At the end of 2019, two pharmaceutical consultants check-in to an Amsterdam hotel. It is one of hundreds of options in the city. It is also where experts from the European Medicines Agency (EMA) are staying for their monthly meeting. Many are their former colleagues.
Senior figures from the agency’s Committee for Medicinal Products for Human Use (CHMP), its most influential body, were gathering at its headquarters in the Dutch capital. They were there to discuss applications from pharmaceutical firms wanting to market their drugs in Europe. It is these experts, plucked from their day jobs at national agencies or health authorities, that the EMA relies on to assess the merits of drugs that could be used by millions.
Discussions are highly confidential. Guido Rasi, the EMA’s director until 2020, says drones were even once sent to spy on proceedings by unknown actors.
The consultants, Tomas Salmonson, a Swede, and Briton Robert Hemmings, previously had a seat at the table. For years they were part of the EMA’s inner circle, in the CHMP and on expert panels. Now, through their joint enterprise Consilium Salmonson & Hemmings, they advise big pharma on how to get the coveted authorisations they once approved.
After setting up Consilium, according to Rasi, the pair would corner experts in the corridors of the hotels. Their presence at hotels did raise eyebrows but the pair deny any wrongdoing. Salmonson said they “repeatedly tried to avoid” staying at the same venue, but it was not always possible. “We tried to avoid meeting CHMP members by, for example, having breakfast after the meeting started.” But it was not “100 per cent successful as on occasions we have accidentally met CHMP members in the lobby.”
Headquarters of the European Medicines Agency in Amsterdam. European Union
However, there are bigger questions that exist for the regulator.
Tomas Salmonson was chairman of the CHMP for six years until September 2018. Three months later he launched his consultancy firm (Consilium Sweden AB). His friend Robert Hemmings, who was a CHMP member for 11 years, incorporated at the same time, Consilium Hemmings (UK) Ltd. “Our best work has been in partnership,” reads the website of their new venture, Consilium Salmonson & Hemmings.
Significantly, as they were not EMA employees, but instead are hired by their national agencies, neither were covered by the EU’s conflict of interest framework. “EMA’s committee members… are not employed by EMA and therefore the EU staff regulations do not apply to them,” a spokesperson said.
This is a “significant loophole”, campaigners argue. “It is incredible that no rules are in place to prevent revolving door cases involving scientific experts,” says Shari Hinds, policy officer at Transparency International. “We urge the EMA to implement effective and comprehensive rules to prevent such cases in the future.”
“It is incredible that no rules are in place to prevent revolving door cases involving scientific experts.”
— Shari Hinds, Transparency International
Investigate Europe has found that the regulator is beset by several conflicts of interest with concerns over professional and financial ties to industry. It is also facing growing criticism for fast-tracking drugs that could pose serious risks to patients.
When it formed in 1995 only 20 per cent of funding came from industry, with the rest from the European Union. This year, more than 90 per cent is forecast to come from industry fees and charges relating to medicine approvals.
“The financing of the agency and the level of fees is agreed by EU lawmakers and codified in EU legislation… Applicants pay for a procedure, but not for the outcome of a procedure,” an EMA spokesperson said. “Using the analogy of a driving test, one needs to pay to take a driving test but there is no guarantee of passing the test.”
In 2022, industry fees accounted for almost 86 per cent of the regulator’s €417 million in revenues its latest accounts show. Investigate Europe can reveal how just 21 companies provided half of these fees received for drug authorisations. Novartis gave almost €20 million, followed by Pfizer, AstraZeneca, Jannsen, Roche and GlaxoSmithKline, all of them paying more than €10 million, data from freedom of information requests reveals.
The regulator’s close ties to industry are widely known, says Yannis Natsis, who was on its board for two and a half years, representing healthcare providers. “The EMA has a long tradition of working closely with the companies that they are supposed to regulate.”
Part of the reason is because authorisation processes have skyrocketed. The regulator has greenlit an average of 85 new drugs a year in the past decade, compared to a yearly average of 49 in the previous 15 years. Its 21 biggest funders were beneficiaries of half of them.
This readiness to authorise is a worry for Beate Wieseler from Iqwig, Germany’s agency for assessing drug safety and efficiency. “The EMA is approving new drugs more quickly, and with less clinical data available,” she says. “It’s becoming more difficult for us to assess their real added benefit compared to existing drugs.”
The EMA has greenlit an average of 85 new drugs a year in the past decade, compared to a yearly average of 49 in the previous 15 years.Shutterstock
There are three ways to get a drug quickly approved by the EMA and 198 drugs have benefited from these accelerated authorisations since 2004 (out of more than 1400 approvals in total). The most used is a procedure called conditional marketing authorisation (CMA), that was given to 91 drugs during the period.
