The defect that reaches the patient is in the pack, not the molecule

Four defects, none of them chemistry

Read this week's quality notices together and a pattern appears that no single notice makes obvious. None of them is about potency, impurities or stability. They are about what surrounds or touches the product.

In the UK, the MHRA issued Class 2 recall EL(26)A/45 for one batch of Accord lercanidipine 10mg tablets after a 20mg strip was found in a carton, following a single complaint of mixed strengths inside a sealed pack. Separately, Advanz Pharma is recalling two batches of Mynzepli aflibercept 40 mg/mL pre-filled syringes as a precaution, after a small variation in finger grip size was found that could affect administration. Baxter has reported incomplete aluminium crimping on Caelyx 50 mg/25 ml single-dose vials, which can leave a vial open when the cap is removed. BfArM published a Rote-Hand-Brief and Slovakia's ŠÚKL a DHPC, both citing possible loss of sterility and unintended occupational exposure. In Canada, one lot of Equate Ultra Strength Antacid chewable tablets was recalled as Type II over possible stainless steel shavings.

Different dosage forms, different regulators, different mechanisms. The common thread is that the last physical steps (filling, closing, cartoning, assembling) are where these controls failed.

My position: presentation is a critical quality attribute

QA and QP teams tend to put their deepest scrutiny into the drug substance, the process and the release specification. Those matter. But a tablet that passes every assay is not a conforming product if it sits in the wrong carton. A syringe with the right solution is not fit for use if the grip behaves differently from what users expect. A sterile product with a crimp that can leave the vial open is not sterile in any way the patient or pharmacist can rely on.

I would go further: the 20mg-strip-in-a-10mg-carton case is a reminder that a sealed pack is not evidence of a correct pack. Certification of a batch rests on the assumption that line clearance, reconciliation and end-of-line checks did their job. A mixed-strength defect surfaced through a complaint, not through release. The records do not say how the other defects were detected, and we should not guess. But the lercanidipine case shows what the safety net can look like when end-of-line controls miss something.

The strongest counter-argument

The fair objection is that these are precisely the system working. The lercanidipine and Mynzepli actions are Class 2, the Mynzepli recall is described as precautionary, and the Equate recall is Type II. The records describe no patient harm in any of these cases. Chasing zero mechanical defects is expensive, and regulators and companies reacted proportionately.

That is true, and I would not argue otherwise. But proportionate recall is a response to a defect, not a control against one. Class 2 labels reflect a risk judgement after the fact. The Caelyx case adds an occupational dimension: the exposure risk falls on the people who handle the vial, not only on patients. And a regulator-led recall of stock held by pharmacies and wholesalers is a heavy operational cost for a defect that a better in-process check might have stopped.

What a mature response looks like

Treat container closure integrity, pack content verification and device-interface dimensions as items with defined critical limits, not as packaging-department housekeeping. Make sure that the QP's batch certification file shows evidence of the final-stage checks, not just analytical results. And take single complaints seriously: one report of mixed strengths in a sealed pack was enough to trigger a recall action.

This is also a supplier-oversight issue. Where closures, grips or components come from third parties, a small dimensional change can matter. Whether that was the root cause in any of these cases is not stated in the records and is not yet known, but the question belongs in every investigation.