AUTH/2310/4/10 - Novo Nordisk v Lilly

Promotion of Byetta

  • Received
    07 April 2010
  • Case number
    AUTH/2310/4/10
  • Applicable Code year
    2008
  • Completed
    02 November 2010
  • Breach Clause(s)
    2, 3.1 (x2), 7.2 (x2) and 9.1 (x2)
  • Sanctions applied
    Undertaking received
  • Additional sanctions
    Advertisement
  • Appeal
    Appeal by respondent
  • Review
    November 2010 Review

Case Summary

Novo Nordisk complained about Lilly's activities associated with the Diabetes UK Annual Professional Conference which took place 3 – 5 March 2010. At issue were presentations given at a Lilly-sponsored symposium held on the eve of the conference which were alleged to have covered, inter alia, the unlicensed use of Byetta (exenatide) with insulin and the development of the once-weekly formulation of exenatide. Novo Nordisk also complained about exhibition panels used by Lilly.

The detailed response from Lilly is given below.

Novo Nordisk noted that the first presentation entitled 'The Association of British Clinical Diabetologists (ABCD) Nationwide Exenatide Audit Update', detailed, inter alia, results from patients using Byetta in combination with insulin. This was an off-licence use of Byetta which should have been emphasized by the external speaker and made clear on the related slides. The implication that Byetta could be used in combination with insulin was misleading since this was inconsistent with its summary of product characteristics (SPC).

In inter-company dialogue Lilly described its symposium as a non-promotional forum for the legitimate exchange of medical and scientific information. Novo Nordisk submitted that it was difficult to consider a Lilly-sponsored symposium, which almost entirely focused on the company's marketed and future GLP-1 agonist products, as non-promotional. Nevertheless the fact that during the symposium, whether promotional or not, neither the speaker nor the slides presented declared that the use of Byetta in combination with insulin was not licensed, constituted a breach of the Code.

The Panel noted Lilly's submission that its symposium was to facilitate the legitimate exchange of medical and scientific information. Supplementary information to the Code stated that the legitimate exchange of medical and scientific information during the development of a medicine was not prohibited provided that any such information or activity did not constitute promotion. The Panel noted that the symposium was alleged to have covered, inter alia, the unlicensed use of Byetta with insulin and the development of a once-weekly formulation of exenatide. That the meeting would perhaps elicit interest in these two topics might not necessarily be unacceptable if the arrangements for the meeting and its content satisfied the supplementary information to the Code.

The Panel noted that the Lilly symposium had taken place on the eve of the Diabetes UK Annual Professional Conference. The symposium had been part of the official conference programme although Lilly had chosen not to have it advertised in the official conference programme. The arrangements for the symposium were supplied to, and agreed by, the conference organising committee in advance. The official application form for sponsorship, exhibition stands etc referred to evening symposia and listed Tuesday, 2 March (6-11pm) as an option. Potential attendees had been invited and offered return travel for the meeting and overnight accommodation. The timing of the return journey was flexible depending on the number of days the invitee planned to attend the main conference. There was nothing on the invitation which indicated that recipients had already arranged to attend the main conference. The invitation was headed 'Lilly Annual Diabetes Medical Satellite Symposium at the Diabetes UK 2010 Annual Professional Conference'. Lilly acknowledged that, although unlikely, some of the attendees might not have subsequently attended the main conference. Lilly's meeting began at 5.45pm with drinks and canapés. The scientific session started at 6.15pm and ended at 8.15pm with pre-dinner drinks followed by dinner at 8.30pm. The briefing material for those members of the sales force that would attend the main conference stated 'No Sales Force to attend the symposium'. It was not clear whether this meant that the sales force could nonetheless attend the pre-symposium drinks and the dinner afterwards.

The symposium had taken place in the context of a major UK scientific/clinical conference. In that regard the Panel considered that such conferences might be an appropriate setting for the legitimate exchange of medical and scientific information. Nonetheless, the Panel considered that just because a symposium took place in association with a major conference did not automatically mean that it would be regarded as the legitimate exchange of medical and scientific information.

