Medicines: EMA closes the door on PDF forms; a study in the Lancet on the duration of therapy after angioplasty

Two unrelated pieces of news, both absent from this morning’s edition, which this newspaper devoted to the Ebola epidemic in Congo. The first is procedural and concerns those who submit applications to the European Medicines Agency. The second is clinical research and should be read with the caveats the authors themselves indicate.

The form

From 1 September 2026, the EMA rejects variation applications relating to centrally authorised medicinal products for human use submitted using the interactive PDF electronic form, except where a valid and justified technical reason applies. The mandatory tool becomes the web form available on the PLM portal. The information appears in the Agency’s communications and is summarised in the regulatory review by Pharmavibes, which reconstructs the EMA and CMDh announcements on the portal.

Advance notice was long: the full announcement dates back to 13 April 2026, and the updated form for variations has already been mandatory for all new human applications since 24 August. Users were asked to verify their portal access and roles in good time and to familiarise themselves with the new interface. An application rejected for format reasons is not a rejection on the merits, but it entails a new submission and lengthens the procedure’s timeline: this is why an apparently bureaucratic change matters to those who follow medicine authorisation timetables in Europe.

The study

The current issue of the Lancet, volume 408, number 10557, dated 29 August 2026, features a study on patients with atrial fibrillation undergoing percutaneous coronary angioplasty guided by intravascular imaging, predominantly for chronic coronary syndrome. The comparison concerns the duration of dual antithrombotic therapy: one month, followed by monotherapy with a direct oral anticoagulant, versus twelve months.

The result reported by the authors is non-inferiority of the short regimen compared with the long one for the composite endpoint of death or thromboembolic events, with a reduction in major or clinically relevant non-major bleeding at twelve months.

The limitations are stated by the authors themselves, and must be reported alongside the result: caution is urged in interpreting the efficacy data, because the observed event rates were lower than expected and because the non-inferiority margin used was fixed and expressed in absolute terms. A margin of this type, with events less frequent than anticipated, makes the demonstration of non-inferiority less stringent than the original design intended. The available abstract does not report the sample size, the list of participating centres or the authors’ conflict-of-interest declarations: these are elements the reader would find in the full text and which cannot be summarised here.

The news currently comes from a single source, the Lancet issue; no independent confirmation is available. This is a peer-reviewed journal publication, not a guideline: no health authority, in the available abstracts, has changed its recommendations on the basis of these results, and this newspaper does not issue therapeutic recommendations of any kind.

What we do not know

On the regulatory front, it is not known how many applications were submitted in PDF format in recent weeks, nor which technical reasons the Agency considers valid for the exception. On the clinical front, we do not know the number of patients enrolled, the total follow-up duration beyond the twelve months indicated, or whether the study was funded by industry sponsors. The result concerns a selected population — atrial fibrillation, angioplasty guided by intravascular imaging, a predominantly chronic coronary picture — and cannot be extended to other patient groups.