| Front page | | Contents | | Previous | | Next |

Guideline on Safety assessment of cosmetic products

Annex 3

Synopsis of efficacy testing of cosmetic products

Synopsis prepared by Prof. Jørgen Serup. M.D., Ph.D.
Ingeborgvej 42, DK-2900 Hellerup, Denmark

Contents

Introduction

Council Directive 93/35/EEC of 14 June 1993 (the Cosmetics Directive) (1), which has been implemented in Denmark, states in Article 7a, 1.g:

The manufacturer or his agent or the person to whose order a cosmetic product is manufactured or the person responsible for placing an imported cosmetic product on the Community market shall for control purposes keep the following information readily accessible to the competent authorities of the Member State concerned at the address specified on the label in accordance with Article 6 (1) (a):
...
g) proof of the effect claimed for the cosmetic product, where justified by the nature of the effect or product.

The Cosmetics Directive defines the nature of the effects of cosmetic products in Article 1, 1:

A "cosmetic product" shall mean any substance or preparation intended to be placed in contact with the various external parts of the human body (epidermis, hair system, nails, lips and external genital organs) or with the teeth and the mucous membranes of the oral cavity with a view exclusively or mainly to cleaning them, perfuming them, changing their appearance and/or correcting body odours and/or protecting them or keeping them in good condition.

Thus, depending on the nature and claimed effect(s) of the product, documentation must be available, organized and presented swiftly to competent authorities upon request to the responsible person or party. To be appreciated fully, the documentation must directly address skin signs and phenomena. Effects may include product-induced changes noticeable during product use, protective effects of the product against various insults or prophylactic effects in which untreated skin suffers damage or spontaneous worsening, whereas treated skin remains unaffected or in a favourable condition.

The European Union and the responsible authorities in Denmark have not yet given any detailed guidance to the cosmetic industry on the principles and methods for proper testing of efficacy of cosmetic products, although the Scientific Committee on Cosmetic Products and Non-Food Products intended for Consumers recently discussed and finally adopted basic criteria for protocols and human testing of skin compatibility (2,3).

Aim

The aim of this synopsis is to outline principles and standards for efficacy testing of cosmetic products that can be used by both responsible authorities in evaluating products and substantiation and the industry in planning efficacy trials. The standards considered primarily apply to the products that, due to their nature, need justification, but the standards are generic and can be applied to cosmetic products in general.

The synopsis focuses on efficacy and does not address safety issues.

Documentation of biomedical efficacy in perspective

In the history of science, efficacy documentation has passed through phases of hypothetical consideration with deduction to clinical situations, empirical experiments in animals and humans, studies of cases and smaller groups, controlled trials and finally randomized controlled trials, introduced more than 50 years ago and now universally accepted as the gold standard for establishing truth in biomedical research with a clinical objective. Further, according to international consensus, a single randomized controlled trial may not suffice to establish a valid general conclusion, and one or more confirmatory trials at additional centres may be required to establish proof in a larger geographical territory.

Several systems have been described for evaluating quality in research. These are condensed into a concept widely used under the name evidence-based medicine.

According to evidence-based medicine, the validity of trials may be ranked as follows (modified from Eccles et al. (4)):

The principles of evidence-based medicine are not limited to evaluating therapies, whether medicinal or non-pharmaceutical, but are also applied to diagnostic tests and prognostic and preventive studies and are used to evaluate adverse reactions. The principles of evidence-based medicine can be directly adopted as a system for evidence-based cosmetic treatment.

The need for systematic reviews in medicine has resulted in the Cochrane Library (http://www.cochrane.co.uk), a huge database started in 1992 by scientific reviewers based on the proposal of Professor Archie Cochrane, with updated and systematic reviews primarily on therapies and prophylaxis. Reviews are based exclusively on randomized controlled trials. A Cochrane Skin Group was started in 1997 (http://www.nottingham.ac.uk/~muzd).

Thus, the randomized controlled trial is applicable to many disciplines in biomedical research testing intervention in body, mind or the environment, and this gold-standard design is directly applicable to the documentation of cosmetic products as well as the development of pharmaceutical products.

The Declaration of Helsinki and protection of subjects participating in trials

The Declaration of Helsinki, adopted in 1964 by the 18th World Medical Assembly, is an ethical recommendation guiding physicians in biomedical research involving human subjects. It has been amended four times, most recently in October 1996. The aim is to protect the integrity and well-being of a subject in a biomedical study by ensuring proper risk assessment and consideration of ethics by an independent review board. Study participants must be well informed and grant written acceptance: informed consent. The Declaration of Helsinki also addresses critical aspects of research such as the study programme and conduct and the educational needs of the research group.