The Panel noted that Lilly's meeting was by invitation only; the attendee list and invitation process was controlled by Lilly. The Panel considered that the overall impression was that Lilly had organised its own stand-alone meeting, albeit on the eve of a national conference. The invitation included prescribing information for Byetta; it thus appeared that Lilly considered the invitation to the symposium to be promotional. The impression given to invitees might be that Lilly considered the symposium to be promotional. The invitation stated that ABCD would present further analysis of their exenatide audit. The meeting would also discuss the benefits of glucose and weight control with both current and future GLP-1 receptor agonists and new data comparing GLP-1 receptor antagonists DPP-4 inhibitors. The emphasis would be on how this new information might enhance attendees' current and future clinical practice. In the Panel's view it was extremely difficult to argue that the symposium could take the benefit of the supplementary information to the Code if Lilly considered any part of it to be promotional, requiring prescribing information. Context was important. In stating that it could take the benefit of the supplementary information Lilly had not explained how the material satisfied the requirement of being 'during the development of a medicine'. Exenatide had a marketing authorization. The long acting version did not. In the opinion of the Panel disseminating data to prescribers which potentially expanded a licensed product's market share might be different to the legitimate exchange of medical and scientific information during the development of a medicine which implied debate which enhanced the current state of scientific knowledge. The status of the audience was relevant: delegates should be able to participate in debate for it to be an exchange of medical and scientific information. The Panel queried whether the invited audience, GPs with an interest in diabetes and diabetes specialist nurses would participate at the requisite level. In the Panel's view, taking all of the circumstances into account, overall the meeting was a promotional meeting for Byetta; on balance it went beyond being the legitimate exchange of medical and scientific information during the development of a medicine.

The Panel noted that the speaker briefing stated that the objective of the presentation was to present the ABCD audit results on exenatide use in the UK and give a fair and balanced interpretation and analysis of the data. Key points to communicate were to clarify and emphasise the Byetta licence and indications for use and to highlight any off-licence use of Byetta. The Panel noted that in a promotional meeting for a medicine there should be no reference to offlicence use of that medicine. The speaker's attention was drawn to the requirements of the Code. Throughout the presentation exenatide was only referred to by its non-proprietary name and no product or company logos were used. Some slides referred to the 'restricted licence for use of exenatide with insulin and glitazones. Also fear of hypoglycaemia in using exenatide with insulin and sulphonylureas'. In the Panel's view this statement did not promote or encourage the use of exenatide with insulin. The Panel noted, however, that some slides at the end of the presentation referred to the use of exenatide plus insulin and detailed some of the clinical results observed. In a statement from the presenter provided by Lilly, it was noted that these were reserve slides with some limited data on the use of exenatide with insulin, they were not used atthe meeting but were available on the ABCD password-protected website for viewing by contributors to the audit.

The Panel considered that Novo Nordisk had to establish on the balance of probabilities that the reserve slides had been used and that the slides used were in breach of the Code. Lilly denied that the reserve slides at issue had been used. Overall, the Panel did not consider that the presentation used at the symposium had been misleading about the licensed use of exenatide nor did it promote Byetta for use in combination with insulin. No breaches of the Code were ruled.

The second presentation, entitled 'Comparison of the Incretin-based Therapies; DDP-4 inhibitors and GLP-1 receptor agonists. An update of recent trial data', referred to exenatide long-acting release (LAR) for once weekly dosing. Exenatide onceweekly was not currently licensed. The new drug application was submitted to the FDA in the US in May 2009. In March 2010 an application was submitted to the European Medicines Evaluation Agency (EMEA). A European licence was not expected for another 12-18 months.

Novo Nordisk noted that this presentation did not clarify (either verbally by the external speaker or on the slides) that exenatide LAR did not have a UK marketing authorization. This misled the health professionals about the regulatory status of the compound. Novo Nordisk suspected that the speaker's had been inadequate and as such Lilly was responsible for the pre-licence promotion of exenatide LAR in breach of the Code.

The Panel noted its comments above regarding the arrangements for and nature of the symposium.

The Panel noted that the speaker briefing stated that the objective of the presentation was to give a fair and balanced presentation of data comparing GLP mimetics vs DPP4 class of therapy. Key points to be communicated were the differentiation of the classes; the presentation of data should be consistent with each medicine's SPC. The speaker was asked to highlight data not considered within the licence and to remind the audience of the licence status if discussing exenatide LAR. The Panel noted Lilly's submission that this was done. The speaker's attention was drawn to the requirements of the Code. Throughout the presentation exenatide was only referred to by its non-proprietary name and no product or company logos were used.

The Panel noted that several of the slides detailed information about exenatide once weekly. The presentation included the results of a study whereby exenatide once weekly demonstrated superior glycaemic control and weight reduction compared with sitagliptin or pioglitazone after 26 weeks' treatment (Bergenstal et al). The Panel considered that, in the context of a promotional meeting, the presentation promoted exenatideLAR prior to the grant of its marketing authorization. A breach of the Code was ruled. None of the slides noted that exenatide LAR was not licensed although Lilly submitted that this information was given verbally by the speaker. On balance the Panel considered that the presentation was misleading with regard to the regulatory status of exenatide LAR. A breach of the Code was ruled. These rulings were appealed. The Appeal Board noted that the title of the symposium organised by Lilly was 'The benefits of GLP-1 Receptor Agonists; current and future therapies'. Invitees were told that the emphasis of the discussions throughout the symposium would be on how the information presented might enhance their present and future clinical practice. In that regard the Appeal Board considered that Lilly appeared to expect the information presented to influence, inter alia, current prescribing practice. The Appeal Board further considered that, given the inclusion of prescribing information on the invitation, most attendees would accept the invitation on the basis that the symposium was promotional. In that regard, the Appeal Board noted that the sales force brief referred to the meeting as the 'Byetta Symposium 2010'.

