MED = multiple epiphyseal dysplasia; NA = not applicable

Genes are listed in alphabetic order.

See Table A. Genes and Databases for chromosome locus and protein.

In individuals with autosomal dominant MED in whom a pathogenic variant in one of the five confirmed genes has been identified. However, the relative proportions are different depending on ethnicity. For example, a study by the European Skeletal Dysplasia Network (ESDN) [Jackson et al 2012] found that in 56 individuals with molecularly confirmed MED, COMP pathogenic variants accounted for 66%, MATN3 for 24%, COL9A2 for 8%, and COL9A3 for 2%. In contrast, a recent study of a Korean cohort identified pathogenic variants in 55 individuals as follows: COMP (43%), MATN3 (55%), and COL9A2 (2%) [Kim et al 2011]. This is in close alignment with a Japanese study that identified pathogenic variants in 19 individuals with MED: COMP (37%), MATN3 (47%), COL9A2 (11%), and COL9A3 (5%). The high prevalence of MATN3 pathogenic variants in these latter populations is believed to be the result of a common founder variant (p.Arg121Trp), but this variant is also common in European populations. None of the three studies identified pathogenic variants in COL9A1.

The proportion of COL9A1-3, COMP, and MATN3 pathogenic variants found in persons with MED is not well established. Previous studies have suggested frequencies of 10%-36% for COMP [Jakkula et al 2005, Kennedy et al 2005b], 10% for MATN3, and 5% for the type IX collagen genes [Briggs & Chapman 2002, Jackson et al 2004]. However, in a later study by the ESDN, the proportion of MED caused by pathogenic variants in COMP increased to 81% when a strict clinical-radiographic review was undertaken before molecular genetic testing was performed [Zankl et al 2007]. The success of this approach was confirmed by Kim et al [2011], when preselection resulted in a variant detection rate of 87%.

See Molecular Genetics for information on variants detected in these genes.

Sequence analysis detects variants that are benign, likely benign, of uncertain significance, likely pathogenic, or pathogenic. Variants may include missense, nonsense, and splice site variants and small intragenic deletions/insertions; typically, exon or whole-gene deletions/duplications are not detected. For issues to consider in interpretation of sequence analysis results, click here.

Gene-targeted deletion/duplication analysis detects intragenic deletions or duplications. Methods used may include a range of techniques such as quantitative PCR, long-range PCR, multiplex ligation-dependent probe amplification (MLPA), and a gene-targeted microarray designed to detect single-exon deletions or duplications.

No whole-gene deletions or duplications involving COL9A1, COL9A2, COL9A3, or COMP have been reported to cause autosomal dominant MED.

A tandem duplication involving exons 2-5 was reported in one individual with MED [Pettersson et al 2018].

Pathogenic variants remain undetected in approximately 20% of individuals with MED. Rarely, individuals with clinical and radiographic features that overlap MED and mild spondyloepiphyseal dysplasia congenita have been found to have heterozygous COL2A1 pathogenic variants [Jackson et al 2012, Dasa et al 2019, Luo et al 2022].