by Kirk, Edwin;
[10 Not Down’s syndrome, by the way. Although the possessive in syndrome names lingers on, it has been considered incorrect since at least the early 1980s. Down didn’t own the syndrome, nor did he have the condition himself. Perhaps the only eponym that still deserves its possessive is Trousseau’s sign. Trousseau, a 19th-century French physician, noticed that people with cancer sometimes have episodes in which clots form in blood vessels under the skin and cause inflammation; the problem moves from place to place on the body, and is thus known as migratory thrombophlebitis. The unfortunate Trousseau diagnosed his own stomach cancer when he developed migratory thrombophlebitis, making this truly ‘Trousseau’s sign’.]
Down points out that ‘the maternal imagination’ might have identified a number of possible causes, which may or may not be relevant. This rings true today: it’s very common that the parents of a child born with a malformation, or diagnosed with a disability, wonder whether some incident during the pregnancy was responsible — be it stress, occupational exposures to chemicals, or fumes from painting the nursery. One of the most useful things I can do is to provide reassurance about this, because it’s very seldom that the source of concern is actually the cause of the child’s problems.
Where Down goes off the rails, from a modern perspective, is in the way he classifies the different conditions he sees: by race. He identifies ‘examples of the Ethiopian variety’ that he calls ‘white negroes, although of European descent’, and others of the ‘Malay variety’; at this remove, it’s not clear which conditions he is referring to. However, it is very clear what he means when he refers to ‘the great Mongolian family’, because the term ‘mongolism’ for Down syndrome was common at least until the 1960s, and probably later. Down saw similarities in the facial appearance of some of his charges with the faces of people of central Asian descent, and concluded that his patients were in fact Asian. Charmingly, he goes on to conclude that, if this is the case, the differences between the races are only superficial, and that his observations ‘furnish some arguments in favour of the unity of the human species’.
With 150 years of scientific progress between us and John Langdon Down, it is easy to be critical of his premises, and his conclusions. But his famous paper (and subsequent writing) includes some useful and accurate clinical information. As he correctly observes, Down syndrome is present from birth, and never results from accidents after birth. He describes some physical characteristics, and also features of the personality that often accompanies the condition: ‘they are humorous, and a lively sense of the ridiculous often colours their mimicry’. He points out that affected individuals are usually able to speak, and he makes the point that, given the opportunity, they can learn manual skills. Down’s is a fundamentally humane approach to his subjects, a standard that has not always been met by his successors.
In non-human biology, there is a rigid rule about naming new species: priority is everything. The first to describe a species gets naming rights. This rule dates back to the efforts of Hugh Edwin Strickland, who in 1837 proposed a set of rules in an attempt to reign in a then-chaotic situation. At the time, the naming of species was in a parlous state, with people attempting to rename species and many species having numerous different names. Strickland, a naturalist, was spurred into action when he read a 1934 proposal to rename the bullfinch to ‘coalhood’. Black though the head of a bullfinch may be, this was the final straw for Strickland, who at once began work on an initial set of 22 rules, with primacy of description taking a prominent place.
It took several years for Strickland to persuade the British Association for the Advancement of Science to adopt the rules, and longer for the scientific community to adopt the standard. Later, Strickland’s rules evolved to become the current International Code of Zoological Nomenclature;11 for well over a century, there have been clear rules in place, not without occasional controversy regarding interpretation but nonetheless accepted by all12 — even when the rules lead to decisions that those of us who are not zoologists might find unfortunate. You may have heard, for instance, that the wonderfully-named dinosaur Brontosaurus (‘thunder lizard’) was deemed to be only a variant of Apatosaurus (‘deceptive lizard’, although it’s hard to see how deceptive an animal could be when it had an average length of over 20 metres and weight of around 20 tonnes). The name Brontosaurus was scrapped because Apatosaurus was described first, a blow to amateur dinosaur-lovers everywhere.13 Those of us who grew up dreaming of being palaeontologists rather than fire fighters may have found this a bitter pill to swallow, but at least there is a consistent set of rules.
[11 There is also an International Code of Botanical Nomenclature.]
[12 Sadly, Strickland did not live to see his triumph. He died in 1853, aged only 42, in a freak railway accident. He had gone to inspect geological strata exposed by the construction of a railway cutting, through which parallel train tracks ran. Seeing an oncoming freight train, he stepped aside onto the other line — only to be hit by an express travelling in the opposite direction.]
[13 A 2015 paper may have resurrected the name, arguing that Brontosaurus truly was a different beast from Apatasaurus. It appears the matter will not be settled until more fossils are discovered.]
The current system for naming syndromes has a certain charm, despite its capriciousness. But perhaps we in genetics should join the rest of biology in taking a more systematic approach to choosing names.