 The Appeal Board noted that the speaker briefings given to the Chairman and to the speaker only referred in detail to certain clauses of the Code. The speaker was asked to highlight data not considered within licence and to remind the audience of the licence status if discussing exenatide LAR. The Chairman was asked to ensure any pre-licence therapies were highlighted in the presentations. In the Appeal Board's view these instructions were ambiguous particularly given that the requirements of Clause 3 had not been referred to in detail.

The Appeal Board noted that a high percentage of the slides in the presentation at issue referred to unlicensed medicines/indications. Further, three members of the marketing team had attended the symposium as well as the drinks and dinner.

The Appeal Board rejected Lilly's submission that the symposium constituted the legitimate exchange of medical and scientific information during the development of a medicine and could thus take the benefit of the exemption described in the supplementary information to the Code. In the Appeal Board's view, the symposium, as arranged, was promotional and in that regard the presentation in question promoted exenatide LAR prior to the grant of the marketing authorization. The presentation was misleading with regard to the regulatory status of exenatide LAR. The Appeal Board upheld the Panel's rulings of breaches of the Code.

Novo Nordisk stated that the third presentation, entitled 'The benefits of GLP-1 Receptor Agonists: An overview of future therapies and their data', was delivered by a Lilly employee who did not state that exenatide LAR did not have a marketing authorization. Thus the presentation wasmisleading in breach of the Code including Clause 2. Novo Nordisk drew parallels with Case AUTH/2234/5/09.

The Panel noted its comments above regarding the arrangements for and nature of the symposium.

The Panel noted that the speaker briefing stated that the objective of the presentation was to provide an overview of current and future data showing the development of GLP-1 receptor agonists and to ensure that the audience knew that exenatide once weekly was currently not licensed. Key points to be communicated were a fair and balanced representation of data around the development of the class and to emphasise that Byetta and Victoza were currently the only licensed GLP analogues available. The speaker's attention was drawn to the requirements of certain clauses of the Code. Throughout the presentation exenatide was only referred to by its non-proprietary name and no company or product logos were used. The presentation gave a positive overview of the development of exenatide once weekly; two slides clearly stated that exenatide once weekly was not currently licensed.

The Panel considered that the presentation promoted exenatide once weekly before the relevant marketing authorization had been granted. The inclusion of statements that the product was not currently licensed were irrelevant in that regard. A breach of the Code was ruled. This ruling was appealed. The Panel considered, however, that the presentation had not been misleading with regard to the regulatory status of exenatide once weekly and in that regard ruled no breach of the Code.

The Panel noted its rulings above that exenatide once weekly had been promoted before the grant of the relevant marketing authorization. The Panel considered that high standards had not been maintained and ruled a breach of the Code. This ruling was appealed. The Panel noted from the supplementary information to Clause 2 that promoting a medicine before the grant of a marketing authorization was an activity likely to be in breach of Clause 2. That clause was reserved as a sign of particular censure. The Panel ruled a breach of Clause 2. This ruling was appealed.

The Appeal Board noted its comments above and that, in its view, the meeting as arranged, was promotional.

The Appeal Board noted the details of the speaker briefing as described above and in particular that there was no mention of the requirements of Clause 3 of the Code.

The Appeal Board considered that the presentation promoted exenatide once weekly before the relevant marketing authorization had been granted. The inclusion of statements that the product was not currently licensed was irrelevant in that regard. The Appeal Board upheld thePanel's ruling of a breach of the Code.

The Appeal Board noted that the symposium included discussions about the future availability of exenatide LAR and mention was made of the unlicensed use of exenatide with insulin. The Appeal Board further noted that the invitation to the symposium stated that the emphasis of the discussions would be on how the data presented might enhance an attendee's current and future clinical practice. The licence application for exenatide LAR was submitted two days after the symposium. The Appeal Board considered that the attendance of three members of the marketing team added to the impression that the meeting was promotional.

Overall, given the arrangements for and the content of the symposium, the Appeal Board considered that high standards had not been maintained. The Appeal Board upheld the Panel's ruling of a breach of the Code.

