Skin irritation testing is a core component of the toxicological safety dossier required to register or file a new cosmetic ingredient with regulatory authorities such as China's NMPA, as well as under international frameworks like REACH and UN GHS classification. This article provides a technical overview of the accepted testing items, sample requirements, and laboratory methods—including reconstructed human epidermis (RhE) in vitro assays and OECD and GB-standardized protocols—used to evaluate the dermal irritation potential of novel cosmetic raw materials.
Technical Overview
Skin irritation is defined as the production of reversible damage to the skin following the application of a test substance for up to four hours, clinically manifested as erythema (redness) and edema (swelling) that fully resolve within a defined observation period. It must be clearly distinguished from skin corrosion, which involves irreversible necrotic damage extending through the epidermis into the dermis, often producing ulceration, bleeding, and scarring. Because new cosmetic ingredients (NCIs) are intended for direct or indirect contact with human skin, evaluating their irritation and corrosion potential is a fundamental element of pre-market safety assessment.
Under China's Cosmetic Supervision and Administration Regulation (CSAR), a new cosmetic ingredient must complete registration or record-filing with the National Medical Products Administration (NMPA) before being placed on the market, and skin irritation and corrosion data form part of the mandatory toxicological package unless a scientifically justified waiver applies. Parallel obligations exist under the EU Cosmetics Regulation, the REACH framework for chemical substances, and the UN Globally Harmonized System (GHS) for classification and labeling of raw materials.
Modern testing follows a tiered, integrated strategy that prioritizes animal welfare. Regulators and accredited third-party laboratories typically begin with a desk review of existing human and animal data, structure-based read-across from analogues, and physicochemical screening—for example, substances with extreme pH values (pH ≤ 2 or ≥ 11.5) may be presumed corrosive without further testing. Validated in vitro methods for corrosion (OECD Test Guidelines 430, 431, and 435) and for irritation (OECD Test Guideline 439) are applied before any consideration of confirmatory in vivo testing under OECD Test Guideline 404 or the corresponding GB/T protocol. All pivotal studies intended for regulatory submission should be conducted under Good Laboratory Practice (GLP).
Testing Samples
Neat New Cosmetic Ingredient
The primary test sample is the new cosmetic ingredient itself, submitted as a representative production batch. Typical submission quantities range from 20 g to 50 g (or the equivalent volume for liquids), accompanied by a Certificate of Analysis documenting purity, batch number, manufacturing date, and shelf life. The sponsor should also provide the full composition and impurity profile, physical form, color, odor, solubility characteristics, and pH value, since these parameters directly influence test design and result interpretation.
Dilutions and Use-Concentration Preparations
Depending on the study protocol, the ingredient may be tested neat, diluted to its maximum intended use concentration in a relevant vehicle, or both. Solid materials are typically moistened with water or a suitable vehicle to ensure adequate skin contact. For hydroalcoholic or highly reactive ingredients, the laboratory will document the selected vehicle and verify that it does not itself produce confounding irritation at the concentrations used.
Controls and Reference Materials
Each study includes concurrent controls to demonstrate assay performance: a negative control such as phosphate-buffered saline or purified water for the in vitro assays, and a positive control such as 5% sodium dodecyl sulfate (SDS) solution for the RhE irritation test. Reference corrosive and non-corrosive chemicals are used periodically to confirm continued method sensitivity in accordance with the applicable OECD performance standards.
Testing Items
The specific testing items for a new cosmetic ingredient safety package are selected according to the registration pathway, anticipated exposure conditions, and available existing data. Core and commonly requested items include:
- In vitro skin corrosion screening using reconstructed human epidermis (RhE) models, the transcutaneous electrical resistance (TER) test, or the membrane barrier test method (OECD TG 431, TG 430, and TG 435).
- In vitro acute skin irritation testing using the reconstructed human epidermis test method (OECD TG 439).
- In vivo acute dermal irritation and corrosion testing in rabbits where in vitro data are inconclusive or where the receiving jurisdiction requires it (OECD TG 404, GB/T 21608, and the methods published in the Safety and Technical Standards for Cosmetics).
- Repeated or cumulative dermal irritation testing, typically involving 14 consecutive days of application, to simulate long-term daily exposure conditions relevant to cosmetic use.
- Supporting endpoints where applicable, such as phototoxicity assessment for UV-absorbing ingredients and skin sensitization testing, which is a separate toxicological endpoint but frequently evaluated in parallel.
Testing Methods
In Vitro RhE Skin Irritation Assay (OECD TG 439)
The RhE test uses a three-dimensional model of human epidermis reconstructed from primary human keratinocytes, reproducing the stratified architecture and barrier properties of native skin. The test chemical (30 µL for liquids or 30 mg for solids) is applied topically to the tissue surface for an exposure period of 42 minutes, followed by a post-treatment incubation of 42 hours. Tissue viability is then quantified by MTT reduction and expressed relative to the concurrent negative control. A mean tissue viability of 50% or below classifies the substance as a GHS Category 2 skin irritant, while viability above 50% supports a no-category (non-irritant) conclusion. Assay acceptance requires a negative control optical density meeting the model-specific minimum, positive control viability at or below 20%, and acceptable within-test variability.
