Dermatological Testing: ROAT and HRIPT

Dermatological testing places the finished cosmetic product on human skin, under controlled conditions and clinical observation, to establish whether it irritates and whether it sensitises. These are different questions with different mechanisms, and they are answered by different tests. The Repeated Open Application Test (ROAT) answers the first. The Human Repeat Insult Patch Test (HRIPT) answers the second.

Both are conducted by Oxford Biosciences under the supervision of our Clinical Pharmacologist. Clinical pharmacology is the science of how substances behave in contact with the human body, and a skin compatibility study is exactly that: a controlled exposure, an observed response, and a reasoned conclusion about the product’s behaviour in use.

Irritation and sensitisation are not the same thing

Irritant contact dermatitis is a direct, non-immunological injury to the skin. It occurs on first exposure if the dose is high enough, its severity is proportional to concentration and duration of contact, it affects anyone whose skin is exposed, and it resolves when exposure stops. Surfactants, solvents, acids, alkalis and some essential oils irritate in this way.

Allergic contact dermatitis is a type IV, delayed-type hypersensitivity reaction mediated by T cells. It requires a prior sensitising exposure, during which the substance, usually a small reactive molecule acting as a hapten, binds to skin proteins, is processed by Langerhans cells and presented to naive T cells in the draining lymph node, generating a population of memory T cells specific to that hapten. On subsequent exposure, even to a very small dose, those memory cells mount an inflammatory response that appears 24 to 72 hours after contact. Once acquired, sensitisation is usually lifelong. Fragrance materials, preservatives such as methylisothiazolinone and formaldehyde releasers, and metals such as nickel are classic sensitisers.

A product that does not irritate may still sensitise. A product that irritates at high concentration may be non-sensitising. This is why the two tests exist, and why a claim that a product is “hypoallergenic” cannot be supported by an irritation study alone.

The Repeated Open Application Test (ROAT)

The ROAT was described by Hannuksela and Salo in 1986 as a provocative use test: instead of applying the product under an occlusive patch, which exaggerates penetration and produces reactions that may never occur in real use, the product is applied openly to a defined site, repeatedly, in the way a consumer would apply it. It is the test that most closely reproduces actual use of a leave-on product and it is our standard method for establishing skin compatibility and the acute irritation potential of a finished product.

Study design

The study is monocentric and simple-blind: the subject does not know which site carries the product and which the control. It is conducted under Oxford Biosciences’ internal procedures, in accordance with the Declaration of Helsinki (1964, as amended) and as closely as possible to ICH E6(R2) Good Clinical Practice. Subject data are processed under Regulation (EU) 2016/679 (GDPR), with each subject identified by a code rather than a name. The methodology follows the COLIPA guidelines on the assessment of human skin compatibility (2nd edition, 1997) and the published cumulative irritation methodology.

A cosmetic tolerance study of this kind is non-interventional: the product has already been assessed as safe, no medicinal product is administered and no diagnostic or therapeutic procedure is applied, so it falls outside the clinical trials framework for medicinal products and does not require Research Ethics Committee approval. The sponsor confirms in writing before the study begins that the product’s safety has been assured.

The legal framework for the study

ROAT and HRIPT reports from some providers still state that the study is non-interventional under Directive 2001/20/EC, the Clinical Trials Directive. That Directive was repealed by Regulation (EU) No 536/2014, the Clinical Trials Regulation, which became applicable on 31 January 2022 when the EU Clinical Trials Information System went live, with a transition period for trials already authorised under the Directive that ended on 30 January 2025. In the EU, the instrument that now defines a clinical trial, and therefore defines what falls outside one, is Regulation 536/2014.

The conclusion is the same under both instruments, but the reasoning has to cite the right one. Article 2(2) of Regulation 536/2014 defines a clinical trial as a clinical study in which the assignment of the subject to a particular therapeutic strategy is decided in advance, a medicinal product is administered, or diagnostic or monitoring procedures beyond normal clinical practice are applied. A cosmetic skin compatibility study administers a cosmetic product, not a medicinal product, so it is not a clinical trial within the meaning of the Regulation and its authorisation, ethics committee and safety reporting provisions do not apply. The obligations that do apply come from the Declaration of Helsinki, ICH E6(R2) as a standard of good practice, GDPR for the subjects’ personal data, and the Cosmetics Regulation itself, which requires that the product have a safety assessment before it is placed on human skin.

In the United Kingdom, Regulation 536/2014 never applied, because it became applicable after the UK left the EU. Clinical trials of medicinal products in the UK continue to be governed by the Medicines for Human Use (Clinical Trials) Regulations 2004, which implemented Directive 2001/20/EC and remain in force as amended, and research involving human participants is governed by the UK Policy Framework for Health and Social Care Research. A cosmetic tolerance study is outside the 2004 Regulations for the same reason it is outside 536/2014: no investigational medicinal product is involved. Our study reports cite the applicable instrument for the market the study is intended to serve. A client presented with a report from elsewhere that still cites Directive 2001/20/EC as the current law should ask for it to be corrected.

