Elsevier

Toxicology in Vitro

Volume 12, Issue 3, 1 June 1998, Pages 305-327
Toxicology in Vitro

The International EU/COLIPA In Vitro Phototoxicity Validation Study: Results of Phase II (Blind Trial). Part 1: The 3T3 NRU Phototoxicity Test

https://doi.org/10.1016/S0887-2333(98)00006-XGet rights and content

Abstract

To date, no standardized international guideline for the testing of chemicals for phototoxic potential has been accepted for regulatory purposes. In 1991, the European Commission (EC), represented initially by the Directorate General XI and later by ECVAM (the European Centre for the Validation of Alternative Methods) and COLIPA (the European Cosmetic, Toiletry and Perfumery Association), agreed to establish a joint EU/COLIPA programme on the development and validation of in vitro phototoxicity tests. The first phase (phase I, 1992–93) was designed as a prevalidation study, to identify in vitro test procedures and test protocols for a formal validation trial under blind conditions. In the second phase (phase II, 1994–95), the formal validation study, the most promising in vitro phototoxicity tests were validated with 30 carefully selected test chemicals in 11 laboratories in a blind trial. The 3T3 mouse fibroblast neutral red uptake phototoxicity test (3T3 NRU PT) was performed as a core test in nine laboratories, since it provided the best results in phase I of the study. The purpose of phase II was to confirm the reliability and relevance of the in vitro tests for predicting phototoxic effects and for identifying phototoxic chemicals. In phase II the phototoxic potential of test chemicals in the 3T3 NRU PT test was either assessed by determining the phototoxicity factor (PIF) by using a cut-off value of 5 as in phase I of the study, or by determining the mean photo effect (MPE) by using a cut-off value of 0.1, as recently proposed by Holzhütter (1997). Results obtained with both approaches in the 3T3 NRU PT test in phase II were reproducible in the nine laboratories, and the correlation between in vitro and in vivo data was very high. Therefore, ECVAM and COLIPA conclude from this formal validation trial under blind conditions that the 3T3 NRU PT test is a scientifically validated in vitro test which is ready to be considered for regulatory purposes for assessing the phototoxic potential of chemicals. A draft OECD Guideline for “In Vitro Phototoxicity Testing”, incorporating the standard protocol of the 3T3 NRU PT test, will be submitted to the OECD test guidelines programme in due course.

Introduction

Photosensitization is defined as a process in which reactions to normally ineffective radiation doses are induced in a system by the introduction of a specific, radiation-absorbing substance, the photosensitizer, which causes another substance, the substrate, to be changed by radiation. When used to describe the reaction of skin to an exogenous chemical and UV or visible radiation, the term includes phototoxic and photoallergic reactions, as well as photomutagenicity and photocarcinogenicity (Spielmann et al., 1994c).

Phototoxicity (=chemical phototoxicity) is the term used for an acute reaction which can be induced by a single treatment with a chemical and UV or visible radiation. In vivo, the reaction can be evoked in all subjects, provided that concentration of chemical and dose of light are appropriate. The term photoirritation is used to describe phototoxic reactions in skin produced by substances applied topically to the skin or via the systemic route and exposure to UV or visible light. Photoallergy is an acquired immunological reactivity. The skin reaction does not occur on first treatment with a chemical or light. Rather, an induction period is required before skin reactivity can occur.

The current toxicity assays for “acute dermal phototoxicity“ are animal tests using guinea pigs, rabbits, rats or mice. Although standard protocols for phototoxicity testing in animals have recently been published (Nilsson et al., 1993; OECD, 1991), no animal phototoxicity test has yet been accepted by the OECD. Instead, OECD experts recommended a sequential approach for phototoxicity testing, involving the use of in vitro assays prior to testing in animals (OECD, 1995). In 1991, DG XI of the EU and COLIPA agreed to conduct a prevalidation study on in vitro phototoxicity tests. It was the goal of the EU/COLIPA validation project on in vitro phototoxicity tests to determine whether currently available in vitro methods were capable of predicting the phototoxic potential to humans of chemicals applied via the systemic route or topically to the skin.

