Participants were required to score the severity of their itching using a 0-10 numerical rating scale (NRS) where 0 represents no itching and 10 indicates the worst imaginable itching. The Worst Daily Itch Score is the most severe (highest) NRS recorded on a given day. Mean Worst Daily Itch score was calculated as the average of the worst daily itch scores provided in the 7 days prior to the Week 16 visit. Baseline is the average of the scores in the 7 days prior to the Week 4 (Visit 3 [V3]). Change from Baseline was calculated as the post-Baseline value minus the Baseline value. Analysis was done using Analysis of covariance (ANCOVA) including treatment group and centered Mean Worst Daily Itch score at Baseline.

Intent-to-Treat (ITT) Population comprised of all randomized participants who received at least one dose of study treatment, had a Baseline and at least one on-treatment assessment. Only those participants with data available at the specified data points were analyzed.

PBC-40 is a disease-specific health-related quality of life (HRQoL) questionnaire for use in PBC participants. It consists of 40 questions arranged in 6 domains with 3 to 11 questions in each domain. Each question is scored from 1 (least impact) to 5 (greatest impact). All questions within a domain are summed to obtain individual domain score. Domains were: Symptoms (7 questions) with score range 7-35, Itch (3 questions) with score range 3-15, Fatigue (11 questions) with score range 11-55, Cognitive (6 questions) with score range 6-30, Emotional (3 questions) with score range 3-15, and Social (10 questions) with score range 10-50. Higher scores for individual domains represent a poor quality of life. Baseline is the assessment performed at Week 4 (V3) which is conducted prior to first dosing of randomized medication that evening. Change from Baseline was calculated as post-Baseline value minus Baseline value. Analysis was performed using ANCOVA including treatment group and Baseline.

ITT Population. Only those participants with data available at the specified data points were analyzed.

Criteria for high risk of PBC progression is defined as serum ALP concentrations more than or equal to (>=)1.67 times upper limit of normal (ULN) range and/or total bilirubin concentrations more than (>)ULN at Day 1. Baseline is the assessment performed at Week 4 (V3), or if missing then Visit 2 (Day 1) or Visit 1 (Screening), excluding unscheduled visits. Change from Baseline was calculated as the post-Baseline value minus the Baseline value. Analysis was performed using ANCOVA including treatment group and Baseline.

High Risk Population comprised of subset of the ITT population who were assigned to the High Risk stratum for randomization (based upon serum ALP concentrations >=1.67 times ULN and/or total bilirubin concentrations >ULN at Day 1 (Visit 2). Only those participants with data available at the specified data points were analyzed.

Number of participants with ALP < 1.67 times ULN and total bilirubin <= ULN at Week 16 is presented. The endpoint was analyzed in Restricted High Risk Population.

Restricted High Risk Population comprised of a subset of the High Risk population, i.e. all those participants assigned to the High Risk stratum for randomization who met the ALP/bilirubin criteria at both Visit 2 (Day 1) and Visit 3 (Week 4).

Baseline is the assessment performed at Week 4 (V3), or if missing then Visit 2 or Visit 1, excluding unscheduled visits. Change from Baseline was calculated as the post-Baseline value minus the Baseline value. Analysis was performed using ANCOVA model including Treatment group and Baseline.

High Risk Population. Only those participants with data available at the specified data points were analyzed.

Baseline is the assessment performed at Week 4 (V3), or if missing then Visit 2 or Visit 1, excluding unscheduled visits. Change from Baseline was calculated as the post-Baseline value minus the Baseline value. Analysis was performed using ANCOVA model including Treatment group and Baseline.

High Risk Population. Only those participants with data available at the specified data points were analyzed.

Baseline is the assessment performed at Week 4 (V3), or if missing then Visit 2 or Visit 1, excluding unscheduled visits. Change from Baseline was calculated as the post-Baseline value minus the Baseline value. Analysis was performed using ANCOVA model including treatment group and Baseline.

High Risk Population. Only those participants with data available at the specified data points were analyzed.

Baseline is the assessment performed at Week 4 (V3), or if missing then Visit 2 or Visit 1, excluding unscheduled visits. Change from Baseline was calculated as the post-Baseline value minus the Baseline value. Analysis was performed using ANCOVA including treatment group and Baseline.

High Risk Population. Only those participants with data available at the specified data points were analyzed.

Baseline is the assessment performed at Week 4 (V3), or if missing then Visit 2 or Visit 1, excluding unscheduled visits. Change from Baseline was calculated as the post-Baseline value minus the Baseline value. Analysis was performed using ANCOVA including treatment group and Baseline.

High Risk Population. Only those participants with data available at the specified data points were analyzed.

