Abstract| Volume 28, SUPPLEMENT 1, S19, April 2020

Biological Agents in OA - Hopes and Disappointments

      Purpose: To review the trial in humans using biologics in osteoarthritis (OA) which can be can be categorized according to the main target: pain, structure or both.
      Methods: A literature review (Medline, Cochrane, Embase) of the most recent trials in OA using biologics
      Results: The analgesic effect of anti-nerve growth factor (NGF) was initially dramatic using high dose and intravenous route of administration, but emerging severe arthropathies (in the targeted knee but not only) have dampened down this enthusiasm. Most recent trials used anti NGF Mab, administered subcutaneously, with lower dose, shorter interval of administration and without concomitant use of NSAIDS. The effect on pain is still superior to placebo but with a much lesser magnitude. Unfortunately, long term follow-up in those trials still shows a concern in terms safety profile. Thus, Patients who will benefit of this anti NGF therapy should be strictly selected and followed up. Biologics agents targeting main cytokines (interleukines 1 and 6, tumor necrosis alpha) have been tested in knee OA and in hand OA. The first randomized controled trial in humans using a single local administration of the antagonist of IL-1, at 2 doses: 50 and 150 mg, failed to show an analgesic effect, except at day 3 with the highest dose. This negative result has been confirmed in trials using systemic administration of monoclonal antibodies directed against only IL1-β or against IL-1 α and Il-1 β in patients with knee OA and in patients with hand OA. However, in the CANTOR trial recruiting patients with cardiovascular disease treated with canakinumab, post hoc analysis indicated that the patients in the verum group, shows a dramatic decrease in the number of total joint replacement (hip and knee) (about 50%). This might open a door for a possible anti IL-1 strategy dedicated to patients with some part of systemic inflammation. Most of trial with anti TNF inhibitors have been performed in patients with erosive digital hand OA and all were negative on pain and on structure (measured by different parameters on magnetic resonance imaging). Finally the last cytokine that has been targeted is IL-6. Results of the RCP trial in hand OA are ongoing. Taking all together, biologics in OA are really disappointing, raising more questions than answers: choice of the targets, choice of selected phenotype of pain, selection of the route of administration, too short time of intra articular residence of the drug etc. News biologics targeting WNT, bradykinin or anti sense micro RNA are under development but so far, the preliminary results needs to be confirmed in large trials in humans. Finally, the last option is to try to stimulate the cartilage repair response which is naturally weak. Fibroblast growth factor 18 (sprifremin), an anabolic and mitogenic factor for the chondrocyte, has been tested against placebo in an ambitious trial over 5 years with 2 years of treatment and 3 years of follow-up. The drug was intra-articularly delivered every 6 months or every 12 months, with 2 different dosages. At the first end point, at year 2, with the highest dose of FGF-18, a statistically significant difference with the placebo was observed on the cartilage volume measured by MRI in the global, medial and lateral compartment on the involved knee. Unfortunately, there was no effect on pain in the ITT population, except in a sub group of patients with a minimum threshold of pain and an initial joint space width between 1.5mm and 3.5 mm (around one third of the population) show a significant effect on pain reduction compared to placebo. If results of this long term trial are confirmed in others trials, FGF-18 may constitute the first real disease modifying drug in OA. Conclusions
      Conclusion: The trials in humans using biologics are disappointing. We should keep in mind that blocking a single mediator will not stop the catabolic process in OA. Thus in the next future, we should probably adapt a drug owing to the evolution of the disease, to the profile of patient, to the radiological aspect of the disease and to the phenotype of pain. The story is just starting ⋯⋯