Subjects

Ten subjects are analysed, with eleven enrolled so that a single withdrawal does not reduce the panel below ten. Panels are balanced by sex, aged 18 to 70, with Fitzpatrick phototypes I to IV so that erythema can be scored reliably against the skin’s baseline colour. All skin types are accepted.

Inclusion requires free, informed, written consent; no previous history of intolerance or allergic reaction to cosmetic products; and willingness to adhere to the protocol. Subjects are not included if they are pregnant, breastfeeding or planning pregnancy; have a cutaneous pathology on the test zone such as psoriasis, eczema, vitiligo, pityriasis versicolor or acne; are taking medication that could interfere with the evaluation of skin tolerance, in the investigator’s judgement; have exposed the test zone to sun or UV in the previous month; have very reactive skin; have marked hairiness, freckles, naevi or a tattoo on the test zone; have a serious or progressive disease; or are enrolled in another study affecting the test zone.

Application

The product is applied undiluted, at a dose of 2 g, to the antecubital fossa, the inner surface of the elbow. This site is chosen because its skin is thin, relatively hairless and readily flexed, which makes it one of the more sensitive sites on the body and gives the test a margin of conservatism over the face or forearm. The contralateral antecubital fossa receives deionised water as the control. Application continues for 5 consecutive days.

Readings

Each site is examined at 30 minutes and 24 hours after the first application and then daily, under standardised lighting, by the same trained assessor throughout. Erythema and oedema are each graded on the following scale at every reading:

Score	Grade	Erythema	Oedema
0	Absent	No erythema	No oedema
0.5	Very mild	Barely detectable: discreet pinkness of part of the test area	Palpable, barely visible
1	Mild	Discreet pinkness of the whole test area, or clearly visible on part of it	Palpable and visible
2	Moderate	Clearly distinguishable dull red erythema covering the whole test area	Obvious oedema, thickness under 1 mm, with or without papules or vesicles
3	Severe	Deep dark or fiery bright red covering the whole test area, or moderate erythema spreading beyond it	Severe oedema, thickness 1 mm or more, or spreading beyond the test area, with or without vesicles or blisters

Changes in skin structure that could be linked to the product, namely dryness, roughness, thickening and reflectivity, are described clinically and graded 0.5 (doubtful), 1 (mild), 2 (obvious) or 3 (important). Any subject with a reaction scored above 1 returns to the centre for further readings until the reaction has fully resolved, and the time to resolution is recorded.

Calculation and classification

For each subject the Cumulative Irritation Index (CII) is the sum of the erythema and oedema scores across all readings divided by the number of readings. The Mean Cumulative Irritation Index (MCII) is the sum of the individual CII values divided by the number of subjects. The maximum possible MCII is 6. The product is classified as follows:

MCII	Classification
Below 0.25	Non-irritating
0.25 to below 0.50	Very slightly irritating
0.50 to below 1.00	Slightly irritating
1.00 to below 2.00	Moderately irritating
2.00 or above	Irritating

The classification is not the whole answer. Individual values are read alongside the mean: a product with an MCII of 0.10 produced by one subject scoring 1 at every reading and nine scoring zero is a different product from one where every subject scored 0.1, and the report says so. The conclusion is written for the product under the conditions of the study, and where the leave-on product is intended for the face, eye area or a site of higher permeability, that is stated as a limit on what the antecubital fossa result can support.

The Human Repeat Insult Patch Test (HRIPT)

The HRIPT is the standard human test for the sensitisation potential of a finished cosmetic product. It descends from the Draize repeat insult test of 1944 and was refined by Marzulli and Maibach in the 1970s into the design used today. Its purpose is not to find out whether an unknown ingredient is a sensitiser; that question belongs to the ingredient safety assessment, where the SCCS has been consistent in its Notes of Guidance that human testing is not an acceptable means of hazard identification. The HRIPT is a confirmatory study on a finished product that the safety assessment has already found to present a low sensitisation risk, and its function is to demonstrate that, under exaggerated and repeated exposure, the product did not induce sensitisation in a human panel. That is the evidence base for a hypoallergenic claim.

Study design

The HRIPT is conducted under the same ethical and quality framework as the ROAT: Declaration of Helsinki, ICH E6(R2) as closely as applicable, informed consent, GDPR, inclusion and non-inclusion criteria, adverse event reporting and follow-up to resolution. The design has three phases.

Induction. A patch containing the product is applied to the same site on the upper back, or on the outer upper arm, and left in place for 24 hours. Semi-occlusive or occlusive patches are used according to the product type: occlusive for rinse-off products and low-viscosity leave-ons, semi-occlusive for creams and balms, with the product applied in the quantity specified for the chamber size, typically 0.02 mL per 8 mm chamber. The patch is removed by the subject or the clinic after 24 hours, the site is read 24 hours later, and a fresh patch is applied to the same site. This is repeated three times a week for three weeks, giving nine consecutive applications to the same site. Any reaction during induction is scored and, if it reaches a defined threshold, the site is moved to adjacent skin so that a cumulative irritant response is not mistaken for sensitisation.