Denmark adopted these principles in 1992 by adopting the Act on a Scientific-Ethical Committee System and the Treatment of Biomedical Research Projects. Such evaluation is required in every biomedical project that includes humans. In Denmark a system with a number of regional ethics committees and one central committee was established. The nature of a cosmetic product, the claim and the study design may determine whether a study is considered biomedical or not. To assess trials, ethics committees in Denmark need a study protocol, a summary for laypeople, the written information provided to subjects entering the study and an informed consent form. The relevant committee can decide based on informal consultation that the trial is not considered biomedical. Thus, a regional ethics committee must be consulted before any human trial on a cosmetic product is started.

Study data are often stored in electronic registers. The Public Authorities’ Registers (Consolidation) Act from 1978 applicable to a broad range of electronic registers states that registers with personal information in which the integrity of the registrants is at risk must be approved by the Data Surveillance Authority, and the register used must be considered secure. Council Directive 95/46/EC of 24 October 1995 (5), which is not yet fully implemented in Denmark, also instituted such protection. In Denmark, under the present rules and administrative practices, any study registered electronically, or manually, containing personal information must be protected under the auspices of the Data Surveillance Authority after evaluation of the study and the security of the register. Thus, the Agency must approve in writing every systematic study of cosmetic products in Denmark. The Agency requests for the purpose of their assessment a brief description of the trial with emphasis on register and data protection.

The ICH tripartite guideline for good clinical practice

The ICH (International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use) guideline for good clinical practice (ICH GCP), which was primarily developed to document pharmaceutical products, is an international ethical and scientific quality standard for designing, conducting, recording and reporting trials that involve the participation of human subjects (6). Compliance with this now internationally implemented standard provides public assurance that the rights, safety and well-being of trial subjects are protected, consistent with the principles that have their origin in the Declaration of Helsinki and that the clinical trial data are credible. The declared objective of the ICH GCP Guideline is to provide a unified standard for the European Union, Japan and the United States as a basis for mutual recognition of clinical data by the regulatory authorities in those three jurisdictions.

The ICH GCP guideline is based on the following general principles.

	Clinical trials should be conducted in accordance with the ethical principles that have their origin in the Declaration of Helsinki, including informed consent and confidentiality covering any record or document.
	Trials should be scientifically sound and described in a clear, detailed protocol, and the anticipated benefits should justify the anticipated risks.
	Trials should be conducted in compliance with a protocol that has received prior approval by an ethics committee.
	A physician or dentist should always be responsible for medical decisions and the medical care given.
	Test personnel should be adequately educated and trained.
	All clinical trial information should be recorded, handled and stored in a way that allows it to be accurately reported, interpreted and verified.
	Investigational products should be manufactured, handled and stored according to good manufacturing practice and used according to the protocol.
	Systems with procedures that assure the quality of every aspect of the trial should be implemented.

The randomized controlled trial in humans is essential in the ICH GCP. The pharmaceutical industry swiftly adopted ICH GCP during 1997, allocated the necessary resources and established elaborate in-house systems with standard operating procedures, quality assurance and quality control. A major educational effort has been undertaken and resulted in an enthusiastic GCP culture in which things are not underdone. Investigators, project leaders, monitors and quality control officers have refined test practices and helped GCP conduct in several ways. The implementation and maintenance of the guideline has become a continuing process with constant refinements and educational activities.

The GCP guideline is directly applicable to the efficacy testing of cosmetic products.

Ambition level for substantiating the efficacy of cosmetic products

The ICH GCP guideline is directly applicable to trials of cosmetic products on humans. This offers responsible authorities and the product users the highest level of certainty that claims made for cosmetic products hold true. The pharmaceutical industry has demonstrated that GCP systems can be practised to a high degree of sophistication, but establishing and maintaining GCP are costly. The pharmaceutical industry has a long tradition of clinical trials based on various national GCP systems introduced some years ago. This platform was important in the successful introduction of ICH GCP during 1997. The same platform is not present in the cosmetic industry, which additionally may have a different company profile, with few large and numerous small and medium-sized enterprises. Thus, an advanced system such as ICH GCP that ensures the credibility of trials in humans is not yet fully and immediately applicable to cosmetic product efficacy testing. Nevertheless, essential elements can be adopted.

