AUTH/3224/7/19 - Pharmacosmos v Vifor

Promotion of Ferinject

  • Received
    14 July 2019
  • Case number
    AUTH/3224/7/19
  • Applicable Code year
    2019
  • Completed
    Interim case report
  • Breach Clause(s)
  • Sanctions applied
    Undertaking received
  • Additional sanctions
    Public reprimand
    Audit of company’s procedures
    Advertisement
  • Appeal
    No appeal

Case Summary

INTERIM CASE REPORT

This is an interim case report because the final report will be delayed due to the Code of Practice Appeal Board’s requirement for an audit of Vifor Pharma UK Limited’s procedures in relation to the Code (Paragraph 11.3 of the Constitution and Procedure refers).

 

*****************************

Pharmacosmos complained that Vifor Pharma was using misleading and inaccurate safety claims with regard to Ferinject (ferric carboxymaltose) and Monofer (iron isomaltoside 100mg/ml solution for injection or infusion). Ferinject and Monofer were indicated for the treatment of iron deficiency where oral iron preparations were ineffective or could not be used.

Pharmacosmos alleged that the comparative safety claim ‘European health authorities have reported differences in the adverse drug reaction profile of the available intravenous irons’ used by Vifor in its promotional materials was misleading and inaccurate. Pharmacosmos submitted that alongside the claim, Vifor representatives had made unsubstantiated, incorrect and misleading safety superiority claims that Ferinject had fewer adverse drug reactions (ADRs) than Monofer. This was evidenced by an email from a nurse to Vifor and a follow up email from Vifor medical information which recognised that the above claim was used promotionally by the representatives and the references chosen to support the claim showed that the representatives had specifically discussed hypersensitivity reactions. Pharmacosmos alleged that the contents of, and the supporting information contained in, the unsolicited medical information letter sent to the nurse was similarly misleading and inaccurate on several grounds.

Pharmacosmos submitted that the claims made by Vifor were inconsistent with and disregarded the current official stance taken by the European Medicines Agency (EMA) and the Medicines and Healthcare products Regulatory Agency (MHRA) ie that no differentiation had been identified between Ferinject and Monofer in terms of hypersensitivity reactions. In that regard, Vifor knew that the claims were incorrect.

Pharmacosmos explained that in 2013 the EMA published a report following a referral requesting that the Committee for Medicinal Products for Human Use (CHMP) gave its opinion on concerns about hypersensitivity and intravenous irons. This procedure included Ferinject and Monofer and was based on all data from marketing authorization holders regarding preclinical and clinical studies, post-marketing reports, and the published literature. The report stated that, inter alia, differentiation between these iron complexes in terms of hypersensitivity reactions could not be identified. Therefore, findings and recommendations were applicable to all the iron complexes. This remained the current position of the EMA and the MHRA.
That the statements made by Vifor’s representative to the nurse were present in written promotional materials, which were made for and used by Vifor representatives to make these inaccurate and misleading promotional comparative claims of a superior hypersensitivity profile with Ferinject over Monofer, was evident by Vifor’s response which made it clear that nothing unusual was considered about a representative’s request for the references to the above claim to be sent to the nurse.

Unfortunately, the inter-company dialogue response from Vifor, received in June 2019, did not address that the use of the claim verbally and in written promotional materials alongside superiority safety claims of fewer ADRs (hypersensitivity reactions) with Ferinject over Monofer was misleading and inaccurate for several reasons, instead it focussed on the medical information response sent, which Vifor claimed was non-promotional and not in breach of the Code. This clearly disregarded case precedent from the Panel ruling in Case AUTH/2828/3/16, upheld on appeal, which found that requests made in response to Vifor representatives raising issues about the safety of a competitor product were not unsolicited and therefore medical information responses to such requests were subject to Code requirements. It was also stated in Case AUTH/2828/3/16 that it was ‘absolutely imperative that communications from medical information were correct’, which Pharmacosmos alleged not to be the case in this instance. Therefore, the medical information letter in this case was solicited and promotional in nature.

