Contraindicated and avoid look like synonyms from outside the agency. Inside it they are two different decisions, and an FDA clinical pharmacology official said the distinction causes enough confusion that she wanted people to carry it out of the room.
“To those who do not work specifically in this area, the words sound identical. Contraindicate, avoid. What on earth is the difference?”
Acting Deputy Director, CDER Office of Clinical Pharmacology
A contraindication means these two drugs should never be given together under any circumstance. The review team does not reach for it lightly. It is used where the consequence is extreme toxicity, morbidity or mortality, and sometimes where a loss of efficacy would stop an essential drug working.
Avoid is the word FDA reaches for when it recognizes that some patients genuinely have no alternative, and that with close medical supervision and monitoring for the specific adverse outcome, concomitant use is possible. Avoid is a strong steer. It is not a prohibition, and the difference is deliberate.
The exchange came during the Q&A panel following a CDER SBIA webinar, September 23, 2026 on communicating drug interaction and QTc information in US prescribing information, built around FDA’s draft guidance on the Drug Interactions section and its guidance on QTc labeling.
Concomitant Use Is Not Co-administration
A second vocabulary distinction got corrected live. Co-administration means giving two or more drugs at the same time. If drug A is taken in the morning and drug B at night, that is not technically co-administration, and yet a clinically significant interaction can still occur.
The correct term for that situation is concomitant use, and the speaker went back to clarify his own presentation to make the point. He corrected a second phrase in the same breath: what Section 7 describes is not the mechanism of action but the mechanism of the clinically significant drug interaction.
These read as pedantry until a labeling supplement turns on them.
The Same Exposure Change, Two Different Conclusions
He also walked through why an exposure number alone does not determine clinical significance.
Take a subject drug used concomitantly with drug B, a narrow therapeutic index drug, where drug B exposure rises 50 percent. That may well be clinically significant. Now take the same subject drug with drug C, not a narrow therapeutic index drug, where exposure also rises 50 percent. Same exposure change, and it may not be clinically significant at all.
The implication for sponsors is that the labeling argument is about therapeutic index and exposure-associated clinical effects, not about the magnitude of the AUC ratio in isolation.
What Goes in Which Section
Several questions circled the same practical problem: where does a given piece of drug interaction information belong, and when does it belong nowhere.
The more striking guidance concerned deletion. Where no drug interaction studies were conducted and that decision was appropriate, FDA does not want a sentence in Section 7 saying so. It wants Section 7 removed, as clearly inapplicable.
The speaker put that in a family of similar deletions: Section 9 on drug abuse and dependence comes out when the drug is not associated with either, and the geriatric use subsection comes out when a drug is approved only in neonates with no safety issue in off-label geriatric use. The principle is that an empty section is worse than no section.
Be as Specific as the Science Allows
Asked whether labeling should say substrate or sensitive substrate, inhibitor or strong inhibitor, the answer was unambiguous. Because labeling does not name individual drugs, the classification language is doing the work, and it should carry as much specificity as is scientifically valid and appropriate.
If an instruction applies only to strong CYP3A inhibitors, the label should say strong CYP3A4 inhibitors. Where justified, there may be separate instructions for moderate and even weak inhibitors. Vagueness here is not caution; it is a gap a prescriber has to fill themselves.
Which connects to another answer, about labeling that is already vague. Where a label says to adjust the dose or dosage without a direction, reduce or increase, FDA generally treats it as something the applicant should address at the next supplement submission rather than as an urgent correction.
When You Can Skip the Study, and What Happens After
On when in vivo interaction studies can be omitted, the answer pointed to ICH M12, the drug interaction guidance FDA posted in August 2024, as the governing framework for any investigational program.
And where a study does lead to a different dose, sponsors asked whether efficacy data are then expected. Rarely, was the answer, and not ruled out but uncommon. In many cases additional pharmacokinetic data are not needed either, because the interaction at a different dose can usually be predicted from the original interaction study, the available clinical pharmacology and the known mechanism. The qualifier was that this is case by case, and that it is a conversation to have during development rather than at submission.
For post-marketing requirements and commitments aimed at interaction questions, the speaker declined to give a revision frequency but explained the logic. A PMR or PMC exists because a risk or gap was identified during review. A timeline attaches to it, the protocol and then the results come in, and the review team decides whether the finding belongs in Section 12, and if actionable, in Section 7. Typically something in the label changes, if only to close the loop that was left open at approval.
When Another Regulator Disagrees
One question came from companies operating globally: what to do when another authority treats an interaction as clinically significant but the evidence has not supported the same conclusion for US labeling.
The answer was straightforward and worth hearing directly. FDA operates under the Federal Food, Drug, and Cosmetic Act and its associated regulations. Other regulators work from different statutes, different laws and different labeling regulations. That difference alone can produce different labeling language for the same drug and the same data. The agency coordinates with its regulatory partners and sometimes works to a consensus, but divergence is not automatically an error to be reconciled.
References
- FDA/CDER SBIA, Communicating Drug Interaction and QTc Information in the U.S. Prescribing Information, webinar held September 23, 2026.
- Part 1 of 2: Drug Interaction Information in the U.S. Prescribing Information (slides, PDF), CDER SBIA.
- FDA draft guidance, Drug Interaction Information in Human Prescription Drug and Biological Product Labeling.
- FDA guidance, QTc Information in Human Prescription Drug and Biological Product Labeling.
- ICH M12, Drug Interaction Studies, posted by FDA August 2024.
- Q&A panel recording (video). Responses are reported by role rather than by name.
- Companion piece: what FDA said about QTc assessment and labeling at the same session
Moe Alsumidaie, MBA, MSF, is founder and Chief Editor of Vanguard Publications, which publishes Clinical Trial Vanguard, Pharma Vanguard and BullScope, and Head of Research at CliniBiz. He has two decades in clinical trial operations and data science, with earlier roles at Genentech, Abbott Vascular and Stanford University Medical Center, and is a guest lecturer in clinical trial sciences at Rutgers University.