Regardless of how they come by their names, there are an awful lot of syndromes. This means that beyond a core set of conditions, plus some extras depending on the skill and experience of the dysmorphologist, if you want to do dysmorphology with success, you are going to need help. That help can come in the form of textbooks, like Smith and Gorlin. It can come in the form of searchable databases, like the London Dysmorphology Database and the Australian database, POSSUM — Pictures of Standard Syndromes and Undiagnosed Malformations. These can be very useful. As a new trainee in clinical genetics, the very first time I was asked to see a child with an unexplained syndrome, I came back from the intensive care unit to the genetics department, plugged the child’s clinical findings into POSSUM, and the correct diagnosis came up. ‘Dysmorphology is easy!’ I thought. It was years before that happened again.
More recently, increasing efforts have been made to use modified facial-recognition software to match patient photographs, including 3D photographs, to databases of patient photographs. The technology is starting to get quite good — just as other technology is coming along that will probably make it redundant (as we saw in chapter 5).
But in the end, the dysmorphologist’s best source of help is often other dysmorphologists. On Monday afternoons, the members of our department get together and discuss every patient we’ve seen in the past week. This review meeting is essential not just for dysmorphology but for all the rare situations we deal in; every patient gets a virtual second opinion (and third, fourth, and fifth). We show photographs of the patients we see: faces, hands, feet, birthmarks, whatever we think might be a clue. Dysmorphology has its own jargon — in describing the shape of the ear alone, we might speak of differences in the helix, the antihelix, the crus, the tragus, the concha … the list goes on. In 2009, Alasdair Hunter and colleagues published a 21-page paper, with over 70 figures, explaining how a dysmorphologist should describe the ear.
This was just one in a series of papers, the ‘elements of morphology’. We talk about handles — yes, I know you can grip someone by the ears, but these are metaphorical handles; features distinctive enough that they are likely to be helpful in reaching a diagnosis. An ear that’s a bit small isn’t much of a handle, because it’s such a common thing, including in people who are otherwise completely well. An ear that looks crumpled, as though it’s been squeezed in someone’s fist while developing — well, that’s a decent handle, and might be a big part of what leads you to a diagnosis.
As is true for any skill, some peo
ple are better dysmorphologists than others. Fortunately for me and for my patients, I work with several people who are very good at it indeed. There have been a number of occasions over the years when I’ve come from clinic without a clue about a particular patient, and someone at that afternoon’s review meeting has made a suggestion that led to the correct diagnosis.14 In particular, my colleague Rani Sachdev is a diagnostic bloodhound; give her a sniff of a diagnosis, and she won’t give up until she runs it to ground. Many a time I’ve come to work on a Tuesday morning to find an email from Rani, sent in the middle of the night, with a couple of journal articles attached and a message saying, ‘Ed, I was thinking about your patient from review yesterday and I was wondering about (syndrome X). Have a look at these papers and see what you think.’
[14 Only once or twice has it gone the other way — but I remember and treasure those exceptions.]
Even with all that expertise in the room, and even with textbooks, databases, and online resources … until recently our hit rate was pretty terrible. So when you think there’s a diagnosis to be made, but you can’t figure out what, and your colleagues can’t figure it out either (even Rani) — then perhaps it’s one for the Dysmorphology Club.
This is undoubtedly one of the world’s oddest clubs. You can’t apply for membership, but only members may attend. It follows that you also can’t resign, although, since attendance is voluntary, and a privilege, you can always stop coming, if you choose. The club meets twice a year. Its rules are few. Be brief. If you bring an unknown (pictures of a patient in whom you are hoping for a diagnosis), you should bring a known (pictures of a patient in whom you have made a diagnosis, preferably of a rare and little-known syndrome). Most importantly, respect the privacy of those who are not present, but who are essential to the club’s success: the patients and their families.
The proceedings of the club are always the same. We meet once a year at the annual scientific meeting of the Human Genetics Society of Australasia, and once in a standalone meeting somewhere in Australia. The attendees are clinical geneticists and trainees in clinical genetics; otherwise, the meeting is closed. There is a fairly strict limit of two patients per person, although, since trainees speak as well, you may be able to sneak extras in via your trainee’s allocation. Time is tight, but not everyone is good at being concise; the key features of the patient’s family history, medical history, examination findings, and previous investigations are summarised, and photographs are shown, all with the consent of the families or individuals concerned. The aim is to educate your colleagues and to get clues to a diagnosis for your patients. Instead of the handful of extra opinions at review meeting, you’re getting scores of expert eyes and minds on the case, often including a visiting international expert. It’s a great institution.
*
These, then, were the tools available to me in trying to make a diagnosis in Dianne. In the end, however, I made the diagnosis using a different set of tools altogether: laziness, and luck.
