Medical Genetics Department

What our medical genetics unit covers — and who it is for

Medical genetics is the specialty of inherited conditions and inherited risk. It answers a different kind of question from the rest of medicine: not what is wrong with you now, but what you were born with, what it means for the years ahead, and who else in your family it concerns.

That last clause is what makes the specialty unusual. A result here rarely belongs to one person. It reaches parents, siblings and children who never asked for it, and much of the work is deciding what to test for, what to report, and how to hand a family information that some of them may not want.

At Acıbadem International the work falls into five strands.

• Genetic counselling — the appointment that comes before any test, establishing what question is being asked and whether a test can answer it.
• Diagnostic testing — finding the cause of a condition already present, in a child or an adult.
• Predictive and carrier testing — testing a healthy person for future risk or for a variant they could pass on.
• Inherited cancer risk — identifying the families in which cancer runs, and turning that into surveillance rather than anxiety.
• Family cascade testing — extending a confirmed result outwards through relatives, which is where most of the benefit of a diagnosis is realised.

Where the borders sit. Testing the tumour rather than the person — the somatic panels that select a cancer treatment — belongs with pathology, which sets that boundary out explicitly, and the treatment itself with medical oncology. Screening in pregnancy, NIPT, invasive testing and the counselling around consanguinity belong with perinatology. Preimplantation genetic testing of embryos belongs with IVF and reproductive health. What BRCA status means for breast surgery and surveillance belongs with breast health. What this unit owns is the decision to test, the interpretation of the result, and the family.

What a genetic counselor does, and why the appointment comes first

A genetic counselor — genetic counsellor in British usage, the same role under either spelling — is the reason genetic testing is not simply a laboratory order. The appointment happens before the test rather than after it, and that sequence is deliberate.

Three things happen in it. A family history is taken properly, going back three generations with ages, diagnoses and ages at diagnosis, because the pattern in that history frequently determines which test is worth doing at all. The question is defined: predicting future risk, explaining a condition already present, and assessing what could be passed to children are three different questions needing three different tests. And the consequences are discussed before the result exists, because that is the only time they can be considered calmly — what a positive result would change, what a negative one would not rule out, and who else would need to know.

The last of those is the one people underestimate. A predictive test result cannot be un-known, and it may have implications for siblings and children who have not been asked whether they want the information. Deciding not to test, after a proper discussion, is a legitimate outcome of a genetics appointment rather than a failure of one.

The counsellor also handles the part that is neither technical nor trivial: a result arriving in a family that has been watching a condition move through it for generations lands differently from a result in a family with no history at all.

How conditions are inherited

Most of the confusion patients arrive with dissolves once the inheritance pattern is clear, so it is worth setting out plainly.

Autosomal dominant

An autosomal dominant condition needs only one altered copy of a gene to cause it. An affected parent has a one in two chance of passing it to each child, independently for each pregnancy, and the condition typically appears in every generation. Huntington disease, Marfan syndrome, familial hypercholesterolaemia and most inherited cancer syndromes follow this pattern.

Two points cause repeated misunderstanding. One in two is a per-pregnancy probability, not a quota — three children can all inherit it, or none can. And a child who does not inherit the variant cannot pass it on: the line stops there, permanently.

Autosomal recessive

An autosomal recessive condition needs two altered copies, one from each parent. Both parents are usually healthy carriers with no family history at all, which is why these conditions appear out of nowhere. Each pregnancy carries a one in four chance of being affected, a one in two chance of being an unaffected carrier, and a one in four chance of neither. Cystic fibrosis, thalassaemia, sickle cell disease and most metabolic conditions follow this pattern.

Because both parents must carry a variant in the same gene, relatedness between parents raises the chance of both carrying the same rare one. That consequence for pregnancy, and the counselling around it, is covered by perinatology.

X-linked

An X linked inheritance pattern follows a gene on the X chromosome. Males, having one X, are affected when they inherit an altered copy; females, having two, are frequently unaffected carriers, although some are affected to a variable degree. The characteristic clue is a condition passing through healthy women to affected men — brothers, maternal uncles, cousins — with no father-to-son transmission, because a father gives his son a Y.

Mitochondrial and de novo

Mitochondrial conditions pass only from the mother, since sperm contribute essentially no mitochondria — an affected father cannot pass one on. And a de novo mutation is one that arose newly in the affected person and is present in neither parent, which explains the many severe genetic conditions in families with no history whatsoever. A de novo finding usually means a low recurrence risk for the parents’ next pregnancy, though not zero.

Penetrance: why a gene is not a diagnosis

This is the single most important idea in the specialty and the one most often lost between the report and the patient.

