The authorities that make a call defensible

No single body governs pharmacogenomics from gene to prescription. CPIC translates genotype into prescribing actions, the FDA curates biomarker information in drug labeling, DailyMed carries the current prescribing text, and DPWG issues its own therapeutic recommendations. A report that leans on one source in isolation is thinner than it looks.

SignalPGx grades every gene-drug pair against multiple named authorities. CPIC guidelines, FDA pharmacogenomic labeling, DailyMed prescribing data, and DPWG recommendations each carry different weight, and the report reconciles them into one evidence-graded call the lab's medical director signs out.

Where CPIC leads

CPIC guidelines are the backbone of clinical PGx. They take an established gene-drug pair and spell out what a given phenotype should mean for prescribing, graded by the strength of the underlying evidence.

That structure is why CPIC anchors most reports. Its recommendations are built to be:

• actionable

• phenotype-based

• evidence-graded

SignalPGx maps each CPIC recommendation to the genotype the lab actually reports, so the guidance shown always matches the result on the bench.

Where FDA labeling and DailyMed matter

FDA pharmacogenomic biomarker labeling reflects what a drug's regulated label says about genetic factors, and DailyMed carries the live prescribing information. These sources ground a recommendation in the document a pharmacist actually dispenses against.

They do not always match CPIC in emphasis, and that gap is informative. Reconciling label language with guideline logic is exactly the work a good report makes visible rather than hides.

Where DPWG adds a second read

DPWG, the Dutch Pharmacogenetics Working Group, publishes its own gene-drug recommendations, and PharmGKB curates much of the evidence that underpins the field.

For many pairs DPWG and CPIC converge; where they differ, seeing both gives the clinician a fuller picture than either source alone. SignalPGx surfaces both reads rather than forcing a single opinion.

Reconciling the sources into one call

Coverage this broad only helps a clinical lab if it resolves to something a physician can act on at the point of care.

SignalPGx reconciles 15+ curated sources across 50 pharmacogenes and roughly 950 medications.

Each gene-drug pair is graded against CPIC, FDA labeling, DailyMed, and DPWG, then rendered as one recommendation with its evidence level attached. SignalAI assists the interpretation and flags gene-drug interactions, but the lab's medical director reviews and signs out every report before it reaches the EHR.

decision = model.predict(input_data)
execute(decision)

Keeping the call current

Guidelines move, and a patient's medication list changes. A recommendation that was defensible last year can quietly go stale in the chart.

Living Reports reanalyze automatically when:

1. a guideline is updated

2. a new medication is added

3. a gene-drug pair is reclassified

When reanalysis changes a call, the report is flagged for a fresh medical-director review rather than updated silently.

Conclusion

Guidelines don't replace the clinician — they make the recommendation traceable back to its source.

A defensible PGx report isn't the one that picks a favorite authority. It shows CPIC, FDA labeling, DailyMed, and DPWG together, grades the evidence, and leaves the final prescribing decision where it belongs: with the ordering clinician.