Inhibition of IL-6 in the treatment of fibrous dysplasia of bone: The randomized double-blind placebo-controlled TOCIDYS trial

Increased interleukin-6 (IL-6) has been observed in the bone tissue of fibrous dysplasia of bone/McCune-Albright syndrome (FD/MAS) and is possibly involved in the increased bone destruction and bone pain characterizing this disease. The TOCIDYS trial was a randomized, placebo-controlled, 1 year, cross-over, proof-of-concept trial, conducted in patients not responding to bisphosphonates, using monthly intra-venous tocilizumab (a monoclonal antibody to the IL-6 receptor) at 8 mg/kg or a matching placebo for 6 months. Over the following 6 months, they received tocilizumab if they first had placebo, and vice-versa. We measured change in serum CTX after 6 months of treatment, compared with baseline (primary endpoint). Other endpoints were the change in bone pain, change in P1NP, bone alkaline phosphatase, osteocalcin and ICTP, and variation of quality of life. The analysis relied on ANOVA, with sequence of treatment, period and treatment as factors and accounting for a potential carry-over effect. We have randomized 8 patients with FD/MAS in each sequence who all completed the first 6 months treatment period. During the second 6 months period, 3 patients stopped therapy, so the efficacy analysis set included 13 patients. We observed no significant change in serum CTX and other biochemical markers of bone turnover between the tocilizumab and placebo groups. There was no significant change in the level of bone pain on tocilizumab, although 3 patients had a sharp decrease in pain while on active drug, with progressive relapse on placebo for 2 of them, but with some degree of improvement in a few patients while on placebo. The SF-36 quality of life scale was not significantly changed. We conclude that tocilizumab does not decrease bone turnover in FD/MAS when administered in patients who fail to respond to bisphosphonates. Tocilizumab does not reduce bone pain in most patients, but a substantial effect in a subset cannot be ruled out in this trial powered for markers but not for pain.