Baseline for AP level was defined as the mean between assessments at Visit 1 (Week -4 to Week 0) and Visit 2 (Week 0).

The modified intent-to-treat (mITT) Analysis Set was defined as any randomized participant who received at least 1 dose of medication and had at least 1 on-treatment AP evaluation.

Percent change in AP serum levels from Baseline to the end of treatment was analyzed. For missing postbaseline data, the last non-missing postbaseline value on treatment was carried forward, with missing assessments imputed by last observation carried forward (LOCF). Baseline for AP level was defined as the mean between assessments at Visit 1 (Week -4 to Week 0) and Visit 2 (Week 0).

Participants in the mITT Analysis Set were analyzed.

Participants were defined as responders for the composite endpoint of AP and total bilirubin if their AP level was <1.67 × upper limit of normal (ULN) and had decreased at least 15% from their Baseline level and their total bilirubin value was within the normal range [0.1-1.1 milligrams/deciliter (mg/dL)].

Participants in the mITT Analysis Set were analyzed.

Percentage change in AST levels from Baseline to the end of treatment was analyzed. For missing postbaseline data, the last non-missing postbaseline value on treatment was carried forward, with missing assessments imputed by LOCF. Baseline for AST level was defined as the last non-missing assessments prior to or on the date of the first study dose.

Participants in the mITT Analysis Set were analyzed.

Percentage change in ALT levels from Baseline to the end of treatment was analyzed. For missing postbaseline data, the last non-missing postbaseline value on treatment was carried forward, with missing assessments imputed by LOCF. Baseline for ALT level was defined as the last non-missing assessments prior to or on the date of the first study dose.

Participants in the mITT analysis set were analyzed.

Percentage change in GGT levels from Baseline to the end of treatment was analyzed. For missing postbaseline data, the last non-missing postbaseline value on treatment was carried forward, with missing assessments imputed by LOCF. Baseline for GGT level was defined as the last non-missing assessments prior to or on the date of the first study dose.

Participants in the mITT Analysis Set were analyzed.

Percentage change in 5'Nucleotidase levels from Baseline to the end of treatment was analyzed. For missing postbaseline data, the last non-missing postbaseline value on treatment was carried forward, with missing assessments imputed by LOCF. Baseline for 5NT level was defined as the last non-missing assessments prior to or on the date of the first study dose.

Participants in the mITT Analysis Set were analyzed.

Percentage change in total bilirubin levels from Baseline to the end of treatment was analyzed. For missing postbaseline data, the last non-missing postbaseline value on treatment was carried forward, with missing assessments imputed by LOCF. Baseline for total bilirubin level was defined as the last non-missing assessments prior to or on the date of the first study dose.

Participants in the mITT Analysis Set were analyzed.

Percentage change in conjugated bilirubin levels from Baseline to the end of treatment was analyzed. For missing postbaseline data, the last non-missing postbaseline value on treatment was carried forward, with missing assessments imputed by LOCF. Baseline for conjugated bilirubin level was defined as the last non-missing assessments prior to or on the date of the first study dose.

Participants in the mITT Analysis Set were analyzed.

Percentage change in unconjugated bilirubin levels from Baseline to the end of treatment was analyzed. For missing postbaseline data, the last non-missing postbaseline value on treatment was carried forward, with missing assessments imputed by LOCF. Baseline for unconjugated bilirubin level was defined as the last non-missing assessments prior to or on the date of the first study dose.

Participants in the mITT Analysis Set were analyzed.

Percentage change in bone-specific AP levels from Baseline to the end of treatment was analyzed. For missing postbaseline data, the last non-missing postbaseline value on treatment was carried forward, with missing assessments imputed by LOCF. Baseline for bone-specific level was defined as the last non-missing assessments prior to or on the date of the first study dose.

Participants in the mITT Analysis Set were analyzed.

Percentage change in TG levels from Baseline to the end of treatment was analyzed. For missing postbaseline data, the last non-missing postbaseline value on treatment was carried forward, with missing assessments imputed by LOCF. Baseline for TG level was defined as the last non-missing assessments prior to or on the date of the first study dose.

Participants in the mITT Analysis Set were analyzed.

Percentage change in TC levels from Baseline to the end of treatment was analyzed. For missing postbaseline data, the last non-missing postbaseline value on treatment was carried forward, with missing assessments imputed by LOCF. Baseline for TC level was defined as the last non-missing assessments prior to or on the date of the first study dose.

Participants in the mITT Analysis Set were analyzed.

Percentage change in HDL-C levels from Baseline to the end of treatment was analyzed. For missing postbaseline data, the last non-missing postbaseline value on treatment was carried forward, with missing assessments imputed by LOCF. Baseline for HDL-C level was defined as the last non-missing assessments prior to or on the date of the first study dose.

Participants in the mITT Analysis Set were analyzed.

Percentage change in LDL-C levels from Baseline to the end of treatment was analyzed. For missing postbaseline data, the last non-missing postbaseline value on treatment was carried forward, with missing assessments imputed by LOCF. Baseline for LDL-C level was defined as the last non-missing assessments prior to or on the date of the first study dose.

Participants in the mITT Analysis Set were analyzed.

Published criteria for response to treatment in PBC included Paris I criteria. For Paris I, the criteria used to measure the PBC participants' response to study treatment was: AP ≤ 3x ULN and AST ≤ 2x ULN and Total Bilirubin ≤ 1 mg/dL. Participants who satisfied the criteria were defined as responders and those who did not were considered nonresponders if all the elements had non-missing assessments at that visit. The number of nonresponders were reported, the missing assessments were imputed by LOCF.