Penetrance is the proportion of people carrying a variant who actually develop the condition. It is rarely complete. Someone can carry a disease-causing variant, pass it to a child who is severely affected, and never develop anything themselves. Expressivity is the companion idea: among those who are affected, severity differs enormously, sometimes within one family carrying one identical variant.

Three consequences follow, and they are practical rather than theoretical.

A positive predictive test states that risk is raised, not that the condition will occur. Reading it as a diagnosis causes real harm — treatment decisions made on a probability presented as a certainty.

An unaffected relative can still be a carrier and can still pass it on, which is why cascade testing follows the gene rather than the symptoms.

And risk figures attached to a variant are population estimates, refined by family history, sex, age and sometimes other genes. Two people with the same variant do not necessarily carry the same risk.

Mosaicism adds a further layer: a variant present in some cells and not others, depending on when it arose during development. It explains why a parent can test negative on a blood sample yet have more than one affected child, because the variant may be present in their egg or sperm cells alone. It is uncommon, and it is the reason a negative parental test does not always reduce recurrence risk to background.

The tests, and what each can and cannot see

Genetic testing is not one test. Each type looks at a different scale, and a negative result means only that nothing was found at the scale examined.

• Karyotype — the chromosomes seen whole under a microscope. It detects extra or missing chromosomes and large rearrangements. It cannot see anything smaller.
• Chromosomal microarray — detects deletions and duplications far below the resolution of a karyotype, and is a first-line test in developmental delay and congenital anomaly. It cannot detect a single-letter change or a balanced rearrangement.
• Targeted single-gene testing — used when the diagnosis is already suspected clinically or when testing a relative for a known family variant.
• Gene panel — many genes associated with one condition or group of conditions, sequenced together. The standard approach for inherited cancer risk, cardiac conditions and hearing loss.
• Whole exome sequencing — the protein-coding portion of every gene, about one to two per cent of the genome, where most known disease-causing variants sit. Used when the presentation does not point to a specific gene.
• Whole genome sequencing — everything, including the regions between genes. Detects some things an exome misses, at the cost of far more uncertainty to interpret.

Two properties of these tests are worth holding on to. A negative result is bounded by the method: an exome that finds nothing has not excluded a deletion a microarray would have seen, and a panel that finds nothing has not excluded a gene outside the panel. And the bigger the test, the more uncertainty it generates — which is the subject of the next section.

The variant of uncertain significance

The variant of uncertain significance — the VUS — is the most misread result in genetics, and understanding it prevents a specific and common harm.

It is also the reason genetic test results are read as a classification rather than as a yes or no. Laboratories classify variants on a five-point scale: pathogenic, likely pathogenic, uncertain, likely benign and benign. A VUS means exactly what it says — the laboratory does not currently know whether this change causes disease. Most turn out to be benign when enough data accumulates.

The critical rule follows directly: a VUS is not acted upon clinically. It does not justify risk-reducing surgery, it does not justify altered surveillance, and it is not a reason to test relatives for reassurance. Management continues to be based on personal and family history as though the variant had not been found — a statement that surprises people and is nonetheless the standard.

Two further points. Classifications are revisited as evidence accumulates, and a VUS can later be reclassified in either direction, which is why staying in contact with the genetics service matters more than the original report suggests. And VUS results are considerably more common in people from populations that are under-represented in genomic databases, simply because there is less comparison data — an inequity in the science rather than a feature of the individual.

Inherited cancer risk

Most cancer is not inherited. A minority arises in families carrying a variant that raises risk substantially, and identifying those families is one of the highest-value activities in preventive medicine — because the answer is surveillance and risk reduction rather than treatment.

What raises suspicion

Cancer diagnosed unusually young. The same cancer in several close relatives. More than one primary cancer in one person. A rare pattern such as male breast cancer. Certain tumour types that carry a high inherited fraction regardless of family history. And specific combinations across a family — breast with ovarian, or bowel with womb.

The main syndromes

BRCA1 and BRCA2 raise the risk of breast and ovarian cancer substantially, and also of prostate and pancreatic cancer. What a positive result means in practice — enhanced surveillance, risk-reducing surgery, timing — is managed by breast health and gynaecology. A BRCA test is therefore rarely ordered alone: the combination of breast and ovarian cancer in a family is what defines hereditary breast and ovarian cancer syndrome, and the panel used reflects that. Beyond BRCA, several moderate-risk genes including PALB2, CHEK2 and the ATM gene appear on panels and carry different, generally lower risks, which is why a panel result requires interpretation rather than a single response.

