Rate of progression of joint space narrowing (JSN) was defined as narrowing in joint space width (JSW) over the course of the study. It was measured radiographically in the medial tibiofemoral of knee of participants with OA. The slope reported in millimeter per year (mm/year) over a 2 year period was used to assess the rate of progression of JSN. Negative values indicating a worsening of osteoarthritis.

FAS included all participants randomized to the study. Here, 'number of participants analyzed' (N) signifies participants evaluable for this outcome measure.

Rate of progression of JSN was defined as narrowing in joint space width over the course of the study. It was measured radiographically in the medial tibiofemoral of knee of participants with OA. The slope reported in mm/year over a 2 year period was used to assess the rate of progression of JSN. KLG system was a method of classifying the severity of knee OA using five grades (0 [no severity] to 4 [maximum severity], higher grade indicating worse knee function). Negative values of slope indicating a worsening of osteoarthritis.

FAS included all participants randomized to the study. Here, 'N' signifies participants with KLG <=2 and evaluable for this outcome measure.

Rate of progression of JSN was defined as narrowing in joint space width over the course of the study. It was measured radiographically in the medial tibiofemoral of knee of participants with OA. The slope reported in mm/year over a 2 year period was used to assess the rate of progression of JSN. KLG system was a method of classifying the severity of knee OA using five grades (0 [no severity] to 4 [maximum severity], higher grade indicating worse knee function). Negative values of slope indicating a worsening of osteoarthritis.

FAS included all participants randomized to the study. Here, 'N' signifies participants with KLG =3 and evaluable for this outcome measure.

The WOMAC was a self-administered, disease-specific instrument which probed clinically important, participant relevant symptoms in the areas of pain, stiffness, and physical function in participants with OA of the knee. The WOMAC composite index was the sum of 24 individual questions regarding subscales of pain, stiffness and physical function (for each item score range: 0 [minimum] to 4 [maximum], higher score indicating worse knee condition). Total score was sum of the 3 subscale scores, giving a possible overall score range of 0 (minimum) to 96 (maximum). Higher score indicating the worse level of pain, stiffness and physical function.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The WOMAC pain subscale was comprised of 5 questions regarding the amount of pain experienced due to OA in the study knee. The WOMAC pain subscale score for each question ranged from 0 (minimum) to 4 (maximum), higher scores signified worse pain. An overall subscale score range of 0 (minimum) to 20 (maximum), with higher scores indicating more pain.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

Stiffness was defined as a sensation of decreased ease in which the participant moved the knee with OA. The WOMAC stiffness subscale was comprised of 2 questions regarding the degree of stiffness experienced in the study knee. The WOMAC stiffness subscale score for each question ranged from 0 (minimum) to 4 (maximum), higher scores signified worse stiffness. An overall score range of 0 (minimum) to 8 (maximum), with higher scores indicating more stiffness.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The WOMAC physical function subscale referred to the participant's ability to move around and perform usual activities of daily living. The WOMAC physical function subscale was comprised of 17 questions regarding the degree of difficulty experienced due to OA in the study knee. The WOMAC physical function subscale score for each question ranged from 0 (minimum) to 4 (maximum), higher scores signified worse physical function. An overall score range of 0 (minimum) to 68 (maximum), with higher scores indicating worse physical function.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

Pain VAS was a self-administered instrument, a 100 millimeter (mm) line marked by participant. Intensity of pain range (over past week): 0 (mm) =no pain to 100 (mm) =worst possible pain. Higher score indicating severe pain.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

Participants answered: "Considering all the ways your arthritis affects you, how are you feeling today?" Participants responded by using the scale ranging from 1 (minimum) to 5 (maximum), where 1 =very good, 2 =good, 3 =fair, 4 =poor and 5 =very poor. Higher scores indicating worse condition.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

