Safety assessments consisted of reports of AEs, vital signs (blood pressure, pulse rate, respiratory rate, and temperature), weight, hemoglobin-oxygen saturation measured by pulse oximetry (SpO2), clinical laboratory tests, somnolence (assessed by the University of Michigan Sedation Scale [UMSS]), conventional 12-lead electrocardiograms (ECGs), and 24-hour digital 12-lead ECGs (Holter monitor). Safety variables were summarized descriptively within age group for the safety population.

The safety population (N=41) was the group of patients who received at least 1 dose of study drug during the study.

The population PK (PopPK) of BTDS buprenorphine in pediatric patients ages 7 to 16 years was described by a 2-compartment model with sequential zero- and first-order absorption from the patch. A fixed allometric relationship was used to describe the effects of changes in ideal body weight (IBW) across pediatric patients on all clearance and volume parameters. Given the sparse sample collections, small sample size, and complexity of the absorption process with the patch formulation, this PopPK model was fit to the pediatric data using nonlinear mixed effects modeling (NONMEM) Bayes method with selective use of priors from previous adult PopPK results. Estimates of fixed effects from the final model were used to calculate the CL/F after patch dosing given the covariate distribution in the PopPK dataset and the typical weights from children in the National Health and Nutrition Examination Survey (NHANES) dataset.

The full analysis population (FAP) for PK consisted of patients who received study drug and had at least 1 valid quantifiable PK blood sample (N=41).Three patients had no quantifiable plasma buprenorphine concentration measurements and were not included in the PK analysis.The final buprenorphine pediatric PK analysis dataset comprised 38 patients.

Pain right now was assessed using the Faces of Pain Scale-Revised (FPS-R). The FPS-R is a horizontal row of 6 faces representing pain intensity, with "no hurt" at the far left and "hurts worst" at the far right; the intensities are scored as 0, 2, 4, 6, 8, or 10. A score of 0=no pain, and 10=very much pain. For screening to week 4, pain right now was assessed at 30 minutes before initial BTDS application on day 1; one hour after initial BTDS application on day 1; thereafter, once daily at approximately 8 PM for the first 4 weeks. Baseline score was the last assessment prior to the first dose. For weeks 1-4, weekly averages of the pain right now scores were calculated using the sum of all available pain right now scores recorded daily during a given week divided by the number of available scores.

The study measured pain right now for weeks 6-24 once a week at approximately 8 PM while on treatment; however, no patients in this age group were treated beyond week 12.

The full analysis population (FAP) was the group of patients who received at least 1 dose of the study drug during the study.

Pain right now was assessed using a 100-mm visual analogue scale (VAS). The 100-mm VAS is a 100-mm line with 1 end marked as "no pain" and the other marked as "pain as bad as it could be." The patient was asked to make a mark on that line indicating his or her level of pain. Pain right now score was defined as the distance (in mm) from the "no pain" end to the patient's mark; 0=no pain and bigger numbers indicate more pain. For screening to week 4, pain right now was assessed 30 minutes before initial BTDS application on day 1; one hour after initial BTDS application on day 1; thereafter, once daily at approximately 8 PM for the first 4 weeks. Baseline was the last assessment prior to the first dose. For weeks 1-4, weekly averages of the pain right now scores were calculated using the sum of all available pain right now scores recorded daily during a given week divided by the number of available scores.

For weeks 6-24, pain right now was measured once a week at approximately 8 PM.

The FAP was the group of patients who received at least 1 dose of the study drug during the study.

The PGIC rating score variable was collected on a 7-point scale ranging from 1 to 7 (where 1 = very much improved; and 7 = very much worse). The PGIC is designed to assess overall satisfaction with the treatment. The PGIC score was summarized using the number and percent of patients in each of the 7 possible response categories overall and by age group. PGIC was assessed by the parent/caregiver at the end of treatment visit or early discontinuation visit.

The full analysis population (FAP) (N=40) was the group of patients who received at least 1 dose of the study drug during the study; however data was provided for only 38 patients for this outcome measure.

The population PK (PopPK) of BTDS buprenorphine in pediatric patients ages 7 to 16 years was described by a 2-compartment model with sequential zero- and first-order absorption from the patch. A fixed allometric relationship was used to describe the effects of changes in ideal body weight (IBW) across pediatric patients on all clearance and volume parameters. Given the sparse sample collections, small sample size, and complexity of the absorption process with the patch formulation, this PopPK model was fit to the pediatric data using nonlinear mixed effects modeling (NONMEM) Bayes method with selective use of priors from previous adult PopPK results. Estimates of fixed effects from the final model were used to calculate the Vc/F after patch dosing given the covariate distribution in the PopPK dataset and the typical weights from children in the National Health and Nutrition Examination Survey (NHANES) dataset.

The full analysis population (FAP) for PK consisted of patients who received study drug and had at least 1 valid quantifiable PK blood sample (N=41).Three patients had no quantifiable plasma buprenorphine concentration measurements and were not included in the PK analysis.The final buprenorphine pediatric PK analysis dataset comprised 38 patients.