The KOOS questionnaire is a commonly used instrument to assess the patient's perception about their knee and associated problems. The original KOOS consists of 5 subscales. One of those is the pain frequency/severity during functional activities consisting of 9 questions with a recall of 7 days. Each question has 5 standardized answer options with a score from 0 to 4. A normalized score (100 indicating no pain and 0 indicating extreme pain) is calculated for each subscale. Change from baseline in KOOS pain score was analyzed using a mixed effects model for repeated measures (MMRM) including all time-points to compare treatment groups. The model includes baseline, treatment, timepoint, baseline-by-timepoints interaction and treatment-by-timepoints interaction as fixed effects. Missing data is assumed to be Missing at Random (MAR).

Full analysis set: All randomized participants that have received any study drug and had a valid measurement at the respective time-point. Based on a primary estimand framework.

Magnetic resonance images (MRI) were obtained from the target knee with dynamic contrast enhancement (DCE) to visualize and quantify changes in k-trans as a marker of the activity of synovial inflammation. In dynamic contrast-enhanced MRI, ktrans is a parameter that reflects how quickly contrast agent moves from blood vessels into the surrounding tissue, capturing both blood flow and vascular permeability. In synovitis, inflamed synovial tissue shows increased perfusion and leaky microvasculature, so higher k-trans values are interpreted as indicating more active synovial inflammation. Change from baseline in synovitis activity level measured from ktrans was analyzed to compare treatment groups. The model included treatment as fixed effect and baseline as continuous covariate. Ktrans is a marker of synovial inflammation in relation to vascularity permeability.

Full analysis set: All randomized participants that have received any study drug and had a valid measurement at the respective time-point. Based on a secondary estimand framework.

HsCRP is a protein produced by the liver in response to inflammation in the body, such as from an infection, injury, or chronic inflammatory conditions. HsCRP was used to assess the effect of DFV890 compared to placebo on systemic inflammatory status.

Full analysis set: All randomized participants that have received any study drug and had a valid measurement at the respective time-point.

ANC measures the number of neutrophils, a type of white blood cell crucial for fighting infection, in a blood sample. ANC was used to judge target engagement of DFV890.

Full analysis set: All randomized participants that have received any study drug and had a valid measurement at the respective time-point.

Cmax is defined as the maximum (peak) observed concentration following a dose. DFV890 plasma concentrations were determined using the actual recorded sampling times and non-compartmental method with Phoenix WinNonlin (Version 8). DFV890 was determined by a validated LC-MS/MS method; the lower limit of quantification (LLOQ) is 1 ng/mL.

The Pharmacokinetic (PK) analysis set included all participants with at least 1 available valid (i.e., not flagged for exclusion) PK concentration measurement, who received DFV890 and with no protocol deviations that impact on PK data.

AUClast is the area under the plasma concentration-time curve from time zero to the time of last quantifiable concentration (tlast). DFV890 plasma concentrations were determined using the actual recorded sampling times and non-compartmental method with Phoenix WinNonlin (Version 8). DFV890 was determined by a validated LC-MS/MS method; the lower limit of quantification (LLOQ) is 1 ng/mL.

The PK analysis set included all participants with at least 1 available valid (i.e., not flagged for exclusion) PK concentration measurement, who received DFV890 and with no protocol deviations that impact on PK data.

AUC(0-12h) is the area under the plasma concentration-time curve from time zero to 12 hours. To calculate AUC0-12h (corresponding to AUCtau, or AUC within a dosing interval), the concentration at 0 hours was used also as a 12 hours time point or, if not feasible, AUC0-12h was extrapolated. DFV890 plasma concentrations were determined using the actual recorded sampling times and non-compartmental method with Phoenix WinNonlin (Version 8). DFV890 was determined by a validated LC-MS/MS method; the lower limit of quantification (LLOQ) is 1 ng/mL.

The PK analysis set included all participants with at least 1 available valid (i.e., not flagged for exclusion) PK concentration measurement, who received DFV890 and with no protocol deviations that impact on PK data.

DFV890 plasma concentrations were determined using the actual recorded sampling times and non-compartmental method with Phoenix WinNonlin (Version 8). DFV890 was determined by a validated LC-MS/MS method; the anticipated lower limit of quantification (LLOQ) is 1 ng/mL.

The PK analysis set included all participants with at least 1 available valid (i.e., not flagged for exclusion) PK concentration measurement, who received DFV890 and with no protocol deviations that impact on PK data.

The KOOS questionnaire is a commonly used instrument to assess the patient's perception about their knee and associated problems. The original KOOS consists of 5 subscales. One of those is other symptoms (eg., stiffness, swelling, clicking) consisting of 7 questions with a recall of 7 days. Each question has 5 standardized answer options with a score from 0 to 4. A normalized score (100 indicating no symptoms and 0 indicating extreme symptoms) is calculated for each subscale. Change from baseline in other symptoms score was analyzed using a MMRM including all time-points to compare treatment groups. The model includes baseline, treatment, timepoint, baseline-by-timepoints interaction and treatment-by-timepoints interaction as fixed effects. Missing data is assumed to be Missing at Random (MAR).