Pharmacosmos submitted that the continued dissemination by Vifor medical information of blatantly promotional, inaccurate and misleading information was a clear breach of the trust placed in medical information across the industry. Furthermore, for Vifor representatives to use misleading claims to proactively raise concerns about the safety of Monofer could be viewed as a deliberate attempt to solicit requests to allow medical information to supply disparaging, biased, and inaccurate information about a competitor product via a route usually exempt from full Code requirements. Such behaviour by Vifor was previously censured in Case AUTH/2828/3/16, and the continuation of such behaviour further compounded the breach of trust between Vifor and health professionals.

Pharmacosmos stated that to undermine the trusted position of medical information through such biased and promotional behaviour was a blatant and shocking action that clearly brought the industry into disrepute, in breach of Clause 2.

Pharmacosmos submitted that Vifor’s conduct as described above should be considered in the context of presenting a clear risk to patient safety.

An article about the management of hypersensitivity reactions recognised that anxiety on behalf of the patient and/or health professional administering the IV iron infusion was a risk factor for a patient having an infusion reaction (Rampton et al 2014). Therefore, Pharmacosmos considered that the anxiety produced through the misleading promotional claims, and the misinformation being provided about Monofer by Vifor, placed patients unacceptably at risk. Not only had the interaction with the Vifor team raised concern about the safety of Monofer with the nurse involved, leading him/her to contact Pharmacosmos medical department for information, but the ruling against Vifor in Case AUTH/2828/3/16 demonstrated that Vifor had a documented history of causing health professionals to doubt the safety of Monofer. In Case AUTH/2828/3/16 Vifor representatives offered misleading comparisons of Monofer with Ferinject, thereby causing the health professionals to feel that scaremongering regarding Monofer safety was occurring.

In summary, Pharmacosmos alleged that the claim ‘European health authorities have reported differences in the adverse drug reaction profiles of the available intravenous irons’ should not be used by Vifor in any promotional or non-promotional materials. Nor should any Vifor employees make claims that suggested that Ferinject had a superior safety profile to Monofer. Given the widely recognised limitations of using data based solely on spontaneous adverse event reporting to compare product safety profiles, Pharmacosmos also considered that Vifor should not use such data without additional necessary data and context, in any of its materials or activities. Pharmacosmos was also extremely concerned about the understanding of Vifor, in particular its medical, marketing and compliance teams, about the scientific validity of data and Code requirements for making comparative claims.

The detailed response from Vifor appears below.

The Panel noted that the nurse had emailed the Vifor representative stating ‘I was also interested in the slide you showed about the reduced ADRs when using ferrinject over Monofer, I have been giving iron for 2 years and helping ward staff administer confidently and so I am very keen on the safety aspects of administering parenteral iron’ and ‘ Are you able to share the slide or the reference from the European drug referencing and WHO demonstrating that there are less ADRs than monofer’. The Panel disagreed with Vifor’s submission that the nurse’s request was ambiguous and referred to alleged claims and materials which simply did not exist. In the Panel’s view, the medical information team had wrongly interpreted the request as a reference to an alternative existing approved claim. The response from medical information stated ‘I understand from my colleague [name] that you have requested references for the statement “European health authorities have reported differences in the adverse drug reaction profiles of the available intravenous irons”’. It thus appeared that medical information had simply ignored the nurse’s references to the slides upon which he/she had based his/her request for information.

The references cited in the medical information response included the Ferinject SPC, the minutes of the EMA Pharmacovigilance Risk Assessment Committee (PRAC) meeting on 6-9 March 2017, EMA list of medicinal products under additional monitoring, the Lareb report, the Spanish Health Authority (AEMPS) report, the WHO Pharmaceuticals newsletter, and the Swissmedic Vigilance-News, Edition 11 newsletter.