Lynch syndrome is caused by variants in the mismatch repair genes and raises the risk of bowel, womb, ovarian, stomach and urinary tract cancer. It is frequently identified from the tumour itself, when mismatch repair testing on a specimen shows deficiency — the point at which pathology hands a family to this unit. Surveillance colonoscopy at intervals shorter than the general population’s is the main intervention.

Familial adenomatous polyposis, Li Fraumeni syndrome — caused by a germline TP53 mutation and raising the risk of several cancers from childhood onwards — von Hippel-Lindau disease and the multiple endocrine neoplasia syndromes MEN1 and MEN2 are rarer and each carries its own surveillance programme, typically beginning in childhood or early adult life. MEN2 is unusual in that a positive result can lead to preventive thyroid surgery in childhood, which is one of the clearest examples of a genetic result changing what happens rather than only what is known.

What a negative result does and does not mean

This distinction matters and is routinely lost. A true negative — testing negative for a variant already identified in the family — genuinely returns that person to population risk. An uninformative negative — testing found nothing, and no variant was ever identified in the family — does not: the family history remains, the cause may be a gene not yet known, and surveillance continues to be based on the history rather than on the report.

Cascade testing: a result belongs to a family

Cascade testing is what happens after a variant is confirmed in one person: relatives are offered testing for that specific variant, and those who test positive enter surveillance while those who test negative are released from it.

It is where most of the benefit of a genetic diagnosis is realised, and it is also where most of it is lost, because a family conversation has to happen for it to occur at all. Genetics services support that conversation — with letters written for relatives, explanations pitched to be shared, and the option of contact being made in some circumstances — but they cannot have it on a patient’s behalf.

Testing a relative for a known family variant is a far simpler exercise than the original test, and the result is far cleaner: a single yes or no about a single defined change, with none of the ambiguity of an open-ended search.

Two situations need care rather than enthusiasm. Testing children is generally deferred for adult-onset conditions where nothing would be done in childhood, so that the person can decide for themselves; where childhood surveillance would begin, testing is offered in childhood. And relatives who choose not to know are entitled to that, which is a position to be supported rather than argued with.

Carrier status, and what it means for you

Being a carrier of a recessive condition means having one altered copy and one working one. The working copy is enough, so carriers are healthy — the finding matters for children rather than for the carrier.

Sickle cell trait is the clearest example and the most misunderstood. It is not a mild form of sickle cell disease. People with the trait are healthy, have a normal life expectancy and do not develop the disease. Two well-documented qualifications belong with that: extreme conditions such as severe dehydration combined with very intense exertion at altitude can occasionally provoke problems, and the trait is associated with a rare kidney tumour and with blood in the urine. Neither changes the central point, and neither is a reason for a carrier to alter their life.

Cystic fibrosis carrier status behaves the same way, as does thalassaemia trait — the last being important in this region, where it is common enough that couples are frequently both carriers without knowing. Thalassaemia trait is also mistaken for iron deficiency on a blood count and treated with iron that does not help, which is a diagnosis worth making correctly.

Where a couple wants to know their position before conceiving rather than react to it afterwards, expanded carrier screening tests both partners across a large number of recessive conditions at once, rather than only the ones suggested by ancestry or family history.

What carrier status changes is reproductive: if both partners carry a variant in the same gene, each pregnancy carries a one in four chance of an affected child, and the options available are discussed before pregnancy rather than during it. Testing in pregnancy belongs with perinatology and embryo testing with IVF and reproductive health.

Inherited metabolic disease

Inborn errors of metabolism are individually rare and collectively not, and they matter disproportionately for one reason: several are treatable, and the treatable ones respond far better when found early.

Pompe disease is the clearest illustration. It is caused by deficiency of an enzyme that breaks down glycogen, and it presents in two very different ways: an infantile form with severe muscle weakness and heart involvement, and a late-onset form that can appear at any age as progressive proximal muscle weakness and breathing difficulty — frequently mistaken for a muscular dystrophy or an unexplained myopathy for years. It is diagnosed by an enzyme assay on a dried blood spot followed by genetic confirmation, and enzyme replacement therapy exists. That combination — a treatable condition that is routinely diagnosed late — is why it appears in the differential of adult unexplained proximal weakness.

Gaucher disease and Fabry disease follow the same shape: lysosomal storage disorders with adult presentations that are missed for years, and available enzyme replacement. Fabry in particular presents with unexplained kidney failure, early stroke, cardiac hypertrophy and neuropathic pain in the hands and feet, and reaches nephrology, neurology and cardiology long before it reaches genetics.

Phenylketonuria is the historical proof of the principle: detected on newborn screening and treated with a strictly controlled diet, it prevents severe intellectual disability entirely — and the dietary management, lifelong and demanding, sits with nutrition and dietetics.