Full analysis set: All randomized participants that have received any study drug and had a valid measurement at the respective time-point. Based on a secondary estimand framework.

The KOOS questionnaire is a commonly used instrument to assess the patient's perception about their knee and associated problems. The original KOOS consists of 5 subscales. One of those is the Function in Daily Living consisting of 17 questions with a recall of 7 days. Each question has 5 standardized answer options with a score from 0 to 4. A normalized score (100 indicating no impact on function in daily living and 0 indicating extreme impact on function in daily living) is calculated for each subscale. Change from baseline in KOOS daily living function score was analyzed using a MMRM including all time-points to compare treatment groups. The model includes baseline, treatment, timepoint, baseline-by-timepoints interaction and treatment-by-timepoints interaction as fixed effects. Missing data is assumed to be Missing at Random (MAR).

Full analysis set: All randomized participants that have received any study drug and had a valid measurement at the respective time-point. Based on a secondary estimand framework.

The KOOS questionnaire is a commonly used instrument to assess the patient's perception about their knee and associated problems. The original KOOS consists of 5 subscales. One of those is the function in sport and recreation consisting of 5 questions with a recall of 7 days. Each question has 5 standardized answer options with a score from 0 to 4. A normalized score (100 indicating no impact on sport and recreation and 0 indicating extreme impact on sport and recreation) is calculated for each subscale. Change from baseline in function in sport and recreation score was analyzed using a MMRM including all time-points to compare treatment groups. The model includes baseline, treatment, timepoint, baseline-by-timepoints interaction and treatment-by-timepoints interaction as fixed effects. Missing data is assumed to be Missing at Random (MAR).

Full analysis set: All randomized participants that have received any study drug and had a valid measurement at the respective time-point. Based on a secondary estimand framework.

The KOOS questionnaire is a commonly used instrument to assess the patient's perception about their knee and associated problems. The original KOOS consists of 5 subscales. One of those is the knee-related quality of life consisting of 4 questions with a recall of 7 days. Each question has 5 standardized answer options with a score from 0 to 4. A normalized score (100 indicating no impact on knee-related quality of life and 0 indicating extreme impact on knee-related quality of life) is calculated for each subscale. Change from baseline in knee related quality of life score was analyzed using a MMRM including all time-points to compare treatment groups. The model includes baseline, treatment, timepoint, baseline-by-timepoints interaction and treatment-by-timepoints interaction as fixed effects. Missing data is assumed to be Missing at Random (MAR).

Full analysis set: All randomized participants that have received any study drug and had a valid measurement at the respective time-point. Based on a secondary estimand framework.

The KOOS questionnaire is a commonly used instrument to assess the patient's perception about their knee and associated problems. The original KOOS consists of 5 subscales. One of those is pain frequency/severity during functional activities consisting of 9 questions with a recall of 7 days. Each question has 5 standardized answer options with a score from 0 to 4. A normalized score (100 indicating no pain and 0 indicating extreme pain) is calculated for each subscale. Change from baseline in KOOS pain score was analyzed using a MMRM including all time-points to compare treatment groups. The model includes baseline, treatment, timepoint, baseline-by-timepoints interaction and treatment-by-timepoints interaction as fixed effects. Missing data is assumed to be Missing at Random (MAR).

Full analysis set: All randomized participants that have received any study drug and had a valid measurement at the respective time-point. Based on a secondary estimand framework.

The Numerical Rating Scale (NRS) Pain is a subjective assessment in which individuals rate their pain on an eleven-point numerical scale. NRS pain score ranges from 0-10 and for analyses were transformed to a 0-100 scale to be consistent with KOOS pain scores. A negative change from baseline implied improvement in pain. The NRS Pain instrument had a recall period of 24 hours and the participants were asked to rate the pain intensity at its worst. Change from baseline in NRS pain score was analyzed using a MMRM including all time-points to compare treatment groups. The model includes baseline, treatment, timepoint, baseline-by-timepoints interaction and treatment-by-timepoints interaction as fixed effects. Missing data is assumed to be Missing at Random (MAR).

Full analysis set: All randomized participants that have received any study drug and had a valid measurement at the respective time-point. Based on a secondary estimand framework.

Neutrophils, a type of white blood cell essential for fighting infection, are measured in blood samples. Instances of neutrophil counts falling below the lower limit of normal were not consistently reported by all Principal Investigators as Adverse Events. To address this, an ad-hoc outcome measure was introduced to capture 'Neutrophil counts below the lower limit' based on laboratory data.

Safety analysis set: All randomized participants that have received any study drug and had a valid measurement at the respective time-point.