Supplement–ABCG2 Assay-Evidence Audit (comp-004)

Question: Do three cited ABCG2/BCRP interaction records—one each for quercetin, curcumin, and EGCG—support a quantitative prediction of intestinal urate-transport inhibition?

Result: No. The calculation that divided nominal bulk gut concentration by IC50 values from drug-substrate assays cannot estimate intestinal urate transport. Its ratios, predicted inhibition percentages, and clinical-risk labels are invalid.

The cited records still carry useful, narrower evidence:

• Quercetin — In Vitro: Cooray et al. used mitoxantrone and BODIPY-FL-prazosin in non-intestinal BCRP systems (PMID 15047179).
• Curcumin — Animal Model: Karibe et al. found an intestinal BCRP interaction in cynomolgus monkeys using sulfasalazine and rosuvastatin, not urate (PMID 29358184).
• EGCG — In Vitro: Farabegoli et al. found reduced mitoxantrone-assayed BCRP activity after EGCG exposure in MCF-7Tam cells without changed BCRP mRNA or protein (PMID 20149610).

These records route the compounds to a direct intestinal assay; they do not rank hazard or predict clinical direction. The discriminating experiment measures free parent compound and metabolites, total and surface ABCG2, ABCG2 attribution, barrier integrity, viability, and basolateral-to-apical urate flux across prespecified exposure times. See validation §1.14.

The bounded machine-readable audit, inputs, provenance, and deterministic summary are in etc/experiments/comp-004-supplement-abcg2-antagonism/.