Skip to content
Parth Ajmera

Bioavailability and bioequivalence

#pk-pd

A brand-name drug goes off patent, and ten generic manufacturers want to make their own version — same active ingredient, same dose, but different fillers, coatings, and manufacturing processes. The regulatory question is simple to state and hard to answer: will these generic versions actually deliver the same amount of drug, at the same rate, to the same effect?

Bioavailability asks how much of a drug reaches the bloodstream, and how fast. An intravenous injection is 100% bioavailable by definition — it goes straight into the blood. An oral tablet has to survive the stomach and cross the gut wall first, so its bioavailability might be 20%, 50%, or 90%, depending entirely on how it's formulated.

Bioequivalence asks whether two formulations of the same drug deliver statistically equivalent bioavailability — under the regulatory standard FDA and most other agencies use, the two products' rate and extent of absorption have to fall within 80–125% of each other.

This is the science that makes the generic drug industry possible at all, and it also governs every post-approval change a manufacturer makes to an already-approved drug — a new tablet size, a new coating, a new manufacturing site — where the company has to demonstrate to a regulator that the change didn't alter how the drug behaves in the body.

Where it sits in development

Bioavailability and bioequivalence (BA/BE) work shows up at two distinct points. Early, it's used inside a single company's own development programme — comparing formulations, checking for food effects, or justifying a formulation change mid-trial. Late, it's the primary requirement for a generic manufacturer's regulatory submission (an ANDA in the US), and it's required again any time a manufacturing change is big enough to trigger a fresh bioequivalence study under FDA's SUPAC framework for post-approval changes.

What's actually settled, and what isn't

For conventional oral small-molecule formulations, this is mature science: the 80–125% standard has held since the 1970s, and the statistical methods behind it are well established. The frontier sits almost entirely in products that don't fit that conventional case:

  • Locally acting drugs — inhaled, topical, ophthalmic — where blood levels don't capture the exposure that actually matters, because the drug is meant to act where it's applied, not systemically. This is a genuinely unsolved regulatory problem, and it's specifically why generic inhalers have taken years longer to reach market than generic tablets.
  • Biosimilars — demonstrating equivalence for a complex biological product, where the generic version is similar but not molecularly identical to the original, requires a different, still-developing "totality of evidence" framework rather than a single blood-level comparison.
  • Highly variable drugs — ones where the same person, on the same drug, on different days, shows more than roughly 30% variation — make the standard bioequivalence test statistically unreliable, and the reference-scaled alternatives aren't universally accepted across regulators yet.
  • In vitro–in vivo correlation (IVIVC) — predicting how a drug will behave in a person from lab dissolution testing alone, without running a clinical study — is an active area, particularly for modified-release formulations.

This is also the one corner of PK/PD closest to what I already know from manufacturing: SUPAC changes, dissolution testing, and the specific ways a formulation decision shows up downstream as PK variability are things I've seen from the operations side, not just read about.