The medical information response included a summary of some of the references provided and hyperlinks to access the full documents for all of the references apart from the Ferinject SPC. The medical information response letter started by explaining that findings by EMA as part of the referral procedure (EMEA/H/A-31/1322) led to revisions in the labels for the entire class of IV iron products and the outcome of the referral procedure concluded that the reviewed data did not allow clear differentiation between IV iron products and their relation to severe hypersensitivity reactions. In that regard, the Panel noted, however, that the EMA Assessment report for iron containing intravenous (IV) medicinal products dated 13 September 2013 stated that as the conclusions of the assessment were mainly drawn from post-marketing data, differentiation between the iv iron complexes in terms of hypersensitivity reactions could not be identified. The medical information response then went on to state that a lab report published by the Netherlands Pharmacovigilance Centre Lareb reported that they have received concerns regarding the safety of IIM [iron isomaltoside] from multiple Dutch hospitals wherein the doctors and nurses have observed an increase in the severity and incidence of allergic reactions after switching from FCM [ferric carboxymaltose] to IIM.4. It further went on to state that the AEMPS issued a warning recently to not start new patients with Monofer due to the risk of severe HSRs. The data available to AEMPS was being reviewed in detail and as a precaution the AEMPS recommended health professionals not to initiate new treatments with Monofer. It then stated that the rate of reporting on severe HSRs with iron isomaltoside was much higher than for other iv iron preparations which was referenced to the AEMPS report and the WHO newsletter.

The letter concluded with please also find attached the Swissmedic Vigilance-News 7 which looks at the risk of intravenous treatment of iron deficiency.

In the Panel’s view, the medical information response thus misrepresented the EMA position and then selectively discussed two reports, one from the Netherlands and one from Spain, both of which unfavourably compared Monofer with Ferinject and ended with a Swiss reference which reported on ADRS with Ferinject but not with Monofer.

The Panel noted that Pharmacosmos had alleged that the claim ‘... European health authorities have reported differences in the adverse drug reaction profiles of the available intravenous irons ...’ was misleading and inaccurate. The Panel noted that the only material before it which included the claim at issue was the medical information letter which began by stating ‘I understand from my colleague [name] that you have requested references for the statement “European health authorities have reported differences in the adverse drug reaction profiles of the available intravenous irons”’. The Panel noted Vifor’s submission that the claim was an approved promotional claim and that the references provided by Vifor medical information in support of the claim all focussed on hypersensitivity reactions. The Panel noted that although individual health authorities in Europe had reported differences between the available IV irons, the current European-wide stance of the EMA was that current data did not allow any differentiation between IV iron preparations in terms of hypersensitivity reactions. In the Panel’s view, the claim was therefore misleading and inaccurate and breaches of the Code were ruled.

The Panel further noted Pharmacosmos’ allegation that the exclusion of key data, including comparative safety randomised control trial data between Monofer and Ferinject which demonstrated broad comparability in terms of hypersensitivity reactions, with any slight differences in favour of Monofer, in the information sent to the nurse was a deliberate attempt by Vifor to selectively favour Ferinject and disparage Monofer. The Panel noted, however, that the medical information letter set out, albeit apparently wrongly given the nurse’s specific questions, to provide references for the statement ‘European health authorities have reported differences in the adverse drug reaction profiles of the available intravenous irons’. The Panel noted its rulings above and comments that the statement was misleading and inaccurate and disparaged Monofer. On that basis, the claim could not be made any more acceptable by the provision of data which was not from European health authorities and so the Panel considered that on this exceptional and very narrow ground there was no breach with regard to the failure to include additional references as cited by Pharmacosmos.
The Panel noted Vifor’s initial submission that it did not have a slide ‘claiming reduced ADRs when using Ferinject vs Monofer’, and that it did not suggest that regulators or the WHO had demonstrated that there were fewer ADRs for Ferinject than for Monofer. Despite Vifor’s submission that the only statement approved for use was ‘… European health authorities have reported differences in the adverse drug reaction profiles of the available intravenous irons…’, that statement did not appear in any of the material supplied by Vifor. The Panel noted that it was only following a request for further information that Vifor provided a copy of the Ferinject Objection Handler (UK-FCM-1900026) entitled ‘Why Ferinject’, which was used by the representatives, and the relevant briefing document. Vifor stated that certain slides of the objection handler were used by the representative at the meeting in question but did not state which slides in particular were used. (The Panel noted that the objection handler had since been withdrawn.)

