# Before trusting a drug-panel colour, find its reference

A red, amber or green drug row is an interface choice. The clinical meaning comes from the evidence and rules underneath it. A useful pharmacogenomic report should let you trace the row to a named gene, medicine, endpoint and interpretation source, rather than asking you to trust the colour alone.

The [FDA’s table of pharmacogenetic associations](https://www.fda.gov/medical-devices/precision-medicine/table-pharmacogenetic-associations) explains its scope and limitations. Inclusion is not a universal instruction to test everyone, and the table does not make every listed association an identical kind of recommendation. The same caution applies when a commercial panel compresses different relationships into one visual scale.

## Ask what the colour describes

Here is a fictional report-reading exercise. One row concerns a medicine’s exposure. Another concerns an adverse-effect endpoint. A third says that the evidence is insufficient for a recommendation. If all three are reduced to a colour, the reader needs the explanatory text to understand why the rows differ.

Write the endpoint beside the row before drawing a conclusion. “Potential concentration difference” and “evidence about a particular adverse reaction” are not interchangeable with “will work” or “will not work”. If the report does not specify the endpoint, record that omission as a question for its provider.

This also prevents a category from becoming a shopping list of medicines to request. A panel may describe genetic information without considering the person’s indication, other medicines, treatment history or current monitoring. A neat display cannot supply missing clinical context.

## The edition can explain a change

Guidelines are updated. The [CPIC thiopurine manuscript titled the 2025 Update](https://pmc.ncbi.nlm.nih.gov/articles/PMC12997511/) was published in 2026 and supersedes an older edition. That concrete example shows why the exact title, publication details and report reference matter when comparing interpretations.

In a fictional case, two reports issued several years apart use different wording for the same recorded assignment. The reader’s first task is to compare their references and methods, not to conclude that the underlying DNA changed. A difference could reflect the interpretation edition, tested coverage or another reporting choice. The report provider should explain which applies.

Keep both original documents while resolving the question. Overwriting the older summary with the newer colour removes the trail needed to understand the change. A note saying “interpretation reviewed on this date using this source” is more useful than silently replacing one permanent badge with another.

## A combined algorithm is not the same as a named guideline

The [2023 CPIC serotonin-reuptake-inhibitor guideline](https://files.cpicpgx.org/data/guideline/publication/serotonin_reuptake_inhibitor_antidepressants/2023/37032427.pdf) discusses the limits of evaluating proprietary combinatorial approaches. A report that combines several genes through an undisclosed method should not be presented as though each final colour is simply a reproduced CPIC recommendation.

This does not settle whether a particular product is useful. It identifies the evidence question. Which rules produced the category? Are those rules public? What population and endpoint support the claimed interpretation? A reader can ask these questions without trying to reverse-engineer a clinical algorithm or decide between treatments.

It is also worth distinguishing a named guideline recommendation from a research association, a product-specific rule and a lack of evidence. Those labels tell a professional what kind of information is being offered. “Guideline-based” should be traceable to an actual edition and relevant section, not only a familiar organisation’s logo.

## Build a reference trail that another person can follow

For the row you want to discuss, save the exact gene assignment, medicine, indication if known, endpoint wording and cited source edition. Include the report provider and date. Where a method or reference is missing, write “not supplied”. Do not substitute the latest-looking internet table for the one the report actually used.

Then ask the pharmacist or prescriber: “Does this source and interpretation apply to my current clinical question?” If the issue is the panel’s construction, ask the report provider for its method first. The clinical professional should not have to guess what an unexplained colour meant to the vendor.

The useful outcome is a traceable discussion, not a self-selected prescription. Once the evidence trail is visible, a colour can remain a navigational aid instead of becoming a verdict about a person or every future use of a medicine. The row is easier to question because its meaning is no longer hidden inside its design.

## Scope and source access

Research accessed 4 October 2026: FDA official indexed text, accessible sections of the updated thiopurine manuscript and the 2023 CPIC guideline PDF. Current CPIC links redirected to JavaScript-only pages, so live tables were not audited. No commercial panel was clinically validated, ranked or endorsed.