21.8 — Rare Diseases

A rare disease is usually defined as one affecting fewer than 1 in 2,000 people.

There are around 7,000 of them.

Which produces the paradox at the centre of this chapter: individually rare, collectively common. Around 300 million people worldwide have a rare disease — roughly 1 in 17.

So "rare disease" is not a rare situation. It is a category that most people will encounter through a family member, a friend's child, or themselves.

The pattern of a rare disease

And it is remarkably consistent, whichever condition it is.

Around 70 percent begin in childhood.

Around 72 percent are genetic.

The average time from first symptom to correct diagnosis is around 4 to 5 years, and longer for many.

Most people see several specialists and receive at least one wrong diagnosis on the way — a phenomenon so consistent it has a name, the diagnostic odyssey.

And only around 5 percent have an approved treatment.

Those five numbers describe the whole problem: it starts early, it is usually genetic, it takes years to name, and once named there is often nothing licensed to give.

Why diagnosis takes so long

Doctors are trained to think common things first, and that training is correct almost all of the time. "When you hear hoofbeats, think horses, not zebras" is good advice that fails precisely in this category — which is why the zebra became the symbol of the rare disease community.

The symptoms are individually ordinary. Fatigue, pain, developmental delay, recurrent infections. The unusual thing is the combination, not any single item.

No individual doctor can know 7,000 conditions. A specialist may see one case of a given rare disease in a career, or none.

And tests are ordered for what is suspected. A test nobody thinks of is a test nobody runs, which is the core of the delay.

What has changed the picture

Genomic sequencing, and this is the genuine revolution in the field.

Whole exome sequencing reads all the protein-coding genes; whole genome sequencing reads everything (Chapter 2.9).

Instead of testing one gene at a time based on a guess, you read them all and look for what is wrong.

Diagnostic yield is around 30 to 50 percent in previously undiagnosed patients, which is remarkable for a group who had already exhausted conventional testing.

And rapid sequencing in critically ill babies now returns results in days rather than months — which has changed treatment in a meaningful proportion of cases, sometimes by identifying a condition with a specific and available therapy.

Newborn screening — a heel-prick blood spot in the first days of life, testing for a panel of conditions where early treatment prevents catastrophe.

The classic is phenylketonuria — an inability to break down the amino acid phenylalanine, which accumulates and causes severe intellectual disability. A restricted diet started in the first weeks prevents that completely. A child identified by screening lives a normal life; the same child missed does not.

Also screened for, depending on country: congenital hypothyroidism, cystic fibrosis, sickle cell disease, and a group of metabolic and immune disorders.

This is one of the highest-value programmes in medicine, and it works because it does not wait for anyone to suspect anything.

Some rare diseases worth knowing

Not because you will diagnose them, but because each illustrates something about how the category works.

Cystic fibrosis — a faulty chloride channel means secretions everywhere are thick and sticky, causing lung infection and pancreatic failure (Chapter 2.8).

And it is the model case for what modern genetics can do. CFTR modulator drugs correct the underlying protein defect rather than treating the consequences. Median survival has risen from childhood in the 1960s to well into adulthood, and the modulators have improved lung function dramatically in those with responsive mutations.

It also illustrates the field's hardest problem: the drugs cost enormous sums, and people with mutations not covered by them are left behind.

Sickle cell disease — abnormal haemoglobin polymerising and deforming red cells (Chapter 7.1). Not rare globally, and under-resourced relative to its burden. Hydroxyurea, transfusion, and now gene therapies that have cured patients in trials.

Huntington's disease — an inherited neurodegenerative condition (Chapter 2.8). It raises the question of predictive testing more sharply than anything else, because you can know decades in advance and there is no treatment. Most people at risk choose not to test, which is a legitimate choice.

Duchenne muscular dystrophy — progressive muscle degeneration in boys from a missing dystrophin protein. Steroids, and newer gene-based treatments beginning to appear.

Spinal muscular atrophy — and this is the field's clearest triumph. A condition where infants with the severe form did not usually survive past two. Three treatments now exist, including a gene therapy, and treated early — ideally before symptoms — many children sit, stand and walk. Which has made adding it to newborn screening urgent, because the treatment works far better before damage occurs.

Marfan syndrome — a connective tissue disorder causing tall stature, long limbs, lens dislocation, and aortic dilatation that can rupture. Recognising it saves lives, because beta blockers and planned surgery prevent the catastrophe.