Colipa Guidelines for the evaluation of the efficacy of cosmetic products

Colipa (the European Cosmetic Toiletry and Perfumery Association) represents European manufacturers of cosmetic products. In August 1997, Colipa published Guidelines for the evaluation of the efficacy of cosmetic products, with general information and some statements about human testing, instrumental evaluation and non-human models (7). The guidelines list essentials such as the need for a formal protocol, a responsible person, the objective and relevance of the trial, schedule, statistics and recording and interpreting the results. Nevertheless, the value of the guidelines is restricted by the lack of detailed information and a lack of opinion about the essential elements needed for a trial or documentation to be conclusive. It can only serve as a loose framework to inspire cosmetic scientists. The guidelines accept studies carried out on non-human models and published data on ingredients as documentation of the final product.

The industry probably has a number of in-house standards for efficacy testing in a company context, but such standards, which are likely to be detailed, have typically not reached the public domain. In-house standards may be relevant for authorities in assessing individual products if they are made transparent.

Study objectives and reported endpoints in trials documenting the efficacy of cosmetic products

A logical prerequisite for substantiation of a claim in a trial is that the claim be identical to a defined endpoint in a trial. Studies may have one primary endpoint covering the main claim of the product and a number of secondary endpoints. People use cosmetic products to achieve cutaneous effects that they can see using the naked eye, feel with their fingers or smell. Endpoints assessed by users themselves or a representative panel of observers have first-line relevance. In-use assessment by specially trained evaluators may produce more precise and reproducible information on a smaller sample. Scoring scales and visual analogue scales are typically used in such assessment. Scores and scales must display the full range of expected effects in a reasonably balanced way.

Claims as understood by lay consumers may easily differ from claims worded on package inserts and other written material and defined endpoints in trials, and it must be critically considered that there be no deviation in meaning or appreciation from the trial endpoint to the claim understood by the consumer in real-world situations.

Surrogate endpoints are study objectives that provide indirect evidence about the postulated effect of the product on a user. Surrogates may be quantitative and measurable with very high precision and allow very detailed and reproducible information on a small sample and may therefore be more operable than direct user endpoints. Surrogates may include skin structure and function variables measured with noninvasive instrumental methods, chemical and physical properties, and others. However, the validity of a surrogate endpoint depends entirely on its relevance to the in-use situation under real conditions.

The outcome of in vitro testing and arguments based on product composition clearly represent non-human data distant from real endpoints in a user situation.

Instrumental evaluation of efficacy in humans using noninvasive techniques

Several biophysical and computerized methods for objectively characterizing skin structure and function without invading the skin or interfering with the function measured have been developed in recent decades. The techniques characterize: surface properties such as scaling and dryness, colour and pigmentation (colorimetry), relief and wrinkles (profilometry); skin structures (high frequency ultrasound, confocal microscopy); functions such as blood flow (laser Doppler scanning), surface lipids and sebum production (sebumetry), water evaporation and sweating (evaporimetry); and other parameters. The methods typically measure selected properties dependent on the measuring principle and not overall endpoints as normally appreciated by product users in their self-assessment. Many of the techniques are highly precise and accurate, and significant changes may be demonstrable in a limited number of individuals. Hydration, blood flow, skin elasticity, skin thickness and dermis structure are examples of parameters that can be quantified noninvasively using techniques but only assessed with difficulty clinically. Various methods have been formally validated in published guidelines describing typical variables such as measuring conditions and needs for preconditioning of the test individual, laboratory room conditions, instrumental variation and calibration.

The critical factors in noninvasive techniques are as follows.

	The parameter measured must be relevant to the effect of the product and the claim.
	The laboratory facility must be adequate.
	Staff must be specially educated and trained in the test method and procedure.
	Instruments and measurement must be described in a standard operating procedure, and existing guidelines should be followed.
	The instrument used must be calibrated and stable and remain so during the trial.
	Test subjects must be well informed and adequately preconditioned before measurement.
	The recorded data and samples taken must be properly labelled, handled and processed.

The relevance of the parameter measured, direct or surrogate, is an important issue that requires special argument in a trial. Because of the nature of the techniques, a narrow feature is typically measured under special conditions (such as magnification, amplification or filtering) and with high precision.

When noninvasive techniques are being used to substantiate the final product as a stand-alone test, the researcher and the sponsoring company have a special obligation to argue the relevance of the test relative to the effect, claim and use of the final product and the precision and validity of the instrument and the test procedure actually used in the trial.

Thus, noninvasive techniques may measure endpoints in the documentation of efficacy of final products in trials in humans and even evaluate the endpoint as a stand-alone test if the test can be argued to be relevant and valid in relation to the effect and the product claim. The format of the basic design must be a randomized controlled trial.

