Prasugrel or Ticagrelor for a New Grandfather's NSTEMI
Two major trials tell two different true stories about P2Y12 inhibition in diabetics — one about a subgroup that benefited more, one about a drug that didn't care whether you had diabetes at all. No trial has ever compared the two agents directly in this population.
A.G., a 60-year-old man, spent his career as a union electrician and became a grandfather for the first time three weeks ago, a fact he mentions before his diagnosis every time a new physician introduces themselves to him. He has had type 2 diabetes for twelve years, managed with metformin and basal insulin, HbA1c 8.2% at his last check — imperfectly controlled, which he attributes candidly to a bad stretch since retiring rather than to any confusion about what he's supposed to be doing. He presented with an NSTEMI, was found to have a critical right coronary lesion, and received a drug-eluting stent; the interventional team now needs to choose his P2Y12 inhibitor.
He has no history of stroke or TIA, his weight is roughly 95 kilograms, and he is 60 years old — none of the specific caution criteria that would push against either prasugrel or ticagrelor apply to him. What genuinely complicates the choice is that the two major trials establishing each drug's benefit tell subtly different stories about diabetic patients specifically. TRITON-TIMI 38's prespecified diabetic subgroup showed prasugrel reducing the composite endpoint from 17.0% to 12.2% compared with clopidogrel, a hazard ratio of 0.70, without the excess bleeding seen in the trial's non-diabetic population — a diabetes-specific magnitude of benefit. PLATO's diabetic subgroup showed a smaller but statistically consistent benefit with ticagrelor, without a significant interaction between diabetes status and treatment effect — meaning ticagrelor's benefit didn't depend on being diabetic at all, a different kind of evidence than a subgroup showing an especially large effect. No randomized trial has ever compared prasugrel against ticagrelor specifically in diabetics.
At the post-PCI antiplatelet decision
TRITON-TIMI 38's diabetic subgroup showed a hazard ratio of 0.70 with prasugrel — a substantially larger relative reduction than the trial's overall population, and without the bleeding excess seen elsewhere in the trial. He has none of the contraindications, and he's P2Y12-naive, so he needs a 60 mg load. I'd give prasugrel 60 mg now, then 10 mg daily.
I'd weigh PLATO's data differently. Ticagrelor's benefit in diabetics was consistent with its overall trial population — no significant treatment-by-diabetes interaction — which is actually a stronger statistical claim than a subgroup with a large point estimate: it means the drug simply works, whether or not he's diabetic, rather than working especially well in a subgroup that a single trial happened to highlight.
PLATO also showed a mortality benefit with ticagrelor in its overall population that TRITON never demonstrated for prasugrel, which I think matters even though neither trial was diabetes-specific for that particular endpoint.
I don't think either of us is wrong about what our preferred trial shows — we're weighting different kinds of evidence, a subgroup magnitude against a consistency argument, and there's no head-to-head trial in diabetics to adjudicate between them. Given his specific profile has no caution flags for either drug, I'd go with prasugrel — 60 mg loading dose, then 10 mg daily — for the magnitude of benefit in patients who look like him specifically, while telling him plainly that ticagrelor would have been an equally defensible choice.
Prasugrel started with a 60 mg loading dose, then 10 mg daily, alongside aspirin 81 mg. No further genetic or platelet-function testing planned given the clear clinical decision already reached.
Not resolved as a general question, and named as such rather than papered over:
Both physicians agreed a diabetic ACS patient with no caution flags for either drug is a genuine case of two guideline-consistent, evidence-supported options pointing in slightly different directions — the final choice here reflected which trial's framing of “benefit in diabetics” the treating cardiologist weighted more heavily, not a clinical feature of A.G. himself that made one option clearly correct over the other.