In Vitro Corrosion Pre-Screening
RhE corrosion protocols under OECD TG 431 apply sequential exposure durations—for example, 3 minutes, 1 hour, and 4 hours—with viability cut-offs that assign GHS corrosivity sub-categories 1A, 1B, or 1C. Alternative validated approaches include the TER test on rat skin disks (OECD TG 430) and the membrane barrier test (OECD TG 435), which measures the time required for a chemical to penetrate a synthetic biobarrier. A positive corrosion result generally eliminates the need for further irritation testing and may trigger severe hazard classification and labeling requirements.
In Vivo Acute Dermal Irritation Test (OECD TG 404 / GB/T 21608)
Where in vivo confirmation is justified and legally acceptable, the classic rabbit patch test is performed on the intact, clipped skin of albino rabbits. A single dose of approximately 0.5 g or 0.5 mL is applied to a gauze patch covering roughly 6 cm² under semi-occlusive conditions for 4 hours. Dermal responses are graded at approximately 1, 24, 48, and 72 hours after patch removal, with observation continued until reversal or up to 14 days to confirm reversibility. Erythema, eschar formation, and edema are each scored on the standardized Draize scale from 0 to 4. Results are expressed as the Primary Irritation Index, calculated from the combined scores divided by the number of observation points, with classifications typically defined as non-irritant (score of 0), mild irritation (0 to 2), moderate irritation (2 to 5), and severe irritation (5 to 8), noting that exact thresholds vary slightly among jurisdictions.
Cumulative Skin Irritation Testing
To reflect realistic repeated-use exposure, cumulative irritation studies apply the test preparation to intact rabbit skin once daily for 14 consecutive days, with daily grading of erythema and edema and calculation of a cumulative irritation index. This item is particularly relevant for ingredients intended for leave-on products or for raw materials with borderline acute findings.
Study Reporting Requirements
A GLP-compliant study report must document the test system, vehicle, dose levels, environmental conditions, individual animal or tissue data with time-point scores, calculations, classification conclusions, and any deviations. Accredited third-party laboratories experienced in cosmetic toxicology can advise on the optimal testing sequence to satisfy both Chinese and international submission requirements while minimizing redundant animal use.
Applications
Skin irritation test results serve multiple purposes across the product lifecycle of a new cosmetic ingredient:
- Regulatory registration and record-filing dossiers for new cosmetic ingredients submitted to authorities such as the NMPA, supported by GLP study reports.
- Quantitative safety assessments, where irritation data contribute to the overall toxicological profile used in margin-of-safety calculations for finished products.
- GHS and CLP classification and labeling of the raw material, including hazard statements, precautionary statements, and safety data sheet preparation for the supply chain.
- Formulation development, helping brand owners set maximum use concentrations, select warning statements, and design follow-up confirmatory testing such as human repeated insult patch testing where appropriate.
- International market access, since OECD-compliant in vitro data are widely accepted across the European Union, North America, and Asia-Pacific jurisdictions, streamlining multi-market submissions.
- Post-market surveillance and defense files demonstrating that adequate safety evaluation was performed prior to commercialization.
FAQ
Is animal testing mandatory for new cosmetic ingredient registration in China?
Not in every case. China's regulatory framework increasingly recognizes validated alternative methods, and existing toxicological data, read-across justifications, and computational assessments can be used where scientifically supported. However, for certain registration categories and where alternative data are insufficient, in vivo studies conducted under GLP at accredited third-party laboratories may still be required. The testing strategy should be confirmed against the current technical guidelines before study initiation.
What happens if the RhE viability result is 50% or below?
A mean tissue viability at or below 50% under OECD TG 439 conditions classifies the ingredient as a GHS Category 2 skin irritant. This does not automatically prohibit cosmetic use; instead, it informs concentration limits, labeling warnings, and the need for confirmatory or risk-management measures depending on the intended product category and exposure level.
How much sample should be submitted for testing?
For a single irritation endpoint, approximately 20 to 50 g of the representative batch is generally sufficient, including allowance for repeat or confirmatory analysis. Sponsors should also provide batch documentation, physicochemical specifications, and safety handling information to support protocol design.
What is the typical turnaround time?
RhE in vitro studies are typically completed within two to four weeks including reporting. In vivo acute studies require approximately three to four weeks plus observation periods, while cumulative 14-day studies may extend to five to six weeks. Additional time should be allocated for protocol approval, sample logistics, and quality assurance audit of the final report.
Can existing data or read-across replace new testing?
Yes, where the data are of adequate quality, relevance, and reliability. High-quality human data, well-documented historical animal studies, and structurally justified read-across from close analogues can support waivers, subject to acceptance by the reviewing authority. A weight-of-evidence rationale should be prepared and documented in the safety assessment file.
Are there situations where irritation testing can be waived?
Waivers may be justified for ingredients with extreme pH or strong buffering capacity that are presumed corrosive, for substances already comprehensively assessed under other regulatory programs, or where exposure is negligible by design. Each waiver requires a documented scientific rationale consistent with applicable regulatory guidance, and accredited third-party laboratories and regulatory consultants can assist in constructing defensible justifications.