Introduction

Sublingual buprenorphine is a standard treatment for opioid use disorder (OUD) in peripartum persons¹ but has disadvantages, including risk of diversion and misuse,² poor adherence,³ and daily peak-trough effects that may inadequately mitigate opioid-related cravings and withdrawal,⁴ especially during pregnancy.⁵ Extended-release buprenorphine helps address these disadvantages and has demonstrated superior illicit opioid abstinence in nonpregnant adults.⁶ A potential concern of extended-release buprenorphine use in pregnancy is the higher buprenorphine exposure (ie, area under the curve) for this formulation, relative to comparable doses of sublingual buprenorphine,⁷ translating into greater fetal buprenorphine exposure. While the published literature includes several case and cohort studies of extended-release buprenorphine in pregnant persons,^8-13 there have been no completed randomized clinical trials.

Methods

Trial Design

This intent-to-treat, 2-group, open-label, noninferiority randomized clinical trial evaluated the effectiveness and safety of extended-release buprenorphine relative to sublingual buprenorphine. The trial was pragmatic in that it was open label; was conducted in geographically varied clinical settings with existing collaborative care models; allowed site-preferred sublingual buprenorphine formulations and site-standard induction and dosing practices; relied heavily on medical record data rather than extensive research-only assessments; and allowed variation in hospital NOWS practices. The trial protocol is in Supplement 1. The protocol followed the Consolidated Standards of Reporting Trials (CONSORT) guideline and was approved by the Food and Drug Administration, the data and safety monitoring board of the National Institute on Drug Abuse Treatment Clinical Trials Network, and by a single institutional review board (IRB) of record, the University of Cincinnati IRB. The trial was conducted at 13 US sites between July 2, 2020, and October 30, 2024; 1 site, which was closed for poor recruitment, never randomized participants. Study sites had close collaboration between prenatal care and addiction treatment clinicians.¹⁴ Most were specialty obstetrics and gynecology programs providing OUD services affiliated with an academic institution.¹⁴ Participants received study medication and attended weekly visits through 12 months post partum. Longer research visits occurred 2 weeks after randomization and then monthly until delivery and at months 1, 3, 6, 9, and 12 during the postpartum phase. Visits could occur onsite or offsite, and some elements (eg, self-report assessments) could be completed via telehealth. Participants provided written informed consent and were compensated for participating.

Participants

Participants were aged 18 to 41 years, with a singleton pregnancy of estimated gestational age (EGA) between 6 and 30 weeks at randomization, meeting Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria for moderate or severe OUD and eligible for buprenorphine treatment and/or already prescribed buprenorphine. They were planning for delivery at a hospital that, based on a completed Better Outcomes Through Research for Newborns survey,¹⁵ met 3 requirements designed to reduce variability in procedures that can impact neonatal opioid withdrawal syndrome (NOWS) severity and outcome measures^16,17: (1) had a written NOWS management protocol; (2) offered rooming-in for infants being observed for NOWS; and (3) did not discharge infants receiving opioid treatment. Exclusion criteria included physiological dependence on alcohol or sedatives requiring medically managed withdrawal; having a psychiatric or medical condition that may make study participation difficult or unsafe; current or pending criminal justice involvement that could interfere with study participation; receiving methadone or naltrexone treatment; or current or planned enrollment in treatment beyond level 3.3 of the American Society of Addiction Medicine Criteria.¹⁸

Per National Institutes of Health requirements, race and ethnicity were assessed. Participants were asked whether they considered themselves to be Hispanic or Latino (yes or no) and what race they considered themselves to be. The participant’s response was recorded by a research staff member with race categorized into the following: American Indian or Alaska Native, Asian, Black or African American, Native Hawaiian or Pacific Islander, White, or some other race; the latter category reflects respect for participant self-identification and autonomy.

Intervention Groups and Medication Dosing

Participants were randomized 1:1 to extended-release or sublingual buprenorphine, balancing on site, EGA at randomization (6-18 weeks vs 19-30 weeks), and whether they were already receiving sublingual buprenorphine (yes or no). Study-purchased generic sublingual buprenorphine was provided as tablets and/or buprenorphine/naloxone film, based on site and participant preference. Braeburn donated extended-release buprenorphine (CAM2038/Brixadi). The target dose was 16 mg/d of sublingual buprenorphine or the equivalent extended-release buprenorphine weekly dose (24 mg), consistent with recommended dosing during pregnancy,¹⁹ but dosing was determined by the clinician. Pregnant and breastfeeding participants in the extended-release buprenorphine group received weekly injections due to concerns about excipient (ie, N-methyl-2-pyrrolidone) in monthly formulations increasing risk for adverse fetal developmental effects²⁰ ; if not breastfeeding, monthly injections were offered. Between December 20, 2021, and May 10, 2022, a supply disruption in extended-release buprenorphine paused site enrollment and led 2 pregnant and 14 postpartum participants to stop study medication; all were offered sublingual buprenorphine. eTable 1 in Supplement 2 details dosing before, during, and after the supply disruption.

