Design: Regulatory database review of FDA approvals
Population: 332 small-molecule new molecular entities approved by the FDA, 2015–2024
Primary outcome: CYP3A-related post-marketing requirements and commitments, and resulting label changes
Effect: 51 drugs (15%) with 91 CYP3A studies; 58% of studies led to label updates

This analysis reviewed all 332 small-molecule new molecular entities approved by the FDA from 2015 to 2024 and identified post-marketing requirements (PMRs) and commitments (PMCs) relating to CYP3A drug interactions.

Fifty-one drugs (15%) had CYP3A-related PMRs or PMCs, totalling 91 studies; about two in three were cancer drugs, 25 of them kinase inhibitors. Thirty-seven drugs were to be studied as substrates, affected by inhibitors or inducers, and 25 as precipitants, affecting other drugs. Initial labels carried interim recommendations for 65% of these studies, and 58% had resulted in label updates.

For prescribers, the practical point is that the interaction section of a newly approved drug's label, especially a cancer drug, is often provisional. The mechanism is familiar: strong CYP3A inhibitors (clarithromycin, azoles, ritonavir) raise exposure of substrates, and inducers (rifampicin, carbamazepine, phenytoin) lower it, sometimes enough to cause treatment failure.

Treat the interaction advice in a newly approved drug's label as provisional; check for updates.
Kinase inhibitors are the class most often affected; review every co-prescription with the oncology team.
Rifampicin, carbamazepine and phenytoin can lower CYP3A substrate levels enough to cause treatment failure.
Clarithromycin, azole antifungals and ritonavir-boosted regimens can raise substrate levels to toxic ranges.
In India, rifampicin co-prescription during TB treatment is a frequent and important trigger.

Why it matters

It shows that the interaction label a prescriber relies on is often still being written after the drug is in use.

The statistics, in plain English

These are counts of regulatory requirements, not measurements of harm. That 58% of studies led to label changes shows how often the interaction picture at launch was incomplete.