Background/Purpose: Previous data showed that neutralization of granulocyte/monocyte colony stimulating factor (GM-CSF) rapidly abolished pain in an experimental OA model suggesting GM-CSF may be a promising therapeutic target in human OA (1).

Objective: Evaluate anti-GM-CSF GSK3196165 treatment on the signs and symptoms of hand OA.

Methods: Subjects with hand OA per ACR criteria, ≥2 swollen and tender IP joints in the same hand, hand pain ≥5 (numeric rating scale [NRS] 0-10, averaged over 7 days), signs of inflammatory changes by MRI and intolerance or unresponsiveness to NSAIDs, were randomized to 5 weekly SC doses of 180mg GSK3196165 or placebo (PBO), followed by 3 further doses every other week. Self-assessed 24h average and worst hand pain was recorded daily by electronic patient-reported outcome (ePRO) for the study duration. The primary outcome was ePRO change from baseline in average hand pain at Week 6. Secondary outcomes included safety, proportion of subjects showing 30% and 50% reductions in hand pain, change in Australian/Canadian Hand Osteoarthritis Index (AUSCAN) 3.1 NRS and changes in number of swollen and tender joints. Exploratory MRI endpoints included change in inflammatory and structural features.

Results: Of 44 treated subjects (91% women; mean age 58.8 years), 21/22 randomised to PBO and 20/22 to GSK3196165 completed the 12 week treatment period. Patient characteristics were well balanced between the groups. Measured GSK3196165 exposures were lower than anticipated from previous studies. Maximum pre-dose concentrations were achieved after 5 loading doses, but dropped significantly during every other week dosing.

Patients receiving GSK3196165 showed numerically larger reductions in hand pain than PBO at all timepoints, difference over PBO at W6: -0.36 (-1.31, 0.58; p=0.442), at W12: -0.89 (-2.06, 0.28; p=0.132); >2 points reduction from baseline seen from W8 to W12.

The proportion of subjects achieving 30% and 50% reductions of average hand pain was higher in the GSK3196165 group vs PBO at each assessment visit. At W12 the difference between GSK3196165 and PBO groups showing 30% reduction was 23% (odds ratio 3.2, 95% CI 0.86, 11.99; p=0.083) and for 50% reduction, was 27% (odds ratio 4.9, 95% CI 1.06, 22.59; p=0.042). AUSCAN pain (0-50) and function (0-90) showed difference vs PBO of -4.7 (-10.1, 0.6; p=0.082) and -8.2 (-19.1, 2.7; p=0.136) at W12 respectively. There was little difference from PBO on tender and swollen joints or MRI endpoints.

GSK3196165 was well tolerated. No serious infections or pulmonary events were observed.

Conclusion: This exploratory 12 week study showed that treatment of patients with inflammatory hand OA with GSK3196165 was well tolerated and while not statistically significant, resulted in reductions in pain accompanied by improvement in functional impairment. These results support findings from preclinical OA models and suggest that GM-CSF may play a role in the pain associated with inflammatory hand OA in man.

(1) Cook, Arthritis Res Ther 2012;14:R199

« Back to 2018 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/a-phase-iia-study-of-anti-gm-csf-antibody-gsk3196165-in-subjects-with-inflammatory-hand-osteoarthritis/