Before digital photography, we would take two cameras to clinic with us — a Polaroid Instamatic and a standard camera. The former was so that we would have at least some photographs available to pass around at review meeting, the latter so that we had higher quality pictures for the patient’s file. At least once a week, we’d go up the road and drop rolls of film off to be developed. The prints would be distributed to the relevant geneticist or trainee, who would then make sure they were put in the correct file.
Or, if the geneticist was indolent or a procrastinator, he would just leave the photographs lying on his desk, awaiting the day that he got round to doing something about them. So it happened that Dianne’s photograph was sitting smiling up at me on top of a pile of paperwork, for … quite a few months. When you look at someone’s photograph several times a day for months on end, you wind up knowing exactly what they look like, which in this case proved to be a good thing.
One day I went to the room where we kept the various genetics journals we subscribed to, in order to look up an article on a completely different topic. I was leafing through the relevant copy of the American Journal of Medical Genetics — and suddenly, there was Dianne, staring back at me. The child in the article looked like her twin! Excitedly, I read the article, which described the condition that came to be known as Cantú syndrome. Affected children were born prematurely, weighing more than expected for their gestation. They had an excess of hair, large hearts, often with a patent (open) ductus arteriosus, and a distinctive facial appearance. It was all an exact match with Dianne’s problems. There was no doubt in my mind that this was the answer.
At the next Dysmorphology Club, I showed pictures of Dianne as a known diagnosis, expecting that I would stump my assembled colleagues.15 There had only been a handful of people with Cantú syndrome described in the medical literature; the name wasn’t even firmly established yet. To my surprise, Stephen Robertson — a New Zealander — spoke up from the back of the room, correctly identifying the diagnosis. He had recently made the same diagnosis in one of his patients, using the more traditional approach of applying actual diagnostic skill, rather than relying on the approach that I had used. Steve and I went on to write a paper together, describing our patients.
[15 Okay … dysmorphology club does have a third goal. Showing off.]
What did this diagnosis mean for Dianne, and for her family? At first, not as much as you might think. There were very few children who had been described with the syndrome, and we were learning far more from Dianne and from the others we started seeing at the time than the families could learn from us. As a result, Dianne has featured in four different scientific papers. Still, there was some useful information we could give her parents. Dianne had been developing a little slowly, and in particular she was slow to sit and to walk. We knew already that this was a common pattern in children with Cantú syndrome, but that at least some of them went on to have normal development and normal intelligence.16 This meant we could provide some cautious reassurance about Dianne’s future.
[16 Later, we were to learn that most people with Cantú syndrome have normal intelligence. I’ve met a psychologist and a doctor with Cantú, and we now know that intellectual disability is the exception, not the rule.]
Parents of a child newly diagnosed with a syndrome typically are concerned to know what caused the problem, and whether it might happen again if they have more children. Until then, it had been thought that this was an autosomal recessive condition: that both parents needed to be carriers, and that the future siblings of an affected child would have a 1 in 4 chance of also being affected. When Steve and I were writing our paper, we read the whole of the medical literature about the syndrome — not that difficult, as ours was only the sixth paper published. Only 12 people had previously been described. The reason this had been thought to be a recessive condition was that in the first paper, describing the syndrome, two affected children had been born to unaffected parents. In another family, the parents of an affected child were cousins — a pointer to recessive inheritance, because we share genes with our relatives (and thus are more likely to share the same faulty gene).
By 1997, Cantú and his group were expressing uncertainty about this, because no more families had been seen with more than one affected child. When we wrote our paper in 1998, I was able to count 39 brothers and sisters of people with Cantú, only one of whom — from that very first paper — had the condition as well. We would have expected about nine or ten, a big difference from just one, and strong evidence that recessive inheritance was unlikely. We now know that this is an autosomal dominant condition — only one of the two copies of the gene needs to be faulty to cause Cantú syndrome. Sometimes, the faulty gene is inherited from an affected parent, who in turn may have inherited it from one of their parents; sometimes, the change has arisen anew in the egg or sperm that joined to form the affected child. So how could two children be born to unaffected parents? Almost certainly
, this was due to one of the parents of these children being a gonadal mosaic (as you’ll recall, mosaicism was described in chapter 5).
The story of Cantú syndrome over the 20 years that have followed Dianne’s first appearance in the scientific literature has been one of international collaboration and friendship, of dazzling scientific insight and slow, patient progress. In 2006, an American geneticist, Kathy Grange,17 made the inspired observation that people with Cantú syndrome have a lot in common with people who have been treated with a drug called minoxidil. Minoxidil is now rarely used for its original purpose, treating high blood pressure, but is still very popular (in cream form) because it makes hair grow, a salve for the balding. Kathy somehow made a link between the side effects of an increasingly obscure treatment and a genetic condition.
[17 Known in her scientific publications and email address as Dorothy K. Grange, but Kathy to her friends.]