The standardization group on noninvasive methods of the European Society of Contact Dermatitis published in the journal Contact Dermatitis various formal guidelines on instrumental evaluation of skin: transepidermal water loss, colour and blood flow. The Handbook of non-invasive methods and the skin provides detailed information on noninvasive methods used for efficacy documentation (8). An informal group named EEMCO (European Group for Efficacy Measurements on Cosmetics and other Topical Products) including industry representatives published various introductory reviews on measuring techniques such as colorimetry and skin elasticity and a guidance paper on clinical evaluation of dryness, which can be used in validating instrumental methods for measuring skin hydration (9).

Information on raw materials and ingredients in substantiating the efficacy of a final product

The efficacy of a final product may refer to a single ingredient, a number of ingredients acting together, the base as such or the complete product. The physicochemical properties and the purity of ingredients, the supplier and the supplied batch of raw material and the whole manufacturing process are known variables influencing the final product. The efficacy of a final product cannot, in general, be proven directly from the recipe listing individual ingredients for the reasons mentioned above, although efficacy may be predicted and argued with some precision depending on the recipe and the background knowledge. Exceptions may occur when, for example, a chemically well characterized ingredient exerts a known, strong effect and is also evaluated with respect to requirements in a formulation and found uncomplicated to formulate or relatively independent of the vehicle. Essentials such as the relationship between dose and concentration or dose and effect, optimum pH and ingredient stability must be known. Peeling agents, acting as corrosive irritants, may represent such an exception, although efficacy always varies somewhat according to vehicle composition and properties.

Thus, strong arguments are needed to substantiate the efficacy of a final product based on information about the recipe and individual ingredients, and proof of efficacy of final products should, with exceptions, depend on the outcome of relevant testing in humans with application of the product according to anticipated use.

Minor modifications of formulations for which claims have already been established and efficacy proven are a special issue. Proof of efficacy may be maintained if arguments are presented that the change is unimportant for the efficacy endpoint.

Products manufactured using the same recipe by different companies or the same company in different plants are not automatically biologically equivalent. Chemical and physical characterization of the original product and the copy may suffice for documenting equivalence, and this may be backed up by demonstrating similar penetration of the product or essential ingredients into artificial or human skin. A range of sophisticated techniques are known from skin pharmacology.

Efficacy testing directly in humans remains the ultimate test.

In vitro efficacy models

A number of in vitro efficacy models exist. These models can be useful tools in the research and development of a product, for example for the purposes of selecting the best ingredient among various options, for estimating the relationship between dose and efficacy in the laboratory and for other purposes that lead to a best-fit product candidate that is ready to be tested in humans. Nevertheless, in vitro testing can only provide a rough estimate and only add marginal information about a final product unless the test in a proper document has been validated in relation to the anticipated human in-use situation and deemed suitable for substantiating the efficacy of such a final product with respect to the specified effect.

Thus, in vitro tests are important research tools. Nevertheless, these tools are generally not acceptable in documenting the efficacy of final products unless their validity with regard to user relevance and the claim of the product is documented.

Reference to literature

Reference to literature is important in explaining the background of a product, discussing the findings of a test and determining the validity of a test method. Thus, critical review and reference to literature is an important instrument in the process of substantiating the efficacy of cosmetic products in various ways but in itself never comprises proof replacing a randomized controlled trial. Literature remains indicative. Referenced literature must be in depth and complete and at an academic level, carefully selected on rational grounds and presented in a non-promotional way with balanced information about advantages and disadvantages.

General principles and conclusion

Any claim that must be verified based on the nature of the product and that can be measured must be documented in a randomized controlled trial in humans who are a representative sample of the anticipated user group.

Trials in humans must be conducted in accordance with the intention and main principles of the ICH harmonized tripartite guideline for good clinical practice, which gives international ethical and scientific quality standards for designing, recording and reporting trials that involve the participation of human subjects, but strict adherence to this guideline is not realistic and therefore not felt mandatory in documenting the efficacy of cosmetic products.

Documentation must be based on testing of the final product identical to the product to be marketed.

Claims must be identical to endpoints measured in trials.

User self-assessment and evaluation of user-relevant effects by panellists or trained observers are gold-standard endpoints in trials. Endpoints representing skin structure and function measured by biophysical methods are also acceptable as endpoints if the methods are validated and demonstrated to be precise and if results are shown to be directly relevant to the claimed effect in users.

