A prospective one-year natural history study of mucopolysaccharidosis types IIIA and IIIB: Implications for clinical trial design

•Robust outcome measures are needed for clinical trials in the mucopolysaccharidoses types 3A and 3B•Leiter nonverbal IQ standard scores significantly decline over 6 month intervals in patients less than 7 years old•Vineland-II composite scores decline significantly over 6 month intervals.•Cognitive and behavioral testing can distinguish declines in patients with MPS over periods relevant to clinical trials

Mucopolysaccharidosis type III is a group of four autosomal recessive enzyme deficiencies leading to tissue accumulation of heparan sulfate. Central nervous system disease is prominent, with initial normal development followed by neurocognitive decline leading to death. In order to define outcome measures suitable for gene transfer trials, we prospectively assessed disease progression in MPS IIIA and IIIB subjects > 2 years old at three time points over one year (baseline, 6 and 12 months). Fifteen IIIA (9 male, 6 female; age 5.0 ± 1.9 years) and ten IIIB subjects (8 male, 2 female; age 8.6 ± 3 years) were enrolled, and twenty subjects completed assessments at all time points. Cognitive function as assessed by Mullen Scales maximized at the 2.5 to 3 year old developmental level, and showed a significant age-related decline over a 6 month interval in three of five subdomains. Leiter nonverbal IQ (NVIQ) standard scores declined toward the test floor in the cohort by 6 to 8 years of age, but showed significant mean declines over a 6 month interval in those < 7 years old (p = 0.0029) and in those with NVIQ score ≥ 45 (p = 0.0313). Parental report of adaptive behavior as assessed by the Vineland-II composite score inversely correlated with age and showed a significant mean decline over 6 month intervals (p = 0.0004). Abdominal MRI demonstrated increased volumes in liver (mean 2.2 times normal) and spleen (mean 1.9 times normal) without significant change over one year; brain MRI showed ventriculomegaly and loss of cortical volume in all subjects. Biochemical measures included urine glycosaminoglycan (GAG) levels, which although elevated showed a decline correlating with age (p < 0.0001) and approached normal values in older subjects. CSF protein levels were elevated in 32% at enrollment, and elevations of AST and ALT were frequent. CSF enzyme activity levels for either SGSH (in MPS IIIA subjects) or NAGLU (in MPS IIIB) significantly differed from normal controls. Several other behavioral or functional measures were found to be uninformative in this population, including timed functional motor tests. Our results suggest that cognitive development as assessed by the Mullen and Leiter-R and adaptive behavior assessment by the Vineland parent interview are suitable functional outcomes for interventional trials in MPS IIIA or IIIB, and that CSF enzyme assay may be a useful biomarker to assess central nervous system transgene expression in gene transfer trials.