Among the assays currently developed for in vitro phototoxicity testing, two main types can be distinguished, namely cellular assays for screening purposes and mechanistic assays to identify specific mechanisms of phototoxicity (Spielmann et al., 1994c). The basic mechanism in phototoxicity can be described as an increase in toxicity of a chemical induced by exposure to UV or visible radiation. Therefore, the phototoxic potential of a chemical can be measured as an increase in cytotoxicity after exposure to UV or visible light. A large variety of test systems have been used to screen for phototoxic potential, including mammalian and non-mammalian permanent cell lines and primary cell cultures (Pape et al., 1994; Spielmann et al., 1994c).The EU/COLIPA in vitro phototoxicity testing programme relies on both cellular and mechanistic assays. The 3T3 NRU PT, an in vitro test for chemical phototoxicity, was the most promising in vitro test to identify chemicals with phototoxic potential in both phase I and phase II of the EU/COLIPA validation study. Therefore, the 3T3 NRU PT will be recommended as the first test in the EU/COLIPA testing strategy for phototoxic potential. To prove that this core test has been properly validated and is ready to be accepted for regulatory purposes, the formal validation of the 3T3 NRU PT is reported here in detail, as Part 1 of two reports on the results of phase II of the EU/COLIPA in vitro phototoxicity validation study, the blind trial. The results obtained with other in vitro phototoxicity tests in phase II of study will be given in Part 2 of the report. How to use in vitro assays within a testing strategy for regulatory purposes is an important question, but is beyond the scope of a formal validation study and will, therefore, not be discussed here.

To co-ordinate the EU/COLIPA validation project, a management team (MT) of six scientists was appointed, three representing the EU (Michael Balls, Georges Pechovitch, Horst Spielmann) and three from COLIPA (Jack Dupuis, Wolfgang Pape, Odile de Silva). In phase II of the study, the MT set up a management structure which basically followed the recommendations of the ECVAM Workshop on Practical Aspects of the Validation of Toxicity Test Procedures (Balls et al., 1995). A Chemicals Task Force (TF) was established, lead laboratories were appointed for each of the tests to be validated, and Standard Operating Procedures (SOPs), including statistically-based prediction models, were prepared and approved by the MT for each in vitro assay. Finally, the distribution and coding of chemicals, as well as the biostatistical analysis of the data to be produced, were contracted out to independent institutions. The biostatistical analysis was performed according to the guidelines of the ECVAM Task Force (TF) on Biostatistics (Holzhütter et al., 1996). COLIPA and ZEBET provided in vivo and in vitro phototoxicity data for selecting test chemicals and for establishing the phototoxicity data base. In collaboration with the lead laboratories, ZEBET helped to establish SOPs for each of the in vitro tests to be validated, and ZEBET acted as lead laboratory for the primary core test of phase II, the 3T3 NRU PT test.

A COLIPA Task Force (TF) on In Vitro Phototoxicity carefully selected 20 chemicals (11 phototoxins, four non-phototoxic and five UV-absorbing non-phototoxic) according to historicalin vivo animal data and human clinical data. It was agreed to compare the performances of in vitro phototoxicity assays established in laboratories of the European cosmetics industry. To standardize the quality of the work, all the participating laboratories agreed to perform an in vitro phototoxicity assay based on the 3T3 mouse fibroblast neutral red uptake (NRU) cytotoxicity test (Borenfreund and Puerner, 1985), which was modified for phototoxicity testing by exposing 3T3 cells to test chemicals in the presence and absence of UVA. In addition, the following established in vitro assays were evaluated in one or more laboratories during phase I of the study (Spielmann et al., 1995), the prevalidation stage: photohaemolysis and haemoglobin oxidation in red blood cells (RBC), histidine oxidation, a Candida albicans assay, a human keratinocyte assay, and two new commercial assays, the physicochemical SOLATEX PTTM assay, and the Skin2 PTTM assay, with reconstructed human skin (Edwards et al., 1994; Liebsch et al., 1995).

Standardization of exposure to UVA was an important technical aspect of phase I of the study. Therefore, all laboratories agreed to use an identical light source in all of the assays.

The results of phase I of the study have been published (Spielmann et al., 1994b, Spielmann et al., 1994c and 1995). The 20 test chemicals covered a representative spectrum of phototoxic and non-phototoxic chemicals. To facilitate testing in simple tissue culture systems, most of the chemicals were water soluble.

Evaluation of the outcome of phase I showed that all of the test chemicals could correctly be identified in the 3T3 NRU PT test, in which a phototoxicity factor (PIF) was used to discriminate between phototoxic and non-phototoxic chemicals.

Taking this encouraging result into account, the MT and the TF decided to use the 3T3 NRU PT test as a core assay in nine laboratories during phase II of the study, that is, for formal validation in a blind trial. In addition to the 3T3 NRU PT test, the following in vitro tests from phase I were included in phase II, although most of them were still at the stage of test development or prevalidation (Curren et al., 1995:

— the RBC PT test, (three labs)

— the SOLATEX PT test (two labs)

— the histidine oxidation test (two labs)

— a protein binding test (two labs)

— a human keratinocyte test (one lab)

— the skin2 ZK1350 PT test (one lab), and

— a complement PT test (one lab).