Baseline is the assessment performed at Week 4 (V3), or if missing then Visit 2 or Visit 1, excluding unscheduled visits. Change from Baseline was calculated as the post-Baseline value minus the Baseline value. Analysis was performed using ANCOVA model including treatment group and Baseline.

High Risk Population. Only those participants with data available at the specified data points were analyzed.

Baseline is the assessment performed at Week 4 (V3), or if missing then Visit 2 or Visit 1, excluding unscheduled visits. Change from Baseline was calculated as the post-Baseline value minus the Baseline value. Analysis was performed using ANCOVA model including treatment group and Baseline.

High Risk Population. Only those participants with data available at the specified data points were analyzed.

An AE is any untoward medical occurrence in a clinical study participant, temporally associated with the use of a study treatment, whether or not considered related to the study treatment. Any untoward event resulting in death, is life threatening, requires hospitalization or prolongation of existing hospitalization, results in persisting disability/incapacity, congenital anomaly/birth defect or any other situation according to medical or scientific judgment is categorized as SAE.

Safety Population. It consisted of all randomized participants who received at least 1 dose of study intervention.

An AE is any untoward medical occurrence in a clinical study participant, temporally associated with the use of a study treatment, whether or not considered related to the study treatment. Any untoward event resulting in death, is life threatening, requires hospitalization or prolongation of existing hospitalization, results in persisting disability/incapacity, congenital anomaly/birth defect or any other situation according to medical or scientific judgment is categorized as SAE.

Safety Population

An AE is any untoward medical occurrence in a clinical study participant, temporally associated with the use of a study treatment, whether or not considered related to the study treatment. Any untoward event resulting in death, is life threatening, requires hospitalization or prolongation of existing hospitalization, results in persisting disability/incapacity, congenital anomaly/birth defect or any other situation according to medical or scientific judgment is categorized as SAE.

Safety Population

Blood samples were collected to measure analyze the following parameters: albumin, calcium, Glomerular filtration rate (GFR) from creatinine, glucose, potassium and sodium. Participants were counted in the worst case category that their value changes to (low, within range [w/in] or no change, or high), unless there is no change in their category. Participants whose laboratory value category was unchanged (for example [e.g.], high to high), or whose value became within range, were recorded in the "To w/in Range or No Change" category. Participants were counted twice if the participant had values that changed "To Low" and "To High", so the percentages may not add to 100 percent (%). Only "To Low" and/or "To High" categories with potential clinical importance data have been presented.

Safety Population. Only those participants with data available at the specified data points were analyzed (represented by n= X in the category titles).

Blood samples were collected to analyze the following parameters: hematocrit, hemoglobin, leukocytes, lymphocytes, neutrophils and platelets. Participants were counted in the worst case category that their value changes to (low, w/in or no change, or high), unless there is no change in their category. Participants whose laboratory value category was unchanged (e.g., high to high), or whose value became within range, were recorded in the "To w/in Range or No Change" category. Participants were counted twice if the participant had values that changed "To Low" and "To High", so the percentages may not add to 100%. Only "To Low" and/or "To High" categories with potential clinical importance data have been presented.

Safety Population. Only those participants with data available at the specified data points were analyzed (represented by n= X in the category titles).

A 12-lead ECG was recorded with the participant in a semi-supine position. 12-lead ECGs were obtained by using an automated ECG machine. Data for abnormal, not clinically significant (NCS) and clinically significant (CS) ECG findings are presented. CS abnormal findings are those which are not associated with the underlying disease, unless judged by the investigator to be more severe than expected for the participant's condition.

Safety Population. Only those participants with data available at the specified data points were analyzed (represented by n= X in the category titles).

SBP and DBP were measured in the semi-supine position with a completely automated device after at least 5 minutes of rest for the participant in a quiet setting without distractions. Baseline is the assessment performed at Week 4 (V3), or if missing then Visit 2 or Visit 1, excluding unscheduled visits. Change from Baseline was calculated as the post-Baseline value minus the Baseline value.

Safety Population. Only those participants with data available at the specified data points were analyzed .

Pulse rate was measured in a semi-supine position after 5 minutes of rest. Baseline is the assessment performed at Week 4 (V3), or if missing then Visit 2 or Visit 1, excluding unscheduled visits. Change from Baseline was calculated as the post-Baseline value minus the Baseline value.

Safety Population. Only those participants with data available at the specified data points were analyzed.

