Change from Baseline in knee pain due to Osteoarthritis were reported by weekly average of average daily pain numeric rating scale (NRS) at Week 12. The pain NRS is an 11-point scale (ranging from 0-10) for self-reporting of average daily knee pain where 0 indicates no pain, and 10 indicates the worst possible pain. For each participant, the weekly average of average daily pain score was calculated using the mean value of available daily pain scores falling in the assessment window for each week. A negative change from baseline indicates an improvement in pain. Participants were asked to complete the pain NRS questionnaire at the same time in the evening each day. Baseline scores were assigned based on an average of 7 days prior to day 1 dosing visit. Posterior mean change from baseline, 95 percent (%) credible interval (CI) was derived using Bayesian mixed model repeated measures. The data presented are the posterior mean with 95% confidence interval refers to 95% credible interval.

The analysis was performed on the full analysis set (FAS) that included all randomized participants who received at least one dose of study intervention. Only those participants who had a measured outcome used in estimation or imputed outcome based on composite intercurrent event strategy at the current time point were included in the overall number of participants analyzed field.

The WOMAC pain subscale have a recall period of 48 hours and includes 5 subscales of pain assessment: 1-walking on flat, 2-going up downstairs, 3-at night while in bed, 4-sitting or lying; 5-standing upright. The total WOMAC pain subscale score ranges from 0-10; where 0 is no pain and 10 is extreme pain. The WOMAC pain subscale score was calculated by taking average of the 5 pain subscales at each visit. Change from baseline was calculated for each visit as the mean WOMAC Pain subscale score minus the mean baseline WOMAC Pain subscale score. A negative change from baseline indicates an improvement in pain. Baseline WOMAC scores for each participant were assigned based on the score measured prior to first dosing on Day1 visit. Posterior mean change from baseline, 95 percent (%) credible interval (CI) was derived using Bayesian mixed model repeated measures. The data presented are the posterior mean change from baseline with a 95% confidence interval refers to 95% Credible Interval.

The analysis was performed on FAS population which included all randomized participants who received at least one dose of study intervention. Only those participants who had a measured outcome used in estimation or imputed outcome based on composite intercurrent event strategy at the current time point were included in the overall number of participants analyzed field.

The WOMAC function subscale have a recall period of 48 hours and include 17 items of daily function assessments. The total WOMAC physical function subscale score ranges from 0-10 scale; where 0 is no difficulty and 10 is extremely difficult. The WOMAC physical function subscale score was calculated by taking the average of the 17 physical function subscales at each visit. Change from baseline was calculated for each visit as the mean WOMAC function subscale score minus the mean baseline WOMAC function subscale score. Baseline scores for each participant were assigned based on the score measured prior to first dosing on Day1 visit. A negative change from baseline indicated improvement. Posterior mean change from baseline, 95 percent (%) credible interval (CI) was derived using Bayesian mixed model repeated measures. The data presented are the posterior mean change from baseline with a 95% confidence interval refers to 95% Credible Interval.

The analysis was performed on FAS population which included all randomized participants who received at least one dose of study intervention. Only those participants who had a measured outcome used in estimation or imputed outcome based on composite intercurrent event strategy at the current time point were included in the overall number of participants analyzed field.

The PtGA is an assessment for disease conditions and intensity of knee osteoarthritis (OA) pain. Participants will respond on a Likert scale ranging from 1-5 based on the question "Considering all the ways in which your knee OA affects you, how do you feel your knee OA is doing today?" and to identify a number from 1 = very good (asymptomatic and no limitation to normal activities) to 5 = "very poor (very severe symptoms which are intolerable and inability to carry out all normal activities). Higher scores indicate worse conditions. Baseline scores for each participant were assigned based on the scores reported prior to first dosing on Day 1 visit. Posterior mean change from baseline, 95 percent (%) credible interval (CI) was derived using Bayesian mixed model repeated measures. The data presented are the posterior mean with a 95% confidence interval refers to 95% Credible Interval.