Newborn screening

Newborn screening is the heel-prick blood spot taken in the first days of life, and it is the most successful genetic programme in existence for a reason that is easy to state: it looks for a small number of conditions in which early treatment prevents severe, irreversible harm and in which nothing visible is wrong at birth.

The conditions differ by country, which matters for families who move. What is universal is the principle: this is screening, not diagnosis. An abnormal screening result means further testing is needed, and most abnormal results turn out not to be the condition — an important thing to know for a parent contacted after a screening test.

It also does not screen for everything genetic. A normal newborn screen excludes the specific conditions on that country’s panel and nothing beyond them, which is why it does not answer a question about family history.

Chromosome conditions

These arise from an extra or missing chromosome, or from a structural rearrangement, and they are usually de novo rather than inherited.

Down syndrome, trisomy 18 and trisomy 13 result from an extra copy of a whole chromosome and are usually diagnosed before or shortly after birth. Turner syndrome, in which one X chromosome is wholly or partly missing, presents with short stature, absent puberty and infertility, and is frequently diagnosed late — sometimes in adolescence or adulthood — which is a diagnostic failure worth naming. Klinefelter syndrome, an extra X in males, is also under-diagnosed and typically presents with infertility or delayed puberty.

One category is inherited and matters for reproduction: a balanced translocation, in which chromosome material is rearranged without being gained or lost. The carrier is entirely healthy, because nothing is missing. But their eggs or sperm can carry an unbalanced arrangement, which is why a balanced translocation surfaces after recurrent miscarriage or a chromosomally abnormal pregnancy. It is invisible on a microarray and requires a karyotype to see — one of the few remaining situations where the older test is the right one.

Fragile X and the repeat expansion disorders

A small group of conditions is caused by a short DNA sequence repeated too many times, and they behave unlike anything else in genetics.

Fragile X syndrome is the commonest inherited cause of intellectual disability and of autism with a known genetic cause. Its distinctive feature is the premutation: an intermediate-length repeat that does not cause the syndrome but can expand when passed on, particularly through mothers — so a healthy woman can have an affected child. The premutation itself is not silent either; it carries a risk of early ovarian insufficiency in women and of a late-onset tremor and ataxia syndrome in older men, both of which are missed because the family is not thought of as a fragile X family at all.

Huntington disease is the other well-known example and the one that shaped how predictive testing is done everywhere. It is autosomal dominant, adult-onset, and currently untreatable in the sense of altering its course — so a predictive test tells a healthy person whether they will develop it, with nothing to offer in response. That is why testing follows a formal protocol with counselling before and after and a deliberate interval to reconsider, and why a substantial proportion of people at risk choose not to be tested. That choice is respected without argument.

Inherited cardiac and connective tissue conditions

These reach genetics through a cardiologist or after a family event, and the reason they matter is that the first symptom is sometimes sudden death in a young person.

Long QT syndrome and the other inherited arrhythmia syndromes cause abnormal heart rhythms in structurally normal hearts. Hypertrophic cardiomyopathy thickens the heart muscle and is the commonest inherited cardiac condition. Both are managed by cardiology; genetics identifies the variant and then tests the family, which is the step that finds the relatives at risk before anything happens to them.

Marfan syndrome affects connective tissue, with consequences in the aorta, the eye and the skeleton. The aortic risk drives management, which sits with cardiovascular surgery, and the lens dislocation with ophthalmology. Vascular Ehlers-Danlos syndrome is rarer and more dangerous, and is distinguished carefully from the far commoner hypermobile type, which currently has no identified genetic cause and is diagnosed clinically — a distinction that matters enormously, because the two carry entirely different risks.

Familial hypercholesterolaemia deserves separate mention as the most under-diagnosed treatable inherited condition there is: markedly raised cholesterol from birth, early cardiovascular disease, effective treatment, and a diagnosis that is missed because a cholesterol result is read as a lifestyle finding rather than a genetic one. It is a strong candidate for cascade testing, because each person identified leads directly to several relatives.

Pharmacogenomics

Pharmacogenomics examines how inherited variation changes the way a person handles a medicine — how fast it is metabolised, whether a prodrug is activated, and whether a severe reaction is likely.

Three uses are established rather than speculative. Certain variants predict severe hypersensitivity to specific drugs and are tested for before prescribing them. Enzyme variants determine whether a drug is metabolised too fast to work or too slowly and accumulates, which is relevant to some psychiatric medicines, some pain medicines and some anticoagulants. And in oncology, variants determine whether a standard dose of certain chemotherapy agents would be severely toxic.