GSRS is a validated scale used to assess gastrointestinal symptoms experienced by participants over the preceding 5 to 7 days. GSRS was measured for all 5 domains: Average Diarrhea Syndrome Score, Average Indigestion Syndrome Score, Average Constipation Syndrome Score, Average Abdominal Pain Syndrome Score, Average Reflux Syndrome Score. All individual domains are scored on a 7-point Likert scale ranging from 1(not at all) to 7(extremely). Higher score indicate more severe symptoms. The Average Total GSRS score was mean of these 5 domains and ranges from 1 to 7. Higher score indicates worst possible degree of symptoms. The responses summarized at each visit are those given during the week prior to the visit, with exception of Day 1. Baseline is the most recent assessment completed by participant prior to randomization. Change from Baseline was calculated as post-Baseline value minus the Baseline value. Data has been presented for each domain along with the average Total GSRS score.

Safety Population. Only those participants with data available at the specified data points were analyzed .

Participants were required to score the severity of their itching using a 0-10 NRS where 0 represents no itching and 10 indicates the worst imaginable itching. The Worst Daily Itch Score is the most severe (highest) NRS recorded on a given day. Mean Worst Daily Itch score was calculated as the average of the worst daily itch scores provided in the 7 days prior to the Week 16 visit. Number of participants with Mean Worst Daily Itch Score of <4 at Week 16 is presented.

ITT Population

Participants were required to score the severity of their itching using a 0-10 NRS where 0 represents no itching and 10 indicates the worst imaginable itching. The Worst Daily Itch Score is the most severe (highest) NRS recorded on a given day. Mean Worst Daily Itch score was calculated as the average of the worst daily itch scores provided in the 7 days prior to the Week 16 visit. Baseline is the most recent assessment completed by the participant prior to randomization. Number of participants with improvement of >= 30% in the Mean Worst Daily Itch Score at Week 16 from Baseline is presented.

ITT Population

Participants were required to score the severity of their itching using a 0-10 NRS where 0 represents no itching and 10 indicates the worst imaginable itching. The Worst Daily Itch Score is the most severe (highest) NRS recorded on a given day. Mean Worst Daily Itch score was calculated as the average of the worst daily itch scores provided in the 7 days prior to the Week 16 visit. Baseline is the most recent assessment completed by the participant prior to randomization. Number of participants with improvement of >=2 in the Mean Worst Daily Itch Score at Week 16 from Baseline is presented.

ITT Population

Percentage of Responder Days with Worst Daily Itch score was calculated as: (number of days response from Visit 3+1 to Visit 6-1 divided by number of days from Visit 3+1 to Visit 6-1 with worst daily itch scores available) times 100. Days for which no worst daily itch score was available did not contribute to either the numerator or the denominator. Analysis was performed using ANCOVA model including treatment group. Percentage of responder days with Worst Daily Itch Score of <4 is presented.

ITT Population

Percentage of Responder Days with Worst Daily Itch score was calculated as: (number of days response from Visit 3+1 to Visit 6-1 divided by number of days from Visit 3+1 to Visit 6-1 with worst daily itch scores available) times 100. Days for which no worst daily itch score was available did not contribute to either the numerator or the denominator. Analysis was performed using ANCOVA model including treatment group. Baseline is the most recent assessment completed by the participant prior to randomization. Percentage of responder days with improvement of >= 30% in the Mean Worst Daily Itch Score at Week 16 from Baseline is presented..

ITT Population

Percentage of Responder Days with Worst Daily Itch was calculated as: (number of days response from Visit 3+1 to Visit 6-1 divided by number of days from Visit 3+1 to Visit 6-1 with worst daily itch scores available) times 100. Days for which no worst daily itch score was available did not contribute to either the numerator or the denominator. Analysis was performed using ANCOVA model including treatment group. Baseline is the most recent assessment completed by the participant prior to randomization. Percentage of responder days with improvement of >=2 in the Mean Worst Daily Itch Score at Week 16 from Baseline is presented.

ITT Population

Mean Daily Sleep Score is defined as the average of the daily sleep scores provided in the 7 days prior to the relevant visit. Participants sleep quality was recorded in an electronic diary each morning using a 0-10 NRS in which 0: good sleep to 10:worst possible sleep. Higher score indicates worse possible sleep. Baseline is the average of the scores in the 7 days prior to the Week 4 (V3) visit. Change from Baseline was calculated as the post-Baseline value minus the Baseline value. Analysis was performed using ANCOVA including treatment group and Baseline.

ITT Population. Only those participants with data available at the specified data points were analyzed.

Mean Daily Fatigue Score is defined as the average of the daily fatigue scores provided in the 7 days prior to the relevant visit. Participants fatigue level was recorded in an electronic diary each evening using a 0-10 NRS in which 0: no fatigue to 10:worst possible fatigue. Higher score indicates worse possible fatigue. Baseline is the average of the scores in the 7 days prior to the Week 4 (V3) visit. Change from Baseline was calculated as the post-Baseline value minus the Baseline value. Analysis was performed using ANCOVA model including treatment group and Baseline.

