Osteoarthritis: Chondroitin sulfate — CONCEPT clear, uncertainties unchanged


osteoarthritis (SYSADOA), such as chondroitin sulfate, should be recommended for the management of knee osteoarthritis (OA) is an area of great contention. In the Chondroitin Versus Celecoxib Versus Placebo Trial (CONCEPT), treatment with either chondroitin sulfate or celecoxib (a cyclooxygenase-2 selective inhibitor) was superior to placebo in reducing pain and improving function in patients with symptomatic OA after 6 months1. Should chondroitin sulfate be considered a first-line pharmacological option, as the investigators of CONCEPT suggest? CONCEPT is the first clinical trial to be conducted in full accordance with the 2015 European Medicines Agency (EMA) guidelines on clinical investigation in OA2. In this sense, CONCEPT could be considered a proofof-concept clinical trial. The approach taken by Reginster et al.1 is of considerable clinical interest as it enables researchers to not only study the efficacy of a drug, but also to test the accuracy of the new guidelines. As well as the EMA guidelines, additional efforts have been made by other investigators to ensure patient safety and data accuracy in OA clinical trials. For example, the Data and Safety Monitoring Board (DSMB) is an independent committee that assures the methodological stringency of a trial by outlining the sampleadjusted statistical analysis to be used and the type of comparison to be conducted, as well should be evident within a few weeks of treatment commencing, and at an earlier time point than 3 months5. Thus, the fact that celecoxib did not show an analgesic effect during the first 3 months of treatment, but demonstrated efficacy after 6 months in CONCEPT, is puzzling. Similarly, a meta-analysis conducted by the Cochrane Collaboration concluded that treatment with chondroitin sulfate improves clinical symptoms in patients with OA in the short-term (<6 months) compared with placebo or a comparator oral medication, with an absolute risk difference of 10%6. In fact, the randomized trials included in this analysis revealed that chondroitin sulfate was better than placebo in studies of 3 months duration or less. As a SYSADOA, chondroitin sulfate would be expected to have an accumulative analgesic effect over time, and not to only show efficacy at the last visit (that is, at 6 months). These reflections led us to look more closely at the study population. Although equally distributed among the three groups, 75% of the patients in CONCEPT had mild disease, as defined by a Kellgren-Lawrence grade of 1 or 2 (REF. 1). Such a population is likely to be more heterogeneous than one containing patients with more severe disease and higher pain levels (indicative of later stages of disease) (FIG. 1), which could result in a more varied drug response. Notably, to be included in the study, patients only needed to have experienced pain for >3 months before enrolment, which might potentially favour the inclusion of patients in the early phases of OA. These potential caveats raise the question of what kind of OA population should be included in a clinical trial simultaneously comparing the analgesic effect of a SYSADOA and a cyclooxygenase-2 selective inhibitor, as each of these drugs is likely to be effective in particular subgroups of patients7. Whether the substantial reduction of pain observed in the three groups at 1 and 3 months in this study could be explained by spontaneous improvements of incidental bone marrow oedema, tendon strains or other lesions associated with knee OA should be taken into consideration. The placebo effect and the apparent efficacy of the drugs could be partially explained by factors inherent to the natural course of the disease in the selected population (FIG. 1). In light of these as providing researchers with a welldefined primary endpoint with a minimum clinically significant difference and rescue medication to consider using3. These different aspects, however, were not completely discussed in the study by Reginster and colleagues. Although a substantial placebo effect has been previously reported in OA trials, the size of the placebo effect in CONCEPT was notable1. The 2015 EMA guidelines contributed critically in this study1 by highlighting the need to initially assess the placebo effect size to indicate what sample size of the population would be needed for clinically significant results, and hence, put the yielded results into context. However, in OA research, we need to be aware that a statistically significant result is not necessarily a strong and robust result. Statistical significance conveys little information when measurements are taken in noisy conditions, and might even overestimate the size of an effect4. In CONCEPT, the efficacy shown for chondroitin sulfate after 6 months of treatment was intriguingly similar to that observed with celecoxib1. Moreover, the same pattern of pain improvement was observed in the three intervention arms, including placebo, after 1 and 3 months of treatment. This finding differs from the expected time curve response of NSAIDs. According to former knee OA clinical trials exploring pain relief using a variety of NSAIDs, the effect of NSAID therapy O S T E OA RT H R I T I S

DOI: 10.1038/nrrheum.2017.131

1 Figure or Table

Cite this paper

@article{HerreroBeaumont2017OsteoarthritisCS, title={Osteoarthritis: Chondroitin sulfate — CONCEPT clear, uncertainties unchanged}, author={G Herrero-Beaumont and Raquel Largo}, journal={Nature Reviews Rheumatology}, year={2017}, volume={13}, pages={576-577} }