General Principles of Toxicology and Risk Assessment
Hazard, risk, and the dose-response relationship
Hazard is the intrinsic capacity of an agent to cause harm; risk is the probability that harm occurs under a defined exposure. Questions frequently blur the two. Risk assessment has four steps: hazard identification, dose-response evaluation, exposure assessment, and risk characterization. Toxicologists rely on the dose-response curve: for most agents a threshold exists, while genotoxic carcinogens are conventionally modeled as non-threshold.
Learn the vocabulary precisely. The LD50 (or TD50) is a population median, not a threshold and not a prediction for an individual patient. Therapeutic index is TD50/ED50; a narrow index (digoxin, lithium, warfarin, theophylline) predicts overdose vulnerability. For non-cancer endpoints, the reference dose is derived as RfD = NOAEL ÷ (uncertainty factors × modifying factor), typically 10 for animal-to-human extrapolation and 10 for human variability.
Traps that recur in questions
- Comparing LD50 values across species, routes, or formulations as if they were interchangeable.
- Assuming an asymptomatic patient is safe: iron, paraquat, amatoxin, acetaminophen, and sustained-release products all have deceptive latency.
- Confusing idiosyncratic (non-dose-dependent) reactions with dose-dependent toxicity.
- Forgetting interaction terms: additivity, synergy, potentiation (a non-toxic agent enhances another), and antagonism.
Worked rule of thumb
A chronic study finds a NOAEL of 5 mg/kg/day. With a composite uncertainty factor of 100, the RfD is 0.05 mg/kg/day; for a 70 kg adult that is 3.5 mg/day. If the exposure assessment shows intake of 0.5 mg/day, the hazard quotient is 0.14 - below 1, so no appreciable risk is anticipated, though this is not proof of zero risk.