ITT Population. Only those participants with data available at the specified data points were analyzed.

The 5-D itch scale had been developed as a brief, single page, instrument for the multidimensional quantification of itch that is sensitive to change over time. It has data to support its validity in a population of participants with pruritus and covers five dimensions of itch experienced by participants: duration, degree, direction, disability and distribution. Each domain was scored on a 5-point scale, ranging from 1 (Not present/resolved/never) to 5 (unbearable/getting worse/always), higher scores indicates worst itching. The scores of each of five domains were achieved separately and then summed together to obtain a total 5-D score. A total 5-D scores potentially ranged between 5 (no pruritus) and 25 (most severe pruritus) where higher score indicates worse possible itching. Baseline is assessment performed at Week 4 (V3) which is conducted prior to first dosing of randomized medication that evening. Change from Baseline was calculated as post-Baseline value minus Baseline value.

ITT Population. Only those participants with data available at the specified data points were analyzed.

Blood samples were collected for evaluating total bile acid concentration as a biomarker of PBC. Baseline is the assessment performed at Week 4 (V3), or if missing then Visit 2 or Visit 1, excluding unscheduled visits. Change from Baseline was calculated as the post-Baseline value minus the Baseline value.

ITT Population. Only those participants with data available at the specified data points were analyzed.

Blood samples were collected for evaluating C4 concentration as a marker of bile acid synthesis. Baseline is the assessment performed at Week 4 (V3), or if missing then Visit 2 or Visit 1, excluding unscheduled visits. Change from Baseline was calculated as the post-Baseline value minus the Baseline value.

ITT Population. Only those participants with data available at the specified data points were analyzed.

Blood samples were collected for measurement of plasma GSK2330672 concentration.

Pharmacokinetic (PK) Population consisted of any randomized participant who had at least one PK sample. Only those participants with data available at the specified data points were analyzed (represented by n=X in category titles).

Participants were required to score severity of their itching each morning and evening using a 0-10 NRS where 0(no itching) and 10(worst imaginable itching). The worst of these 2 scores was Worst Daily Itch Score. For each week, mean Worst Daily Itch Score was calculated to form Mean Worst Daily Itch Score. The Monthly Itch Score was defined as worst weekly score (e.g., Mean Worst Daily Itch Score) for that month. The monthly itch score ranges from 0 to 10, higher score indicates worst imaginable itching. Baseline is average of scores in the 7 days prior to Week 4 (Visit 3 [V3]). Change from Baseline was calculated as post-Baseline value minus Baseline value. Mean change from Baseline in monthly itch score over 12 week treatment period is presented. Analysis was performed on change in Monthly Itch Scores over 12 week treatment period using Mixed model repeated measures (MMRM) with Baseline itch, treatment group, visit and a treatment group*visit interaction as covariates in the model.

ITT Population. Only those participants with data available at the specified data points were analyzed.

Blood samples were collected for evaluation of total bile acid concentration as a biomarker of PBC. Baseline is the assessment performed at Week 4 (V3), or if missing then Visit 2 (Day 1) or Visit 1 (Screening), excluding unscheduled visits. Ratio to Baseline was defined as the geometric mean of post-Baseline visit value divided by the geometric mean of Baseline value. Values were log-transformed and mean change from Baseline on the log-scale was calculated over the 12 week treatment period. Analysis was performed on change in total serum bile acid concentration on the log-scale over the 12 week treatment period using Mixed model repeated measures (MMRM) with log-transformed Baseline total serum bile acid, visit, treatment group, a log-transformed Baseline total serum bile acid*visit interaction, and a treatment group*visit interaction used as covariates in the model. Afterwards, values were back-transformed to the original scale. Ratios of geometric means are presented.

ITT Population. Only those participants with data available at the specified data points were analyzed.

Blood samples were collected for evaluation of C4 concentration as a marker of bile acid synthesis. Baseline is the assessment performed at Week 4 (V3), or if missing then Visit 2 (Day 1) or Visit 1 (Screening), excluding unscheduled visits. Ratio to Baseline was defined as the geometric mean of post-Baseline visit value divided by the geometric mean of Baseline value. Values were log-transformed and mean change from Baseline on the log-scale was calculated over the 12 week treatment period. Analysis was performed on change in C4 on the log-scale over the 12 week treatment period using Mixed model repeated measures (MMRM) with log-transformed Baseline C4, visit, treatment group, a log-transformed Baseline C4*visit interaction, and a treatment group*visit interaction used as covariates in the model. Afterwards, values were back-transformed to the original scale. Ratios of geometric means are presented.

ITT Population. Only those participants with data available at the specified data points were analyzed.