukraine at this time. Nevertheless, compared to recent reports from high-income countries, where 6-month OAT retention ranged between 22–50% (Bharat et al., 2021; Hochheimer & Unick, 2022; K. Y. Xu et al., 2024), retention on buprenorphine in Ukraine remains high.

Participants who discontinued OAT constituted a heterogeneous group. Only 38% of them ended treatment voluntarily, whereas the remaining participants discontinued due to death, incarceration, relocation, rules violations, or loss to follow-up (Supplementary Table S2). Although this represents the conventional approach to measuring treatment retention, it is important to note that some individuals may have continued OAT at other facilities or could have remained in treatment had OAT been widely available within Ukrainian jails and prisons. Participants who discontinued voluntarily may have returned to illicit opioid use, or alternatively, may have used BUP-XR as part of a planned tapering strategy within a broader recovery pathway (Boyett et al., 2023; Craft et al., 2022; Hayes et al., 2025).

The 12-month cumulative probability of transitioning to BUP-SL observed in our study (30%) was remarkably similar to 27% estimated in a large US cohort (Ivasiy et al., 2025). The majority of switching patients did so due to personal preference (63%), while others indicated side effects as the reason (37%). Previous studies reported similar findings where patients indicated personal circumstances, practical and relational concerns, side effects, or achievement of individual goals as reasons for patient-initiated discontinuation (Clay et al., 2023; Parkin et al., 2024). Our study did not include a systematic clinical assessment of side effects, but patient-reported data suggest that non-specific opioid withdrawal symptoms, as well as mental health issues (anxiety, depression), reported by most patients switching to BUP-SL could be perceived as side effects of BUP-XR. Another reason for BUP-SL preference indicated by more than a half of transitioning participants was related to the need for more frequent contact with clinical staff, which gives patients a sense of greater accountability and structure to their treatment, as was observed in other OAT studies (Yarborough et al., 2016). This finding emphasizes the importance of provider adaptation to novel treatments and practices, including attention to interpersonal relations, continuity of care and psychosocial support (Lancaster et al., 2024; Neale & Strang, 2024). At the same time, patients, who desire greater autonomy, may prefer having the medication in their possession and controlling their daily intake. One of the key perceived benefits of BUP-XR, the reduced time for transportation, was apparently not as tangible in the Ukrainian OAT context, where over 90% of patients are receiving their medications for up to 10 days of take-home dosing.

For the purposes of this study, treatment retention was defined as receipt of all scheduled BUP-XR injections. Although flexibility was permitted regarding deviations from the recommended inter-injection intervals, participants who discontinued BUP-XR for any reason were not followed beyond the point of discontinuation. Nevertheless, reinitiation of BUP-XR following a period of discontinuation was permissible, and anecdotal evidence indicated that some participants resumed BUP-XR treatment after interruption. These observations highlight the need for a more nuanced conceptualization and measurement of treatment engagement—one that captures its dynamic nature as a process characterized by periods of connection and disconnection, rather than a static, binary outcome (Gendera et al., 2025; Parkin et al., 2023).

Despite the BUP-SL being a more attractive option for the majority of buprenorphine patients, BUP-XR may be optimal for a small proportion of patients who could benefit from this formulation, and in more specific contexts. For example, among incarcerated individuals, where the availability of OAT is limited, monthly injections are an easier (Dunlop et al., 2022) and satisfactory (White et al., 2025) way to provide treatment. In addition to prison settings, BUP-XR could be beneficial for hospitalized patients with OUD (Cuperfain et al., 2024) and before discharge from hospitalizations for injection-related conditions (Seval et al., 2023). Finally, BUP-XR may also be a promising option for pregnant people living with OUD. To the best of our knowledge, no clinical trials were conducted in this population to date, but early case reports offer promising results (Ramage et al., 2025), while current evidence suggests that buprenorphine is safer for a fetus than methadone (Suarez et al., 2022).

Our analysis highlighted that the predictors of the two main reasons of BUP-XR discontinuation were different. The hazard of switching to BUP-SL was associated with baseline negative expectations about BUP-XR and with a lower initial BUP-XR dose. In this study, we did not investigate the origins and formation of these expectations. Future studies are needed to provide insight into the dynamics and effects of negative beliefs about BUP-XR on patient experiences. The association between the highest initial dose (128mg) and the lower hazard of transition to BUP-SL may not be causal and requires further investigation. In our study, we observed no BUP-SL transitions in this group, but the sample size was small (n=22). For OAT dropout, the correlates were consistent with other studies and included injecting drug use (Farrell et al., 2022), female sex (Farnum et al., 2021), receiving BUP-SL for take-home dosing – an indirect marker of patient stability (Ivasiy et al., 2024), and receiving supplemental BUP-SL (Ivasiy et al., 2025). About 30% of patients in our sample had elevated anxiety level (however lower than the general population in Ukraine during the full-scale war (Levin et al., 2023; W. Xu et al., 2023)), which predicted OAT dropout.

