Fixing False Positives in Cosmetic Testing

In this post I’m going to talk about ongoing innovations in the field of in vitro genotoxicity testing for cosmetics, particularly how Cosmetics Europe has been tackling the issue of misleading positives in these tests. With animal testing banned for cosmetics in many regions, improving the reliability of in vitro methods has never been more critical.

In case you missed my introductory blog post, you can read it here where I explain what I’m pursuing in this blog: tracking the latest advancements in toxicology, in vitro testing, and regulatory science as they relate to cosmetic and personal care products. Today, we’ll look at a fascinating article from Cosmetics Europe, which you can read here, detailing their Genotoxicity Task Force’s work over the past several years.

Tackling Misleading Positives: The “False Positives” ProjectThe in vitro micronucleus (MN) assay has been plagued by high false-positive rates, leading to potentially safe chemicals being eliminated early in development. The “False Positives” project sought to address this through two critical improvements:

1. Cell Type Selection: Human cells (TK6, HepG2, HuLy) showed better specificity than commonly used rodent cells (CHO, CHL, V79). Human cells also retain important metabolic characteristics missing in rodent models.

2. Toxicity Measurements: Switching from Relative Cell Count (RCC) to Relative Population

Doubling (RPD) and Relative Increase in Cell Count (RICC) significantly reduced misleading positives by better selecting appropriate test concentrations.

These changes dramatically improved the test’s specificity without sacrificing sensitivity.

3D Skin Models: Closer to Real Use Conditions

Since cosmetics are applied dermally, Cosmetics Europe invested in validating reconstructed skin (RS) models using both the MN and Comet assays. These 3D models mimic real human skin more closely and offer a viable alternative for evaluating genotoxicity without in vivo follow-up.

The RSMN (Reconstructed Skin Micronucleus) Assay went through a rigorous three-phase validation:

• Phase 1: Protocol optimization and lab transfer.

• Phase 2: Demonstrated inter-lab reproducibility.

• Phase 3: Ongoing, but already shows improved specificity with ~90% correct prediction for non-carcinogens.

The RS Comet Assay followed a similar path. While still in development, it’s progressing well. A major innovation was the use of “underdeveloped” RS tissues (EPI-201), which recovered better from shipping stress and reduced experimental variability.

Metabolism Project: Validating the Biotransformation Capacity of RS ModelsTo ensure RS models behave like real skin, researchers compared their enzyme activity to native skin and liver. Notably:

• Skin has low levels of CYP1–3 enzymes (critical for many bioactivations), but strong presence of GST Pi and other phase II detox enzymes.

• RS models demonstrated similar metabolic profiles to native skin, supporting their validity.

Interestingly, compounds like BaP and cyclophosphamide showed differing metabolic activation patterns, which helped researchers understand the strengths and limitations of the models.

Conclusion: Looking Forward in Genotoxicity Testing

Cosmetics Europe’s comprehensive approach — targeting assay design, biological relevance, and metabolic capability — represents a major step toward more accurate and ethical genotoxicity testing.

With improvements in specificity, reproducibility, and predictive accuracy, fewer safe ingredients will be wrongly rejected, helping innovation continue without compromising safety or ethics.