Plasma chitotriosidase activity versus CCL18 level for assessing type I Gaucher disease severity: protocol for a systematic review with meta-analysis of individual participant data

BACKGROUND Gaucher disease (GD) is an autosomal recessive lysosomal storage disorder caused by deficiency in acid beta-glucosidase. GD exhibits a wide clinical spectrum of disease severity with an unpredictable natural course. Plasma chitotriosidase activity and CC chemokine ligand 18 (CCL18) have been exchangeably used for monitoring GD activity and response to enzyme replacement therapy in conjunction with clinical assessment. Yet, a large-scale head-to-head comparison of these two biomarkers is currently lacking. We propose a collaborative systematic review with meta-analysis of individual participant data (IPD) to compare the accuracy of plasma chitotriosidase activity and CCL18 in assessing type I (i.e., non-neuropathic) GD severity. METHODS Eligible studies include cross-sectional, cohort, and randomized controlled studies recording both plasma chitotriosidase activity and CCL18 level at baseline and/or at follow-up in consecutive children or adult patients with type I GD. Pre-specified surrogate outcomes reflecting GD activity include liver and spleen volume, hemoglobin concentration, platelet count, and symptomatic bone events with imaging confirmation. Primary studies will be identified by searching Medline (1995 onwards), EMBASE (1995 onwards), and Cochrane Central Register of Controlled Trials (CENTRAL). Electronic search will be complemented by contacting research groups in order to identify unpublished relevant studies. Where possible, IPD will be extracted from published articles. Corresponding authors will be invited to collaborate by supplying IPD. The methodological quality of retrieved studies will be appraised for each study outcome, using a checklist adapted from the Quality Assessment of Diagnostic Accuracy Studies-2 tool. The primary outcome will be a composite of liver volume >1.25 multiple of normal (MN), spleen volume >5 MN, hemoglobin concentration <11 g/dL, or platelet count <100 × 109/L. Effect size estimates for biomarker comparative accuracy in predicting outcomes will be reported as differences in areas under receiver operating characteristic curves along with 95% confidence intervals. Effect size estimates will be reported as (weighted) mean differences along with 95% confidence intervals for each biomarker according to outcomes. IPD meta-analysis will be conducted with both one- and two-stage approaches. DISCUSSION Valid and precise accuracy estimates will be derived for CCL18 relative to plasma chitotriosidase activity in discriminating patients according to GD severity. SYSTEMATIC REVIEW REGISTRATION PROSPERO 2015 CRD42015027243.