The analysis was performed on FAS population which included all randomized participants who received at least one dose of study intervention. Only those participants who had a measured outcome used in estimation or imputed outcome based on composite intercurrent event strategy at the current time point were included in the overall number of participants analyzed field.

AEs, SAEs, and AESIs were collected. An AE is any untoward medical occurrence in participant, temporally associated with use of study intervention, whether or not considered related to medicinal product. Any untoward event resulting in death, life threatening, requires hospitalization or prolongation of existing hospitalization, results in disability/incapacity, congenital anomaly/birth defect, medically important were categorized as SAEs. The AESIs of the study drug included serious and opportunistic infections, tuberculosis (TB) and TB reactivation, serious hypersensitivity reactions and Injection site reactions.

The analysis was performed on safety population which included all participants who received at least 1 dose of study treatment.

The laboratory measurements included hematology and clinical chemistry. The parameters evaluated were Basophil, Eosinophil, Erythrocyte Mean Corpuscular Hemoglobin, Erythrocyte Mean Corpuscular Volume, Erythrocytes, Hematocrit, Hemoglobin, Lymphocyte, Monocyte, Neutrophils, Platelets and Reticulocytes, Alanine Aminotransferase, Albumin, Alkaline phosphatase, Aspartate Aminotransferase, Bilirubin, Calcium, Creatinine, Direct Bilirubin, Glucose, Potassium, Sodium and Urea. Worst case grade (G) increase from baseline grade was provided for all the laboratory tests that were gradable by National Cancer Institute Common Terminology Criteria (NCI CTCAE, Version 5.0). Data for any participants with the worst-case grade changes (Increase or equal to Grade 3 or Increase to Grade 4) post-baseline are reported. Missing baseline grade was assumed as grade 0.

The analysis was performed on safety population which included all participants who received at least 1 dose of study treatment. The overall number of participants analyzed represents only those participants available at the specified time points and evaluable for this outcome measure.

Blood samples were collected at the indicated time points for pharmacokinetic (PK) analysis of GSK3858279 . Pharmacokinetic analysis was conducted using a model based analysis using all available data.

The analysis was performed on PK set which included all randomized participants in the Safety Population who had at least 1 dose of GSK3858279 and at least 1 non-missing PK assessment (Non-quantifiable values were considered as non-missing values). Only those participants with data available at specified time points have been analyzed.

Tmax predicted from the population PK model fitted to GSK3859279 serum concentration time data collected at the indicated time points for PK analysis.

The analysis was performed on PK set which included all randomized participants in the Safety Population who had at least 1 dose of GSK3858279 and at least 1 non-missing PK assessment (Non-quantifiable values were considered as non-missing values). Only those participants with data available at specified time points have been analyzed.

Ctau predicted from the population PK model fitted to GSK3859279 serum concentration time data collected at the indicated time points for PK analysis.

The analysis was performed on PK set which included all randomized participants in the Safety Population who had at least 1 dose of GSK3858279 and at least 1 non-missing PK assessment (Non-quantifiable values were considered as non-missing values). Only those participants with data available at specified time points have been analyzed.

Cavg predicted from the population PK model fitted to GSK3859279 serum concentration time data collected at the indicated time points for PK analysis.

The analysis was performed on PK set which included all randomized participants in the Safety Population who had at least 1 dose of GSK3858279 and at least 1 non-missing PK assessment (Non-quantifiable values were considered as non-missing values). Only those participants with data available at specified time points have been analyzed.

AUC(0-tau) predicted from the population PK model fitted to GSK3859279 serum concentration time data collected at the indicated time points for PK analysis.

The analysis was performed on PK set which included all randomized participants in the Safety Population who had at least 1 dose of GSK3858279 and at least 1 non-missing PK assessment (Non-quantifiable values were considered as non-missing values). Only those participants with data available at specified time points have been analyzed.