The search identified 531 publications, of which 63 were removed as duplicates before title/abstract screening and 267 were excluded following title/abstract screening (Fig. 1). The full texts of the remaining 201 publications were screened against the PICOS criteria, following which 111 published studies were considered eligible for inclusion. Six abstracts were also identified from the congress searches, and three studies were included on request. In total, data were extracted from 120 studies, of which 88 were full publications and 32 were abstracts. Of the 120 studies, 22 were interventional in design (all of which were single-armed) and the remaining 98 were observational. All non-randomized, interventional studies were assessed with the ROBINS-I tool and determined to be of moderate quality.

PRISMA diagram. aThe reference lists from these publications were cross-checked against lists of included references to ensure that all relevant data were identified. Data from SLRs/meta-analyses were not extracted. bFour references were data-extracted, but data were either not reported separately for patients with MLD, or no relevant data were reported in the publication. cOf the 120 publications from which data were extracted, 108 related to the natural history and burden of illness of MLD (i.e., non-treatment related). MLD: metachromatic leukodystrophy; PRISMA: Preferred Reporting Items for Systematic review and Meta-Analysis; SLR: systematic literature review

The studies selected for data extraction used patient data from countries across six continents. The countries with the largest number of studies reporting patient data were the USA, Germany, and Italy (20, 12, and 12 studies, respectively). Six studies included data from multiple countries/continents, and the country was not reported or unclear for 13 studies. We report a subset of studies (n = 108) that describe the natural history and burden of illness of MLD. Of these, 70 were natural history studies. Note that any data from the remaining observational, retrospective, or interventional studies (n = 38) were related to patients who were not receiving treatment in the form of hematopoietic stem cell transplantation, gene therapy, or enzyme replacement therapy, except where otherwise stated.

The birth incidence and birth prevalence of MLD were reported in eight [9,10,11,12,13,14,15,16] and four [13, 14, 16, 17] studies, respectively (Table 2). For birth incidence, only postnatal diagnoses were considered, whereas for birth prevalence, both prenatal and postnatal diagnoses were considered. The data from these studies spanned 10 countries over a wide time frame and used different methodologies.

The Czech Republic was reported to have the lowest birth prevalence of MLD at 0.69 per 100,000 births [16], and Poland was reported as having the highest birth prevalence of 4.1 per 100,000 births [13]. Sweden was reported to have the highest incidence of MLD at 2.5 per 100,000 births [10], and Japan was reported to have the lowest incidence at 0.16 per 100,000 births [16].

The percentage of MLD cases within leukodystrophies was reported in nine studies [11, 18,19,20,21,22,23,24,25] and ranged from 8.0% (country not reported) [18] to 42.4% (Saudi Arabia) [22]. One study reported a proportion lower than this range (3.0%), but it only included individuals who did not have diagnostic testing in another healthcare system and who had no previous family members with the same diagnosis [25]. The reported proportion of MLD within LSDs was reported in seven studies [12, 14, 16, 26,27,28,29] and ranged from 3.3% (Japan) to 47.6% (Tunisia); however, the inclusion criteria varied between the studies. Those reporting a higher proportion of MLD within leukodystrophies or LSDs tended to have a smaller study population or were conducted in countries with higher rates of consanguinity. Three studies reported the proportion of MLD in inborn errors of metabolism, which ranged from 1.4% to 18.2% [29,30,31]. Three studies reported the prevalence of MLD in lipidoses, which ranged from 18.0% to 30.2% [12, 17, 32].

Age at symptom onset was reported in 51 studies [5, 9, 10, 22, 24, 32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77]. Across all studies, the range of ages at symptom onset (for individual patients) was 0.5–3 years for late-infantile MLD [9, 10, 34, 35, 38, 44, 45, 49,50,51, 53, 55, 57,58,59, 61,62,63,64,65, 72, 75, 77], 2–16 years for juvenile MLD (one study exclusively recruited early-symptomatic, early-juvenile patients) [9, 10, 34, 35, 45, 47, 49, 51,52,53, 55, 57, 63,64,65, 71, 72, 75, 77, 78], and 17–35 years for adult MLD [49, 72]. In most studies, mean age at symptom onset ranged from 16 to 18 months for the late-infantile subtype of MLD. For juvenile MLD, mean age at symptom onset ranged from 6 to 10 years [5, 9,