Accepted answer
Two administrations of 2.5 mg instead of one of 5 mg — the same 5 mg a week either way. Total weekly exposure is unchanged; what changes is the peak-to-trough ratio, and for an agent with a half-life measured in days the trough barely moves because the dosing interval is already short relative to it. The arithmetic is the easy part: 5 ÷ 2 = 2.5. Whether it is worth doing is a pharmacokinetic question, and nothing here is medical advice.
The honest answer is that reported tolerability benefits are real to the people reporting them and are not well explained by the exposure profile.
For a drug with elimination half-life t½ dosed at interval τ, the peak-to-trough ratio at steady state is 2^(τ/t½). With a one-week half-life dosed weekly, τ/t½ = 1, so the ratio is 2 — a factor of two between peak and trough, which is already flat by pharmacological standards.
Halving a dose accurately requires the reading resolution to support it. A 10-unit dose split into two 5-unit doses is at the coarse end of any barrel.
Titration schedules in the trial programmes were designed to manage gastrointestinal tolerability and are the evidenced approach to that problem.
The caveat is that no split schedule has been evaluated in a trial, so the whole discussion is inference plus report.
Work out 2^(τ/t½) for your agent before arguing about this. It usually settles it.
edited 20 Jul 2024 by orla_ferriter — added a caveat about sampling