Background Pain is the principal clinical symptom of osteoarthritis (OA), and development of safe and effective analgesics for OA pain is needed. Trials of new analgesics for OA pain is impaired by substantial placebo-response (PR), and data describing characteristics and pain sub-types particularly associated with placebo-response is needed.
Objectives The purpose of this post-hoc analysis was to investigate clinical characteristics and pain subtypes associated with PR as measured by the five pain questions of the Western Ontario and McMasters Arthritis (WOMAC) Index of the target (T)- and the non-target (NT) knees in the pooled data of two phase III randomized clinical trials of an oral treatment in OA.
Methods Pooled data from the placebo-groups of two phase III randomized clinical trials (NCT00486434 and NCT00704847) in patients with knee OA followed for two years were analyzed. Main inclusion criteria for T knees were a Kellgren-Lawrence grade of 2 or 3, Joint-Space Width (JSW) of minimum of 2.0 mm at the medial tibio-femoral joint, and WOMAC pain ≥150 out of 500 mm. All analyses were performed on data from the T knees (N=771) and a sub-group of NT knees matching the pain inclusion criteria of the T knees (N=256). All subjects in the analysis belonged to the Per-Protocol population.
Pain of the T and NT knees was measured using the WOMAC pain sub-index, five questions: Q1; during walking on a flat surface, Q2; using stairs, Q3 at night while in bed, Q4; sitting or lying and Q5; while standing. Changes in percent from baseline pain were calculated at months 1, 6, 12 and 24, and differences between individual questions were assessed using a repeated measures ANOVA. Pain by category (“under load”; WOMAC questions 1, 2 and 5 and “idle”; WOMAC questions 3 and 4) were similarly analyzed for change over time, and compared in a repeated measures ANOVA. Selected patient baseline characteristics; pain by WOMAC question, Joint-space width (JSW), age, and BMI were assessed for association with PR defined as change in percent from baseline to year two using Spearman's correlation.
Results Baseline pain was significantly higher in WOMAC Q2, compared to other pain questions (p<0.001). This was observed for both T- and NT knees. WOMAC Q2 was furthermore associated with the highest PR for both T, and NT knees (36.8% ±52.4 (SD) and 28.0% ±44.4, respectively) at year two. The observed PR was higher in the T knee, compared to the NT knee at all corresponding time-points.
The level of PR was significantly higher in questions related to pain while under load (Q1, Q2, Q5) compared to pain while idle (Q3, Q4), and this was statistically significant in the T knee (p<0.005) at month 1 and onwards (Fig. 1), but did not reach statistical significance for the NT knee.
Baseline BMI, age and JSW was not associated with PR. This indicates while no common clinical characteristics were associated with PR, other, perhaps psychological, factors may affect the susceptibility to PR.
Conclusions These results indicate that the level of PR is significantly different between pain subtypes, as pain under load is significantly more susceptible to lead to PR than pain while idle.
Disclosure of Interest A. Bihlet Shareholder of: Nordic Bioscience, Employee of: Nordic Bioscience, M. Karsdal Shareholder of: Nordic Bioscience, Employee of: Nordic Bioscience, I. Byrjalsen Employee of: Nordic Bioscience, A.-C. Bay-Jensen Shareholder of: Nordic Bioscience, Employee of: Nordic Bioscience, C. Christiansen Shareholder of: Nordic Bioscience, Employee of: Nordic Bioscience, J. Andersen Shareholder of: Nordic Bioscience, Employee of: Nordic Bioscience, H. Gühring: None declared, C. Ladel: None declared, M. Michaelis: None declared, B. Riis Shareholder of: Nordic Bioscience, Employee of: Nordic Bioscience, I. Valter: None declared