Data on raw materials and ingredients in a product, pure laboratory testing and reference to the literature are generally informative but not acceptable for documenting the efficacy of a final cosmetic product.

Documentation must always be scientifically sound and relevant to the claim and the intended use of the product.

The documentation must not make claims or be laid out is such a way that various statements dilute, scatter or modify or deviate in opinion or impression from what was actually documented through experimentation.

Notes on design and critical elements in human testing of the efficacy of cosmetic products

As stated above, the elaborate ICH GCP system for human testing is not considered to be realistic for the cosmetic industry at this time. Elements that may be sought implemented immediately in efficacy testing of cosmetic products are addressed in the following.

Evidence-based documentation of the efficacy of cosmetic products

For cosmetic products, as a general rule, the acceptable rank of evidence should be category 1: a minimum of one well-conducted randomized controlled trial. Category 2a may also be acceptable if proper blinding and randomization cannot be conducted for technical or ethical reasons, for example in the documentation of a peeling agent when visible irritation is closely linked with efficacy. Category 2b is acceptable if the test method actually used is validated and its user relevance documented, including determination of the minimum user-relevant difference.

Ethical testing

The spirit of the Declaration of Helsinki would indicate that written information and informed consent should be standard in cosmetic product testing, although it may not be formally required depending on evaluation of a particular trial by a local review board. In Denmark the Data Surveillance Authority must approve a study. Ethics and testing of cosmetic products were addressed previously in this synopsis.

Sponsor and dossier on efficacy

A sponsoring company, institution or person should be assigned to take the main responsibility, including liability, for test subjects. This sponsor should organize a dossier on efficacy documentation, properly secured and with all relevant trial material readily available. British Standard BS 5454:2000 (10) addresses storage conditions and secure storage of source data and documents.

Who can be accepted as responsible for cosmetic product testing?

Cosmetic products are not medicines and differ formally and by their nature and effect-to-safety ratio from medicinal products, and a researcher with documented relevant education and qualification can therefore take responsibility for a test and a research team. Qualifications can be documented in a curriculum vitae and an educational record. Supervision by a physician is not felt necessary unless a medical problem arises or is likely to arise.

Trial protocol

A detailed trial protocol must be prepared before the trial is conducted and properly dated and signed off by the investigator and the sponsor.

Trial design

A randomized controlled trial in humans is the gold standard to be used whenever applicable, independent of the trial endpoint and the method of recording.

Comparator product

A placebo control should be chosen whenever possible. When efficacy is supposed to be related to a single or a few ingredients that can be omitted in the formulation without changing its physical properties and odour, such a placebo formulation without active ingredient(s) is preferable. Special attention should be focused on the need for similarity between test product and comparator with respect to fragrance and odour, viscosity and colour. It may sometimes be impossible to identify or omit active ingredients. A marketed, neutral product claimed to be a moisturizer or a cream base or an otherwise relevant but inert marketed product similar to the test product with respect to odour, viscosity and colour may be chosen as the comparator. The relevant fragrance may be added to the comparator product, if necessary, to approximate similarity. An experimental formulation may also be used, but this must be formulated as an inert reference without any anticipated endpoint effect, positive or negative, that could invalidate it as a comparator formulation. The lack of difference between investigational and comparator products in physical properties may be documented in the laboratory and confirmed in a pilot study in humans that also addresses odour.

Production of test product

Test products should be produced under quality conditions according to the Council of Europe Guidelines for Good Manufacturing Practice of Cosmetic Products (GMPC) from 1995 (11), ideally after scale up in the production line used for the final production.

Calculation of sample size to estimate the size of the groups needed

The state-of-the-art requirement in trials is a statistical power calculation by a statistician. True differences are overlooked if groups are too small, especially if the expected endpoint difference among groups compared is minor and the variation in the measured endpoint is major. It is in the interest of the sponsor that test groups not be too small and that true effect as a basis for claims not be overlooked. Groups smaller than 20 are not usually considered convincing in substantiating cosmetic products that normally induce minor changes, in contrast to medicinal products used for skin diseases. Depending on the study and products tested, major statistically significant differences obtained in groups smaller than 20 might arouse suspicion that the study suffers from a major bias or imperfection of blinding, for example, related to the selection of test subjects, the cosmetic products applied or the measuring methods.

Selecting test subjects according to defined inclusion and exclusion criteria

Study groups must realistically represent the range of anticipated users of the product with respect to such factors as sex, age and social class. Participants with skin diseases or sun-damaged skin may invalidate the results. Positive (inclusion) and negative (exclusion or non-inclusion) criteria should be defined in the trial protocol.