By Patrick Hayden
A major milestone for in vitro skin irritation testing was achieved with the recent release of the Statement on the Validity of In-Vitro Tests for Skin Irritation by the European Centre for the Validation of Alternative Methods (ECVAM) Scientific Advisory Committee (ESAC). The statement endorsed the use of in vitro skin models and the MTT endpoint as a replacement for the in vivo Draize (rabbit) Skin Irritation Test (OECD TG 404 & Method B.4 of Annex V to Directive 67/548/EEC) for the purpose of distinguishing between R38 (skin irritating) and no-label (non-irritating) chemicals. Thanks and congratulations are in order to all of the many dedicated individuals involved in the long and difficult efforts to make this day a reality, particularly those who served on the validation study Management Team, independent experts of the Peer Review Panel, the sponsoring organization ECVAM/Joint Research Centre, as well as the testing laboratories.
The success of the validation effort notwithstanding, several important challenges and issues for in vitro skin irritation testing remain. Three such challenges/issues will be discussed in this brief commentary: 1) Validation of additional in vitro skin models in order to establish the solid foundation of in vitro skin model supply that will be required to support their use in the regulatory arena; 2) The impact of the Globally Harmonized System of Classification and Labeling of Chemicals (GHS) on the regulatory acceptance and use of the ESAC endorsed in vitro skin irritation test; and 3) The predictive ability of the in vivo Draize rabbit test with respect to human skin irritation hazard, and the use of Draize rabbit data as the gold standard for skin irritation validation studies.
Before beginning, it should be disclosed that the commentator is employed by MatTek Corporation, producer of the EpiDerm™ in vitro skin model.
At present, the ESAC statement fully endorses only one in vitro model (EPISKIN®) as a stand alone replacement for the in vivo Draize test. The EpiDerm™ model is currently endorsed as a validated component in a tiered testing strategy (as described, for example, by OECD TG 404). As history has shown (e.g. setbacks caused by the demise of the SKIN2 model from the marketplace) it will be important to ensure that multiple in vitro models are available to fill the need for in vitro skin irritation testing in the event of availability problems from any particular producer. Furthermore, it is critically important that validated in vitro models are able to deliver long-term reproducibility that will provide consistent, good quality data over time, not just during the validation process (Rispin et al., 2004).
Additional in vitro skin models currently exist, and new models will undoubtedly be introduced into the marketplace. These will need to be validated in future "catch-up" studies. Likewise, additional endpoints (e.g. IL-α) may be considered in future validation studies. Ideally, a "common protocol" that can be applied to any model would be desirable for ease of regulatory acceptance. Therefore, this approach was taken in the ECVAM skin irritation validation study. However, in hindsight, one lesson that was derived from the recent validation study is that a certain amount of flexibility should be possible to accommodate the multiple model systems that will be required in the marketplace to provide a firm foundation to support regulatory use of in vitro models.
As a case in point, the protocol utilized in the recent ECVAM skin irritation validation study, originally developed for use with the EPISKIN® model, and later optimized and applied to the EpiDerm™ model, ultimately did not function adequately with the EpiDerm™ model. That this situation occurred despite the best efforts and intentions of everyone involved in the validation effort underscores the subtle pitfalls and difficult nature of the validation process. It is now recognized that, due to differences in barrier properties between the EPISKIN® and EpiDerm™ models, a modified protocol providing a longer treatment time with test chemicals is required with the EpiDerm™ model in order to achieve similar levels of sensitivity (Kandarova et al., 2007). As a remedy, a follow-up validation of EpiDerm™ is now being conducted with the modified protocol, sponsored by the Centre for Documentation and Evaluation of Alternatives to Animal Experiments (ZEBET) of the (German) Federal Institute for Risk Assessment.
As the preceding example demonstrates, a flexible approach that can accommodate differences in model systems will facilitate successful development and validation of protocols for new models and/or endpoints. Availability of multiple validated in vitro skin model systems will be necessary to avoid past setbacks caused by sudden disruptions or unreliable in vitro skin model supply. Furthermore, demonstration of long-term reproducibility will be necessary to support regulatory use of in vitro skin models.
Regarding the impact of the Globally Harmonized System of Classification and Labeling of Chemicals (GHS), while the ESAC statement on in vitro skin irritation tests is indeed a major achievement, we are not out of the woods yet with regard to universal acceptance of in vitro tests for skin irritation.
