Pharmacogenomics and precision therapeutics — formative SAQs

SAQ 1 — A TPMT and NUDT15-guided thiopurine plan for a child with Crohn disease (10 marks, 15 minutes)

Stem: A ten-year-old with newly diagnosed moderate Crohn disease is about to start azathioprine. The team asks you to design the precision-prescribing plan, state the tests to order, and defend the action if the child is a poor metaboliser. ^[1]

Model answer

Tests to order (3 marks). Before the first dose, order thiopurine methyltransferase (TPMT) and NUDT15 genotype together, because either gene can cause thiopurine toxicity and their effects are additive. Pair the genotypes with a baseline full blood count and liver enzymes so the monitoring curve starts from a known point. The genotype is a once-in-a-lifetime result, so it must be recorded in a place the next prescriber finds and flagged on the chart. ^[1]

Translating the result (3 marks). Translate the genotype into a metaboliser phenotype using the CPIC table. A normal metaboliser uses the standard azathioprine starting dose under gastroenterology guidance. An intermediate metaboliser starts at a reduced dose and is titrated to the full blood count and disease response. A poor metaboliser carries two loss-of-function copies and should not receive a standard dose. ^[1]

Action if the child is a poor metaboliser (3 marks). A poor metaboliser accumulates active thioguanine nucleotides and is at high risk of severe myelosuppression. The 2018 CPIC update recommends avoiding a standard starting dose and either starting at a greatly reduced dose — about 10 per cent of standard — under intensive full-blood-count monitoring, or choosing an alternative immunosuppressant such as mycophenolate or a biologic. The child still has Crohn disease, so the question is not whether to treat but how to treat safely. ^[1]

Communication and follow-through (1 mark). Record the genotype, the chosen dose, and the reason permanently on the chart and in the family medicines record, and explain the result to the family in plain language — because the most common and most dangerous failure of pharmacogenomics is the positive result filed and the prescription never changed. ^[1]