Physician assessed the overall impact of arthritis on the participant's daily life. Participant's condition was rated by the physician using the scale ranging from 1 (minimum) to 5 (maximum), where 1= very good, 2= good, 3= fair, 4= poor and 5= very poor. Higher scores indicating worse condition.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The pain VAS following a 50 foot walk was a single-item, self-administered instrument. Participants were asked to assess the pain due to OA in their study knee after a 50-foot walk. Participants responded on a VAS scale ranging from 0 (no pain) to 100 (severe pain). Higher scores indicating more pain.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The OA pain and assessment tool-knee joint is also known as the intermittent and constant osteoarthritis pain (ICOAP) scale. The OA pain assessment tool-knee joint was an 11-item scale, with each item scored from 0 (minimum) to 4 (maximum), where 4 indicated worst health condition. The total score was calculated by adding the 11 items and rescaled to a 0 (minimum) to 100 (maximum) scale, where higher scores indicating worse health.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluated for this outcome measure at given time points for each group.

The OA pain assessment tool-knee joint constant pain subscale was a 5 item scale, with each item scored from 0 (minimum) to 4 (maximum), where 4 indicated worst health condition. Overall subscale score was calculated by adding the 5 items and rescaled to a 0 (minimum) to 100 (maximum) scale, where higher scores indicating worse constant pain.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The OA pain assessment tool-knee joint intermittent pain subscale score a 6 item scale, with each item scored from 0 (minimum) to 4 (maximum), where 4 indicated worst health condition. Overall subscale score was calculated by adding the 6 items and rescaled to a 0 (minimum) to 100 (maximum) scale, where higher scores indicating worse intermittent pain.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The OARSI knee function survey was an 11-item scale with each item scored 0 (minimum) to 4 (maximum), where 4 indicated worst health condition. The total score was the sum of the 11 items and ranged from 0 (minimum) to 44 (maximum), where higher scores indicating worse health condition.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The KOOS-PS was used to rate participant's opinions about the difficulties they experienced with activity due to problems with their knee. It was a 7-item scale, each item scored from 0 (minimum) to 4 (maximum), where 4 indicated worst health condition. Total score was calculated by adding the responses to 7 items and rescaled to a 0 (minimum) to 100 (maximum) scale, where higher scores indicating worse health condition.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The SF-36 was a participant administered scale assessing general quality of life. It consisted of self-administered 36-item questionnaire that measured 8 health domains: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. These 8 domains were also summarized as physical and mental component scores. The score for each domain and component score was the mean of the individual question scores, which were scaled from 0 (minimum) to 100 (maximum), where higher scores indicated highest level of health/functioning. Linear transformations were performed to transform scores and rescaled to a score range of 16.18 (minimum) to 57.11 (maximum), with higher scores indicating better physical functioning.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The SF-36 was a participant administered scale assessing general quality of life. It consisted of self-administered 36-item questionnaire that measured 8 health domains: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. These 8 domains were also summarized as physical and mental component scores. The score for each domain and component score was the mean of the individual question scores, which were scaled from 0 (minimum) to 100 (maximum), where higher scores indicated highest level of health/functioning. Linear transformations were performed to transform scores and rescaled to a score range of 18.45 (minimum) to 56.62 (maximum), with higher scores indicating better role-physical.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The SF-36 was a participant administered scale assessing general quality of life. It consisted of self-administered 36-item questionnaire that measured 8 health domains: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. These 8 domains were also summarized as physical and mental component scores. The score for each domain and component score was the mean of the individual question scores, which were scaled from 0 (minimum) to 100 (maximum), where higher scores indicated highest level of health/functioning. Linear transformations were performed to transform scores and rescaled to a score range of 19.23 (minimum) to 60.88 (maximum), with higher scores indicating lower bodily pain.