Iodine-131

I-131 has been treating thyroid disease at gigabecquerel activity for decades, and among the roster's beta therapeutics its 8-day half-life is the longest clock. That combination, high activity sustained over the longest beta clock, makes it the most persistent beta peroxide burden modeled here: production barely tapers across a typical shelf life. (Ac-225's alpha chain runs longer still, but at far lower, single-digit-megabecquerel activity.)

The mean beta energy is modest, close to Lu-177's, so intensity per decay is not the story. Duration is. Where a short-lived tracer finishes generating peroxide in hours, I-131 keeps generating across more than a week, so the burden not only reaches a high absolute level but stays there through QC release, shipping, and dispensing.

Radiolysis of the water throughout an I-131 vial generates superoxide and hydrogen peroxide continuously, at random points in the solution and in every direction. DuraRad lines the vial wall with a sol-gel coating that carries immobilized manganese superoxide dismutase and catalase, so the coating acts as a distributed sink: SOD converts the superoxide that reaches the wall into hydrogen peroxide, catalase decomposes that peroxide into water and oxygen, and together they lower the steady-state concentration of superoxide and peroxide in the vial. The enzymes are fixed in the coating rather than dissolved into the drug product, so they are not an ingredient in the formulation. Clearing peroxide at the wall keeps its concentration low across I-131's long shelf life and lowers the oxidative burden on the drug.

I-131's decay profile suits a sink that operates on this timescale. Its beta emission carries only modest energy per decay, close to Lu-177, so no single decay is intense; the difficulty is duration, because an 8-day half-life keeps peroxide production running through QC release, shipping, and dispensing rather than finishing in hours. Enzymatic turnover happens on a scale of minutes, so across the week and more that I-131 keeps generating peroxide the wall runs many clearance cycles per increment of production. In the model this holds peroxide well below the uncoated accumulation, and the suppression is large, but it remains a modeled result: the Q4 2026 radioactive validation at Washington University in St. Louis is designed to test it.