One molecule, two organisms
When you treat a cancer, the target lives in your own cells, so every blow lands partly on you. Anti-infectives are luckier. The pathogen is a separate organism, and the dream is to find a drug that strikes something the microbe has and you do not. This is selective toxicity — the single idea underneath every antibiotic and antiviral ever made.
Paul Ehrlich called it the magic bullet: a compound that flies past human tissue and detonates only on the invader. Whenever selective toxicity is high, the therapeutic window is wide and the drug feels safe. Whenever the pathogen looks too much like us, selectivity collapses and side effects climb.
Where the differences live
Bacteria are a goldmine for selectivity because they build things we do not. They wrap themselves in a peptidoglycan cell wall (we have no cell wall), read their genes with bacterial-specific ribosomes and enzymes, and run metabolic shortcuts our cells skipped. Each of these is a clean target whose mechanism of action hits the bug and ignores you.
Measuring the gap
Selective toxicity is not a slogan; it is a ratio you can measure. Compare the dose that harms the pathogen with the dose that harms the host, and the wider the spread, the safer the agent. A drug that kills bacteria at a hundredth of the concentration that bothers your cells has a strong on-target signal and few off-target surprises.
Hold onto this lens. Every guide that follows — β-lactams, MIC, resistance, antivirals — is really one long answer to the same question: what does the pathogen do that we don't, and how do we hit it?