A Study of LY3023703 Testing Pain Relief After Wisdom Teeth Removal

Pain intensity was rated by the participant on a 100-mm VAS: 0 mm (no pain) and 100 mm (worst pain imaginable). The participant marked the line at the point that corresponded with his or her perception of pain. Weighted mean change from baseline was calculated as: \[the area under the change in pain intensity versus time curve\] / 8 hours (h). The baseline pain intensity was the pain assessment prior to dosing of study medication (0 h). Least Squares (LS) mean were calculated using a Bayesian analysis of covariance analysis (ANCOVA) adjusted for treatment as a fixed effect and baseline pain VAS as a continuous covariate. The measure of dispersion reported is 95% Credible Interval (CrI) not Confidence Interval (CI). A negative direction indicates a pain reduction from baseline.

Time frame: 0 to 8 h post-dose

Population: Full Analysis Set (FAS): Part A participants who were randomly assigned to treatment who received at least 1 dose of study medication and who had at least 1 postdose efficacy assessment. Pain assessments after rescue medication were imputed using last observation carried forward (LOCF) from the pain assessment prior to rescue therapy.

PGI-I is a participant-rated instrument that measures the improvement of the participants symptoms on a 7-point scale: 1 (very much improved), 4 (no change), and 7 (very much worse). LS mean was calculated using Mixed Effect Model Repeated Measures (MMRM) adjusted for treatment, time, the interaction of treatment and time and baseline pain VAS and fixed effects.

Time frame: 2, 4, 8, 12 and 24 h post-dose

Population: Part A participants who were randomly assigned to treatment who received at least 1 dose of study medication and who had at least 1 post-dose PGI-I assessment. Participants who received rescue medication were not included in the specific timepoints post administration of rescue medication.

Time to onset of the first perceptible pain relief is defined as the time from study drug administration to the measured onset of first perceptible pain relief in hours as reported by the participant. Participants who received rescue mediation prior to first perceptible pain relief were censored at the time the rescue medication was received.

Time frame: Study drug administration to first perceptible pain relief (0 to 24 h post-dose)

Population: FAS: Part A participants randomly assigned to treatment who received at least 1 dose of study medication and who had at least 1 post-dose efficacy assessment. Participants censored: Part A: LY3023703=11, Celecoxib=2, Placebo=9.

Time to onset of meaningful pain relief is defined as the time from study drug administration to the measured onset of meaningful pain relief in hours as reported by the participant. Participants who received rescue mediation prior to meaningful pain relief were censored at the time the rescue medication was received.

Time frame: Study drug administration to meaningful pain relief (0 to 24 h post-dose)

Population: FAS: Part A participants randomly assigned to treatment who received at least 1 dose of study medication and who had at least 1 post-dose efficacy assessment. Participants censored: Part A LY3023703=24, Celecoxib=2, Placebo=15.

The summed (time-weighted) pain intensity difference to baseline (SPID) at 4, 6, 8, 12, and 24 h post-dosing, as measured by a participant-rated 4-point categorical scale of 0 (no pain) to 3 (severe pain) and was calculated as: the area under the change in pain intensity versus time curve. Total scores range: -24 (best) to 8 (worst) for SPID 0 to 8 h. Score ranges for SPID(0-4h), SPID(0-6), SPID(0-12) and SPID(0-24) are -12 to 4, -18 to 6, -36 to 12 and -72 to 24 respectively. Participants were required to have moderate (score=2) or severe (score=3) pain at baseline in order to be eligible for randomization.LS mean were calculated using ANCOVA and was adjusted for treatment as a fixed effect and baseline pain intensity as a continuous covariate. The measure of dispersion reported is 95% CrI not CI. A negative direction indicated a pain reduction from baseline. There were no planned efficacy analysis for Pre-Part B per protocol.

Time frame: 0 to 4, 0 to 6, 0 to 8, 0 to 12, and 0 to 24 h post-dose

Population: FAS: All data from all participants randomly assigned to treatment who received at least 1 dose of study medication and who had at least 1 post-dose efficacy assessment. Pain assessments after rescue medication were imputed using LOCF from the pain assessment prior to rescue therapy.

Time to first use of rescue medication is defined as the time from study drug administration to the measured first use of rescue medication in hours. Participants were censored at 24 h post-dose if no rescue medication was administered. Pre-Part B used 3 participants in order for the study site to develop proficiency in the dialysate placement, collection, and maintenance techniques. There were no planned efficacy analysis for Pre-Part B per protocol.

Time frame: Study drug administration to first use of rescue medication (0 to 24 h post-dose)

Population: FAS: All data from all participants randomly assigned to treatment who received at least 1 dose of study medication and who had at least 1 post-dose efficacy assessment. Participants censored: Part A: LY3023703=5, Celecoxib=18, Placebo=12, Part B: LY3023703=0, Placebo=2.

TOPAR was calculated as the area under the pain relief versus time curve of the participant reported pain relief scores from the 5-point pain relief scale of 0 (no pain relief) to 4 (complete pain relief). LS mean were calculated using ANCOVA adjusted for treatment as a fixed effect. The measure of dispersion reported is CrI not CI. A negative direction indicated a pain relief from baseline. Pre-Part B used 3 participants in order for the study site to develop proficiency in the dialysate placement, collection, and maintenance techniques. There were no planned efficacy analysis for Pre-Part B per protocol.

Time frame: 0 to 4, 0 to 6, 0 to 8, 0 to 12, and 0 to 24 h post-dose

Population: FAS: All data from all participants randomly assigned to treatment who received at least 1 dose of study medication and who had at least 1 post-dose efficacy assessment. Pain assessments after rescue medication were imputed using LOCF from the pain assessment prior to rescue therapy excluding Pre-Part B.

Pain intensity was rated by the participant on a 100-mm VAS: 0 mm (no pain) and 100 mm (worst pain imaginable). The participant marked the line at the point that corresponded with his or her perception of post oral surgery pain. Weighted mean change from baseline was calculated as: \[area under the change in pain intensity versus time curve\] / \[time period that is (i.e.) 24 h for 0 to 24 h endpoint\]. The baseline pain intensity was the pain assessment prior to dosing of study medication. LS mean were calculated using ANCOVA adjusted for treatment, time and interaction of treatment as a fixed effect and baseline pain VAS as a continuous covariate. The measure of dispersion reported is the 95% CrI not CI. A negative direction indicated a pain reduction from baseline. Pre-Part B used 3 participants in order for the study site to develop proficiency in the dialysate placement, collection, and maintenance techniques. There were no planned efficacy analysis for Pre-Part B per protocol.

Time frame: Part A and B: 0 to 4, 0 to 6, 0 to 12, and 0 to 24 h post-dose and Part B 0 to 8 h post-dose

Population: FAS: All data from all participants randomly assigned to treatment who received at least 1 dose of study medication and who had at least 1 post-dose efficacy assessment. Pain assessments after rescue medication were imputed using LOCF from the pain assessment prior to rescue therapy.