Generic Drug Labeling and Bioequivalence Claims: What FDA Auditors Check Before Your ANDA Clears

About 90% of all prescriptions dispensed in the United States are filled with generic drugs. FDA cites that figure routinely, and IQVIA’s data backs it up. But behind each of those prescriptions sits a labeling document — a prescribing information package or drug facts panel — that had to survive one of the most meticulous regulatory reviews in pharmaceutical development. Get the labeling wrong, and it doesn’t matter how clean your bioequivalence data looks. The application stalls.

This post walks through what FDA auditors and experienced regulatory compliance consulting teams actually examine in a generic drug labeling package — specifically around bioequivalence claims, carve-out language, and therapeutic substitution statements. It’s not a theoretical framework. It’s a practical account of where ANDA submissions generate deficiencies, based on the patterns visible in Complete Response Letters and FDA’s OGD communications under GDUFA III.

What “Bioequivalent” Really Means — and What It Doesn’t Cover

Bioequivalence means the generic formulation delivers its active ingredient to the bloodstream at a rate and extent not statistically different from the Reference Listed Drug (RLD). FDA defines the standard BE window as 80–125% for both AUC (area under the curve) and Cmax (peak plasma concentration). A typical study enrolls 24 to 36 healthy adult volunteers in a fasting or fed crossover design, depending on the product’s biopharmaceutical characteristics.

But bioequivalent is not the same as identically labeled. This is where many ANDA sponsors make their first — and most expensive — mistake. The generic label must, in most cases, mirror the RLD’s approved prescribing information. But it can’t always do so word-for-word, because some RLD language is actively protected by unexpired patents or regulatory exclusivities. That’s where 21 CFR 314.94(a)(8)(iv) enters the picture: the “section viii carve-out.”

A section viii statement allows an ANDA applicant to omit a patented indication or condition of use from its labeling while still pursuing approval for the non-patented portions. FDA calls this a “skinny label.” The carve-out must be carefully drafted. Too narrow, and you leave market share on the table. Too broad, and you invite an induced infringement argument in Hatch-Waxman litigation. FDA’s Office of Generic Drugs (OGD) reviews each proposed carve-out for regulatory sufficiency — and auditors examining ANDA packages look for carve-out language that is internally consistent across the entire label, not just within the Indications and Usage section.

The 5 Labeling Elements FDA Auditors Focus On First

When FDA reviewers and third-party regulatory compliance consulting teams examine an ANDA labeling package, they work through a defined checklist. Based on the deficiency patterns visible in OGD’s Complete Response Letters across the first three GDUFA cycles, these are the five areas generating the most labeling-related CRLs:

1. PLR format compliance under 21 CFR 201.56 and 201.57. The Physician Labeling Rule format applies to most prescription drug labels submitted after June 2006. Generic sponsors must match the RLD’s PLR structure exactly — including section headings, subheadings, and cross-reference language. Misaligned section numbering, even by a single digit in a nested subsection, gets flagged in OGD’s formatting review.

2. Carve-out cross-reference consistency. If a section viii carve-out removes a patented indication, every downstream reference to that indication across the Warnings and Precautions, Dosage and Administration, Clinical Pharmacology, and Clinical Studies sections must also be removed or appropriately modified. Missing a single cross-reference is a near-automatic deficiency. Experienced auditors run a full cross-reference trace through the entire document before signing off on a labeling package. Most sponsors don’t.

3. Boxed warning fidelity. Boxed warnings must be reproduced character-for-character from the RLD’s current approved label, unless FDA has separately approved a modification. Sponsors sometimes inadvertently reformat a boxed warning during document processing — different line breaks, font handling artifacts introduced during Word-to-PDF conversion, or spacing changes introduced by a document management system. Those formatting deviations are flagged.

4. Nonproprietary name size and placement. Under 21 CFR 201.10, the established (nonproprietary) name must appear with at least half the type size of any proprietary name on prescription drug labeling. Auditors verify this on both the container label and the carton — and they check it on the actual submitted artwork, not the sponsor’s internal style guide. Differences between digital mockup and print production have caused submissions to fail this check.

5. Patient labeling and Medication Guide conformity. When the RLD has an FDA-approved Medication Guide (MedGuide) or patient package insert (PPI), the ANDA must include a conforming version. Any language in a MedGuide instructing patients not to substitute products — particularly relevant for narrow therapeutic index drugs — must be reproduced accurately. A careless edit to a MedGuide signals to reviewers that the sponsor’s labeling review process isn’t mature. That perception doesn’t help the rest of the application.

Therapeutic Substitution Claims and the Compliance Tripwire Most Teams Miss

The phrase “therapeutically equivalent” carries a precise regulatory meaning. The Orange Book assigns a two-letter code to each approved drug product: an “A” rating means therapeutically equivalent to the RLD; a “B” rating means it is not currently considered so. Generic manufacturers cannot make therapeutic equivalence claims in their labeling that exceed what the Orange Book code explicitly permits.

