Genetic Prevention: Screening and Engineering

The key contrast is somatic versus germline: editing body cells treats a patient without changing what they pass to children. Keep these words handy.

Every baby born in NSW is offered a heel-prick blood test between 48 and 72 hours of age. A few drops of blood on a card are tested for more than twenty serious conditions, including phenylketonuria, congenital hypothyroidism, cystic fibrosis and MCAD deficiency. All of them share one feature: early treatment works, and late treatment does not.

Why PKU is the classic case

In phenylketonuria a mutation in the PAH gene leaves the enzyme phenylalanine hydroxylase missing or non-functional. Phenylalanine from ordinary dietary protein cannot be converted to tyrosine, so it accumulates in the blood and brain, where high concentrations damage developing neurons and cause severe, permanent intellectual disability.

A low-phenylalanine diet started in the first weeks keeps blood phenylalanine below the toxic range, and the child develops normally. Notice what has and has not changed. The genotype is untouched, so the faulty allele is still there and is still passed on. What has been prevented is the phenotype, by removing the substrate that the broken pathway cannot handle.

Most people who carry a recessive disease allele are perfectly healthy and have no family history at all, because one working copy is enough. The risk only appears when two carriers have children together. For an autosomal recessive condition such as cystic fibrosis, two carriers face a one in four chance of an affected child in every pregnancy.

Reproductive carrier screening tests prospective parents before or early in pregnancy. Australia's Mackenzie's Mission study offered free screening for cystic fibrosis, spinal muscular atrophy and fragile X syndrome, and Medicare now funds testing for those three. Roughly one in twenty people screened turns out to carry a variant for at least one of them.

Screening only produces a benefit if what follows is genuinely useful, and that is the job of genetic counselling. It is deliberately non-directive: the counsellor explains the inheritance pattern, the actual numerical risk and every available option, then supports whichever decision the family reaches. Telling people what to choose would be both unethical and, in practice, ineffective.

Each technology plays a different role. Screening and PGT find or select, while engineering tries to change genetic material, and each has its own limits.

Preimplantation genetic testing only exists inside an IVF cycle. Eggs are collected and fertilised in the laboratory, and the embryos grow for about five days to the blastocyst stage. A handful of cells is then taken from the trophectoderm, the outer layer that would become placenta rather than the baby, and their DNA is tested.

The version that matters for inherited disease is PGT-M, testing for a single known monogenic variant such as the cystic fibrosis or Huntington's allele already identified in that family. PGT-A instead counts chromosomes to detect aneuploidy, and PGT-SR checks for the unbalanced products of a parental structural rearrangement such as a translocation.

What PGT can and cannot claim

No embryo's DNA is altered at any point. Embryos without the variant are identified and transferred, which is selection, not modification. That distinction sets the limits. PGT needs the family's exact variant to be known in advance, requires a full IVF cycle with its cost and physical burden, produces a limited number of embryos that may all be affected, and raises genuine ethical questions about which conditions justify selection.

Some genetic technologies are established for detection or selection, while many engineering approaches remain limited to specific diseases or research contexts. Strong HSC answers avoid claiming CRISPR is a widespread prevention method for all genetic disease.

Gene therapy is no longer hypothetical, but the approved treatments are narrow and specific. The usual method loads a working copy of a gene into a modified virus, most often an adeno-associated virus stripped of its ability to replicate. The virus delivers the gene to the target tissue, where the cells transcribe and translate it to make the missing protein.

Luxturna treats an inherited retinal dystrophy caused by two faulty copies of RPE65. The vector is injected under the retina, so it reaches exactly the cells that need the gene, and treated patients regain useful vision in dim light. Zolgensma treats spinal muscular atrophy by delivering a working SMN1 gene intravenously to motor neurons, and it works only if given before those neurons are lost.

Why this is not general prevention

Each therapy targets one gene in one accessible tissue, and the gene must be small enough to fit inside the vector. A single Zolgensma dose costs around 2.5 million dollars, which Australia funds through a national program for a small number of infants. The immune system can also attack the vector, which limits repeat dosing.

CRISPR-Cas9 is a targeting system. A short guide RNA is designed to be complementary to a chosen DNA sequence, and it carries the Cas9 nuclease to that exact site, where Cas9 cuts both strands. The cell then repairs the break, and the repair is what produces the edit, either by disrupting the gene or, less reliably, by copying in a supplied template.

The first approved CRISPR medicine shows the realistic scale. Casgevy treats sickle cell disease and beta-thalassaemia by removing a patient's own blood stem cells, editing them outside the body to switch fetal haemoglobin production back on, and returning them. It is somatic, it is done outside the body where the edit can be checked, and it treats one patient at a time.

Two technical problems keep it from becoming general prevention. Off-target effects occur when the guide RNA binds a similar sequence elsewhere and Cas9 cuts the wrong gene. Mosaicism occurs when editing an embryo succeeds in some cells and not others, leaving an individual who is a patchwork of edited and unedited tissue.