The executive summary in the briefing document stated, inter alia, ‘We want you to actively differentiate between IV irons using information in your sales aid’. The stated proactive growth and active differentiation strategy included getting health professionals to understand how IV irons were different and use Ferinject preferentially.

Representatives were to have reactive differentiation discussions if a health professional stated that they considered all IV irons were the same in terms of, inter alia, tolerability. The briefing document summary stated ‘Only use the ‘Ferinject objection handler’ reactively when a health professional considers that all IV irons are the same; requests a comparison of tolerability of irons; or requests a comparison of the efficacy of IV irons. In the Panel’s view, it was clear that Vifor intended to clearly, and favourably, differentiate Ferinject from other IV irons including on grounds of tolerability.

The objection handler included a number of slides comparing the tolerability of Ferinject vs Monofer, the majority of which focussed on hypersensitivity reactions. The first slide within the Tolerability profile section was titled ‘IV irons and hypersensitivity reactions: a European assessment and gave a timeline which according to the briefing document showed why Ehlken et al (2018) was commissioned. The next slide reported on Ehlken et al, a retrospective pharmacoepidemiologic study using data from the European Union Drug Regulating Authorities Pharmacovigilance database and data from the WHO VigiBase database both of which showed that hypersensitivity reactions were more common in patients receiving iron isomaltoside 1000 vs Ferinject (10.7 x and 8.4 x respectively). Whilst the relative frequencies appeared quite different, a pop-up graph (which might or might not have been used at the meeting) showed that the absolute numbers of reported severe hypersensitivity reactions, adjusted for exposure, per 100mg dose equivalent, per 100,000 administrations were still very small in both groups and no more than 5 in the iron isomaltoside 1000 group. There was no indication in the objection handler as to whether the apparently meaningful difference between Ferinject and iron isomaltoside 1000 was clinically or statistically significant. Ehlken et al was sponsored by Vifor and was described in the briefing notes as presenting ‘important information’ about Ferinject and iron isomaltoside 1000 (Monofer and Diafer). It was thus likely that representatives would be keen to show this slide. Diafer (iron isomaltoside 50mg/ml solution for injection) was also marketed by Pharmacosmos and was indicated in adults for the treatment of iron deficiency in patients with chronic kidney disease on dialysis, when oral iron preparations were ineffective or could not be used.

Another slide followed claiming in bold type a 75% lower risk of severe hypersensitivity reactions with Ferinject vs Monofer (p<0.0001) (Mulder et al 2018). In smaller font below it was explained that hypersensitivity reaction occurred in 18/836 (2.1%) of Ferinject and 43/496 (8.7%) of Monofer administrations. Mulder et al was a single centre Dutch study and although the briefing notes referred to study limitations and that the study indicated that there might be differences in the clinical profiles of the two IV irons; the slide in the objection handler itself was unequivocally headed ‘A recent study showed Ferinject is associated with a lower risk of severe hypersensitivity reactions than Monofer’.

A third slide detailed Bager et al (2017) which was a single centre Danish study which showed that Monofer was associated with a higher incidence of hypersensitivity reactions (10.7%) compared with Ferinject (2.5%) (p<0.01). A pop-up slide showed the absolute difference of 11 reactions in the Monofer group and 4 in the Ferinject group. Although it was stated in the objection handler that all hypersensitivity reactions were grade 1 or 2, there was no information as to how many of each grade occurred in either group. The Panel further noted that the briefing document stated that ‘When using Monofer they observed a relatively high number of [hypersensitivity reactions] and for safety reasons, they switched back to Ferinject …. However, a drawback when using Ferinject was a higher rate of Hypophosphataemia compared with Monofer’.

The Panel considered, on the balance of probabilities, that the slides described above prompted the nurse to refer to ‘the slide you showed about the reduced ADRs when using Ferinject over Monofer’ and to ask for ‘the slide or reference from European referencing and WHO demonstrating that there are less ADRs than Monofer’. In that regard it was wholly disingenuous for Vifor to have stated that it did not have a slide claiming reduced ADRs when using Ferinject vs Monofer and that it had not suggested that regulators or the WHO had demonstrated that there were fewer ADRs with Ferinject than with Monofer. The matter was not complicated and even a cursory glance at the objection handler would show that Vifor’s original response was incorrect.