Assessment

The primary outcome was illicit opioid abstinence during pregnancy, defined as the proportion of urine drug screens negative for fentanyl, morphine, codeine, ethylmorphine, heroin, hydrocodone, hydromorphone, methadone, and oxycodone. Consistent with a CAM2038 phase 3 trial,⁶ results missing for any reason were imputed as positive for illicit opioids. Urine samples were collected weekly throughout the study. Postpartum illicit opioid abstinence was the maternal key secondary outcome. Infant key secondary outcomes were opioid treatment for NOWS (yes or no) and if yes, the number of days of opioid treatment, both extracted from medical records. Maternal secondary outcomes were collected during pregnancy and post partum, except Kotelchuck’s Adequacy of Prenatal Care Utilization index,²¹ which was based on medical records and combines prenatal care start timing with the ratio of observed-to-expected visits. Buprenorphine adherence was calculated as days adherent divided by active treatment study days (regardless of study or medication discontinuation); for participants in the extended-release buprenorphine group, days affected by the supply disruptions were excluded. Adherence was assessed weekly and included both study-provided and nonstudy buprenorphine (ie, clinically provided outside of the study). Extended-release buprenorphine adherence was defined as 7 days of adherence per weekly injection and 28 days per monthly injection. Sublingual buprenorphine adherence was defined as receipt of study dispensation or prescription and self-reported use, reduced by 7 days for each negative buprenorphine and norbuprenorphine urine result. Drug and alcohol abstinence was defined as the proportion of urine drug screens negative for illicit opioids, cocaine, methamphetamine, amphetamine, cannabis, benzodiazepines, barbiturates, phencyclidine, methylenedioxymethamphetamine, and ethyl glucuronide. According to the package inserts for the urine drug screen tests used in this study, comparisons with confirmatory gas chromatography–mass spectrometry (MS) or liquid chromatography–tandem MS testing showed a 0% false-positive rate. Although false positives can occur in clinical settings, this risk is minimal and did not justify the added cost of confirmatory testing for this trial. Outcomes collected during research visits were the (1) Opioid Craving Scale (each of 3 items scored from 0-10 and averaged, with higher scores indicating more craving)²² and (2) Short Opiate Withdrawal Scale–Gossop (10 items each scored from 0-3 to give total scores ranging from 0-30, with higher scores indicating more withdrawal).²³

Of the 6 secondary infant outcomes, 3 NOWS-related outcomes and 2 discharge status outcomes came from medical records. NOWS outcomes included length of hospital stay (days old at discharge), peak NOWS score on the Modified Finnegan²¹ (the only scoring method with sufficient data for analysis), and for morphine-treated infants, total milligrams’ morphine received. Hospital discharge outcomes included maternal custody and whether a case was open with child protective services (CPS). The sixth infant outcome indicated any potential delay at 12 months in communication, gross motor, fine motor, problem solving, or personal-social development assessed by the 12-month version of the Ages and Stages Questionnaire, Third Edition, which scores each item as yes (10 points), sometimes (5 points), or not yet (0 points), with total scores compared with cutoffs to identify delays.²⁴

Maternal safety measures assessed weekly included reported adverse events (AEs) and injection site examinations for extended-release buprenorphine participants, with the latter not captured as AEs. AEs were classified using Medical Dictionary for Regulatory Activities (MedDRA) version 27.1. Maternal safety measures at research visits included self-reported opioid overdoses—defined as an overdose causing unresponsiveness or inadequate breathing, resulting in naloxone rescue and/or emergency medical care or hospitalization²⁵—and the Hospital Anxiety and Depression Scale, which has 7 items each for depression and anxiety scored from 0 to 3, with total scores ranging 0 to 21, and higher scores indicating more severe anxiety and depression symptoms.²⁶ Other maternal safety outcomes—including primary cesarean delivery; abnormal fetal presentation; labor complications; and use of pain medication during labor and delivery, postpartum hospitalization, or prescribed at discharge—were obtained from medical records.

Infant safety measures assessed weekly post partum included participant-reported serious AEs and infant sedation signs, such as not waking for feeding, difficulty breathing (beyond a stuffy nose), or being limp when held. Additional medical record outcomes were live birth, EGA at delivery, birth head circumference, weight, length, Apgar scores at 1 and 5 minutes, presence of abnormal conditions (eg, respiratory distress, feeding issues), related interventions (eg, resuscitation, assisted ventilation), preterm status (<37 weeks), and discharged alive.

Statistical Analysis

The statistical analysis plan (Supplement 1) used the intent-to-treat sample and was finalized before analyses were undertaken. The Bonferroni method set the family-wise α level. Primary outcome noninferiority and superiority tests used a 1-tailed α = .025. The 3 key secondary analyses used α = .05/3 = .0167, and the 15 secondary analyses used α = .05/15 = .0033. Safety analyses applied α = .05 to detect safety differences. SAS version 9.4 (SAS Institute) was used to conduct analyses.

For primary and maternal key secondary outcomes, data were truncated at the supply disruption for affected extended-release buprenorphine participants, analyzed via mixed-model regression with treatment cohort as the main effect and random effects for site and baseline EGA of less than 19 weeks. If noninferiority was demonstrated, superiority testing followed. Infant key secondary analyses excluded infants of mothers who experienced supply disruptions for extended-release buprenorphine and used regressions adjusting for treatment cohort and fixed effects (site and baseline EGA if significant).

Secondary and safety outcomes included data during the supply disruption, with regressions adjusting for treatment cohort, fixed effects (site and baseline EGA if significant), and baseline outcome values. Sensitivity analyses incorporated postdisruption data and assessed covariate impact per the statistical analysis plan. Fisher exact tests compared AEs between groups.

The target sample size was 300 based on a primary outcome noninferiority margin of 0.11, 65% sublingual buprenorphine illicit opioid abstinence, and a 0.11 variance in both groups.²⁰ During early implementation, the noninferiority margin was deemed too conservative and the power calculations were repeated with a margin of 0.15²⁷; the sample size was reduced to 200. Due to slow enrollment during the COVID-19 pandemic and paused enrollment during the extended-release buprenorphine supply disruption, sample size was reestimated in January 2022 and again in March 2023. The first reestimation used the variance from the first 64 participants, and the second used the abstinence rate and variance from the 52 completed sublingual buprenorphine pregnancies; type I error rate was not impacted. Both reestimations demonstrated at least 84% power to detect a noninferiority margin of 0.15 with a sample size of 126.