Phase II was conducted according to the recommendations of European experts on validation (Balls et al., 1995), as a blind trial in 11 laboratories in Europe and the USA. The test chemicals were mainly from those selected by a panel of experts at an ECVAM Workshop on In Vitro Phototoxicity Testing, who took into account high-quality human data from clinical trials (Spielmann et al., 1994c).

The MT of the study has decided to publish the results of phase II in two parts. In Part 1, the present report, the results of the formal validation of the 3T3 NRU PT test will be described in detail. In Part 2, the prevalidation data obtained from the remaining seven in vitro phototoxicity tests will be presented and analysed.

The participating laboratories and the names of all those who actively contributed to the study are listed in Table 1.

Section snippets

Selection of chemicals for the blind trial

In 1993, an ECVAM Workshop on In Vitro Phototoxicity Testing (Spielmann et al., 1994c) selected a list of reference chemicals for phototoxicity validation studies, which was entirely based on human data. At the ECVAM workshop, data were presented from a clinical trial in which a standardized photopatchtest was used to evaluate the acute skin phototoxic potential and photoallergy potential of drugs and chemicals to the human skin (Hölzle et al., 1991). The list of chemicals recommended for in

Quality check and processing of the raw data

Raw data from the 3T3 NRU PT test were submitted by participating laboratories on standard MS EXCEL spreadsheets to the independent biostatistician, where they were carefully registered and checked for consistency. This check included:

completeness of the data

correct numerical format of all sheet entries

identification of apparently wrong concentration units

identification of deviations from the SOP

The participating laboratories reported all problems detected during this quality check to the

PIF values: correlation with phototoxicity in vivo

The PIF values (including type “<”, and type “∗1” results) obtained in nine laboratories in two determinations with each of the test chemicals are shown in Table 3. One of the laboratories provided data produced according to the SOP for only 16 test chemicals. According to the classification scheme whereby PIF >5=“phototoxic” in the NRU PT test, the discordance rate column (DR) of Table 3 gives a summary of the false positive and negative classifications in comparison with the in vivo

Biostatistical analysis

The two versions of the prediction model proposed for the 3T3-NRU PT test, PIF and MPE, provided an accurate prediction of in vivo phototoxicity for almost all the test chemicals. Only one chemical (no. 14) could not be correctly classified by the majority of the participating laboratories. As furosemide was incorrectly classified as “non-phototoxic” by the majority of laboratories, the in vivo data should to be carefully evaluated again. There was no indication that the remaining in vitro/in

Influence of solubility

To test whether the aqueous solubility of chemicals might limit the predictive power of the 3T3 NRU PT assay, three phototoxic chemicals were tested in phase II in the form of free acids or bases and also as salts (acridine, nalidixic acid and protoporphyrin IX). The data obtained with both the PIF and the MPE approaches revealed that the three chemicals were correctly identified to be phototoxic, irrespective of their solubility. We therefore conclude that the test is able to identify

Acknowledgements

This study was supported by contracts from the European Commission via DG XI/E/2 (Brussels, Belgium) and ECVAM (European Centre for the Validation of Alternative Methods, part of the Environmental Institute at the JRC, Ispra, Italy), and by the BgVV (Federal Institute for Health Protection of Consumers and Veterinary Medicine, Berlin, Germany). The authors are indebted to COLIPA (the European Cosmetic, Toiletry and Perfumery Association, Brussels, Belgium) for continuous support, and to all

References (24)

  • Efron B. and Tibshirani R. J. (1993) An Introduction to the Bootstrap. Chapman & Hall,...
  • H.G. Holzhütter

    A general measure of in vitro phototoxicity derived from pairs of dose response curves and its use for predicting in vivo phototoxicity of chemicals

    ATLA

    (1997)
  • Cited by (245)

    • Catalase-loaded polymersomes as a promising safe ingredient to active photoprotection

      2021, Journal of Photochemistry and Photobiology
      Citation Excerpt :

      Regarding the photocytotoxicity profile of the samples (Fig. 2, Table 1), the results indicated the blank Ps as not phototoxic. Based on the validation study by Spielmann (1998), substances presenting PIF < 2 or MPE < 0.1 are classified as non-phototoxic; 2 < PIF < 5 or 0.1 < MPE < 0.15 as probably phototoxic; and PIF > 5 or MPE > 0.15 as phototoxic [18,31] (OECD, 2004; SPIELMANN et al., 1998). As an endogenous enzyme with an essential role in photodamage repair and no target structural alert for phototoxicity, catalase did not reduce cell viability with or without radiation, as expected.

    View all citing articles on Scopus
    1

    Author for correspondence.

    View full text