The Panel noted its comments above and considered overall that the claims in the objection handler used by the representative at the meeting in question were misleading; some of the data was not sufficiently complete to allow readers to form their own opinion of the therapeutic value of Ferinject vs Monofer. The objection handler did not reflect the fact that, in a review of data, the EMA had been unable to clearly differentiate between IV irons in terms of hypersensitivity reactions. The objection handler also did not reflect the fact that the SPCs for Monofer and Ferinject were almost identical in terms of hypersensitivity reactions; both SPCs listed anaphylactoid/anaphylactic reactions as rare (≥ 1/10 000 to <1/1000). The Monofer SPC listed hypersensitivity including severe reactions as uncommon (≥ 1/1000 to <1/100) and the Ferinject SPC stated that hypersensitivity was uncommon. The Panel therefore ruled breaches of the Code including that the objection handler disparaged Monofer.

The Panel noted its comments above and considered that the medical information response to the nurse could not take the benefit of the exemption to promotion as set out in the Code. In the Panel’s view, the response from medical information was not in reply to an unsolicited enquiry, it did not relate solely to the subject matter of the enquiry and the content was not accurate, balanced or fair. In that regard, the nurse had been sent a promotional email which was subject to the requirements of the Code. The Panel noted its comments above and considered that the medical information letter was misleading and disparaged Monofer. Breaches of the Code were ruled.

The Panel noted its comments and rulings above and considered that Vifor had failed to maintain high standards in breach of the Code.

The Panel noted its comments above and was particularly concerned that the medical information letter was misleading and disparaging. In that regard the Panel noted Pharmacosmos’ comment that reducing a health professional’s confidence in Monofer might increase a patient’s anxiety and lead to an infusion reaction. It was thus absolutely imperative that communications from medical information were accurate, fair and balanced. In the Panel’s view, the medical information letter at issue was poor and, in that regard, it reduced confidence in the pharmaceutical industry. A breach of Clause 2 was ruled.

The Panel was very concerned to note that despite being asked by the case preparation manager to provide copies of relevant material including the slides used by the representative in question with the nurse and copies of the current relevant representative’s briefing which referred to Monofer and side-effects, Vifor did not provide any of this information.

The Panel was concerned to note that it was only in response to a request for further information that Vifor provided a copy of the Ferinject Objection Handler (UK-FCM-1900026) certain slides of which it stated were used by the representative at the meeting in question and the briefing document (ref UK-FCM-1900027) which clearly set out differences in the occurrence of hypersensitivity reactions between Ferinject and Monofer in favour of Ferinject and specifically referred to the European Union Drug Regulating Authorities Pharmacovigilance Database and the WHO VigiBase Database which showed that hypersensitivity reactions were reported more frequently in patients receiving iron isomaltoside 1000 vs Ferinject. The Panel queried why this information was not provided initially. The Panel noted that self-regulation and the reputation of the industry in that regard, relied upon full and frank disclosure at the outset.

The Panel was extremely concerned to note that its request for further information appeared to mark a complete turn-around by Vifor. Having previously provided none of the relevant material and vigorously denying all allegations it now acknowledged potential breaches of the Code including Clause 2; the company only appeared to take this more open and transparent approach following the engagement of external advisors. Although Vifor’s abrupt and complete reversal of its position had clarified the matters in hand, the Panel considered that the company’s original response appeared obstructive and uncooperative despite its submission that it was committed to self-regulation. In that regard, the Panel decided to report Vifor to the Code of Practice Appeal Board in accordance with Paragraph 8.2 of the Constitution and Procedure.