Results

Participants

One hundred forty participants were randomized to extended-release buprenorphine (n = 69) or sublingual buprenorphine (n = 71) (Figure). Pregnancy and postpartum phases were completed by 98% and 81% of participants (137 and 114 participants), respectively, with no group differences. Table 1 provides participant demographic and clinical characteristics. The mean (SD) age was 31.2 (4.6) years; there were 10 Black participants (7.1%), 10 Hispanic participants (7.1%), 116 (82.9%) White participants, and 14 participants (10.0%) who belonged to additional groups; and 47 (33.6%) were employed. Mean (SD) baseline EGA was 21.1 (6.5) weeks. All but 2 participants were receiving sublingual buprenorphine at randomization.

Figure. Flowchart of Study Enrollment and Retention

Table 1. Baseline Characteristics of Randomized Participants by Study Treatment Group

Medication Exposure

During pregnancy, for the dose closest to delivery or at medication disruption, 41 participants in the extended-release buprenorphine group (59%) were receiving 32 mg, the highest dose, which is approximately equivalent to 18 to 24 mg of sublingual buprenorphine. For the sublingual buprenorphine group, 49 (69%) received buprenorphine-only tablets and 61 (86%) dosed multiple times per day; the mean (SD) daily dose was 18.8 (6.7) mg. Medication was discontinued by 8 participants in the extended-release buprenorphine group (12%) and 6 in the sublingual buprenorphine group (9%) during pregnancy. During the postpartum phase, for the last dose or the dose at medication disruption, 34 participants in the extended-release buprenorphine group (58%) were receiving the monthly injection. For those receiving weekly injections, the highest dose (32 mg) was the most frequent, accounting for 44% of the sample (11 of 25 participants). For those receiving monthly injections, the highest dose (128 mg) was the most frequent, accounting for 65% of the sample (22 of 34 participants). For the sublingual buprenorphine group, 28 of 65 (43%) received buprenorphine-only tablets and 54 (83%) dosed multiple times per day; the mean (SD) daily dose was 18.2 (6.8) mg. Medication was discontinued by 25 extended-release buprenorphine (37%) and 23 sublingual buprenorphine (33%) participants post partum. eTable 2 in Supplement 2 provides additional dosing information.

Effectiveness Outcomes

Both the maternal primary and key secondary outcomes met prespecified criteria for noninferiority (Table 2). Superiority analyses demonstrated superiority of extended-release buprenorphine for illicit opioid abstinence during pregnancy (primary outcome: mean [SE] negative urine samples, 82.5% [4.2] vs 72.6% [4.2]; mean difference, 9.84 [95% CI, 1.72 to 17.95] percentage points; P = .009) but not post partum (key secondary outcome: mean [SE] negative urine samples, 60.2% [4.2] vs 59.5% [4.1]; mean difference, 0.65 [98% CI, −12.72 to 14.02] percentage points; P = .45). Significant group differences were not found for any other maternal effectiveness outcomes (Table 2). Table 3 shows infant effectiveness results. There were no significant group differences at the preestablished α levels for any outcome, including the key secondary outcomes of need for opioid treatment for NOWS (mean [SE] proportion treated with opioids: 30.2% [5.8] vs 26.5% [5.4]; relative risk, 1.14 [98% CI, 0.54 to 1.99]; P = .64) and mean (SE) days of opioid treatment (10.9 [2.2] vs 14.8 [3.0] days; relative risk, 0.73 [98% CI, 0.36 to 1.51]; P = .28). All sensitivity analyses produced comparable results.

Table 2. Maternal Effectiveness and Significant Safety Outcomes as a Function of Study Treatment Group

Table 3. Infant Effectiveness and Significant Safety Outcomes as a Function of Study Treatment

Safety Outcomes

Serious and nonserious event summaries (Table 4) showed that fewer participants in the extended-release buprenorphine group had serious AEs during pregnancy (6 [8.7%] vs 19 [26.8%]; P = .007) and post partum (4 [6.0%] vs 13 [18.6%]; P = .04). Only 1 serious medication-related event occurred (sublingual buprenorphine group; hospitalization after return to opioid use). All 135 infants with a medical record safety outcome had a live birth and were discharged alive. Infant serious events mainly involved hospitalizations, occurring for a variety of reasons (eg, pneumonia, poor weight gain, and others) (eTable 3 in Supplement 2) and congenital anomalies (eTable 4 in Supplement 2). With the exception of 1 hospitalization (infant in the sublingual buprenorphine group for NOWS), no hospitalization or congenital anomaly was related to study drug. One infant death (sublingual buprenorphine group, unrelated to medication) occurred during a cosleeping event. Maternal nonserious adverse events did not differ between groups, but during pregnancy, these were more frequently deemed medication-related in the extended-release than sublingual buprenorphine group (18 [26.1%] vs 5 [7.0%]; P = .003). These medication-related AEs were rated as mild or moderate, with more gastrointestinal issues in the extended-release buprenorphine group (eTable 5 in Supplement 2). eTable 6 in Supplement 2 provides AEs by MedDRA System Organ Class. Injection site reactions were generally mild (eTables 7 and 8 in Supplement 2). Other maternal safety results in Table 2 show fewer extended-release buprenorphine participants received any pain medication (mean [SE], 85.7% [4.4] vs 98.5% [1.5]; P = .03) and opioids specifically (30.2% [5.8] vs 50.0% [6.2]; P = .02) during postdelivery hospitalization. Extended-release buprenorphine participants had significantly lower postpartum anxiety scores. All other maternal safety analyses were not significant (eTable 9 in Supplement 2). Nonfatal overdoses occurred in 1 participant in the sublingual buprenorphine group during pregnancy, and 3 participants in the extended-release buprenorphine group post partum; data were insufficient for statistical analysis. The participant in the sublingual buprenorphine group reported inconsistent medication use, and all participants in the extended-release buprenorphine group stopped the medication for several months before the overdoses. Other infant safety results (Table 3) significantly differed only in mean (SE) infant birth head circumference, which was larger among infants exposed to extended-release buprenorphine (34.0 [0.2] vs 33.4 [0.2] cm; P = .049); the non–statistically significant results are provided in eTable 10 in Supplement 2.

