Considerations when designing a disease-modifying osteoarthritis drug (DMOAD) trial using radiography.

Objective: Using placebo data from a recently completed disease-modifying osteoarthritis (OA) drug trial, we seek to inform study design of future radiographic studies.

Methods: Eligible patients aged ≥40 years, with body mass index (BMI) 25-40kg/m(2) and symptomatic knee OA diagnosed by modified Kellgren and Lawrence grade (KLG) 2 or 3 and pain/stiffness and/or use of medication for knee pain in the past year, were assessed by radiography using a modified Lyon-schuss (mL/S) protocol for joint space narrowing (JSN) (primary outcome variable) at baseline and weeks 48 and 96. Multifaceted quality control was conducted throughout. Repeat images were requested when the medial tibial plateau (MTP) was not aligned (inter-margin distance [IMD] >1.5mm) or for other quality issues. Data are given mean ± standard deviation.

Results: Patients (74.9% female; 61.3 ± 9.1 years) had BMI 31.6 ± 4.1kg/m(2) at baseline; 222 (173 females) had KLG2, 264 (191 female) KLG3. A significant loss in joint space width (JSW) from baseline to week 48 (-0.13 ± 0.36mm) and to week 96 (-0.22 ± 0.45mm) was observed for all randomised placebo patients (p < 0.001 for both), and at both time points when stratified by KLG2 or KLG3. Standard deviations were small relative to mean changes, providing standardised response means for all placebo patients of 0.35 (week 48) and 0.48 (week 96).

Conclusions: Using a tightly controlled radiographic technique, JSN is a viable outcome variable for determining disease progression in mild-to-moderate knee OA. The mL/S protocol is a sensitive and feasible method for OA studies aiming to assess rate of JSN in the knee.