(19) Dosing to effect with Subcutaneous and Sublingual Buprenorphine for Opioid Use Disorder

Background & Introduction: Over 1 million Americans have opioid use disorder (OUD), a chronic illness which caused over 83,000 opioid overdose deaths in the US in 2022.[1] CAM2038 is a subcutaneous buprenorphine depot shown to be efficacious in the treatment of OUD.[2] Compared to sublingual buprenorphine (SL-BPN), CAM2038 treatment has been associated with superior reduction of overall illicit opioid use, and non-inferior likelihood of being a responder or having negative urine test results.[2] A range of doses are available for both CAM2038 and SL-BPN, which allows flexible dosing to address the individual treatment needs of patients. However, further research is needed to better understand the association of buprenorphine dose with treatment outcomes for CAM2038 and SL-BPN.[3] The aim of this post-hoc analysis was to assess whether dose of CAM2038 and SL-BPN was associated with treatment outcomes.

Methods: This was a post-hoc analysis of a phase 3, 24-week, randomized, double-blind, double-dummy, active-controlled, multicenter trial. Treatment-seeking adults with moderate-to-severe OUD were randomized to receive CAM2038 and daily SL placebo (subcutaneous buprenorphine [SC-BPN] group) or SC placebo injections and daily SL-BPN with naloxone (SL-BPN/NX group). CAM2038 or placebo injections were given weekly for 12 weeks then monthly for 12 weeks.

Doses were flexible and individualized per clinical judgement at each site based on current practice guidelines (e.g. dose increased if participant showed ongoing illicit opioid use or withdrawal).[4] Starting in the second month of treatment, the effect of each dose given was evaluated at the end of that dosing interval by examining four outcomes: 1) urine drug screen result (UDS); 2), Clinical Opiate Withdrawal Scale (COWS, 0–48); 3) Subjective Opiate Withdrawal Scale (SOWS, 0–64); and 4) need- and desire-to-use opioid visual analog scales (VAS, 0–100). UDS results (mean percentage of urine samples negative for opioids, with missing urine samples handled as positive) and mean COWS, SOWS and VAS scores were calculated for weeks 5–24. Associations between dose and the four outcomes were assessed by descriptive statistics.

Results: Mean age of participants (n=428) was 38.4 years. Primary self-reported opioid at initiation was heroin for 303 (70.8%) participants and prescription opioid pain reliever for 125 (29.2%) participants.

No association of dose with mean percentage of negative urine samples was identified for either weekly CAM2038 (16 mg, n=9: 76.4%; 24 mg, n=99: 47.0%; 32 mg, n=74: 53.4%), monthly CAM2038 (64 mg, n=9: 55.6%; 96 mg, n=75: 52.0%; 128 mg, n=63: 57.9%; 160 mg, n=7: 57.1%), or SL-BPN (8 mg, n=4: 87.5%; 16 mg, n=83: 34.9%; 24 mg, n=161: 36.6%; 32 mg, n=84: 39.9%).

No clinically significant differences were observed in mean COWS scores across different doses for weekly CAM2038 (16 mg, n=9: 2.0; 24 mg, n=100: 2.1; 32 mg, n=76: 3.0), monthly CAM2038 (64 mg, n=11: 0.5; 96 mg, n=77: 2.0; 128 mg, n=63: 2.7; 160 mg, n=9: 3.9), or SL-BPN (8 mg, n=4: 0.4; 16 mg, n=85: 2.1; 24 mg, n=168: 2.5; 32 mg, n=85: 3.1). These scores were reduced from a mean of 12 at baseline in both the overall CAM2038 and SL-BPN/NX groups.[2]

The previous n values reported for COWS scores apply for all subsequent outcomes. Increasing dose was not associated with any clinically significant differences in mean SOWS scores for weekly CAM2038 (16 mg: 4.3; 24 mg: 5.0; 32 mg: 8.3), monthly CAM2038 (64 mg: 1.3; 96 mg: 4.7; 128 mg: 7.3; 160 mg: 10.2), or SL-BPN (8 mg: 0.1; 16 mg: 5.1; 24 mg: 6.0; 32 mg: 7.0). These scores were reduced from a mean of 32 and 31 at baseline for CAM2038 and SL-BPN/NX groups, respectively.[2]

Mean need-to-use opioid VAS scores did not show any clinically significant differences across different doses for weekly CAM2038 (16 mg: 2.9; 24 mg: 12.2; 32 mg: 16.7), monthly CAM2038 (64 mg: 3.8; 96 mg: 8.9; 128 mg: 16.1; 160 mg: 36.0), or SL-BPN (8 mg: 1.7; 16 mg: 12.7; 24 mg: 14.4; 32 mg: 20.0). There were also no clinically significant differences observed in mean desire-to-use opioid VAS scores with increasing dose for weekly CAM2038 (16 mg: 4.9; 16 mg: 15.5; 24 mg: 19.8), monthly CAM2038 (64 mg: 2.8; 96 mg: 10.3; 128 mg: 19.0; 160 mg: 32.7) or SL-BPN (8 mg: 1.7; 16 mg: 13.1; 24 mg: 16.1; 32 mg: 22.2).

Conclusion & Discussion: Although higher rates of negative urine samples were observed with CAM2038 than SL-BPN,[2] efficacy was observed across all doses of CAM2038 and SL-BPN. No apparent association of dose with outcomes was identified after titrating dose to effect. This supports current practice guidelines that emphasize the importance of individualizing dosing based on clinical response.[4]

No clear relationship was identified between CAM2038 or SL-BPN dose and percentage of negative urine samples. There were relatively few patients in some dose groups, which may have impacted the possibility of identifying trends. Notably, the highest percentages of negative urine samples were observed in the lowest dose groups. This was expected, as doses would not be escalated for those whose OUD was well managed at a low dose, and few patients were maintained at those doses, consistent with clinical guidelines.[4]

Similarly, no clinically significant associations of dose with treatment outcomes were identified for withdrawal or cravings. Mean COWS and SOWS scores for all doses of both treatments indicated mild withdrawal (COWS: ≤12, SOWS: ≤10), demonstrating the efficacy of both CAM2038 and SL-BPN treatments.[2,5]

Overall, our findings were consistent with previous reports of the efficacy of CAM2038 and support individualization of OUD treatment to clinical effect.[2]

References: 1. Ahmad FB. National Center for Health Statistics 2023; 2. Lofwall MR. JAMA Intern Med 2018;178:764–73; 3. Hjelmström P. Drug Dev Ind Pharm 2020;46:1–7; 4. Substance Abuse and Mental Health Administration. Treatment Improvement Protocol (TIP) 63 2021; 5. Handelsman L. Am J Drug Alcohol Ab 1987;13:293–308; 6. Schuckit MA. N Engl J Med 2016;375:357–68.