Table 4. Serious and Nonserious Adverse Events Occurring During the Study^a

Discussion

To our knowledge, this is the first randomized clinical trial to test extended-release buprenorphine for OUD in pregnancy. Participants receiving weekly extended-release buprenorphine had a significantly higher illicit opioid abstinence rate during pregnancy compared with those receiving sublingual buprenorphine. Other maternal and infant outcomes were generally similar between groups, save for fewer serious AEs during pregnancy and post partum, less maternal opioid pain medication receipt during postdelivery hospitalization, lower postpartum anxiety scores, and higher medication-related AEs during pregnancy in the extended-release buprenorphine group compared with sublingual group. Larger birth head circumference in extended-release buprenorphine–exposed neonates, relative to those exposed to sublingual buprenorphine, was the only statistically significant infant safety finding.

Illicit opioid abstinence was significantly greater among those receiving extended-release buprenorphine (82.5%) than sublingual buprenorphine (72.6%) during pregnancy, which was higher than the 65% reported in prior research.²⁸ Opioid craving and withdrawal were minimal, and buprenorphine adherence was high (approximately 85%) in both groups during pregnancy. The superior illicit opioid abstinence observed in the extended-release buprenorphine group potentially is related to its pharmacokinetic advantages (eg, no daily peak-trough).

Only 28% of infants received opioid treatment for NOWS, lower than the 39% to 48% reported in the literature.^29-31 This may be due to the majority of participants having adequate or better prenatal care (Kotelchuck index²¹), and already being established on buprenorphine at the time of randomization. Some of the hospitals used the Eat, Sleep, Console method for scoring NOWS, which has been shown to result in a lower proportion of infants being started on opioids.³² However, there was not a significant difference in the proportion receiving opioid treatment for the 30 infants scored with this system vs the 105 infants scored with another or unknown system (23.3% vs 29.1%; P = .65 [data not shown]). Postpartum, illicit opioid abstinence rates dropped and were similar in both groups (60.2% vs 59.5%). Postpartum individuals face additional challenges, including the demands of infant care for those with infant custody (90% of the sample)¹ and, for those with an open CPS case (36% of the sample), potential concern about continuing treatment³³; the high rate of open CPS cases is consistent with the finding that opioid-medication treatment is associated with an increased risk of CPS referral often due to state laws requiring reporting of opioid-medication exposure.³⁴ The combination of high rates of custody and open CPS cases may account for the observed increase in medication discontinuation postpartum (35%) relative to during pregnancy (10%). While not significant, postpartum buprenorphine adherence was higher in the sublingual buprenorphine (75.5%) than extended-release buprenorphine (69.3%) group. Breastfeeding participants who received extended-release buprenorphine continued weekly injections (81%), whereas sublingual buprenorphine participants received buprenorphine covering a longer period, which may have been more convenient. Recent animal research suggests that N-methyl-2-pyrrolidone, an excipient in monthly extended-release buprenorphine, does not have adverse fetal-infant effects⁸ and Food and Drug Administration labeling was updated to remove the previously stated potential effect; thus, the monthly formulation might be considered for use when breastfeeding.³⁵

Importantly, participants in the extended-release group had fewer serious AEs during pregnancy and post partum. However, they had more nonserious AEs during pregnancy rated as medication related. Notably, these were mostly mild, and none were severe. Research has found that attributing AEs to study medication can be unreliable, particularly for gastrointestinal AEs (ie, constipation, nausea, and vomiting),³⁶ and AEs are overestimated in open-label trials.³⁷ The absence of treatment group differences in AE rates (ie, regardless of medication relatedness) in the present trial, along with gastrointestinal AEs attributed to extended-release buprenorphine (n = 14) at double the rate for sublingual buprenorphine (n = 6) suggests a potential bias in attributing AEs to the more novel medication. Fewer participants in the extended-release group received opioid pain medication after delivery, and they had modestly lower postpartum anxiety scores. Sublingual buprenorphine has been shown to be efficacious for pain³⁸ and to have anxiolytic effects.³⁹ These effects might be more pronounced for extended-release buprenorphine, which provides more stable buprenorphine levels and higher exposure. Extended-release injection site reactions were primarily mild and moderate, consistent with past research.⁴⁰ Overdose events were rare in both groups and associated with no longer taking the medication or inconsistent medication use.

