Multifactorial disease and pharmacogenomics
Introduced| English | Meaning |
|---|---|
| liability threshold model | liability threshold model |
| empiric recurrence risk | empiric recurrence risk |
| Carter effect | Carter effect |
| pharmacogenomics | pharmacogenomics |
It runs in the family, but not by Mendel's rules
- Diabetes, high blood pressure, asthma, cleft lip and neural-tube defects all cluster in families, but never in 1/2 or 1/4 ratios.
- They are multifactorial: many genes of small effect plus the environment. One model makes "it runs in the family" into a number.
The liability threshold model
- In the liability threshold model 易患性阈值模型, each person's liability is the sum of many small genetic and environmental effects, so across a population it forms a bell-shaped curve.
- People whose liability passes a threshold have the condition.
Relatives sit further along the curve
- Close relatives share genes with an affected person, so their whole curve is shifted towards the threshold, and more of them cross it.
- The empiric recurrence risk 经验再发风险 is measured in family studies: for most multifactorial malformations it is 2–5% after one affected child, against 25% recessive and 50% dominant.
What is the typical empiric recurrence risk for a multifactorial malformation after one affected child?
Difficulty 1/5 · 100 coins + a small bonus on first completion today
Family studies give 2–5% for most such conditions; 25% and 50% are single-gene numbers.
What raises the risk
- The risk rises with more affected relatives and with a more severe form in the affected person.
- It also rises when the affected person is of the less often affected sex: the Carter effect 卡特效应. Pyloric stenosis is about four times commoner in boys, so an affected girl must carry more liability, and her relatives' risk is higher.
Pyloric stenosis is commoner in boys. Whose relatives have the higher recurrence risk?
Difficulty 1/5 · 100 coins + a small bonus on first completion today
This is the Carter effect. The less often affected sex has a higher threshold, so an affected girl's family carries more liability.
The recurrence risk is the same after one affected child as after two.
Difficulty 1/5 · 100 coins + a small bonus on first completion today
A second affected child shows the family sits further along the liability curve; neural-tube risk rises from about 3% to about 10%.
The environment is part of the sum
- Folic acid before conception cuts the recurrence of neural-tube defects by about 70%.
- After an affected pregnancy, women are advised a high dose, 4–5 mg a day, from before conception into early pregnancy, whatever the computed risk.
A recurrence risk of 3% is cut by 70% with folic acid. What is the new risk, as a percentage?
Difficulty 2/5 · 140 coins + a small bonus on first completion today
3% × (1 − 0.7) = 3% × 0.3 = 0.9%.
Pharmacogenomics: CYP2C19
- Pharmacogenomics 药物基因组学 uses genotype to choose drugs and doses. Clopidogrel, a heart drug, is a prodrug that the liver enzyme CYP2C19 must activate.
- Poor metabolisers have two loss-of-function alleles: about 15% of East Asians against about 2% of Europeans. For them clopidogrel barely works.
Why does clopidogrel work poorly in CYP2C19 poor metabolisers?
Difficulty 1/5 · 100 coins + a small bonus on first completion today
Without working CYP2C19, little of the active drug is made.
About 15% of East Asians are CYP2C19 poor metabolisers. Out of 1000 East Asian patients starting clopidogrel, about how many?
Difficulty 2/5 · 140 coins + a small bonus on first completion today
1000 × 0.15 = 150, which is why genotyping before treatment matters more in East Asia.
Screening before the first dose
- HLA-B57:01 predicts a severe hypersensitivity reaction to abacavir, an HIV drug; testing first has almost removed the reaction. HLA-B15:02, common in Han Chinese and Southeast Asians, predicts severe skin reactions to carbamazepine.
- TPMT deficiency makes standard doses of thiopurine drugs such as azathioprine dangerously toxic.
Match each gene to the drug it guides.
Difficulty 2/5 · 140 coins + a small bonus on first completion today
Each test turns a genotype into a prescribing decision.
Worked example: the recurrence quote
- (Fictional) A couple's first child had a neural-tube defect. The empiric recurrence risk is about 3%; after two affected children it is about 10%.
- High-dose folic acid cuts the 3% by about 70%: $3\% \times 0.3 \approx 0.9\%$. The quote comes from population tables, not a Punnett grid.
Worked example: the genotype before the stent
- (Fictional) A 58-year-old has a coronary stent fitted, and a genotype test shows CYP2C19 poor metaboliser status.
- Clopidogrel would barely be activated, so a drug that does not need CYP2C19, such as ticagrelor, is chosen from the first day.
Three common slips
- Quoting 25% or 50% for a multifactorial condition. Those are Mendel's numbers; use empiric tables.
- Adding genetic and environmental risks as percentages. They add on the liability scale, not as risks.
- Reading a pharmacogenomic result as a disease. It is an instruction about a drug.
Recap
- Liability (genes + environment) is bell-shaped; disease lies beyond a threshold.
- Empiric recurrence is 2–5% after one case, rising with more cases, more severity and the less often affected sex.
- Folic acid cuts neural-tube recurrence by about 70%.
- CYP2C19 (clopidogrel), HLA-B57:01 (abacavir), HLA-B15:02 (carbamazepine), TPMT (thiopurines): test before prescribing.