Rest. No product is applied for 10 to 14 days. This interval allows any irritant reaction from the induction phase to resolve and, critically, allows time for the immunological events of sensitisation, if they have occurred, to generate memory T cells.

Challenge. A single patch containing the product is applied for 24 hours to a naive site that has never received the product, and, in most designs, to the original induction site as well. The sites are read at 24, 48 and 72 hours after patch removal, and at 96 hours where any reaction is present. A reaction at the naive site that appears or intensifies at 48 to 72 hours is the signature of a delayed-type hypersensitivity response and is the finding the test exists to detect. A reaction that appears at 24 hours and fades is irritant. Where a challenge reaction is equivocal, the subject is re-challenged after a further rest period to confirm or exclude sensitisation.

Panel size

The standard panel is 50 subjects, which is the minimum on which a negative HRIPT result is considered meaningful. For hypoallergenic claims, for products intended for the United States market where 100-subject or larger panels are customary, and for products with a fragrance load or a known weak sensitiser at a level judged acceptable by the safety assessment, larger panels are advised, and we design the study to the claim it has to support.

Scoring

Reactions at each reading are graded on a five-point scale from 0 (no reaction) through erythema, erythema with oedema, erythema with oedema and vesicles, to bullous reaction, using the International Contact Dermatitis Research Group convention. The report presents the score for every subject at every reading in induction and challenge, the number of subjects completing the study, the reason for every withdrawal, and the investigator’s conclusion on whether any subject showed evidence of sensitisation.

ROAT and HRIPT compared

	ROAT	HRIPT
Question answered	Does the product irritate in use?	Does the product sensitise?
Mechanism assessed	Non-immunological irritation	Type IV delayed hypersensitivity
Exposure	Open application, no patch	Patch, occlusive or semi-occlusive
Site	Antecubital fossa	Upper back or upper arm
Duration	5 days	6 to 7 weeks including rest and challenge
Applications	Daily	Nine induction patches plus challenge
Panel	10 subjects	50 subjects standard, 100 or more where required
Control	Deionised water on the contralateral site	Naive site at challenge; vehicle where relevant
Output	Mean Cumulative Irritation Index and class	Incidence of sensitisation in the panel
Claims supported	Non-irritating, skin compatibility tested, dermatologically tested	Hypoallergenic, non-sensitising, dermatologically tested
Reproduces real use	Yes, by design	No, by design; exposure is exaggerated to be conservative

A product intended for a “dermatologically tested” claim needs at least one of the two, and the CPSR states which. A product intended for a “hypoallergenic” or “suitable for sensitive skin” claim needs the HRIPT, and usually both.

Claims and the Common Criteria

In the EU and UK, Regulation (EU) No 655/2013 lays down the Common Criteria that every cosmetic claim must meet: legal compliance, truthfulness, evidential support, honesty, fairness and informed decision-making. “Dermatologically tested” is a truthful claim only if a dermatological study was conducted on the finished product; “hypoallergenic” is honest only if the product has been formulated to minimise allergenic potential and tested to confirm it; and “non-irritating” or “suitable for sensitive skin” requires evidence on a panel appropriate to the claim. The Commission’s Technical Document on Cosmetic Claims (2017) addresses “hypoallergenic” and “free from” claims specifically and is applied in our review of every claim the study is intended to support. The ROAT and HRIPT reports are written so that they stand as the evidential support in the Product Information File, and the CPSR cross-references them.

What the report contains

• Study parameters page: objective, methodology, kinetics, product, application conditions, population and result
• Experimental protocol: subject characteristics, inclusion and non-inclusion criteria, product specification
• Methodology: equipment, dose, site, duration, control, reading conditions and scoring scales
• Individual reading tables for every subject at every time point
• Calculation of CII and MCII, or sensitisation incidence, and classification
• Results and conclusion signed by the investigator
• Authentication page and quality assurance statement
• Ethical requirements and regulatory standards appendix: adverse event definitions, notification, premature termination, confidentiality and data validation

What to send

For a ROAT, 50 g or mL of product from a single batch. For an HRIPT, 200 g or mL. Both in final packaging where possible. The full formulation with percentages is required before the study, because a product is not placed on a human panel until the safety assessment has confirmed it is appropriate to do so. Studies are scheduled on receipt of product and formulation.

How it fits with the CPSR

Skin compatibility data support the undesirable effects and the conclusion on safety in Part B of the safety report, and they are the substantiation for any tolerance claim made on the label or in marketing. Where we prepare the CPSR, the study is designed to the product and the claim, and the assessment is written from the raw data. Where a client brings a study from another provider, we check the panel, the site, the exposure, the scoring method and the calculation before relying on it, and we say plainly when a “dermatologically tested” certificate cannot bear the claim placed on it.