What happened

• Complaint 1 (Jan 2022 webinar): A virtual “Early Stage Think Tank” meeting (described as non-promotional) included discussion of Tagrisso (osimertinib) and gefitinib trial data; the Panel considered the meeting promotional and found issues including lack of certification/PI and use of “cure” on slides.
• Complaint 2 (LinkedIn posts, Mar/May 2021): Two LinkedIn posts about Tagrisso and acalabrutinib were alleged to be public promotion; AstraZeneca said they were posted before the individual joined AstraZeneca.
• Complaint 3 (LinkedIn likes/repost): UK-based AstraZeneca employees “liked” and reposted a third-party research institute’s LinkedIn posts describing capivasertib’s “remarkable benefits” and linking to more trial information; capivasertib did not have a UK marketing authorisation at the time.
• Complaint 4 (LinkedIn like): A UK-based employee “liked” a LinkedIn post sharing an article and survival graph about Tagrisso; the Panel treated the “like” as proactive dissemination to the employee’s connections, including members of the public.
• Complaint 5 (MSLs/account reviews): Allegations that MSLs were pressured to promote Imfinzi (durvalumab) off-label during quarterly account review meetings and that Medical activity was being directed by commercial.
• Complaint 6 (Early Access Programme): Allegations that the OlympiA olaparib Early Access Programme was used as a seeding/disguised promotion programme and that Medical teams were encouraged to promote it prior to authorisation/reimbursement.

Outcome

• Complaint 1: Breach of Clause 5.1 (x3); Breach of Clause 6.1 (x2); Breach of Clause 8.1. No breach of Clause 2, Clause 3.1, Clause 5.2, Clause 15.6 (and no breach findings also recorded for Clause 5.1 and Clause 6.1 as listed in the case summary).
• Complaint 2 (2019 Code): No breach of Clause 2, Clause 9.1, Clause 12.1, Clause 14.1, Clause 26.1.
• Complaint 3: Breach of Clause 2; Breach of Clause 3.1; Breach of Clause 5.1. No breach of Clause 26.1.
• Complaint 4: Breach of Clause 5.1; Breach of Clause 26.1. No breach of Clause 2.
• Complaint 5: No breach of Clause 5.1; No breach of Clause 17.2; No breach of Clause 17.9.
• Complaint 6: No breach of Clause 5.1; No breach of Clause 11.1; No breach of Clause 17.9.

Clauses

• Breach Clause(s): 2; 3.1; 5.1; 6.1; 8.1; 26.1
• No breach Clause(s): 2; 3.1; 5.1; 5.2; 6.1; 9.1; 11.1; 12.1; 14.1; 15.6; 17.2; 17.9; 26.1

Sanctions

• Undertaking received
• Additional sanctions: Advertisement

ABPI signatory lens

Why this matters

• “Non-promotional” labels don’t control classification: The Panel treated the meeting as promotional based on content and overall impression, triggering certification and PI expectations.
• Social media engagement is dissemination: “Liking” and reposting can be treated as proactive dissemination and therefore promotion, including to the public.
• Pre-authorisation exposure is high-risk: Even third-party posts can become in-scope when employees amplify them, leading to Clause 3.1 and potentially Clause 2 where seniority and repeated patterns are factors.
• Language like “cure” is especially sensitive in oncology: Unqualified “cure” messaging can be misleading by implication, particularly for mixed MDT audiences.

Where teams slip up

• Assuming a medical education format/CPD accreditation means the activity is non-promotional.
• Using product names, trial conclusions, and benefit language in “education” decks without treating them as promotional material requiring certification and PI.
• Allowing unqualified outcome language (e.g., “cure”) to appear on slides without clear context and substantiation.
• Senior staff engaging with third-party product-related posts on personal LinkedIn accounts, contrary to internal policy, and underestimating the compliance impact of a “like”.

Control that would have prevented it

• Clear decision tree for meeting classification (non-promotional vs promotional) with mandatory escalation to signatories when product data/claims appear.
• Mandatory certification workflow for any deck that includes product references or could reasonably be seen as promotional.
• Slide-level claims governance (restricted terms list for oncology such as “cure”) plus medical/legal review of implied outcomes.
• Social media guardrails: explicit prohibition and technical controls/training for engaging with third-party product-related content (including “likes”, reposts, comments).

What I’d check in an audit

• Evidence of how the ESTT meeting was classified (and by whom), including rationale against the Code’s broad definition of promotion.
• Certification records for the meeting slide deck(s) and any associated materials; if not certified, documented justification.
• Whether prescribing information was included/available where required for promotional materials.
• Claims substantiation files for any implied outcomes (especially “cure”) and whether slides were designed to “stand alone”.
• Social media SOP content, training completion, and monitoring/enforcement (including how “likes” and reposts are addressed).
• Incident management: speed and completeness of takedown actions and whether root-cause CAPA was implemented.
• For EAPs and cross-functional meetings: documentation showing reactive-only provision, handling of unsolicited requests, and separation of Medical vs Commercial direction (even though no breach was found for Complaints 5 and 6).

What the sanctions tell you

• The combination of an undertaking and an advertisement indicates the PMCPA viewed the upheld issues as significant enough to warrant public corrective action beyond simply closing the case.
• Public-facing risks (social media dissemination; misleading implications such as “cure”) and governance failures (misclassification leading to no certification/PI) tend to attract escalated outcomes.

3 questions to ask your team this week

• When we run “medical education” meetings, what objective criteria do we use to decide whether the content is promotional—and who signs off that decision?
• Do our social media rules explicitly cover “likes”, reposts and comments on third-party content, and do senior leaders follow the same restrictions?
• Which words/phrases are on our “high-risk claims” list for oncology (e.g., “cure”), and what is our process to ensure slides are not misleading by implication?

Key facts

Download the full case report (PDF)

Reviewed by Dr Anzal Qurbain (FFPM) — ABPI Final Signatory