The ESAC endorsed test will apply to chemicals marketed in the European Union. However, the United States and many other countries throughout the world will continue to follow the GHS, which at present continues to rely on the in vivo Draize skin irritation test. EU companies marketing chemicals in the US and other countries that adopt the GHS system will need to meet GHS requirements.
The ESAC endorsed in vitro test specifies a chemical as a skin irritant by criteria corresponding to an in vivo Draize score of =2 (OJEC, 1967). In contrast, the GHS specifies a chemical as a skin irritant by criteria corresponding to an in vivo Draize score of =2.3. Additionally under the GHS system, "for those authorities, such as pesticide regulators, wanting more than one designation for skin irritation, an additional mild irritant category (in vivo Draize score =1.5<2.3) is provided" (OSHA, 2006).
According to US Government guidelines for the GHS, "test data already generated for the classification of chemicals under existing systems should be accepted when classifying these chemicals under the GHS, thereby avoiding duplicative testing and the unnecessary use of test animals" (OSHA, 2006). Thus, determination of a non-irritant classification for a chemical under the recently validated EU system should presumably provide reliable evidence that, at the very worst, the chemical can be expected to produce only slight, if any, in vivo rabbit (or human) skin irritation according to the GHS classification. However, development and validation of protocols designed to classify slight irritants under GHS is a remaining need.
This brings us to a final issue, the predictive ability and reliability of the Draize rabbit skin test with respect to human skin irritation, and the use of Draize rabbit data as a gold standard for skin irritation validation studies. While the Draize rabbit skin irritation test has been in regulatory use since 1944, it has frequently been criticized for its poor quality historical data and poor ability to predict human skin responses, being in fact, over-predictive of human skin irritation (Robinson, et al., 2002; Welss, et al., 2004). Nevertheless, Draize rabbit data are utilized as the "gold" standard for in vitro validation studies.
The poor quality of this standard in vivo data set has confounded efforts to validate alternative methods. Therefore, human patch test data, generated from ethical human patch test protocols have been suggested for use as standards for validation studies of alternative methods (Basketter, et al., 2004). Recent studies comparing in vivo human data to in vivo Draize data confirm the over-prediction by the later (Jirova, et al., 2007). Furthermore, it seems that with proper protocol design, in vitro models show good correlation with human patch test data (Jirova, et al., 2007). Thus, existing data obtained from highly standardized human patch tests may provide a more relevant and reliable standard for the development and validation of in vitro methods.
In closing, release of the recent ESAC Statement on the Validity of In-Vitro Tests for Skin Irritation represents a major breakthrough for in vitro toxicology that should allow for replacement of the Draize rabbit skin irritation test in Europe in the near future. This event is indeed cause for celebration and renewed optimism for further reduction of animal tests in the future. However, additional investigative and validation studies will likely be required to meet requirements of the GHS. In that view, utilization of existing human patch test data as a gold standard should be considered for cases where the animal and in vitro data disagree. This approach might improve prospects for successful development and validation of alternative methods capable of discriminating irritants and slight irritants as required by GHS. Widespread and reliable availability of in vitro human skin models that can demonstrate reproducible and consistent results over the long term will also be required to support regulatory use of the alternative methods in the marketplace.
References
Biography
Dr. Hayden received a B.S. in Chemistry from the University of MD at College Park in 1984 and a Ph.D. in Chemistry from Clarkson University, Potsdam, NY in 1991. His Ph.D. thesis work on biotransformation of halogenated ethylenes was conducted in the laboratory of Dr. James L. Stevens at the W. Alton Jones Cell Science Center in Lake Placid, NY. As a Post-Doctoral (PRAT) Fellow in the laboratory of Dr. Colin F. Chignell at the NIEHS, he studied ESR spectroscopy, spin-trapping of free radicals and free radical toxicity in skin cells.
Dr. Hayden first began working with in vitro skin models while employed by the Gillette Company (1995-1999), where he experimented with the Skin2™ and EpiDerm™ models in relation to irritation screening of consumer products.
Since 1999, Dr. Hayden has been employed by MatTek Corporation, where he has led the development of MatTek's full-thickness skin model (Epiderm-FT™) and manages the production and ongoing research related to the EpiDerm-FT™, EpiAirway™ and other in vitro human models.
Patrick J. Hayden
MatTek Corporation
200 Homer Avenue
Ashland, MA 01721
E-mail: phayden@mattek.com
© 2007 Patrick Hayden
|
