Attribution works with any regimen: the engine assesses every drug the trial's protocol names for the patient, whatever the combination. Burna AI also carries pre-built oncology regimen profiles, each listing the component drugs, the schedule and the adverse events each drug is known for, so common combinations resolve by name, such as FOLFOX (including mFOLFOX6), FOLFIRI, AC-T, R-CHOP, carboplatin with pemetrexed and pembrolizumab monotherapy. If the protocol, the enrollment or the regimen cannot be resolved for a patient, the engine does not guess: it stops and names the missing link, because a grade computed against drugs the patient is not taking would look correct and be wrong.
More on multi drug attribution.
- What is multi-drug attribution?
- How does Burna AI compute adverse event attribution?
- What are the WHO-UMC causality categories and the Kramer algorithm?
- What is dechallenge and rechallenge?
- Can Burna AI grade single-agent trials too?
- How does Burna AI handle protocols where the trial drug profile is sensitive sponsor IP?