The only significant infant safety difference between groups was a larger head circumference at birth in the extended-release buprenorphine group, although both groups’ averages were within normal range.⁴¹ This finding’s clinical significance is unclear, but it is reassuring given the higher exposure from extended-release buprenorphine.⁷ The lack of adverse safety outcomes in extended-release buprenorphine–exposed infants, relative to sublingual buprenorphine–exposed infants, is consistent with research finding that higher doses of buprenorphine during pregnancy are not associated with adverse infant outcomes.⁴² The observed lack of significant treatment group differences in infant sedation is consistent with a recent study of extended-release buprenorphine, which found that, similar to the results from studies of sublingual buprenorphine, buprenorphine plasma concentrations in breastfed infants were below the lower limit of quantification.³⁵ Many of the other maternal and infant safety outcomes are consistent with normal ranges and rates in the general US population (ie, Apgar scores,⁴³ rates of primary cesarean delivery,⁴⁴ and medical complications⁴⁵) or with findings from buprenorphine-exposed neonates (ie, length and weight,⁴⁶ rates of preterm birth,³⁰ abnormal conditions,⁴⁷ neonatal intensive care unit admissions³⁰). The rate of approximately 16% of infants scoring below the cutoff for a developmental area is consistent with the 12% to 16% prevalence of developmental disabilities in US children.²⁴

Strengths and Limitations

This trial’s strengths include a randomized design, 13 geographically diverse sites, comprehensive effectiveness and safety assessments, a sample with polysubstance use, and high medication adherence and study completion rates.

It also has limitations. Some outcomes may have been underpowered despite adequate power for the primary measure; outcomes relying on medical records had more missing data and potential quality issues, including likely variation in Modified Finnegan scoring across hospitals. The sample was clinically stable relative to many pregnant persons with OUD, as most participants were receiving sublingual buprenorphine treatment prior to randomization; thus, it is not clear whether the findings will generalize to a more unstable OUD population or to patients being initiated on buprenorphine. However, the sample did have high rates of current cannabis and stimulant use disorders and nicotine use, which is common among patients receiving buprenorphine treatment. Finally, the sample was largely White and non-Hispanic, and it is unclear whether the findings will generalize to a more ethnically and racially diverse patient population.⁴⁸

Conclusions

In this trial, pregnant adults with OUD who were randomized to receive extended-release buprenorphine had higher rates of illicit opioid use during pregnancy than those randomized to receive sublingual buprenorphine. However, there was not between-group difference in the postpartum period. These results support the use of weekly extended-release buprenorphine in pregnant individuals with OUD.

Introduction

Sublingual buprenorphine is a common treatment for opioid use disorder (OUD) in pregnant and postpartum individuals. However, it has drawbacks such as the potential for misuse, difficulty with consistent use, and daily fluctuations in medication levels that may not fully prevent cravings or withdrawal symptoms, especially during pregnancy. Extended-release buprenorphine is designed to overcome these issues and has shown better results in preventing illicit opioid use in non-pregnant adults. A concern with using extended-release buprenorphine during pregnancy is that it leads to higher overall buprenorphine levels in the body, which means the fetus would also be exposed to more of the medication. While some studies have looked at extended-release buprenorphine in pregnant individuals, a full randomized clinical trial has not been completed until now.

Methods

This study was a non-inferiority randomized clinical trial comparing extended-release buprenorphine to sublingual buprenorphine for effectiveness and safety. It was designed to be practical, involving diverse clinical settings and existing care models. The study used medical records for most data collection and allowed for variations in hospital practices for Neonatal Opioid Withdrawal Syndrome (NOWS). The trial followed established research guidelines and ethical approvals from the Food and Drug Administration and institutional review boards. It took place across 13 US sites from July 2020 to October 2024.

Individuals included were 18 to 41 years old, 6 to 30 weeks pregnant with a single fetus, met diagnostic criteria for moderate or severe OUD, and were eligible for or already on buprenorphine. Hospitals where deliveries were planned had specific NOWS management protocols to reduce variability in infant outcomes. Exclusions included physiological dependence on alcohol or sedatives requiring withdrawal management, interfering medical or psychiatric conditions, criminal justice involvement that could affect participation, or certain other treatments.

Participants were randomly assigned to either extended-release or sublingual buprenorphine. Doses were determined by the clinician, with a target of 16 mg/day for sublingual or the equivalent weekly extended-release dose (24 mg). Pregnant and breastfeeding participants in the extended-release group received weekly injections due to prior concerns about certain inactive ingredients in monthly formulations; monthly injections were offered if not breastfeeding. A supply disruption affected some participants during the study.

The main goal was to measure illicit opioid abstinence during pregnancy, based on the proportion of negative urine drug screens. Key secondary outcomes included postpartum illicit opioid abstinence and infant NOWS outcomes (whether opioid treatment was needed and its duration). Other maternal and infant outcomes, such as medication adherence, craving, withdrawal symptoms, adverse events, birth outcomes, and developmental assessments, were also collected.

The statistical analysis plan used the intent-to-treat sample and was finalized before analyses began. The study employed specific alpha levels for primary, key secondary, and secondary outcomes. Mixed-model regression and other statistical tests were used to analyze the data. Sample size estimations were adjusted over time, ultimately aiming for 126 participants to ensure sufficient power to detect noninferiority.

Results

A total of 140 individuals were randomized: 69 to extended-release buprenorphine and 71 to sublingual buprenorphine. Most participants completed the pregnancy phase (98%) and postpartum phase (81%). The average age was 31.2 years, and the group was predominantly White (82.9%). Most participants (all but two) were already receiving sublingual buprenorphine at the start of the study.

During pregnancy, 59% of the extended-release group received the highest weekly dose (32 mg), while the average daily sublingual dose was 18.8 mg. Medication discontinuation rates were low during pregnancy (12% for extended-release, 9% for sublingual). Postpartum, dosing patterns varied, with similar discontinuation rates between groups (37% for extended-release, 33% for sublingual).

