Nosology and classification of genetic skeletal disorders: 2015 revision

The purpose of the nosology is to serve as a “master” list of the genetic disorders of the skeleton to facilitate diagnosis and to help delineate variant or newly recognized conditions. This is the 9th edition of the nosology and in comparison with its predecessor there are fewer conditions but many new genes. In previous editions, diagnoses that were phenotypically indistinguishable but genetically heterogenous were listed separately but we felt this was an unnecessary distinction. Thus the overall number of disorders has decreased from 456 to 436 but the number of groups has increased to 42 and the number of genes to 364. The nosology may become increasingly important today and tomorrow in the era of big data when the question for the geneticist is often whether a mutation identified by next generation sequencing technology in a particular gene can explain the clinical and radiological phenotype of their patient. This can be particularly difficult to answer conclusively in the prenatal setting. Personalized medicine emphasizes the importance of tailoring diagnosis and therapy to the individual but for our patients with rare skeletal disorders, the importance of tapping into a resource where genetic data can be centralized and made available should not be forgotten or underestimated. The nosology can also serve as a reference for the creation of locus‐specific databases that are expected to help in delineating genotype–phenotype correlations and to harbor the information that will be gained by combining clinical observations and next generation sequencing results. © 2015 Wiley Periodicals, Inc.

The purpose of the nosology is to serve as a "master" list of the genetic disorders of the skeleton to facilitate diagnosis and to help delineate variant or newly recognized conditions. This is the 9th edition of the nosology and in comparison with its predecessor there are fewer conditions but many new genes. In previous editions, diagnoses that were phenotypically indistinguishable but genetically heterogenous were listed separately but we felt this was an unnecessary distinction. Thus the overall number of disorders has decreased from 456 to 436 but the number of groups has increased to 42 and the number of genes to 364. The nosology may become increasingly important today and tomorrow in the era of big data when the question for the geneticist is often whether a mutation identified by next generation sequencing technology in a particular gene can explain the clinical and radiological phenotype of their patient. This can be particularly difficult to answer conclusively in the prenatal setting. Personalized medicine emphasizes the importance of tailoring diagnosis and therapy to the individual but for our patients with rare skeletal disorders, the importance of tapping into a resource where genetic data can be centralized and made available should not be forgotten or underestimated. The nosology can also serve as a reference for the creation of locus-specific databases that are expected to help in delineating genotypephenotype correlations and to harbor the information that will be gained by combining clinical observations and next generation sequencing results. Ó 2015 Wiley Periodicals, Inc.

INTRODUCTION
The publication of a nosology of skeletal dysplasias started 45 years ago in Paris and has seen multiple revisions [1970, 1971a,b, 1979, 1983, 1998Hall, 2002;Lachman, 1998;McKusick and Scoot, 1971;Rimoin, 1979;Spranger, 1992;Superti-Furga and Unger, 2007;Warman et al., 2011] The current nosology revision took place in Bologna, Italy just prior to the 11th International Skeletal Dysplasia Society meeting organized by Professor Luca Sangiorgi. In the 2015 version of the nosology, the number of conditions has decreased while the number of genes has increased dramatically. This is a reflection of consolidation of repeat entries into a single one when there is no discernible phenotypic difference while at the same time acknowledging the discovery of new genes. The inclusion of MIM numbers is maintained as this invaluable database is often a first reference for clinicians. There is not a complete concordance between MIM and the nosology because of different inclusion and review criteria and thus MIM retains some obsolete diagnoses and duplicates others (under differing names or eponyms).
This version of the nosology is the 9th edition and while it contains several new disorders, it is not radically different from its predecessor [Warman et al., 2011]. The groups of disorders remain a hybrid mix as they are defined either by a single gene or group of related genes (e.g., FGFR3 chondrodysplasia group and sulphation disorders group), or by a particular phenotypic feature (e.g., dysplasias with multiple joint dislocations), or by some radiological finding (e.g., metaphyseal dysplasia group and slender bone dysplasia group).
When the concept of the skeletal dysplasia families was first elaborated, it was hoped that there would be a limited number of molecular based groups with each group containing multiple allelic disorders [Spranger, 1985]. However, the biology of the skeletal dysplasias has turned out to be much richer, and more complex than anticipated. So while it makes sense to have a type 2 collagen disorder group where there is some similarity between conditions but enough phenotypic difference to warrant separate diagnoses (e.g., Stickler syndrome versus achondrogenesis type 2), there are many other genes that, to the best of our knowledge, are not associated with a "skeletal dysplasia family," those with no wide spectrum (e.g., SEDL (Spondyloepiphyseal dysplasia tarda) or Spondyloepimetaphyseal dysplasia with joint laxity-leptodactylic type). For these genes and conditions, it still makes sense to group them with clinically or radiographically similar disorders. Table I has been simplified with the columns "locus" and "gene" merged into one. For some disorders, the etiology is a copy number disturbance and thus they are not single gene disorders in the classic sense. For those disorders with a known causative gene, the chromosomal location of that gene is often not important (especially if it is an autosome), and when necessary, the information can be readily retrieved from public databases.
The criteria used for inclusion of disorders are unchanged from the previous revision [Warman et al., 2011]. They are: 1) Significant skeletal involvement, corresponding to the definition of skeletal dysplasias, metabolic bone disorders, dysostoses, and skeletal malformation and/or reduction syndromes. 2) Publication and/or listing in MIM (observations, even those by experts in the field should not find their way into the nosology before they have achieved peer-reviewed status). 3) Genetic basis proven by pedigree or very likely based on homogeneity of phenotype in unrelated families. 4) Nosologic autonomy confirmed by experimental analysis.
We have included conditions in which only one family has been described but for which the gene has been identified. For e.g., the heterozygous mutations in FZD2 in dominant omodysplasia [Saal, et al., 2015].
The total number of diseases has gone down (from 456 to 436) thanks to grouping of phenotypically indistinguishable entities and despite the appearance of several new conditions (e.g., MAGMAS related skeletal dysplasia) [Mehawej et al., 2014].
A few groups have changed names in this edition and the overall number has increased from 40 to 42. The short-rib dysplasia (with or without polydactyly) group has become the ciliopathies with major skeletal involvement group. Due to the increasing number and complexity of the brachydactylies, the group has now been made into two separate categories: brachydactylies without extraskeletal manifestations and brachydactylies with extraskeletal manifestations. The ectrodactylies have been given their own group.
The field of osteogenesis imperfecta (OI) continues to expand with multiple new genes. We have chosen to stick with the Sillence classification that was phenotypically and not molecularly based [Sillence and Rimoin, 1978;Sillence et al., 1979]. For this reason, OI type 5 is included as it is radiologically distinguishable from types 1 through 4. OI is the archetype of a skeletal dysplasia for which molecular diagnosis relies on next generation sequencing but prognosis is based on the careful phenotypic observations collected over the last four decades [Van Dijk and Sillence, 2014]. Examples are also available from other domains of medical genetics (spino-cerebellar ataxia or Meckel-Gruber syndrome).

DISCUSSION
The pace of disease related gene discovery has accelerated phenomenally in recent years thanks to the development of nextgeneration sequencing technologies and increasing availability of whole exome sequencing. This has led to both expansion and contraction of the nosology. It has expanded to incorporate new genes and new conditions but also contracted as we recognize our limits in differentiating by phenotype. While each patient may be unique, there are clear advantages both medical and human to belonging to a group of similar individuals [Superti-Furga, 2014]. It is truly an exciting time as we struggle to correctly interpret the         Note: the particularly complex genetic basis of Fanconi anemia and its complementation groups is acknowledged but not further listed in this Nosology. The (Continued)