Participants in the mITT Analysis Set with available data were analyzed. Only participants who did not meet the criteria for response at Baseline were included in the analysis. Baseline for published PBC response criteria (Paris I) was defined as the last non-missing assessments prior to or on the date of the first study dose.

Published criteria for response to treatment in PBC included Paris II criteria. For Paris II, the criteria used to measure the PBC participants' response to study treatment was: AP ≤ 1.5x ULN and AST ≤ 1.5x ULN and Total Bilirubin ≤ 1 mg/dL. Participants who satisfied the criteria were defined as responders and those who did not were considered nonresponders if all the elements had non-missing assessments at that visit. The number of nonresponders were reported, the missing assessments were imputed by LOCF.

Participants in the mITT Analysis Set with available data were analyzed. Only participants who did not meet the criteria for response at Baseline were included in the analysis. Baseline for published PBC response criteria (Paris II) was defined as the last non-missing assessments prior to or on the date of the first study dose.

Published criteria for response to treatment in PBC included Toronto I criteria. For Toronto I, the criteria used to measure the PBC participants' response to study treatment was: AP ≤ 1.67x ULN. Participants who satisfied the criteria were defined as responders and those who did not were considered nonresponders if all the elements had non-missing assessments at that visit. The number of nonresponders were reported, the missing assessments were imputed by LOCF.

Participants in the mITT Analysis Set with available data were analyzed. Only participants who did not meet the criteria for response at Baseline were included in the analysis. Baseline for published PBC response criteria (Toronto I) was defined as the last non-missing assessments prior to or on the date of the first study dose.

Published criteria for response to treatment in PBC included Toronto II criteria. For Toronto II, the criteria used to measure the PBC participants' response to study treatment was: AP ≤ 1.76x ULN. Participants who satisfied the criteria were defined as responders and those who did not were considered nonresponders if all the elements had non-missing assessments at that visit. The number of nonresponders were reported, the missing assessments were imputed by LOCF.

Participants in the mITT Analysis Set with available data were analyzed. Only participants who did not meet the criteria for response at Baseline were included in the analysis. Baseline for published PBC response criteria (Toronto II) was defined as the last non-missing assessments prior to or on the date of the first study dose.

The UK-PBC Risk Score was used to estimate the risk that a PBC participant would develop liver failure that would require liver transplantation within 5, 10 or 15 years from diagnosis. The UK-PBC Risk Score calculation was adapted based on the AP assessment, transaminases (refers to the ALT, where available, otherwise the AST assessment) and total bilirubin assessment respectively at end of treatment divided by its upper limits of the corresponding normal ranges; "albumin" and "platelet" refer to the baseline values of these parameters divided by their corresponding lower limits of normal ranges. Missing assessments at end of treatment was imputed by last observation carried forward (LOCF). The UK-PBC risk scores for 5, 10 and 15 years was calculated for each treatment group

Participants in the mITT Analysis Set with available data were analyzed.

The 5-D itch scale was used to for the multidimensional quantification of pruritus over time. It assessed five aspects of itching: degree, duration, direction, disability, and distribution.The 5-D itch Scale is a measure of itching that has been validated in patients with chronic pruritus to detect changes over time. A 5-D itch scale score typically ranges from 5 to 25 with a higher score indicating greater itch severity, where 5 represents no pruritus (itching) and 25 represents the most severe pruritus; Each dimension is scored separately, and the scores are added together to get a total 5-D itch score. Higher scores indicate worsening of itching. The total 5-D itch Scale score change from Baseline was calculated. Baseline for 5-D itch scale was defined as the last non-missing assessments prior to or on the date of the first study dose. Missing assessments at end of treatment was imputed by LOCF.

Participants in the mITT Analysis Set with available data were analyzed.

VAS was used to measure pruritis in participants with PBC. Participants placed a mark representing their level of itching since the previous measurement on a 100-mm VAS with 0 indicating no itching and 100 mm indicating the worst possible itching. Higher scores indicated worsening of itching The change from Baseline was presented by treatment group. Baseline for VAS score was defined as the last non-missing assessments prior to or on the date of the first study dose. Missing assessments at end of treatment was imputed by LOCF.

Participants in the mITT Analysis Set with available data were analyzed.

The PBC-40 questionnaire was used to evaluate health-related quality of life measures, specifically fatigue. The PBC-40 questionnaire is a disease-specific tool developed to specifically measure the psychometric profile of PBC patients. It consists of 40 items divided into the six domains relevant to PBC including Cognitive, Social, EmotionalFunction, Fatigue, Itch, and Other Symptoms/General Questions. Items are scored from 1 to 5 and individual items scores are summed to give a total domain score. PBC 40 QoL domain score ranges are:Cognitive (6-30), Emotional Function (3-15),Fatigue (11-55), Itch (3-15),Social (10-50),Symptoms (7-35).Higher scores represent high impact and lower scores low impact of PBC on quality of life.The change and percentage change from Baseline for individual domain score at end of treatment were reported. Baseline for PBC-40 questionnaire was the last non-missing assessments prior to or on the date of the first study dose.Missing assessments imputed by LOCF.

Participants in the mITT Analysis Set with available data were analyzed.