The core science — scoring scales, statistical methodology, inclusion/exclusion criteria — is essentially unchanged between the two drafts. The most important change is a structural one: the 2026 revision splits what was previously a single combined study design into two distinct pathways.
1. Biggest Change: Irritation-Only Studies Now Have Their Own Pathway
Revision 1 (2023) treated irritation and sensitization as a single combined evaluation. Section IV was literally titled "Combined Evaluations of Skin Irritation and Sensitization," and all I/S studies were expected to enroll at least 200 evaluable subjects, with irritation-only sample sizing only loosely referenced.
Revision 2 (2026) splits this into two distinct designs, each with its own sample-size requirement:
- Irritation-only study: minimum of 40 evaluable subjects (or enough to power the study at 0.80 or higher, whichever is greater). Requires only a single 21-day induction phase — no rest period or challenge phase needed.
- Combined irritation + sensitization study: minimum of 200 evaluable subjects. Requires the full design — 21-day induction phase, followed by a 14- to 17-day rest period, followed by a challenge phase (as in the 2023 version).
This is a meaningful cost and design consideration. Many sponsors may now be able to run a much smaller irritation-only study if their product does not trigger the sensitization-testing criteria described below.
2. Clearer Trigger for When a Sensitization Study Is Needed
Revision 1 (2023) allowed sensitization testing to be skipped based on general "adequate justification," or where FDA determined it was "unnecessary or unethical" (for example, where the active ingredient is already a known sensitizer).
Revision 2 (2026) adds a concrete, actionable trigger: a combined irritation-and-sensitization study is recommended specifically when the T product contains an inactive ingredient that:
- is not contained in the RLD and is known to be sensitizing, or
- is used at an amount not previously used in an FDA-approved drug product with a similar context of use.
Outside of this trigger, an irritation-only study is generally considered sufficient. This gives sponsors a clearer, more defensible basis for deciding — and justifying to FDA — which study pathway applies to their product.
3. Reorganized and Expanded Section IV
Revision 2 breaks the previously monolithic Section IV into clearly labeled subsections:
- Study Design Recommendations
- Study Conduct Considerations
- Recommended Inclusion and Exclusion Criteria
- Scoring Scales for Dermal Responses
- Calculating and Formatting Dermal Response Scores
This is largely an organizational improvement rather than a substantive change, but it makes the new irritation-only vs. combined-study distinction much easier to follow throughout the document.
4. Terminology Shift
Revision 1 uses the combined shorthand "I/S" (irritation/sensitization) consistently throughout the document, treating the two assessments as one unit. Revision 2 largely retires this shorthand and refers to irritation and sensitization as more distinct assessments and analyses — consistent with the new bifurcated framework introduced in the study design section.
5. Minor Administrative and Reference Updates
- FDA contact person changed from Melissa Mannion to Susan Levine.
- Updated cross-reference to the *Controlled Correspondence Related to Generic Drug Development* guidance, from the December 2020 edition to the March 2024 edition.
- Vehicle/positive-control TDS wording tightened slightly (for example, "0.1% sodium lauryl sulfate" in Revision 2 versus "less than or equal to 0.1 percent" in Revision 1).
What Stayed the Same
- Scoring scales: The two dermal-response scoring scales (Scale 1: skin appearance, scored 0–7; Scale 2: other effects, scored A–H) are identical in both versions.
- Statistical framework: The mean irritation score (MIS) remains the primary endpoint, with a noninferiority margin of 0.20 and the same hypothesis-testing approach.
- Inclusion/exclusion criteria: Subject eligibility requirements are substantively unchanged.
- Study conduct rules: Randomization, blinding expectations, TDS detachment/replacement rules, and allowances for bathing/showering during the study are the same.
- Vehicle and positive-control TDS approach: Substantively unchanged.
- Partial (cut) TDS provisions: Substantively unchanged.
- Data submission format: SAS transport data sets in XPT format, with define files, remain the required submission format.
The differences are generated by AI models and can contain mistakes. Please double-check the content.