1.4 Route Selection, Dosing Regimens & Treatment Duration

Why route and duration decide whether you see the toxicity

Handbook I.A.1 D lists dose/concentration, duration of treatment, life-stage, endpoints, route of administration, and alternative in vitro and ecotoxicological studies as the levers for identifying toxic responses. Two studies of the same molecule can reach opposite NOAELs if one uses a dietary mix at a true 90 days and the other uses a daily gavage bolus for 28 days. DABT will ask you to match the human exposure scenario, not the laboratory’s favorite pump.

Match the route to the human (or ecological) contact

ICH M3(R2) expects the clinical route for pharmaceuticals. An oral drug still needs oral toxicology; an inhaled corticosteroid needs inhalation toxicology. Route-to-route extrapolation is a risk-assessment tool after the fact, not a license to skip the portal of entry when that is where injury will occur.

For chemicals, workplace inhalation plus dermal is common. Running only gavage because inhalation chambers are expensive answers a different question (high-dose oral systemic hazard) and can miss bronchiolar injury, olfactory lesions, or particle overload.

In vitro assays and ecological tests belong in this same decision tree. If the question is mutagenicity, OECD 471 may be the right “route.” If the question is aquatic invertebrate hazard, a Daphnia study may beat another rodent gavage. Those alternatives still do not replace a mammalian inhalation study when workers inhale the dust.

Oral gavage versus diet versus drinking water

Gavage delivers a known mg/kg at a known time. That is ideal when you need a steep maximum plasma concentration (Cmax) comparison, a controlled vehicle, or a drug that will be taken as a discrete dose. Costs of that precision: reflux and aspiration, esophageal trauma, a bolus that does not mimic sipping or grazing, and vehicle volume limits (often on the order of 10 mL/kg for aqueous gavage in rats; oil vehicles usually lower). Gavage error can kill animals and mimic treatment-related mortality.

Diet mixing is closer to some human food and feed exposures. Achieved dose equals dietary concentration times food consumption divided by body weight. If the chemical is bitter, food intake falls, body weight falls, and you may be looking at starvation dressed as toxicity. Diet studies demand homogeneity and stability data; a hot spot in the mixer is a dose error. Photolabile or volatile agents may not survive in open feeders.

Drinking water shares diet’s “consumed dose” problem plus leakage, evaporation, and light. Insoluble or hydrolytically unstable compounds do not belong in water just because gavage staffing is scarce.

Choose gavage when dose precision and bolus kinetics matter. Choose diet or water when the human intakes the agent that way and you can keep achieved dose inside the target range.

Dermal, inhalation, and parenteral regimens

Dermal. Clip the site, define percent body surface area, choose a vehicle that does not itself irritate, and decide occlusion versus non-occlusion (occlusion usually increases absorption). Prevent oral ingestion from grooming with collars or other restraints, or you have an unplanned oral study. Local irritation can limit dose below any systemic MTD.

Inhalation.

• Nose-only: animals are restrained in tubes and breathe a characterized atmosphere. You control chamber concentration, particle mass median aerodynamic diameter (MMAD) and geometric standard deviation, and estimated inhaled dose. Restraint is a Refine/stress issue.
• Whole-body: less restraint, more confounding. Material deposits on fur, then is groomed into the gut, and can absorb across skin. The “inhaled” dose is a mixture of inhalation, oral, and dermal.
• Intratracheal or intranasal instillation: a liquid or particle bolus into the airway. Useful for pulmonary mechanistic or instilled-particle studies. It is not equivalent to a 90-day OECD 413 inhalation exposure and should not be sold as one.

IV, SC, IM. Intravenous administration gives complete bioavailability and is the clinical route for many hospital drugs; infusion rate matters (dogs are prone to histamine-type reactions with some vehicles). Subcutaneous and intramuscular routes create depots and local-tolerance questions—standard for many biologics. Do not “convert” an IV clinical drug to oral gavage toxicology without a scientific reason (for example identical first-pass metabolite burden already shown).

Duration classes and life-stage

Names are not interchangeable. Calling a 28-day study “subchronic” or a 90-day study “chronic” is a vocabulary error that will cost items. ICH M3 duration matching is a separate clock: the toxicology duration must support the proposed clinical duration, which is why a 14-day nonclinical package can be enough for a 14-day first-in-human trial and still be the wrong answer for a pesticide 90-day data requirement.

Life-stage is not the same as duration:

• Prenatal / embryo-fetal: dams dosed during organogenesis (OECD 414, ICH S5); fetuses examined. A 90-day adult 408 does not cover this.
• Juvenile: developing nervous, reproductive, and renal systems; pediatric drug plans (ICH S11 territory).
• Adult: the default starting age for general toxicity (often 6–8-week-old rats at first dose).
• Geriatric animals are for geriatric questions, not a stealth way to raise background tumors in a standard 408.

Limit doses

For repeated-dose oral OECD studies, the usual limit dose is 1000 mg/kg/day, unless anticipated human exposure is high enough that a higher dose is needed to provide a margin. If frank toxicity appears at 150 mg/kg/day, you do not climb to 1000 to “hit the limit.” You set high dose at a maximum tolerated dose (MTD) or maximum feasible dose that still lets animals survive with interpretable pathology.

Do not confuse that repeated-dose 1000 mg/kg/day figure with acute oral limits, which are often 2000 mg/kg in GHS/OECD acute protocols. Inhalation and dermal guidelines have their own limit concentrations or applied doses; copying 1000 mg/kg onto a nose-only aerosol is nonsense.

Diet percent inclusion can hit nutritional disruption before 1000 mg/kg/day is reached. Feasibility, not the limit slogan, then caps the high dose.

Realistic exam scenario

Workers will inhale a poorly soluble dust 8 hours a day for years. A sponsor proposes a 90-day rat gavage study at 1000 mg/kg/day “because that is the OECD oral limit and chambers cost more.” Two failures sit in that sentence. First, the human route is inhalation; OECD 413 (90-day inhalation) or a justified inhalation design with particle characterization is the study that can see portal-of-entry lung injury. Gavage at a limit dose tests oral systemic overload and GI local effects. Second, 1000 mg/kg/day is a cap, not a target. If a palatable dietary or an inhalation atmosphere produces toxicity far below a limit, you stop there. If the dust is so insoluble that oral absorption is negligible, a high gavage dose may even under-predict lung particle toxicity.

A better package: inhalation 90-day with MMAD in the respirable range, a limit concentration if no toxicity occurs, satellite TK or lung-burden measurements, and only then a route-to-route discussion for incidental oral ingestion. If the decision question were instead an oral drug tablet, that same inhalation insistence would be the wrong 3Rs spend.

Traps

• Treating whole-body chamber concentration as equivalent to inhaled dose.
• Calling instillation “inhalation.”
• Using 14-day data and labeling it subchronic.
• Climbing to the 1000 mg/kg/day oral limit after MTD is already exceeded.
• Ignoring prenatal or juvenile life-stage when the human population includes fetuses or children.