Importantly, only 13% (n=24) of our study participants were prescribed supplemental BUP-SL, which is in stark contrast with more than 50% in the US (Ivasiy et al., 2025) and lower than 17% in one Australian study (Farrell et al., 2022). The high self-reported prevalence of withdrawal symptoms in the exit interviews, and the fact that half of the supplemental prescriptions were issued by one physician indicates that other providers might have taken a more conservative approach, potentially further reducing retention on BUP-XR. Supplemental BUP-SL was prescribed by physicians on a case-by-case basis with no specific recommendations available in the clinical and manufacturer guidelines or the study protocol. We observed no OAT dropouts among patients receiving supplemental BUP-SL but no effect of supplemental BUP-SL on transitions to BUP-SL.

A unique finding of our study was that the majority of patients starting on lower doses of CAM2038/Buvidal^® switched to a higher dose during the observation period (74% of those starting on 64mg). This is in contrasts with RBP-6000/Sublocade^® studies that generally followed the recommended decrease from 300mg to 100mg (Farrell et al., 2022; Iacono et al., 2024). This finding warrants further investigation and may suggest that the BUP-SL – CAM2038/Buvidal^® dose matching guidelines may need a revision. Use of supplemental BUP-SL, especially during the first several months after switching to BUP-XR, may help in managing withdrawal symptoms (Guillery et al., 2023). Importantly, the time-varying dose of BUP-XR and the fact of dose increase were not associated with OAT dropout in the multivariable models. The lack of association, however, cannot be interpreted as lack of causality, because dose increases were not random and were likely prescribed in response to suboptimal treatment experiences. This finding warrants further investigation, as extensive evidence suggests a strong positive effect of OAT medication dose (buprenorphine and methadone) on retention (Dumchev et al., 2017; Farnum et al., 2021; Ivasiy et al., 2024; Stringfellow et al., 2025). In our study, none of the patients who started on the highest dose of BUP-XR (128mg) transitioned to BUP-SL.

This study had several limitations. First, because of unanticipated recruitment challenges, the study sample size and resulting statistical power were limited. This could prevent us from detecting important associations, particularly in subgroups with no type-specific dropout events, such as patients starting on 128mg of BUP-XR (for transition to BUP-SL) and those receiving supplemental BUP-SL prescriptions (for OAT dropout). Second, our participants, on average, had a substantial treatment history with BUP-SL, which affects the generalizability of findings. Although we did not find an association between the OAT duration before BUP-XR and either reason for discontinuation, it could still play a role in determining medication preferences. Additionally, the lengthy previous treatment experience in our sample precludes a direct comparison of OAT dropout rates with OAT-naïve cohorts.

Conclusion

Our study contributes to the expanding knowledge of real-world implementation of BUP-XR by providing one of the first results from a middle-income country. Despite initial interest, BUP-XR uptake was low, and may be context dependent due to the broad availability of take-home dosing. The retention rates on BUP-XR in Ukraine were comparable to other BUP-XR studies but lower than was expected in the volatile wartime context among patients who volunteered to transition to this formulation. Two key reasons – transition to BUP-SL and OAT dropout – contributed equally to BUP-XR discontinuation. Further studies should investigate patient expectations and preferences in different contexts to inform BUP-XR scale-up in real-world settings and specific populations that may benefit the most. Importantly, our findings suggest that revising clinical guidelines on initial BUP-XR (in particular, CAM2038/Buvidal^®) dosing, dose adjustment, and supplemental BUP-SL prescriptions may reduce the occurrence of opioid withdrawal and improve retention.

Highlights.

• Despite initial interest, BUP-XR enrollment in Ukraine was low.

• 12-months BUP-XR retention was 44%; 30% switched to BUP-SL; 26% discontinued OAT.

• Most patients starting on low BUP-XR doses increased their dose during the study.

• 13% of participants received supplemental BUP-SL, mostly after the first injection.

• Correlates of the two main reasons of BUP-XR discontinuation were different.