Lysosomal storage disorders — Gaucher, Fabry, Pompe and others, where a missing enzyme lets material accumulate inside cells. Several have enzyme replacement therapy, given as regular infusions.

Primary immunodeficiencies — the immune system missing a component (Chapter 13.6). Suspect them with unusually frequent, severe or unusual infections. Treated with immunoglobulin replacement, and some with bone marrow transplantation, which is curative.

Amyloidosis — misfolded proteins depositing in organs. Presents in many disguises, and treatments have improved substantially.

Ehlers-Danlos syndromes — connective tissue disorders with joint hypermobility, skin changes and, in the vascular type, arterial rupture. The vascular type is the one that must be identified, because it changes surgical and obstetric management.

Mitochondrial diseases — faults in the cell's energy machinery, affecting the most energy-hungry tissues: brain, muscle, heart and eye (Chapter 1.6). And they can be inherited through mitochondrial DNA, which comes only from the mother — one of the few clean exceptions to ordinary inheritance.

What to do when nothing fits

And this is the practical heart of the chapter, because it is where most people in this situation actually are.

Keep a detailed record. A written timeline of symptoms, dates, tests, results and doctors seen. This is genuinely useful, because each new specialist otherwise starts from an incomplete account, and patterns visible across years are invisible in a fifteen-minute appointment.

Photograph and video things that come and go, because the rash, the movement or the swelling is never present in the consulting room.

Get copies of your own results.

Ask for genetic testing where the pattern suggests it — early onset, multiple systems affected, a family history, or features present since birth.

Ask about specialist centres. Most countries have designated centres for undiagnosed and rare disease, and referral is possible.

Find the patient organisation. For rare disease this is not a support-group nicety — the patient organisations frequently hold more accumulated knowledge about a specific condition than any single clinician, and they know which centres and which specialists to approach.

Consider research studies, which frequently offer sequencing and expert review that is otherwise hard to access.

And keep looking. New genes are identified continuously, so a negative sequencing result is not permanent. Reanalysing stored genomic data every couple of years produces new diagnoses in a meaningful proportion of previously unsolved cases, and it costs nothing but a request.

Why diagnosis matters even without a cure

And this is worth stating, because "there is no treatment" is frequently taken to mean "the diagnosis doesn't matter".

It ends the odyssey. The uncertainty and the repeated implication that nothing is really wrong cause real harm.

It gives a prognosis, which allows planning.

It identifies what to watch for. Many rare conditions carry specific risks — a heart rhythm problem, an aortic risk, a cancer predisposition — that are manageable once known and dangerous when not.

It informs family members, who may be at risk or planning children.

It gives access to the right specialists and to the right patient community.

It enables reproductive choices, including prenatal and pre-implantation testing.

And it makes trials possible. You cannot be offered a trial for a condition nobody has named.

The economics, honestly

Developing a drug for 500 patients worldwide costs roughly what it costs to develop one for 500,000.

Orphan drug legislation addressed this — market exclusivity, tax credits and regulatory support — and it worked. Rare disease drug approvals rose substantially after these laws were introduced.

And it created a second problem: prices. Some rare disease treatments cost hundreds of thousands of dollars a year, and the most expensive gene therapies are priced in the millions for a single administration.

The manufacturers' argument is that the development cost has to be recovered from very few patients. The counter-argument is that much of the underlying science was publicly funded and that the prices exceed what that reasoning justifies.

Both positions contain truth, and the result is that effective treatments exist which many health systems cannot fund — which is an uncomfortable place for medicine to be, and it is where the field currently is.

The direction of travel

And it is genuinely encouraging.

Sequencing costs have fallen by orders of magnitude, so genomic diagnosis is becoming routine rather than exceptional.

Gene therapy is delivering actual cures in a growing list of conditions — some immunodeficiencies, spinal muscular atrophy, haemophilia, sickle cell disease.

Antisense oligonucleotides and RNA-based drugs can target a specific mutation, which opens the possibility of treatments designed for very small groups — in a few documented cases, for a single patient.

And the knowledge gained from rare diseases feeds back into common ones. PCSK9 inhibitors, which lower cholesterol powerfully, came from studying families with rare inherited cholesterol abnormalities. Statins themselves trace back to work on familial hypercholesterolaemia.

Rare diseases are, repeatedly, where the mechanism of a common process is revealed — because a single broken gene shows you exactly what that gene was doing.

What the next Part covers

Part 22 is the medicine cabinet — how drugs actually work, what the common ones do, what to keep at home, and how to take them safely.