But there’s a subtler trap that catches teams who aren’t paying close attention. Some ANDA sponsors — particularly those handling complex drug-device combinations, modified-release formulations, or locally acting drugs — include language in their package inserts or patient-facing materials that implies broader substitutability than FDA has actually evaluated for that product. This can pull in the Office of Prescription Drug Promotion (OPDP) in addition to OGD. Having two divisions simultaneously involved in your labeling deficiency is not a situation you recover from quickly.

For biosimilars, this tension plays out even more sharply. The 351(k) pathway under the Biologics Price Competition and Innovation Act (BPCI Act) carries its own interchangeability standard: demonstrating that the biosimilar product can be substituted for the reference biologic without the intervention of the prescribing healthcare provider. An “interchangeable” designation comes with specific labeling permissions and prohibitions that differ materially from a standard biosimilar label. Auditors reviewing biosimilar BLAs treat the interchangeability section with the same rigor as a carve-out in a small-molecule ANDA.

The Orange Book check is one step many sponsors treat as administrative rather than substantive. It isn’t. Verifying that your proposed labeling language — including what you’ve omitted — aligns with the assigned therapeutic equivalence code is part of what a serious pre-submission labeling audit covers.

How AI-Augmented Labeling Reviews Are Changing Pre-Submission Audit Prep

The traditional approach to ANDA labeling review involves a regulatory affairs specialist running a manual redline comparison between the ANDA label draft and the current RLD label pulled from DailyMed. For a 40-page prescribing information document, that comparison takes a skilled reviewer 8 to 12 hours. Add a carve-out cross-reference trace, a MedGuide conformity check, and a container label review against the print-ready artwork, and you’re looking at 15 to 20 hours of document review before you’ve even reached any substantive scientific questions.

AI-augmented document review tools are compressing that cycle meaningfully. Models trained on FDA labeling corpora can identify structural mismatches against PLR format, flag cross-references that weren’t carried through a carve-out, and surface promotional language patterns that have historically drawn OPDP scrutiny. Some tools run a parallel comparison against the Orange Book therapeutic equivalence code to verify that the labeling’s substitution language — or the absence of it — is consistent with the assigned rating. That’s exactly the kind of systematic check that’s easy to miss under time pressure.

At Aurora TIC, our regulatory compliance consulting engagements for ANDA sponsors increasingly incorporate AI-assisted pre-submission labeling reviews as a core element of audit preparation. What previously required a two-week manual review cycle can now be structured as a 72-hour AI-assisted document sweep followed by targeted expert review of flagged items — the work where human judgment actually matters. The result is a higher-confidence submission package and a measurably lower rate of labeling-related CRLs going into the first review cycle.

And it’s worth being direct about what AI doesn’t replace here: the expert judgment required to evaluate whether a proposed carve-out is legally defensible under current Hatch-Waxman case law, whether a MedGuide modification will satisfy FDA’s patient comprehension standard under 21 CFR Part 208, or whether a particular container label design creates a medication error risk under ISMP guidance. Those decisions require a trained regulatory professional with knowledge of the specific product, its patent landscape, and FDA’s recent precedent on similar filings. The document-comparison work — the high-stakes task of verifying 400 cross-references for consistency after a carve-out — is where AI adds the most value without substituting for expertise.

A Pre-Submission Labeling Audit Checklist

If you’re preparing an ANDA package or approaching a pre-ANDA meeting with FDA, these five steps address the most common deficiency patterns from the last three GDUFA performance cycles:

1. Pull the current RLD label from DailyMed the week before you finalize your draft — not three months earlier. Labeling supplements for the RLD can be approved without advance notice, and submitting against an outdated version of the reference label is an avoidable error.
2. Run and document a full cross-reference trace for every indication, condition of use, or patient population affected by any section viii carve-out. Trace through every section of the label and record what was modified or removed.
3. Verify your Orange Book therapeutic equivalence code for the RLD and confirm that your label’s substitution language — or its absence — is consistent with that rating.
4. Check container label type size ratios under 21 CFR 201.10 on the actual print-ready artwork or a printed mockup, not solely in the digital file. Rendering differences between screen and print have caused otherwise compliant labels to fail.
5. Compare your MedGuide against the version currently approved for the RLD, not the version that existed when you began development. Use DailyMed, not your internal document archive.

These steps won’t catch every possible deficiency — labeling is creative in ways that keep regulatory professionals employed — but they directly address the patterns OGD has flagged most consistently since GDUFA I. GDUFA III’s performance goals target a 90% first-cycle review completion rate for standard ANDAs within the 12-month goal date. A labeling CRL that forces a second review cycle is a 3 to 6 month commercial setback at minimum. The pre-submission work to avoid it is measurably cheaper.

Written by Sam Sammane, Founder & CEO, Aurora TIC | Founder, Qalitex Group. Learn more about our team

Reserve early access to our AI audit tools Contact us

Related from our network

• USP and Compendial Testing for Generic Drug ANDA Analytical Packages — Qalitex Laboratories provides ISO 17025-accredited compendial and method validation testing to support ANDA submissions with FDA-ready documentation.
• Canadian Generic Drug Registration and Health Canada GMP Compliance — Androxa supports pharmaceutical manufacturers navigating Health Canada’s generic drug approval pathways and GMP audit requirements.