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The SF-36 was a participant administered scale assessing general quality of life. It consisted of self-administered 36-item questionnaire that measured 8 health domains: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. These 8 domains were also summarized as physical and mental component scores. The score for each domain and component score was the mean of the individual question scores, which were scaled from 0 (minimum) to 100 (maximum), where higher scores indicated highest level of health/functioning. Linear transformations were performed to transform scores and rescaled to a score range of 16.75 (minimum) to 63.72 (maximum), with higher scores indicating better general health.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The SF-36 was a participant administered scale assessing general quality of life. It consisted of self-administered 36-item questionnaire that measured 8 health domains: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. These 8 domains were also summarized as physical and mental component scores. The score for each domain and component score was the mean of the individual question scores, which were scaled from 0 (minimum) to 100 (maximum), where higher scores indicated highest level of health/functioning. Linear transformations were performed to transform scores and rescaled to a score range of 22.02 (minimum) to 69.92 (maximum), with higher scores indicating better vitality.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The SF-36 was a participant administered scale assessing general quality of life. It consisted of self-administered 36-item questionnaire that measured 8 health domains: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. These 8 domains were also summarized as physical and mental component scores. The score for each domain and component score was the mean of the individual question scores, which were scaled from 0 (minimum) to 100 (maximum), where higher scores indicated highest level of health/functioning. Linear transformations were performed to transform scores and rescaled to a score range of 13.38 (minimum) to 56.40 (maximum), with higher scores indicating better social functioning.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The SF-36 was a participant administered scale assessing general quality of life. It consisted of self-administered 36-item questionnaire that measured 8 health domains: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. These 8 domains were also summarized as physical and mental component scores. The score for each domain and component score was the mean of the individual question scores, which were scaled from 0 (minimum) to 100 (maximum), where higher scores indicated highest level of health/functioning. Linear transformations were performed to transform scores and rescaled to a score range of 10.25 (minimum) to 55.68 (maximum), with higher scores indicating better role-emotional.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The SF-36 was a participant administered scale assessing general quality of life. It consisted of self-administered 36-item questionnaire that measured 8 health domains: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. These 8 domains were also summarized as physical and mental component scores. The score for each domain and component score was the mean of the individual question scores, which were scaled from 0 (minimum) to 100 (maximum), where higher scores indicated highest level of health/functioning. Linear transformations were performed to transform scores and rescaled to a score range of 8.02 (minimum) to 63.43 (maximum), with higher scores indicating better mental health.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The SF-36 was a participant administered scale assessing general quality of life. It consisted of self-administered 36-item questionnaire that measured 8 health domains: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. These 8 domains were also summarized as physical and mental component scores. The score for each domain and component score was the mean of the individual question scores, which were scaled from 0 (minimum) to 100 (maximum), where higher scores indicated highest level of health/functioning. Linear transformations were performed to transform scores and rescaled to a score range of 22.88 (minimum) to 58.69 (maximum), with higher scores indicating better physical health.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The SF-36 was a participant administered scale assessing general quality of life. It consisted of self-administered 36-item questionnaire that measured 8 health domains: physical functioning, role-physical, bodily pain, general health, vitality, social functioning, role-emotional, and mental health. These 8 domains were also summarized as physical and mental component scores. The score for each domain and component score was the mean of the individual question scores, which were scaled from 0 (minimum) to 100 (maximum), where higher scores indicated highest level of health/functioning. Linear transformations were performed to transform scores and rescaled to a score range of 11.11 (minimum) to 61.67 (maximum), with higher scores indicating better mental health.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