Extended-release buprenorphine demonstrated superiority in preventing illicit opioid use during pregnancy (82.5% negative urine samples versus 72.6% for sublingual buprenorphine; P = .009). However, no significant difference was observed postpartum (60.2% versus 59.5%; P = .45). Other maternal effectiveness outcomes showed no significant differences. For infants, there were no significant differences in the need for opioid treatment for NOWS (30.2% versus 26.5%; P = .64) or the duration of treatment.

The extended-release buprenorphine group experienced significantly fewer serious adverse events (AEs) during both pregnancy (8.7% versus 26.8%; P = .007) and postpartum (6.0% versus 18.6%; P = .04). Only one serious AE was medication-related (in the sublingual group). Non-serious AEs did not differ overall, but medication-related non-serious AEs were more frequent in the extended-release group during pregnancy (26.1% versus 7.0%; P = .003), mostly gastrointestinal and mild to moderate. Extended-release buprenorphine participants also received less pain medication (85.7% versus 98.5%; P = .03) and fewer opioids specifically (30.2% versus 50.0%; P = .02) during postpartum hospitalization, and had lower postpartum anxiety scores. The only significant infant safety finding was a larger average birth head circumference in the extended-release buprenorphine group (34.0 cm versus 33.4 cm; P = .049), though both averages were within normal ranges.

Discussion

This study represents the first randomized clinical trial to assess extended-release buprenorphine for opioid use disorder (OUD) during pregnancy. Individuals receiving weekly extended-release buprenorphine showed a significantly higher rate of illicit opioid abstinence during pregnancy compared to those receiving sublingual buprenorphine. Other outcomes for both mothers and infants were largely similar between the groups. Notable differences included fewer serious adverse events in the extended-release buprenorphine group during pregnancy and postpartum, less need for opioid pain medication during postpartum hospitalization, and lower postpartum anxiety scores. However, the extended-release group reported more non-serious medication-related adverse events, primarily mild gastrointestinal issues, which might reflect a potential bias in reporting for a newer medication. A larger average birth head circumference was observed in infants exposed to extended-release buprenorphine, but this finding's clinical importance remains unclear, as both groups were within normal ranges.

The superior illicit opioid abstinence observed with extended-release buprenorphine during pregnancy may be linked to its more stable medication levels throughout the day. The overall low rate of infants requiring opioid treatment for NOWS in this study (28%) was lower than typically reported. This might be due to participants generally receiving good prenatal care and already being stable on buprenorphine before the study. Postpartum, illicit opioid abstinence rates decreased in both groups and became similar, possibly due to the increased demands of infant care and potential involvement with child protective services, which often accompanies opioid treatment. The continued use of weekly injections for breastfeeding participants receiving extended-release buprenorphine, while sublingual buprenorphine offered longer coverage, might have impacted postpartum adherence.

This trial's strengths include its randomized design, diverse clinical settings, thorough assessments, and high rates of participant adherence and completion. However, limitations include the possibility of insufficient power for some secondary outcomes and potential data quality issues from medical records. Most participants were clinically stable and already on sublingual buprenorphine, which may limit the generalizability of these findings to individuals with less stable OUD or those initiating buprenorphine treatment. Additionally, the study population was primarily White and non-Hispanic, meaning the findings may not apply broadly to more diverse populations.

Conclusions

This trial showed that pregnant adults with opioid use disorder who received extended-release buprenorphine had higher rates of illicit opioid abstinence during pregnancy compared to those who received sublingual buprenorphine. However, this difference was not observed during the postpartum period. These findings support the use of weekly extended-release buprenorphine as a treatment option for pregnant individuals with opioid use disorder.

Introduction

Sublingual buprenorphine is a common treatment for opioid use disorder (OUD) in pregnant individuals. However, it has drawbacks such as the risk of the medication being misused or diverted, difficulty for patients to stick to their treatment plan, and daily highs and lows in medication levels. These daily fluctuations might not fully reduce opioid cravings and withdrawal symptoms, especially during pregnancy. Extended-release buprenorphine helps to overcome these issues and has shown better results in preventing illicit opioid use in adults who are not pregnant. A potential concern with extended-release buprenorphine in pregnancy is that it results in higher overall exposure to the medication, which means greater exposure for the fetus. While some smaller studies have looked at extended-release buprenorphine in pregnant individuals, a large, organized clinical trial has not yet been completed.

Methods

This study was a randomized clinical trial designed to compare the effectiveness and safety of extended-release buprenorphine with sublingual buprenorphine. It was an open-label trial, meaning participants and researchers knew which treatment was being given. The trial was conducted in various clinical settings across the US, using existing care models and relying mostly on medical records rather than extensive research-specific tests. It involved 13 sites and ran from July 2020 to October 2024. Participants were adults aged 18 to 41 years with a single pregnancy between 6 and 30 weeks gestation. All met criteria for moderate or severe OUD and were either eligible for buprenorphine treatment or already receiving it. Hospitals where participants planned to deliver had specific protocols for managing Neonatal Opioid Withdrawal Syndrome (NOWS). Participants were randomly assigned to receive either extended-release or sublingual buprenorphine. The target daily dose for sublingual buprenorphine was 16 mg, or an equivalent weekly dose of 24 mg for extended-release buprenorphine, with clinicians adjusting doses as needed. During pregnancy and breastfeeding, weekly injections of extended-release buprenorphine were given due to initial concerns about an ingredient in monthly formulations, although this guidance was later updated. The main goal was to measure illicit opioid abstinence during pregnancy using weekly urine drug screens. Key secondary outcomes included illicit opioid abstinence after delivery and, for infants, whether opioid treatment was needed for NOWS and how long it lasted. Many other maternal and infant health and safety measures were also tracked. Statistical analyses were performed to compare the two treatment groups, with careful consideration for the number of comparisons made and the study's ability to detect differences.