The Appeal Board noted the Panel’s comments and rulings of breaches of the Code including its decision to report Vifor to the Appeal Board. The material at issue was the same as that in Case AUTH/3199/5/19. The allegations were different in that Case AUTH/3224/7/19 dealt with the claims whereas Case AUTH/3199/5/19 dealt with the breach of undertaking. The Appeal Board noted that Vifor had provided brief details about its plan to address the issues and had apologised.
The Appeal Board was very concerned that the responses from Vifor (dated 29 July 2019 and 19 September 2019) in addition to the response to Case AUTH/3199/5/19 dated 3 June 2019, could be described as obstructive, defensive, aggressive, dissembling and untrue. Vifor’s follow up letter (29 November 2019) referred to the cases as a ‘fishing exercise by Pharmacosmos’ to obtain Vifor promotional material. Subsequently, Vifor’s response to a request for further information (dated 6 December 2019) which referred to Case AUTH/3199/5/19 and Case AUTH/3224/7/19 admitted that the company’s initial response was inaccurate and provided the Ferinject Objection Handler (UK-FCM-1900026) and the briefing document (ref UK-FCM-1900027). This version of the Objection Handler had been withdrawn in September 2019. The form of undertaking provided in Case AUTH/3224/7/19 stated that the Objection Handler was last used on 29 September 2019.

The Appeal Board noted that Vifor accepted full responsibility for its initial responses to Case AUTH/3224/7/19 dated 29 July and 19 September 2019.

Vifor stated that it became apparent that its initial response to Case AUTH/3199/5/19 dated 3 June 2019 was not appropriate. Further the request was clearly in relation to statements made within the objection handler. Concerns about the content of the initial response in Case AUTH/3199/5/19 were subsequently investigated by Vifor. The investigation resulted in a change in the senior leadership team and following its advice and that of an external agency resulted in Vifor’s reversal of position declared to the PMCPA. The Vifor representatives at the report hearing also referred to very recent changes in leadership at the global level. In response to a question the Vifor representatives confirmed that the letter of 6 December 2019 was written by the external third party contracted to provide compliance advice.

The Appeal Board was concerned about the length of time it had taken for Vifor to change its approach to the complaint. Vifor representatives explained that the delay, in part was caused by the need to follow certain internal processes and that the lteam (which included legal and medical representation) insisted that Vifor continued to submit that the case be dismissed rather than provide the materials including the Objection Handler. The Appeal Board also noted Vifor’s subsequent response to the PMCPA and Vifor’s admission of errors and that it accepted responsibility for the breaches of the Code.

The Appeal Board noted the submission from Vifor’s representatives at the report hearing that Vifor was now committed to change how it would promote its medicines. First line managers had been briefed in late February 2020. The Appeal Board welcomed the change in approach from Vifor and its plans to ensure that such issues did not recur. However, the Appeal Board was very concerned about the prevailing company culture within which the initial response was submitted. The Appeal Board noted that company culture took time to change. The Appeal Board noted that self-regulation relied, inter alia, upon the provision of complete and accurate information from pharmaceutical companies.

The Appeal Board decided that in accordance with Paragraph 11.3 of the Constitution and Procedure, Vifor should be publicly reprimanded for its failure to provide accurate and truthful information to the Panel and its disingenuous approach to responding to the complaint. The Appeal Board also decided to require an audit of Vifor’s procedures in relation to the Code in the present case. The Appeal Board required Vifor to provide a comprehensive and detailed corrective and preventive action (CAPA) plan with timelines in time for it to be considered at the next Appeal Board meeting on 11 March. On consideration of the CAPA plan the Appeal Board would confirm the date of the audit. In any event this audit would take place at the same time as that required in Case AUTH/3199/5/19. On receipt of the report of the audit the Appeal Board would consider whether further sanctions were necessary.

On receipt of the requested CAPA plan from Vifor, the Appeal Board noted the timelines in the CAPA plan. There were completion dates between February and August 2020. The Appeal Board queried whether the timelines were sufficient and reflected the urgency of the situation given its comments about the seriousness of these cases. In that regard it noted that Vifor was still yet to brief all employees about the current cases. The Appeal Board decided that the audits should take place in September 2020 by which time it expected to see substantial progress. On receipt of the report for the audits it would decide whether further sanctions were necessary.