EQ-5D: participant rated questionnaire to assess health-related quality of life. Health state profile component assessed level of current health for 5 domains: mobility, self-care, usual activities, pain and discomfort, anxiety and depression. EQ-5D mobility domain score scale ranged from 1 (minimum) to 3 (maximum), where 1 =better health (no problem), 2 =moderate health (some problems) and 3 =worst health (confined to bed). Higher scores indicating worse health condition. Participants with EQ-5D mobility domain score were reported in this measure.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

EQ-5D: participant rated questionnaire to assess health-related quality of life. Health state profile component assessed level of current health for 5 domains: mobility, self-care, usual activities, pain and discomfort, anxiety and depression. EQ-5D self-care domain score scale ranged from 1 (minimum) to 3 (maximum), where 1 =better health (no problems with self-care), 2 =moderate health (some problems) and 3 =worst health (unable to wash or dress). Higher scores indicating worse health condition. Participants with EQ-5D self-care domain score were reported in this measure.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

EQ-5D: participant rated questionnaire to assess health-related quality of life. Health state profile component assessed level of current health for 5 domains: mobility, self-care, usual activities, pain and discomfort, anxiety and depression. EQ-5D usual activity domain score scale ranged from 1 (minimum) to 3 (maximum), where 1 =better health (no problems), 2 =moderate health (some problems) and 3 =worst health state (unable to perform usual activities). Higher scores indicating worse health condition. Participants with EQ-5D usual activity domain score were reported in this measure.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

EQ-5D: participant rated questionnaire to assess health-related quality of life. Health state profile component assessed level of current health for 5 domains: mobility, self-care, usual activities, pain and discomfort, anxiety and depression. EQ-5D pain and discomfort domain score scale ranged from 1 (minimum) to 3 (maximum), where 1 =better health (no pain and discomfort), 2 =moderate health (moderate pain and discomfort) and 3 =worst health state (extreme pain and discomfort). Higher scores indicated worse health condition. Participants with EQ-5D pain and discomfort domain score were reported in this measure.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

EQ-5D: participant rated questionnaire to assess health-related quality of life. Health state profile component assessed level of current health for 5 domains: mobility, self-care, usual activities, pain and discomfort, and anxiety and depression. EQ-5D anxiety and depression domain score scale ranged from 1 (minimum) to 3 (maximum), where 1 =better health (not anxious, depressed), 2 =moderate health (moderately anxious, depressed) and 3 =worst health (extremely anxious, depressed). Higher scores indicating worse health condition. Participants with EQ-5D anxiety and depression domain score were reported in this measure.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The EQ-5D VAS score was a participant rated questionnaire to assess health-related quality of life in terms of a single index value. It was a visual analogue scale that ranged from 0 (minimum) to 100 (maximum), with higher scores indicating a better health condition.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

Increase in total analgesic medication use for OA in the study knee was a comparison back to baseline of an increased and sustained use of standard background and/or rescue medication for more than 28 days as measured at the Month 12 and 24 visits. Only medications for OA knee pain were considered.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

Decrease in total analgesic medication use for OA in the study knee was a comparison back to baseline of a decreased and irregular use of standard background and/or rescue medication for more than 28 days as measured at the Month 12 and 24 visits. Only medications for OA knee pain were considered.

FAS included all participants randomized to the study. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

Patient global impression of change was a participant-rated instrument that measured change in participant's overall status on a 7-point scale ranging from: 1 =very much improved, 2 =much improved, 3 =minimally improved, 4 =no change, 5 =minimally worse, 6 =much worse and 7 =very much worse. Higher scores indicating worse condition.

FAS included all participants randomized to the study. Here, 'N' signifies participants evaluable for this outcome measure.

The OMERACT-OARSI responder index was used to determine whether participants may be considered responders to treatment. An OMERACT-OARSI responder was a participant who had a better response on the WOMAC pain subscale score, a better response on the WOMAC physical function subscale score or improvement on at least two of the three domains: WOMAC pain subscale score (overall score range of 0 [minimum] to 20 [maximum], higher scores indicating more pain), WOMAC physical function subscale score (overall score range of 0 [minimum] to 68 [maximum], higher scores indicating worse physical function) and patient global assessment of arthritic condition score (overall score range of 1 [minimum] to 5 [maximum], higher scores indicating worse condition). Number of participants who were OMERACT-OARSI responder were reported in this measure.

FAS included all participants randomized to the study. Here, 'N' signifies participants evaluable for this outcome measure.

JSN progressor was defined as a participant with a decrease in joint space width that was greater in magnitude than the smallest detectable difference (0.199 mm).

FAS included all participants randomized to the study. Here, 'N' signifies participants evaluable for this outcome measure.

A virtual joint replacement candidate was defined as a participant whose last two WOMAC pain subscale scores (overall score range of 0 [minimum] to 20 [maximum], higher scores indicating more pain) were at least 8, last two WOMAC physical function subscale scores (overall score range of 0 [minimum] to 68 [maximum], higher scores indicating worse physical function) were at least 28 and was a joint space narrowing progressor (a participant with a decrease in JSW that was greater in magnitude than the smallest detectable difference =0.199 mm).