The honest limits matter as much. Broad pharmacogenomic panels sold directly to consumers cover many drug-gene pairs for which the evidence is thin, and a result is one input among several — age, kidney and liver function, other medicines and the clinical response all still apply. No result from such a panel is a reason to stop or change a prescribed medicine independently; that decision belongs to the doctor who prescribed it.

Direct-to-consumer tests, and what they do not do

People increasingly arrive with a result from a consumer testing service, and the conversation that follows is a recurring part of this unit’s work.

Direct to consumer genetic testing is genuinely informative about ancestry and about some traits. Its health claims need qualification on three points.

Most such tests use genotyping rather than sequencing: they check specific known positions rather than reading the gene. A service reporting BRCA typically checks a small number of specific variants, so a reassuring result excludes those and nothing else — and someone with a strong family history who is reassured by such a report has been misled by a technically accurate answer to a question they did not ask.

Findings are not clinically confirmed. Anything of consequence is repeated in an accredited diagnostic laboratory before it is acted on, and consumer results are not accepted as a basis for surgery or surveillance.

And risk estimates for common conditions combine many small effects into a score that is far weaker than family history, and considerably less accurate in people whose ancestry is under-represented in the underlying data.

None of that makes these tests worthless. It makes them a reason to come to a genetics appointment rather than a substitute for one.

What genetics cannot do

It cannot tell you what will happen. Penetrance is incomplete and expressivity varies, so a positive predictive result raises probability rather than fixing a future.

It cannot exclude a genetic cause. Every test is bounded by what it examines, and a negative result means nothing was found at that scale, in those genes, with today’s knowledge.

It cannot resolve a variant of uncertain significance by wanting to. A VUS is not acted on clinically, and re-testing does not clarify it — accumulating evidence over time does.

It cannot make a family talk to each other. Cascade testing is where the benefit lives, and it depends on a conversation that no service can have on a patient’s behalf.

It cannot treat most of what it finds. Some conditions have enzyme replacement or targeted management; many have surveillance; several have neither, and saying so plainly before a predictive test is part of consenting to it.

It cannot make a consumer report into a clinical one. Anything of consequence is confirmed in an accredited laboratory before anything is done about it.

Your multidisciplinary team

The clinical geneticist assesses the patient and the family, selects the test, interprets the result in its clinical context and coordinates surveillance. The genetic counsellor takes the family history, defines the question, prepares the family for what a result would mean and supports the cascade conversation afterwards. The laboratory geneticist and bioinformatician run and interpret the sequencing and classify variants against the evidence — the step that determines whether a finding is pathogenic, benign or uncertain. The cytogeneticist reports the chromosome studies that a microarray cannot replace, balanced translocations among them.

Around them: pathology, where a tumour result raises an inherited question and where the somatic-versus-germline boundary is drawn, medical oncology and breast health for inherited cancer risk, gynaecology for gynaecological surveillance and risk reduction, gastroenterology for Lynch and polyposis surveillance, perinatology for everything in pregnancy, IVF and reproductive health for embryo testing, cardiology and cardiovascular surgery for inherited cardiac and aortic disease, neurology for the neurogenetic conditions, nephrology for inherited kidney disease, paediatrics and paediatric surgery for children with syndromic conditions, and nutrition and dietetics for the metabolic diets.

The international patient journey

Genetics travels unusually well, because most of what happens is conversation and laboratory work rather than examination.

What to send before anything else. A family history is the single most valuable document, and it is worth preparing properly: three generations, with each relative’s relationship, current age or age at death, any diagnosis and the age at which it was made. Cancer diagnoses need the site and the age. Guessing is worse than leaving a gap.

Existing genetic reports, in full. The complete laboratory report rather than a summary — the gene name, the exact variant nomenclature, the classification and the methodology all matter, and a letter saying “genetic testing was normal” is close to useless without knowing which test was done. This is the commonest gap in an international referral.

Confirmation of the family variant. If a relative has tested positive, their report is what makes testing simple: a single defined change to look for, rather than an open search. Without it, a family member’s test is a much larger and less conclusive exercise.

The relevant clinical records. Pathology reports for cancers in the family, cardiac investigations, and any previous microarray or karyotype.

Three practical notes. Much of this is done remotely — counselling, review of an existing report, and a plan for what to test — and a sample can frequently be arranged without travel. Where a condition has already been diagnosed in a relative, testing the affected relative first is almost always more informative than testing the unaffected one who is asking, and that ordering is worth understanding before travel is planned. And a result may take considerably longer than other laboratory work, because interpretation rather than analysis is the slow step.