Design: Three open-label, fixed-sequence drug–drug interaction studies
Population: 54 healthy adults, 18 per arm, given aildenafil 60 mg alone and with a CYP3A modulator
Primary outcome: Aildenafil peak concentration and area under the curve, with and without the modulator
Effect: Clarithromycin AUC ratio 2.58 (90% CI 2.42–2.76); rifampicin 0.01 (0.01–0.02); cimetidine 1.17 (1.10–1.25)

Three open-label, fixed-sequence studies in healthy volunteers measured what happens to aildenafil, a phosphodiesterase type 5 inhibitor cleared mainly by cytochrome P450 3A4, when it meets a strong inhibitor, a strong inducer, or a drug often assumed to be an inhibitor but which is not a CYP3A one.

Clarithromycin raised the area under the curve 2.58-fold and peak concentration 1.67-fold. Cimetidine did nothing measurable — geometric mean ratios sat at 0.99 and 1.17, both with confidence intervals that exclude any clinically useful effect. Rifampicin was the striking arm: exposure fell to 0.01 of control, with peak concentration at 0.02. That is not a dose adjustment. That is the drug being removed.

The specific molecule matters less than the shape. Aildenafil is a Chinese-developed phosphodiesterase type 5 inhibitor not licensed in most markets, and few readers will prescribe it. The induction arm is the transferable finding, because rifampicin is one of the most powerful and most widely given CYP3A inducers in the world, and in India it is given for months at a time to very large numbers of people.

When a CYP3A substrate is unexpectedly ineffective, ask about tuberculosis treatment before increasing the dose
Induction takes one to two weeks to reach full effect and persists a similar time after rifampicin stops — so both starting and stopping it need a plan
Cimetidine's reputation as a broad inhibitor does not extend to CYP3A; do not use it to explain a raised concentration
With a strong CYP3A inhibitor, titrate on clinical response and tolerability rather than applying a fixed fractional dose
Record the inducer in the problem list, not just the drug chart — the interaction outlives the prescription

Why it matters

Most interaction teaching is about inhibition and toxicity; this is the other direction, where the failure mode is a drug that silently does nothing.

The statistics, in plain English

A geometric mean ratio of 0.01 with a 90% confidence interval of 0.01 to 0.02 means exposure fell roughly a hundredfold, and the interval is narrow enough that the size of the effect is not in doubt. Contrast cimetidine, where the ratios were 0.99 and 1.17 with intervals spanning 1.0 — a genuine absence of effect rather than a study too small to find one. These were healthy volunteers at a single dose, so the absolute numbers will not transfer exactly to patients on repeated dosing, but the direction and order of magnitude will.