Enforcement moves ahead of guidance

MHRA says it has already seen the failure mode: one inspection response it received "totalled over 90 pages and did not address the deficiencies identified," and other submissions cited MHRA guidance that doesn't exist. The agency's response is procedural, not just a warning, it says it "may reject the response or return it for another attempt," weight it toward "higher risk" in future inspection prioritisation, or refer the firm to the Inspection Action Group. A single invented citation in an AI-drafted response is now a documented path to that outcome.

Add an explicit AI-use control to your quality system now. MHRA's stated bar for any submission is three-part: "factually accurate and verifiable," "technically reviewed by appropriately experienced people," and "signed off by someone with authority and accountability." Build that into your SOP for AI-assisted CAPA, deviation and inspection-response drafts, and prohibit unverifiable citations outright. If you have already submitted AI-assisted responses, audit them for fabricated references before your next inspection.

CDER's re-issued compliance program instructs investigators to "audit and verify raw data at the facility that are associated with the product" specifically to "authenticate the data submitted in the CMC section of the application as relevant, accurate, complete, and reliable." That's a standing instruction, not a for-cause trigger, every new-application or supplement PAI now includes it. A gap between filed data and source records becomes a first-visit finding that can delay approval.

Before any PAI, run your own version of Objective 3: pull the raw chromatograms, notebooks and audit trails behind every result filed in your CMC section and confirm each one is "complete, accurate, and reliable" against the submission, with no orphan or unexplained reprocessed injections. Identify who's accountable for any discrepancy you find and fix it under change control before FDA's investigator runs the same trace.

FDA's revised guidance states sponsors may now rely on "one rigorous, adequate and well-controlled pivotal clinical investigation, plus confirmatory evidence, to demonstrate substantial evidence of effectiveness", language that reframes a two-trial default into an explicit single-trial option. Paired with an Expedited IND pilot built to "shorten the time it takes from drug identification to first-in-human study," the evidence bar and the on-ramp to the clinic are moving on the same timeline. Both dockets are open now, so sponsors can still shape the final text.

Assign owners to the two open dockets and file comments: the Expedited IND RFI closes 22 July, and the revised "Demonstrating Substantial Evidence of Effectiveness" draft guidance (docket FDA-2019-D-4964) closes 22 September. Re-examine your pipeline's pivotal-trial and confirmatory-evidence strategy against the single-trial standard. Brief clinical operations and CMC on the Expedited IND pilot and the new Phase 1 IND Navigator.

EDQM confirmed Ph. Eur. Issue 12.3 became applicable across 39 countries on 1 July 2026. That's a hard force date, not a guidance target, from that date, EU/CEP-facing QC and bioanalytical labs must run methods and hold specifications against the 12.3-revised texts. Any lag between the effective date and method implementation is a live compliance gap an inspector can cite directly against release testing.

Confirm that every affected monograph and general chapter in 12.3 is implemented in your methods, specifications and CoA templates as of 1 July, under change control with training records. For CEP holders, action any dossier updates tied to the revised monographs, EDQM's own newsroom post doesn't itemize which CEPs are affected, so check your specific certificate against the 12.3 text rather than assuming. Do not release EU product against a superseded text.

FDA's investigator didn't infer this failure from records, they watched it happen. The letter states that after plates showed growth, "the next day, our investigator observed microbial plates with the same identification information with no growth." Separately, "investigators also found laboratory forms were pre-filled with microbial testing information", results recorded before testing occurred. FDA had already placed Wizcure's products on Import Alert 66-40 on 23 December 2025, ahead of this 24 June warning letter. For India-based sterile sites serving the US, this is a direct read on what an inspector is trained to catch.

Audit your microbiology data chain end to end: are plate reads recorded contemporaneously with a second witness, are results captured before any disposal, is there photographic or system evidence, and is custody traceable? Verify aseptic-zone segregation against your qualification, FDA's letter states that "the absence of a physical barrier separating the ISO 5 (Grade A) filling zone from the ISO 7 (Grade B) surrounding environment creates an unacceptable risk to product sterility." Treat pre-filled forms and any post-hoc plate handling as a critical deviation.

FDA's draft guidance (docket FDA-2026-D-6539) sets out "recommendations on the appropriate use of a quantitative systems pharmacology (QSP)-based approach for determining minimum anticipated biological effect level (MABEL) dose in first-in-human (FIH), phase 1 trials." That gives sponsors of first-in-class biologics a mechanistic alternative to a purely empirical starting dose, but a QSP model is only as defensible as the bioanalytical and target-engagement data feeding it. This raises the bar on the data package behind FIH entry, not just the modelling.

If you run FIH programmes, map your dose-selection SOPs to the QSP approach in docket FDA-2026-D-6539 and identify where your bioanalytical and target-engagement data would need strengthening to support a model-based MABEL. Ensure assay validation and PK/PD characterisation are fit to feed a QSP model, and document every assumption. File comments while the guidance is in draft.

Chapters 2.9.42 (dissolution test for lipophilic solid dosage forms) and 2.9.43 (apparent dissolution) were revised and published in Ph. Eur. Issue 12.3, effective 1 July 2026 alongside the rest of the issue. Dissolution methodology underpins BCS-based biowaivers, batch release and in-vitro bioequivalence bridging, so any QC method, release specification or dossier that cites these two chapters needs to be checked against the revised text, and any biowaiver argument built on the prior version may need re-justification.

Inventory every method, specification and dossier that references Ph. Eur. 2.9.42 or 2.9.43. Both chapters went through public consultation in Pharmeuropa 36.3 before finalization, confirm your methods match the final 12.3 text rather than the consultation draft, update where needed under change control, and re-check any BCS biowaiver or in-vitro BE argument that relies on them. Prioritise EU/CEP-facing products and anything with a pending filing.

FDA approved Norwich Pharmaceuticals' generic baloxavir marboxil on 17 June 2026, the first generic of a single-dose oral antiviral. Iilun Murphy, M.D., Director of the Office of Generic Drugs in FDA's Center for Drug Evaluation and Research, called it "a meaningful milestone for the treatment of influenza." First-generic approvals of single-dose products effectively define the accepted bioequivalence pathway for the class: study design, comparator sourcing and endpoint strategy that competing ANDA sponsors will follow into the 2026–27 respiratory season.

If baloxavir or single-dose oral antivirals are in your pipeline, review the FDA product-specific guidance and the accepted BE design behind Norwich Pharmaceuticals' approval. For BA/BE units, treat single-dose PK studies of this type as a reference model for design, sampling schedule and statistical approach, and track the product-specific guidance for any revision.

Until now, data-integrity expectations for labs lived in FDA/MHRA guidance and PIC/S documents, persuasive but not compendial. Chapter 5.38 was published in Ph. Eur. Issue 12.3 and, per EDQM's own description reported by industry trade press, addresses digital information generated or used by quality control laboratories, with particular focus on the "life cycle management of digital data." That moves data quality from guidance into an official compendial text an EU inspector can cite directly against chromatography systems and audit trails.

Read chapter 5.38 in full, EDQM's own newsroom page was not directly accessible for this review, so verify the final scope text at source before citing specifics. Map its coverage of "life cycle management of digital data" against your laboratory data-governance controls: chromatography data systems, audit-trail review, access control and record retention. Close gaps under change control and update your data-integrity SOPs to reference the new compendial chapter.