In the rapidly evolving landscape of product safety and drug development, the traditional “test everything” approach is becoming obsolete. Regulatory bodies, including the FDA and ECHA, are increasingly pivoting toward more efficient, humane, and scientifically advanced methods to assess chemical safety. At the heart of this shift is the OECD Guidance on Grouping of Chemicals (https://www.oecd.org/en/publications/guidance-on-grouping-of-chemicals-third-edition_b254a158-en.html).
Whether you are a researcher, a regulatory affairs specialist, or a product developer, understanding how to group chemicals and perform “read-across” is essential. This guide breaks down the core principles of the OECD framework, transforming complex guidance into actionable learning notes.
The Core Concept: What is Grouping?
Grouping is the process of identifying chemicals that share common features, allowing them to be treated as a “category” rather than as individual, isolated substances.
When chemicals are grouped, we can use existing data from one or more “source” chemicals to predict the properties of a “target” chemical—a process known as read-across. This eliminates the need for redundant animal testing and accelerates the safety assessment process.
The Four Pillars of Similarity
To place chemicals in the same group, you must prove they are “alike.” The OECD identifies four primary “bridges” to establish this similarity:
Structural Similarity: The molecules share common functional groups or “skeletons” (e.g., all members are long-chain fatty alcohols).
Metabolic Similarity: Different substances are transformed by the body into the same breakdown products (e.g., different esters that all hydrolyze into the same acid and alcohol).
Physicochemical Similarity: The chemicals share similar physical properties, such as boiling points, vapor pressure, or their ability to dissolve in fats (LogP).
Mechanistic Similarity: The chemicals interact with the body in the same way, hitting the same biological targets or triggering the same Adverse Outcome Pathway (AOP).
The 6-Step Workflow for Regulatory Success
If you want a regulatory agency to accept your grouping strategy, you must follow a disciplined, stepwise procedure.
Step 1: Define Group Membership
Start by listing every chemical you want to include in the group. You must provide clear chemical structures, CAS numbers, and—crucially—purity profiles. Sometimes, a “clean” chemical behaves differently than a “technical grade” one due to impurities.
Step 2: Create a Data Matrix
This is your “gap analysis.” Build a table where the rows are your chemicals and the columns are the required safety tests (e.g., skin irritation, genotoxicity, oral toxicity). Mark where you have data and where you have “holes.”
Step 3: Establish a Hypothesis
You must explicitly state why these chemicals belong together. A simple statement like, “These three chemicals all metabolize into Phenol within one hour,” provides a clear scientific rationale for the regulator to evaluate.
Step 4: Assess Similarity via Tools
Use standardized software to support your claim. The OECD QSAR Toolbox is the industry standard. It helps identify similar molecules in global databases and predicts how your chemicals might be metabolized.
Step 5: Fill the Data Gaps
Now, you perform the read-across.
Analogue Approach: Using one source chemical for one target.
Category Approach: Using a trend across multiple chemicals. For example, if toxicity increases predictably as the carbon chain grows from C8 to C12, you can accurately estimate the toxicity of the C10 version.
Step 6: Identify Outliers and Justify
Be your own toughest critic. Is there a member of the group that is significantly more soluble than the others? Does one have an extra “branch” that might change its toxicity? If a chemical doesn’t fit the pattern, it is an “outlier.” You must either explain why it’s still safe to include or remove it from the group.
The Modern Edge: NAMs and the 2026 Shift
As of 2026, regulatory agencies are placing a higher value on New Approach Methodologies (NAMs) to strengthen grouping justifications. If your structural argument is slightly weak, you can use “bridge data” from:
In Vitro Assays: Testing the chemicals on human cell lines to see if they trigger the same genes.
Organ-on-a-Chip: Comparing how the chemicals affect a “liver-on-a-chip” or “lung-on-a-chip” to prove human relevance.
By combining the logic of the OECD grouping guidance with the precision of NAMs, we create a “Weight of Evidence” (WOE) that is often more reliable for predicting human safety than traditional animal tests alone.