My drug is highly variable (CV > 30%) and keeps failing the 80–125% bioequivalence window — what do I do?
دوائي شديد التغايُر (معامل التغاير > 30%) ويرسب باستمرار في نافذة التكافؤ 80–125% · ماذا أفعل؟
Why this happens
For a highly variable drug the reference's own within-subject variability can breach the fixed 80–125% window even when the two products are genuinely equivalent — so a standard 2-way crossover fails on chance, not on true difference.
في الأدوية شديدة التغاير قد يتجاوز التغاير الداخلي للمنتج المرجعي نفسه نافذة 80–125% الثابتة حتى مع تكافؤ المنتجين فعلًا · فيرسب التصميم المتقاطع الثنائي بالصدفة لا لفرق حقيقي.
What to do, step by step
Confirm high variability: within-subject CV of the reference (CVwr) ≥ 30% for the affected metric (usually Cmax), from a pilot or the literature.
Switch to a replicate design — partial-replicate (3-period) or full-replicate (4-period) — so the reference's own variability is measured directly.
Apply reference-scaled average BE: FDA RSABE, or the EMA scaled approach, to widen the Cmax limits in proportion to CVwr — EMA caps the widening at 69.84–143.19% (for CVwr up to 50%).
Keep AUC at the conventional 80.00–125.00%; scale only the highly variable metric.
Pre-specify the replicate design and the scaling method in the protocol so the assessor accepts it up front.
أكّد التغاير العالي: معامل التغاير الداخلي للمرجع (CVwr) ≥ 30% للمقياس المتأثّر (غالبًا Cmax)، من دراسة تمهيدية أو الأدبيات.
انتقل إلى تصميم تكراري · تكرار جزئي (3 فترات) أو كامل (4 فترات) · ليُقاس تغاير المرجع نفسه مباشرةً.
طبّق التكافؤ المتوسط المُقاس على المرجع: RSABE لدى FDA أو منهج EMA المُقاس، لتوسيع حدود Cmax تناسبًا مع CVwr · ويحدّ EMA التوسيع عند 69.84–143.19% (لـCVwr حتى 50%).
أبقِ AUC عند 80.00–125.00% التقليدية؛ وطبّق التوسيع على المقياس شديد التغاير فقط.
حدّد التصميم التكراري وطريقة القياس مسبقًا في البروتوكول ليقبلها المُقيّم.
Cite as: AUPAM Experts' Office, My drug is highly variable (CV > 30%) and keeps failing the 80–125% bioequivalence window — what do I do? — AUPAM Solutions Bank, retrieved 2026-08-31, https://aupameo.org/datasets/solutions/q13-my-drug-is-highly-variable-cv-30-and-keeps-failing-the-.html