The Appeal Board noted from the supplementary information to Clause 2 that promoting a medicine before the grant of a marketing authorization was an activity likely to be in breach of Clause 2. That clause was reserved as a sign of particular censure. The Appeal Board noted its comments above and upheld the Panel's ruling of a breach of Clause 2.

Novo Nordisk noted that Lilly's exhibition panels featured two graphs from Klonoff et al (2008). The first graph showed the HbA1c improvement from the core phase of three randomized, controlled trials and their 3-year long, uncontrolled, observational extension period. The graph contained a suppressed zero y-axis to exaggerate the 1% HbA1c decrease revealed by the study. Regardless of no comparator on the graph, this was misleading, and did not maintain high standards.

Novo Nordisk noted that shortening the y-axis exaggerated the observed glycaemic improvement. Lilly's view that health professionals would be able to interpret such results suggested that this type of presentation was acceptable in every case when there was no comparator on the graph. This was clearly not so as this presentation did not give a clear, fair, balanced view of the matter. Further, it had not been stated on the exhibition panel that the analysis was post-hoc. This was an important piece of information to interpret the results correctly and its omission was misleading.

Novo Nordisk noted that more importantly Lilly had not stated that this post-hoc analyzed patient subgroup (n=217) represented only 22.5% of the total patient population exposed to exenatide during the core randomized, controlled phases of the study (n=963). Klonoff et al reported that the intention to treat (ITT) population that entered the extension phase was 527, but even in this case the reported graphs represented only 41% of thestudy population. Knowing this piece of information, one could easily conclude that the paper reported the results from the responders and in fact most patients needed to be switched to other therapies due to the inadequate response to exenatide during the study period. Conversely, without knowing this information, one could conclude that the 1% HbA1c improvement could be sustained with exenatide for 3 years in the general type 2 diabetes population. Clearly the missing pieces of information were highly important and the graphs on the exhibition panels (HbA1c improvement and weight change) misled and failed to maintain high standards.

Novo Nordisk alleged that the graphs were a deliberate attempt to mislead the participants at the largest diabetes scientific event of the UK in breach of Clause 2.

The Panel noted that Lilly's exhibition panel included a graph of the 'Change in HbA1c from baseline in 3 year completer population'. The heading to that section of the exhibition panel was 'Choose BYETTA to provide sustained HbA1c improvement over 3 years'. The x axis plotted weeks of treatment and the y axis was labelled HbA1c (%). The y axis was shortened between 0 to 5% and then showed 5 to 9%. The Panel noted Lilly's submission that the y axis represented a physiological range of HbA1c. The results obtained for Byetta showed that from a baseline of 8.2%, HbA1c fell sharply within the first 26 weeks, and that an initial 1% fall was maintained at week 156. A claim to the right of the graph stated 'Almost half (46%) of patients achieved HbA1c ≤7%. The graph and the claim were derived from Klonoff et al. Only data for Byetta was shown; there was no comparison with any other medicine.

The Panel noted that clinicians would be familiar with the physiological range of HbA1c and that they would treat patients to a target HbA1c of around 7%. It considered that to shorten the y axis between 0 to 5% did not mean that a suppressed zero was used in a misleading way. The decrease in HbA1c was clearly stated and not exaggerated. The Panel did not consider that the graph was misleading or exaggerated as alleged. In that regard the Panel did not consider that high standards had not been maintained. No breaches of the Code were ruled.

 The Panel noted that Klonoff et al had taken patients from three placebo controlled trials and their open-label extensions and enrolled them into one open-ended, open-label clinical trial. There had been 527 patients in the ITT population from the three studies; only 217 completed 3 years of exenatide therapy ie only 41% of the original patients. The Panel noted the claim that 'Almost half (46%) of patients achieved HbA1c ≤7%' referred only to the 3 year completers and so in that regard it was 46% of 41% ie approximately 19%. The Panel considered that the claim implied that almost half of all diabetic patients would achieve HbA1c ≤7% with exenatide therapy whereas withthe population studied it was only about 19%. Similarly, claims were made regarding the percentage of patients who would lose weight whilst on exenatide therapy. The Panel considered that with regard to the data from Klonoff et al, important information had been omitted from the exhibition panel; the material was not sufficiently complete such as to allow clinicians to form their own opinion of the therapeutic value of exenatide. The Panel considered that the exhibition panel was misleading as alleged. High standards had not been maintained. Breaches of the Code were ruled.

The Panel noted its rulings above and considered that the exhibition panel, although misleading, was not such as to bring discredit upon or reduce confidence in the pharmaceutical industry. No breach of Clause 2 was ruled.