This comprehensive set of in vitro tools helps us predict outcomes of consumer skin care products with almost 100% accuracy

Did you know that your skin is your largest organ, stretching to about 2 m2 (22 square feet) if it were to be all spread out, and accounting for about 15% of your body weight? 

If, like me, you are part of Team-Sensitive-Skin, then every day can be a battle of wits with this temperamental organ. Something as simple as applying a layer of sun cream, or changing laundry detergent can kick off days of being driven to distraction by itching! Countless products make fancy claims of being for “sensitive skin” using all kinds of baffling and clever sounding sciency buzz-words, but my own experience tells me to KEEP CLEAR of this aisle in the store. 

What help is there for out there for me? 

In the past, testing of consumer products to see if they cause allergic reactions was routinely conducted on animals. Fortunately, now in the UK and elsewhere in Europe legislation exists to prevent this from happening. A number of alternative, in vitro methods exist, and more are being developed all of the time. Used in combination, these in vitro skin sensitization tests can effectively replace animal sensitization testing, and accurately predict human outcomes.

To understand skin sensitization testing, first we must understand what causes an allergic reaction. This is called the skin sensitization Adverse Outcome Pathway (AOP).

What is an Adverse Outcome Pathway?skinblog

An AOP is a structured representation of biological events leading to adverse effects and is considered relevant to risk assessment. AOPs link in a linear way existing knowledge along one or more series of causally connected key events (KE).

Your skin has three main layers: the epidermis (including the stratum corneum), dermis and subcutaneous hypodermis (shown on right). The stratum corneum is comprised of tightly packed layers of dead, flattened corneocytes, and acts as the skin’s main barrier to the outside world. The rest of the epidermis is where proliferating keratinocytes (epidermal cells) are found. The dermis, below the epidermis, contains tough connective tissue, hair follicles, and sweat glands. Underneath all that is the subcutaneous tissue (hypodermis) made of fat and connective tissue.

Our story starts in the stratum corneum. 

  1. Skin Penetration

In order to cause an allergic reaction, a substance must be bioavailable. Therefore, step one in the AOP is skin penetration. A substance must penetrate the stratum corneum barrier. Skin penetration by substances is a well-established science, and skin absorption testing for regulatory purposes is outlined in OECD Guideline No. 428.

  1. Binding to peptides in the skin

Once the substance has made its way through the stratum corneum, the next step in the AOP is our first molecular initiating event (Key Event 1): covalent binding of the substance to proteins in the epidermis (i.e. cysteine and lysine residues).  Most skin sensitizing agents are electrophiles, i.e. they will accept an electron pair from a nucleophile to form a covalent bond. The amino acids cysteine and lysine are thought to be the nucleophiles most frequently modified in proteins during sensitization, and the ability of small molecules to react with these amino acids forms the basis of the Direct Peptide Reactivity Assay (DPRA). DPRA is a simple chemical test, where solutions of the substance in question and synthetic peptides containing either cysteine or lysine are prepared, and the amount of unbound peptide is quantified by high performance liquid chromatography. OECD Guideline No. 442C outlines the DPRA test.

  1. Activation of signalling cascades and antioxidant response elements

It’s getting more complex now! Once the substances have bound to the peptides, they may activate biochemical pathways in keratinocytes and dendritic cells, including the pathways that respond to oxidative stress or electrophiles. These include the antioxidant response element (ARE) dependent pathways, and mitogen activated protein kinase signalling pathways (Key Event 2).  An in vitro test (ARE-Nrf2 Luciferase Test) has been developed using a genetically modified keratinocyte cell line where a luciferase marker gene has been spliced onto an ARE, allowing quick and easy quantification of the level of ARE activation (i.e. upregulation of antioxidant response genes). One such cell line is called KeratinoSens™. The relevant OECD guideline covering this test is OECD 442D.

  1. Immune recognition of allergens

The next stage in the AOP is the recognition of these allergens by immune cells, and dendritic cell activation by expression of cell surface markers and cytokines (interleukins, Key Event 3). The relevant in vitro test here for the AOP is called hCLAT (human Cell Line Activation Test). Here, a human leukemia cell line called THP-1 is exposed to the test substance, and changes in the expression of CD86 and CD54 cell surface markers (markers of dendritic cell activation) are measured, using flow cytometry following cell staining with fluorescein isothianocyanate.  Again, we have a relevant OECD guidance document, this time it is OECD 442E. An alternative test here is the U-Sens™ test method. This also measures dendritic cell activation, and currently has a draft OECD guideline out for review.

  1. Next, dendritic cells move into the lymph nodes, where they activate T-Cells (Key Event 4).

And finally……. (drum roll please!)

  1. Sensitivity, dermal inflammation, and allergic contact dermatitis

No single in vitro test alone is capable of fully predicting skin sensitization outcome. However, a strategic battery of multiple tests, along with appropriate in silico modelling, has been proposed to do exactly this, in a Weight of Evidence approach, such as the 2 out of 3 model proposed by Bauch et al. (2012)1. 

The REACH chemical testing strategy takes this a step further, and lays out clear guidance for chemical testing for sensitization (Note: this is for chemicals, not consumer products, but I think that lessons can be learned from this).

The REACH Strategy for Sensitization Testing of Mono-Constituent Substances

STEP 1: In silico modelling (such as DEREK) + DPRA + ARE-Nrf2 Luciferase test

STEP 2: Depending on the outcome, U-Sens™ / hCLAT

STEP 3: Weight-of-Evidence assessment

For chemicals: in case of equivocal result from in silico/in chemico/in vitro testing, in vivo testing such as the local lymph node assay (LLNA) may be used to classify or assess potency.


ARE-Nrf2 Luciferase







Sensitizer (potency evaluation required)**



hCLAT or U-Sens™

2 out of 3 approach to determine sensitizer / non-sensitizer

If sensitizer, further potency evaluation may be required**





* If one or both tests gives equivocal results

** For REACH testing, potency evaluation would often be done using the LLNA, but as this is an animal test, it would not apply for consumer products.

So there you have it. 

The AOP on skin sensitization is one of the most comprehensive pieces of toxicology guidance out there for in vitro scientists. It is not yet complete, and the current in vitro tests for skin sensitization are not 100% accurate at predicting human outcomes, but if used in an appropriate tiered strategy they can come very close. As more in vitro sensitization tests are developed, and toxicologists develop a greater understanding of how they all fit together in pathways such as this, I hope that in the very near future our supermarket shelves will be bursting with safe, and non-allergenic products that actually do what they say on the tin.

Could my days of playing “Sun-scream”-Roulette be numbered……………?



  1. Bauch et al. (2012). Regul Toxicol Pharmacol;63(3); 489-504