LAUSR

Background Uricosurics (benzbromarone, probenecid, or sulfinpyrazone) are recommended for second-line treatment of gout if treatment target cannot be achieved with allopurinol or if allopurinol cannot be tolerated. While some uricosurics have been available since the 1970s, only two head-to-head trials have been conducted, both of which compared benzbromarone with probenecid. A new uricosuric, lesinurad, has recently been developed, but also has no head-to-head data against current uricosurics. There is therefore a need to compare the efficacy of uricosurics relative to each other via network meta-analysis (NMA), which requires high-quality evidence from well conducted trials, and also for evidence-based recommendations. Objectives To identify and assess the quality of the randomised controlled trial (RCT) evidence for uricosurics currently licensed for the treatment of gout, and to assess the feasibility of developing an NMA based on these trials. Methods A systematic review was conducted to identify RCT evidence for uricosurics, followed by an assessment of the quality of the comparisons possible based on this evidence using the GRADE system. This assessment was then used in a feasibility assessment for NMA based on the comparability of the trials in terms of population, comparators, and outcomes. Results In total, 21 publications reporting on 13 unique studies were identified by the SR, two of which had no common comparator arm with other trials and were excluded from further assessment. Eleven studies were included in the quality assessment informing comparisons between treatments for two outcomes: proportion of patients achieving sUA <6 mg/dL (or 0.36 mMol/L) and change from baseline in sUA. Issues with two or more elements of the methodological quality of the studies were present in all comparisons other than lesinurad plus allopurinol vs placebo plus allopurinol, including open-label design, incomplete reporting and sparse data mainly of safety assessment, and inadequate randomisation. The outcomes assessed in these RCTs were generally comparable across trials, but follow-up time ranged from 1 week to a mean of 19.6 months and few trials reported data at the same endpoint. In addition, line of therapy was unclear/not reported in four trials, with uricosurics used in a first-line population in two trials. The final GRADE quality rating was 5 (high) for lesinurad plus allopurinol vs placebo plus allopurinol and –1 to 1 (very low) for all other comparisons (febuxostat plus lesinurad was not included in comparisons). Based on this assessment, it was concluded that a NMA or other statistical comparison is not possible because of substantial heterogeneity between trials for methodological quality, outcomes, and population. Conclusions The quality of the randomised controlled trial evidence supporting the use of uricosurics other than lesinurad for gout is poor, and there is substantial heterogeneity in reported follow-up times and study populations. It is therefore not possible to make valid comparisons between these treatments based on this data, which may have implications for decision-makers in assessments of future treatments for gout. Disclosure of Interest None declared