Results

A total of 140 participants were randomly assigned to either extended-release buprenorphine (69 participants) or sublingual buprenorphine (71 participants). Most participants completed both the pregnancy and postpartum phases of the study. The average age was around 31 years, and most participants were already receiving sublingual buprenorphine when the study began. During pregnancy, a significant number of participants in both groups received consistent medication doses. For the primary outcome, extended-release buprenorphine showed superior results, with a higher rate of illicit opioid abstinence during pregnancy (82.5% of urine samples were negative) compared to sublingual buprenorphine (72.6% negative). However, after delivery, the rates of illicit opioid abstinence were similar between both groups, with no significant difference found. For infants, there were no significant differences between the groups regarding the need for opioid treatment for NOWS or the duration of that treatment.

Regarding safety, participants receiving extended-release buprenorphine experienced fewer serious adverse events both during pregnancy and after delivery. While overall non-serious adverse events were similar between groups, medication-related non-serious adverse events were more often reported in the extended-release group during pregnancy, though these were mostly mild and often gastrointestinal issues. Notably, fewer participants in the extended-release group required any pain medication, especially opioids, after delivery during their hospital stay, and they reported lower anxiety scores postpartum. The only significant difference in infant safety was that infants exposed to extended-release buprenorphine had a slightly larger head circumference at birth, though both groups were within normal range.

Discussion

This trial is the first randomized study to investigate extended-release buprenorphine for OUD in pregnant individuals. A key finding was the significantly higher rate of illicit opioid abstinence in the extended-release buprenorphine group during pregnancy compared to the sublingual buprenorphine group. This superior abstinence rate might be due to the stable medication levels provided by extended-release buprenorphine, avoiding the daily peaks and troughs seen with sublingual forms. The overall rate of infants needing opioid treatment for NOWS was lower than in other studies, possibly because most participants had good prenatal care and were already stable on buprenorphine.

After delivery, illicit opioid abstinence rates decreased in both groups and became similar. This decline might be linked to the increased challenges faced by new parents, such as infant care demands and interactions with child protective services. Fewer serious adverse events were observed in the extended-release group, which is an important safety benefit. While more non-serious medication-related adverse events were reported for extended-release buprenorphine during pregnancy, these were generally mild, suggesting a possible bias in attributing common symptoms like gastrointestinal issues to the newer medication. The extended-release group also experienced less need for pain medication after delivery and reported lower postpartum anxiety, possibly due to more consistent and higher buprenorphine levels. The slightly larger head circumference observed in extended-release buprenorphine-exposed infants was within normal limits, and its clinical importance is unclear. Overall, the lack of adverse safety outcomes for infants in the extended-release buprenorphine group is reassuring and aligns with other research indicating that higher buprenorphine doses during pregnancy do not necessarily lead to worse infant outcomes.

Strengths and Limitations

The study's strengths include its randomized design, participation from diverse sites, comprehensive measurement of effectiveness and safety, and high rates of medication adherence and study completion. However, some limitations exist. The study may not have had enough participants to detect differences in certain secondary outcomes. Data taken from medical records sometimes had missing information or potential quality issues, such as varying methods for scoring NOWS in different hospitals. Also, most participants were already receiving sublingual buprenorphine and were clinically stable, so the results may not apply to pregnant individuals with more severe or unstable OUD or those just starting buprenorphine treatment. Finally, the study population was mostly White and non-Hispanic, meaning the findings may not be generalizable to more ethnically and racially diverse groups.

Conclusions

In this trial, pregnant adults with opioid use disorder who received extended-release buprenorphine had higher rates of abstinence from illicit opioid use during pregnancy compared to those who received sublingual buprenorphine. No difference was observed between the groups in the postpartum period. These results support the use of weekly extended-release buprenorphine for pregnant individuals managing opioid use disorder.

Introduction

Opioid use disorder (OUD) in pregnant and recently pregnant individuals is often treated with sublingual buprenorphine. However, this treatment has some drawbacks. These include the risk of the medication being diverted or misused, people not consistently taking it, and daily changes in drug levels that may not fully prevent opioid cravings and withdrawal, especially during pregnancy. Extended-release buprenorphine, given as an injection, helps to overcome these issues and has been shown to be more effective in preventing illicit opioid use in non-pregnant adults. A concern with using extended-release buprenorphine during pregnancy is that it results in higher overall exposure to buprenorphine, potentially meaning greater exposure for the fetus. While some smaller studies on extended-release buprenorphine in pregnant individuals exist, a full-scale randomized clinical trial was needed to compare its effectiveness and safety.

Methods

This study was a randomized clinical trial designed to compare the effectiveness and safety of extended-release buprenorphine injections with sublingual buprenorphine. The study took place at 13 sites across the United States. Participants included pregnant individuals aged 18 to 41, between 6 and 30 weeks of pregnancy, who had moderate or severe OUD. Most participants were already receiving sublingual buprenorphine treatment before joining the study. They were randomly assigned to receive either weekly extended-release buprenorphine injections or daily sublingual buprenorphine. The main goals were to see how well each treatment prevented illicit opioid use during pregnancy and after childbirth. Researchers also looked at outcomes for the newborns, such as the need for opioid treatment for neonatal opioid withdrawal syndrome (NOWS) and how long that treatment lasted. Additionally, the study tracked any side effects or safety concerns for both the mothers and their infants.