FAS included all participants randomized to the study. Here, 'N' signifies participants evaluable for this outcome measure.

An AE was any untoward medical occurrence in a participant who received study drug without regard to possibility of causal relationship. An SAE was an AE resulting in any of the following outcomes or deemed significant for any other reason: death; initial or prolonged inpatient hospitalization; life-threatening experience (immediate risk of dying); persistent or significant disability/incapacity; congenital anomaly. Treatment-emergent were events between first dose of study drug and up to 7-10 days after last dose that were absent before treatment or that worsened relative to pre-treatment state. Adverse events included both serious and non-serious adverse events.

The safety analysis set included all participants in the FAS who received at least 1 dose of the randomized study medication.

Atrial (enlargement, fibrillation, premature beat), axis deviation, atrioventricular (accelerated conduction, first/second degree block), left anterior and posterior hemiblock, left atrial hypertrophy, left and right (complete/incomplete bundle branch block, ventricular hypertrophy), QRS (high/low voltage, nonspecific, prolongation greater than [>]140 milliseconds [msec]), junctional/paced rhythm, intraventricular conduction delay (>120 msec), early repolarization, ventricular premature contraction and beat, prolonged QTC, sinus (arrhythmia, bradycardia/tachycardia), supraventricular extra systole and premature beat, short PR syndrome. Abnormal Q-wave (>=30 msec), P-wave left/right atrial abnormality, T-wave flattened/inverted abnormality, U-wave abnormality, ST-T indeterminate abnormality, ST-T nonspecific changes, ST-T changes compatible with ischemia and pericarditis. ECG findings were judged by investigators for qualitative evaluation of abnormalities.

The safety analysis set included all participants in the FAS who received at least 1 dose of the randomized study medication. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

Criteria for laboratory abnormalities: Hemoglobin (Hgb), hematocrit (hct), red blood cell(RBC) count: less than(<)0.8*lower limit of normal(LLN), platelet: <0.5*LLN or greater than (>)1.75*upper limit of normal (ULN), white blood cell (WBC): <0.6*LLN or >1.5*ULN, lymphocyte, neutrophil:<0.8*LLN or >1.2*ULN, basophil, eosinophil, monocyte:>1.2*ULN; total bilirubin >1.5*ULN, aspartate aminotransferase, alanine aminotransferase, gammaglutamyl transferase, alkaline phosphatase:> 3.0*ULN, total protein, albumin: <0.8*LLN or >1.2*ULN; blood urea nitrogen, creatinine:>1.3*ULN, uric acid >1.2*ULN; sodium <0.95*LLN or >1.05*ULN, potassium, chloride, calcium, magnesium, bicarbonate: <0.9*LLN or >1.1*ULN, phosphate <0.8*LLN or >1.2*ULN; glucose <0.6*LLN or >1.5*ULN, lipase >1.5*ULN; urine (specific gravity <1.003 or >1.030, pH <4.5 or >8, glucose, ketones, protein, blood/Hgb greater than or equal to [>=]1); pancreatic amylase >1.5*ULN.

The safety analysis set included all participants in the FAS who received at least 1 dose of the randomized study medication. Here, 'N' signifies participants evaluable for this outcome measure.

Blood pressure (BP) was measured by sphygmomanometer while participant was in supine position. Conditions were kept constant from visit to visit including observer, participant's same arm, cuff size, supine position, location, temperature, noise level. The same size BP cuff which was properly sized and calibrated, was used to measure BP each time.

The safety analysis set included all participants in the FAS who received at least 1 dose of the study medication. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

BP was measured by sphygmomanometer while participant was in supine position. Conditions were kept constant from visit to visit including observer, participant's same arm, cuff size, supine position, location, temperature, noise level. The same size BP cuff which was properly sized and calibrated, was used to measure BP each time.

The safety analysis set included all participants in the FAS who received at least 1 dose of the study medication. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.

The safety analysis set included all participants in the FAS who received at least 1 dose of the study medication. Here, 'n' signifies participants evaluable for this outcome measure at given time points for each group.