The reflex on a lot of nonclinical teams is to treat photosafety testing as a line item every new chemical entity needs, somewhere between genotoxicity and the first-in-human package. ICH S10 does not work that way. The guideline opens with a screen, not a study, and for a meaningful share of compounds the screen is the entire photosafety evaluation — nothing further is warranted, and nothing further gets built into the timeline.

The screen comes before the study

S10's starting question is whether the parent compound, and any major metabolites, absorb light in the range relevant to sunlight and indoor lighting exposure — roughly 290 to 700 nm. If a compound absorbs in that range with a molar extinction coefficient below 1,000 L·mol⁻¹·cm⁻¹, the guideline does not treat that as raising a photosafety concern on its own, and a dedicated nonclinical phototoxicity study is generally not warranted. That threshold is not an arbitrary cutoff industry adopted after the fact — it came out of the same evidentiary process ICH used for every quantitative trigger in the S-series, and it is the reason a compound can clear early development without a 3T3 neutral red uptake assay ever entering the plan.

Nov 13, 2013
ICH S10 reached Step 4; EMA/CHMP adopted it the following month with a June 2014 effective date.
290–700 nm
The UV-visible absorption range S10's initial screen checks the parent compound and major metabolites against.
<1,000 L·mol⁻¹·cm⁻¹
The molar extinction coefficient below which absorption is not considered a photosafety concern on its own.

Scope: phototoxicity and photoallergy, not every photobiological question

S10 is deliberately narrow about what it covers. The guideline addresses phototoxicity (an acute light-induced tissue response) and photoallergy (an immune-mediated light-induced response) — not photocarcinogenicity, not every conceivable photobiological endpoint a toxicologist could theorize about. It builds on, rather than replaces, the timing framework in ICH M3(R2) and the oncology-specific nonclinical scope in ICH S9: those guidelines say when a photosafety question needs to be answered relative to clinical trials, and S10 says how to answer it.

  • New chemical entities and new formulations. S10's assessment strategies are built for compounds entering development, not drugs already on the market.
  • Parent compound and major metabolites. A metabolite that absorbs in range, even if the parent does not, can still trigger the exposure analysis.
  • Topical and systemic routes alike. Direct skin or eye application is the most obvious exposure path, but systemic distribution to skin or ocular tissue counts too.
  • Not already-marketed products. S10 states it does not apply to photosafety assessment of components of marketed products unless a new concern about the active ingredient or an excipient arises.
The screen tells you whether a photosafety question exists. It does not tell you the question is answered — exposure still has to be real before a positive spectrum becomes a study. Why the screen and the exposure assessment are two separate steps

Where sponsors get the scoping wrong

The more common error is not skipping a study S10 required — it is running one the guideline never called for, because a program defaulted to 'anything that absorbs UV light gets tested' instead of applying the extinction-coefficient threshold and the exposure analysis together. The less common but costlier error runs the other way: a compound clears the absorption screen on the parent molecule alone, but a major metabolite absorbs well above threshold, or the formulation concentrates the compound at the skin surface in a way the systemic-exposure screen alone would miss. S10's logic assumes someone is actually running the metabolite and exposure analysis, not stopping at the parent compound's spectrum because that is the easiest data point to generate. The same discipline applies to the genotoxicity and safety pharmacology questions sitting next to it in the nonclinical package — see our articles on genotoxicity weight-of-evidence and the ICH S7A core battery for the same pattern: a guideline that rewards a documented, evidentiary call over a reflexive default in either direction.

A photosafety scoping sequence you can run this quarter
  1. Run the screen before scoping any study. Generate the UV-Vis spectrum for the parent compound and major metabolites and check it against the 1,000 L·mol⁻¹·cm⁻¹ threshold first.
  2. Pair the screen with a real exposure assessment. Route, formulation, and systemic distribution data belong in the same review as the spectrum, not a follow-up question.
  3. Choose nonclinical or clinical assessment deliberately. Let the exposure profile — not CRO habit — decide between a nonclinical phototoxicity study and a clinical photosafety assessment.
  4. Document the waiver, not just a positive result. A reviewer will ask why testing was not performed as readily as why it was.

None of this is exotic nonclinical science — it is a bounded, two-step decision that a lot of development timelines skip past by defaulting to the test menu. Companies that treat ICH S10 as a nonclinical development scoping question, built on a documented screen and exposure analysis, are the ones who either skip an unnecessary study with a defensible file or catch a metabolite-driven exposure risk before it becomes a clinical-hold conversation. If your nonclinical package has been defaulting every new compound into the full photosafety test menu — or skipping the screen altogether — that is the place to start.

Frequently asked questions

Does every new drug need a dedicated photosafety study under ICH S10?

No. ICH S10 opens with a screening step, not a testing mandate. A compound (and its major metabolites) that does not absorb light between 290 and 700 nm, or that absorbs with a molar extinction coefficient below 1,000 L·mol⁻¹·cm⁻¹, is not considered to raise a photosafety concern on that basis alone, and a dedicated nonclinical phototoxicity study is generally not warranted.

What triggers a photosafety evaluation under ICH S10?

A positive UV-Vis absorption screen above the extinction-coefficient threshold, combined with meaningful exposure of skin or eyes to the parent compound or a photoreactive metabolite — through systemic distribution, direct application, or formulation design — is what moves a program from screening into S10's nonclinical and, where relevant, clinical assessment strategies.

Does ICH S10 apply to already-marketed drugs?

Not retroactively. S10 states it does not apply to the photosafety assessment of components of already-marketed pharmaceutical products unless a new safety concern arises about the active ingredient or an excipient. New chemical entities and new formulations of existing actives are the guideline's intended scope.

Sources & further reading

  1. ICH. S10 Photosafety Evaluation of Pharmaceuticals (Step 4, 13 November 2013). database.ich.org
  2. EMA. ICH guideline S10 on photosafety evaluation of pharmaceuticals, Step 5 (EMA/CHMP/ICH/752211/2012). ema.europa.eu
  3. FDA. S10 Photosafety Evaluation of Pharmaceuticals — Guidance for Industry. hhs.gov

This article is provided for general informational purposes and reflects the regulatory landscape as of November 2026. It is not legal or regulatory advice. Confirm current ICH S10 implementation guidance with FDA, EMA, or qualified counsel before scoping a photosafety program.