Results

A total of 140 participants completed the study. During pregnancy, those receiving extended-release buprenorphine had a significantly higher rate of avoiding illicit opioid use compared to those on sublingual buprenorphine (82.5% versus 72.6%). However, after childbirth, the rates of illicit opioid abstinence were similar between both groups, around 60%. For infants, there were no significant differences between the two treatment groups regarding the need for opioid treatment for NOWS or the duration of that treatment. In terms of safety, individuals on extended-release buprenorphine experienced fewer serious adverse events during both pregnancy and after childbirth. While they did report more minor medication-related adverse events during pregnancy (such as digestive issues or reactions at the injection site), these were generally mild or moderate. Infants exposed to extended-release buprenorphine had a slightly larger head circumference at birth, though averages for both groups remained within the normal range. All infants were born alive and discharged from the hospital.

Discussion

The findings suggest that weekly extended-release buprenorphine is more effective in preventing illicit opioid use during pregnancy than sublingual buprenorphine. This could be due to the extended-release form providing more stable medication levels without the daily ups and downs. The generally low rate of infants needing NOWS treatment in the study might be related to participants receiving good prenatal care and already being stable on buprenorphine. After childbirth, the decrease in illicit opioid abstinence rates and the similar rates between groups may reflect the added challenges new parents face, such as infant care and involvement with child protective services, which sometimes leads to people stopping their medication. The observation of fewer serious adverse events for mothers on extended-release buprenorphine is a positive finding. The larger infant head circumference, while statistically significant, is not clearly important clinically, but it is reassuring that no other negative safety outcomes were observed for these infants. The study's strengths include its randomized design and comprehensive assessments, but it is limited by a sample that was generally stable and mostly white and non-Hispanic, meaning the results may not apply to all populations with OUD.

Conclusions

This study found that pregnant individuals with opioid use disorder who received weekly extended-release buprenorphine had higher rates of illicit opioid abstinence during pregnancy compared to those receiving sublingual buprenorphine. Most other outcomes for mothers and infants, including safety measures, were similar between the two treatment groups. These results support the use of weekly extended-release buprenorphine as a treatment option for pregnant individuals with OUD.

Introduction

The usual medicine for pregnant people with opioid addiction is buprenorphine taken under the tongue. But this medicine has some downsides. It can be misused, people might not take it as they should, and daily doses may not fully stop cravings or withdrawal, especially during pregnancy. A new kind of long-acting buprenorphine shot can help with these issues. This shot has been shown to help adults who are not pregnant stay away from illegal opioids better. One concern with the shot for pregnant people is that it leads to more buprenorphine in the body, which means the baby also gets more of the medicine. Before this study, no large, fair studies had directly compared the shot to the under-the-tongue medicine in pregnant people.

Methods

This study aimed to see if the long-acting buprenorphine shot was as good as, or better than, the under-the-tongue buprenorphine for pregnant people. The study was open, meaning everyone knew which treatment they were getting. It took place in different clinics and hospitals. The study included pregnant people between 18 and 41 years old, who were 6 to 30 weeks pregnant and had opioid addiction. They needed to plan to deliver their babies at hospitals that were ready to care for newborns affected by opioids. People were divided into two groups by chance. One group received the long-acting buprenorphine shot every week. The other group received buprenorphine under the tongue, given as tablets or film. Doctors decided the exact dose for each person. The study followed participants for 12 months after giving birth. Researchers checked if people stopped using illegal opioids during and after pregnancy, and how their babies were doing, such as if they needed medicine for withdrawal. They also looked for any side effects.

Results

A total of 140 people took part in the study, with most finishing all phases. During pregnancy, people who got the long-acting buprenorphine shot were more successful at not using illegal opioids (82.5% stayed off) compared to those who took buprenorphine under the tongue (72.6% stayed off). This was a clear improvement for the shot group. However, after the baby was born, both groups had similar rates of not using illegal opioids, around 60%. Babies in both groups had similar needs for withdrawal treatment; about 30% needed medicine, which is lower than what some past studies have found. People who received the long-acting shot had fewer serious health problems during and after pregnancy. But during pregnancy, they did report more minor side effects linked to the medicine, mostly stomach issues. Those in the shot group also needed less pain medicine after birth and had lower anxiety levels after birth. Babies born to mothers who received the shot had slightly larger head sizes at birth, but these were still within the normal range. All babies in the study were born alive and were discharged from the hospital.

Discussion

This study is the first of its kind to test the long-acting buprenorphine shot for opioid addiction during pregnancy. The shot helped pregnant people stop using illegal opioids more effectively during pregnancy than the under-the-tongue medicine. This better outcome might be because the shot provides a more steady level of medicine in the body. After birth, the advantage of the shot faded, and both groups had similar results. This could be due to new mothers facing more challenges, such as caring for their infants or dealing with child protective services, which can make it harder to continue treatment. While the shot group had fewer serious health problems overall, they reported more minor, medication-related side effects like stomach issues during pregnancy. This might be partly because people were more likely to link side effects to the newer, less familiar shot. The fact that babies of mothers who received the shot were just as safe is good news, even though the shot leads to more buprenorphine in the body. The small difference in head size was considered normal and not a concern.

Conclusions

This study shows that pregnant adults with opioid addiction who received the weekly long-acting buprenorphine shot stayed off illegal opioids more often during pregnancy than those who took buprenorphine under the tongue. After the baby was born, both treatments worked about the same for keeping people from using illegal opioids. These findings support using the weekly long-acting buprenorphine shot as a safe and helpful treatment option for pregnant people with opioid addiction.