SEARCH

SEARCH BY CITATION

An anniversary always provides some opportunity for reflection on what has been done so far and how it could be done better in the future. Naming is at the core of our daily practice. Looking at the hundreds of studies published by the BJD over the past 125 years, it is clear that naming is also central to much of what we write about the diseases we diagnose and treat. Giving a name to their ailment is what our patients expect from us; this is also what we are aiming at because diagnosis is key to proper treatment design. At first, it may seem that there is nothing more trivial than naming things, people or… diseases… After all, this is something we all do on a daily basis from an early age. However, naming can be a complex process, and occasionally the source of much confusion. The terms ‘mouse’ or ‘mailbox’ are nowadays used more often to designate electronic and quasi-virtual entities rather than a small mammal or a familiar reddish box ornamenting the entrance to our houses. Words very often carry much more than their primary meaning. In fact, the more they carry, the more confusing they can be. And what is true of daily words is, of course, also true of disease names, which is why the more we know about a disease, its clinical spectrum of manifestations or its pathogenesis, the more difficult it is to give it a single, unequivocal and clinically useful name. And as diagnosis is all about giving a name to a constellation of clinical features, novel and quite dramatic advances in our understanding of the biological basis of diseases are paradoxically threatening the very foundations of the diagnostic process.

Recent developments in genodermatology are most remarkably illustrating this situation, as evidenced at each level of the four phases of the dermatological diagnostic process and exemplified by the following short stories.

History taking

  1. Top of page
  2. History taking
  3. Physical examination
  4. Histopathology
  5. Genetic testing
  6. References

In the past, a most diagnostically useful piece of information you needed to retrieve from any patient suspected to be affected by a genetic skin disease was its mode of inheritance. Unfortunately, it has now been shown that a wide range of modes of inheritance exists for most cutaneous traits so that inheritance, like other medical history elements, is becoming less and less useful to define and name diseases. For example, keratin disorders have been considered as prototypic examples of autosomal dominant disorders (with mutations in keratin genes typically exerting a dominant negative effect).[1, 2] It is now clear that these disorders can be inherited in a recessive,[3] semi-dominant[4] and even digenic[5] manner. Co-inheritance of mutations in more than one gene seems, in fact, to be the rule rather than the exception in what appears to have been wrongly named the ‘monogenic’ diseases.[6-8]

Physical examination

  1. Top of page
  2. History taking
  3. Physical examination
  4. Histopathology
  5. Genetic testing
  6. References

Physical signs can also be deceiving. Clinical entities designated under distinct names have now been recognized to reflect different manifestations of the same pathophysiology, which, in turn, predict therapeutic responses, as with the various syndromes featuring in different combinations acne, pyoderma gangrenosum, colitis, hidradenitis suppurativa and arthritis, all caused by mutations in the PSTPIP1 gene, and, as such, all likely to respond to anti-interleukin-1 agents.[9-12] Conversely, a well-defined phenotype, such as epidermolysis bullosa with mottled pigmentation, which was originally thought to reflect a specific defect in keratin 5 function,[13] has now been found to be caused by mutations in other keratin[14] and even nonkeratin[15] genes. These and many similar data cast doubt as to the role of anamnestic data and physical features in the diagnosis of genodermatoses. Once considered as the cornerstone of this process, they should now be recognized for what they really are, often useful, as well as potentially misleading, hints.

Histopathology

  1. Top of page
  2. History taking
  3. Physical examination
  4. Histopathology
  5. Genetic testing
  6. References

Histopathological features are no better. Take, for example, psoriasis and pytiriasis rubra pilaris, two disorders we have been used to consider as characterized by very divergent histological features. Recent data indicate that both disorders are characterized by overexpression of CARD14, an activator of nuclear factor-kappa B,[16-19] which predicts, in turn, a good therapeutic response to drugs targeting this signalling pathway. However, diseases associated with sometimes radically different aetiologies can share the very same histological features. For example, subcorneal acantholysis has been reported in staphylococcal scalded skin syndrome, caused by a bacterial toxin, in pemphigus foliaceus, caused by autoantibodies and in sinobrachial allergic mycosis syndrome, caused by mutations in the desmoglein 1 gene.[20, 21]

Genetic testing

  1. Top of page
  2. History taking
  3. Physical examination
  4. Histopathology
  5. Genetic testing
  6. References

Even molecular data may be confusing. Mutations in the same gene can lead to widely different phenotypes: as a recent example, mutations in the ENPP1 gene can lead, at the same time, to Cole disease,[22] a disorder featuring punctate keratoderma and hypopigmented patches, as well as to various forms of hereditary ectopic mineralization.[23, 24] In some cases, the fact that different phenotypes result from mutations in the same gene has been explained on the basis of the domain affected by the causative mutations such as with ichthyosis hystrix of Curth Macklin, resulting from mutations affecting the tail domain of keratin 1, a gene otherwise associated with epidermolytic ichthyosis.[25] The mechanism of action of the causative mutations has also been invoked to account for some of these discrepant observations: mutations in KRT5 cause either epidermolysis bullosa simplex when exerting a dominant negative effect or Dowling-Degos disease when leading to haploinsufficiency.[26] However, these explanations do not help us understand why, in many other cases, the same mutation in the same gene can have a strikingly different effect: the p.R388H mutation in KRT14 can either be silent (i.e. not cause any clinical phenotype) or result in epidermolysis bullosa simplex;[3] the same mutation in GALNT3 has been shown in individuals sharing an identical genetic background to cause either familial tumoral calcinosis (a skin disease characterized by the formation of subcutaneous calcified deposits)[27] and hyperostosis–hyperphosphatemia syndrome (a bone disorder).[28] Clearly, in some of these cases, co-inheritance of modifier gene mutations may underlie phenotypic variability as previously shown in X-linked ichthyosis[29] and in dystrophic epidermolysis bullosa.[30]

Collectively, the data reviewed above tend to suggest that the way we are currently used to naming and cataloguing diseases may be outdated and that it may be time for us to move from a morphology-focused to a pathogenesis-based approach to disease classification. Not only would such a new classification scheme facilitate the diagnostic process, it is also likely to benefit our patients by directly suggesting treatment strategies of potential benefit. Indeed, it should make little difference for a clinician whether a patient's phenotype better fits a diagnosis of psoriasis or of pityriasis rubra pilaris at the clinical or histopathological levels; what does matter is the functional defect underlying the patient's disease, as it is the best predictor of therapeutic response.

Change is in the air, and some initial steps have already exemplified the re-grouping of numerous clinical entities under new appellations such as the ‘RASopathies’[31, 32] and the ‘p63 syndromes’.[33] I have no doubt that the next anniversary of the BJD will provide a great opportunity to reflect on the effect that the rewriting of our dermatological classification systems has had on the way we treat our patients. For the time being, we should enjoy being part of the process…

Funding sources

No external funding.

Conflicts of interest

None declared.

References

  1. Top of page
  2. History taking
  3. Physical examination
  4. Histopathology
  5. Genetic testing
  6. References