Background The safety and efficacy of galcanezumab, a monoclonal antibody directed against CGRP, were assessed in a phase 2 clinical trial NCT02192190 in patients with moderate to severe osteoarthritis (OA) knee pain. Patients were randomized to placebo, galcanezumab (5, 25, 120 and 300 mg subcutaneously every 4 weeks, at weeks 0 and 4) or celecoxib (200 mg once daily) for 16 weeks in a 2:1:1:1:1:1 ratio. The study was terminated after an interim analysis due to inadequate efficacy for OA pain.
Objectives This study assessed the correlation of baseline plasma CGRP concentrations with signs, symptoms and radiographic severity of OA, and response to galcanezumab and celecoxib treatments.
Methods Plasma samples were collected at baseline and weeks 4, 8, 12 and 16 after study drug treatment. CGRP concentrations were determined by a validated high sensitivity (HS) assay. Correlation of baseline CGRP levels to WOMAC scores, PGA and radiographic Kellgren-Lawrence (K-L) grades were assessed using Spearman's correlation and Wilcoxon test. Patients were stratified into high vs low groups by baseline CGRP concentrations and post-treatment changes from baseline WOMAC scores evaluated by mixed effect model repeated measures for each subset.
Results At the interim analysis, baseline plasma CGRP samples were available for 262 patients with 54 patients providing samples at study termination through the week 8 visit. The median CGRP concentration at baseline was 1.07 pg/ml, range <0.78 to 33.91, and 31% of patients were below the level of quantitation (BLQ, <0.78 pg/ml). Median baseline CGRP levels were 1.0 pg/ml for K-L grade 2 (N=178), and 1.2 pg/ml for K-L grade 3 (N=84) (p=0.06). Correlations of WOMAC or PGA scores with baseline CGRP levels were all r <0.01 (showed no significant correlations). In OA patients receiving galcanezumab 300mg SC at week 0 and week 4, those with high baseline CGRP levels demonstrated a 14mm improvement in WOMAC Pain response at week 12, (95% CI 0, 29mm). The pain response to galcanezumab 300mg did not reach the magnitude of celecoxib response and no effects were seen at 5–120mg doses. Celecoxib treatment had larger pain reduction among patients with high baseline CGRP compared to low baseline CGRP levels. Treatment with celecoxib did not alter plasma CGRP concentrations.
Conclusions At baseline, CGRP levels in OA patients were not associated with WOMAC or PGA scores. There was a modest association to radiographic K-L grade. Subgroup analyses of patients with high (>median) CGRP levels at baseline suggested a potential response to galcanezumab for the highest dose, 300mg, but not lower doses. Celecoxib response was greater in those with higher CGRP levels. However, interpretation was limited by small samples sizes at the latter time points. Further studies may determine if enriching the OA population for higher CGRP levels at baseline, or if increased or longer dosing of galcanezumab would improve pain responses or if CGRP blockade is relevant in relieving OA knee pain.
Disclosure of Interest T. Mcnearney Shareholder of: Eli Lilly stock, Employee of: Eli Lilly and Company, C. Smith Shareholder of: Eli Lilly stock, Employee of: Eli Lilly and Company, R. Brown Shareholder of: Eli Lilly and Company, Employee of: Eli Lilly and Company, A. Camporeale Shareholder of: Eli Lilly stock, Employee of: Eli Lilly and Company, M. Deeg Shareholder of: Eli Lilly stock, Employee of: Eli Lilly and Company, D. Montieth Shareholder of: Eli Lilly stock, E. Collins Employee of: Eli Lilly and Company, T. Schnitzer Consultant for: Eli Lilly and Company, A. Kivitz: None declared, J. Talbot Shareholder of: Eli Lilly stock, Employee of: Eli Lilly and Company, Y. Jin Shareholder of: Eli Lilly stock, Employee of: Eli Lilly and Company