Allometric Dose Scaling

The Reagan-Shaw Method

Reagan-Shaw, Nihal & Ahmad (2008) revisited the FDA Guidance for Industry on Estimating the Maximum Safe Starting Dose in clinical trials. The core insight is that drug clearance scales with body surface area, not body weight, across species.

For each species, Km = body_weight_kg / body_surface_area_m². This ratio is approximately constant within a species and reflects the fact that smaller animals have a much larger surface-to-volume ratio than larger animals.

• Mouse: Km = 3
• Hamster: Km = 5
• Rat: Km = 6
• Guinea pig: Km = 8
• Rabbit: Km = 12
• Cat: Km = 12
• Monkey: Km = 12
• Dog: Km = 20
• Human (adult, 60 kg): Km = 37

HED (mg/kg) = animal_dose (mg/kg) $\times$ (animal_Km / human_Km).

For mouse → human: HED = mouse_dose $\times$ (3/37) $\approx$ mouse_dose $\times$ 0.081. For rat → human: HED = rat_dose $\times$ (6/37) $\approx$ rat_dose $\times$ 0.162.

This calculator generalizes the formula to any pair of species in {mouse, rat, human}.

When BSA Scaling Falls Apart

BSA scaling is a *starting point*, not a substitute for proper pharmacokinetic modeling. Conditions under which the method underestimates or overestimates the true HED:

Drugs cleared by saturable enzymes (e.g. high-affinity CYP3A substrates near Vmax) do not scale linearly with BSA. The mouse may clear the drug efficiently while the human is enzyme-saturated at the BSA-equivalent dose.

For monoclonal antibodies and other biologics, clearance is dominated by target binding and internalization. BSA scaling does not capture this — use receptor occupancy modeling instead.

Some CYP isoforms (CYP2D6, CYP2C19) have very different activity profiles between rodents and humans. A drug that is 90% cleared by CYP2D6 in humans may be cleared by an unrelated isoform in mice.

If the parent compound is a prodrug, BSA scaling on the parent dose can dramatically misestimate the active metabolite exposure. Always verify with PK data when active metabolites are involved.

Use BSA scaling for: (1) initial cohort sizing, (2) IACUC dose justification, (3) sanity-checking published doses, (4) FDA IND HED estimation per the 2005 Guidance for Industry. Do NOT use it as the final dose for a clinical trial without empirical PK data.