A CMA helps patients who have limited options and time to access new drugs, the regulator says. It involves only two of the three test phases typically required. In exchange, companies agree to additional clinical trials at a later date.
Two-thirds of drugs approved via CMA belong to the 21 firms that paid most fees to the EMA. There is no indication that the payments influenced decisions.
Investigate Europe has found that 51 CMAs were recommended by EMA in the last five years – more than half of those granted since the special procedure was introduced in 2006. To date, only nine drugs with a CMA are not on the market anymore.
“It’s scandalous,” says Pierre Chirac, director of La Revue Prescrire, a French medical journal who told us that not having enough evidence about the benefit puts patients at “unnecessary risk”.
One of them is Ocaliva, a treatment for primary biliary cirrhosis, an autoimmune liver disease. It was approved under a CMA in 2016. Four years later, US firm Intercept Pharmaceuticals finally submitted the promised test for its drug. The results found that not only did Ocaliva not work, it also caused serious side effects, and sometimes death. The EMA began a review in October 2023. A month later, the regulator renewed Ocaliva’s conditional marketing authorisation. In a press release in June this year, Intercept – which sold the European rights to the drug in 2022 to Advanz Pharma – said “improvement in survival or disease-related symptoms has not been established” with Ocaliva.
The EMA are currently reviewing the latest data, assessing the overall benefit-risk balance of Ocaliva with a view to making a recommendation on whether the medicine’s marketing authorisation in the EU should be amended. Meanwhile Ocaliva continues to circulate throughout the EU.
Two-thirds of drugs approved via CMA belong to the 21 firms that paid most fees to the EMA.Shutterstock
Guido Rasi, EMA’s former director, who now teaches at a university in Rome and advises the Italian Health Ministry, defends fast-track procedures. “Conditional marketing authorisation is the way forward,” he says, as long as robust tests are done at a later date.
But researchers at King’s College London found that in half of all cases between 2013 and December 2018, the required tests had not been provided more than seven years after conditional authorisation.
“For 30 years we’ve been told that post-marketing studies will fill in the gaps,” says Dr Courtney Davis, a medical sociologist at the university. “But this is not the case. We don’t get this evidence.”
There are tools to “robustly support the development and authorisation of medicines” the EMA said. “Medicines under CMA are constantly being monitored. Their benefit/risk balance is reviewed on an annual basis.” The regulator added that Covid-19 vaccines “might shed some light on the use of this procedure in the last few years”. In fact, these only account for seven out of the 51 recommended between 2019 and 2023.
Translarna, a drug for Duchenne muscular dystrophy developed by PTC Therapeutics International, received conditional approval in 2014. When renewing its CMA in 2023, the regulator’s expert advisory group decided that Translarna should be removed from the market. Side effects reported include serious heart conditions, and even some deaths have been linked to the drug. The EMA submitted its recommendation for non-renewal to the European Commission, which has the final say on approvals and withdrawals. Yet in May this year,the proposal was rejected.
In response to Investigate Europe, a Commission official said they had raised concerns about the “composition” of the advisory group and questioned whether it met the “principle of objective impartiality”. The Commission has now asked the EMA to review its decision with a new opinion.
Even once a drug is proven to be dangerous or ineffective, obtaining a withdrawal can prove difficult. Researchers at Leiden University found several cancer drugs withdrawn by the FDA, the EMA’s parallel in the US, were still available in some European countries.
“Your country pays for years for a drug that ultimately proves ineffective, and even when it’s withdrawn you have to keep paying because patients have to finish their treatment,” says the university’s Sahar Barjesteh van Waalwijk van Doorn-Khosrovani.
The EMA has approved more than 1,400 drugs in total since 2004. Shutterstock
Nathalie Dallard is painfully aware of the dangers, her 80-year-old mother was given Pradaxa while in hospital. She later suffered a fatal haemorrhage. The memory of nurses emptying basins of blood and the panic among staff is still vivid 12 years on, as are the words of one doctor present: “I’m going to file a pharmacovigilance report on this new anticoagulant we’ve given her.”
It turned out her mother was not alone. At least 374 deaths across Europe and the US have been linked to Pradaxa, a drug designed to prevent blood clots and strokes.
Eager to market their breakthrough drug, the German firm Boehringer Ingelheim which produces Pradaxa, sought advice to support its view that only a ‘single test’ was sufficient and allow it to forgo the double-blind’ testing procedure. Expert opinion can be provided by the EMA – at an extra cost to companies – to help advance applications. One expert selected was a French cardiologist who, after giving a positive opinion on downgrading the tests, became a consultant at Boehringer.
Another was the late Eric Abadie. Investigate Europe has learnt that Boehringer wrote to the regulator requesting Abadie be on the three-person panel. He was selected from the thousands of EMA experts. Abadie left before the process was complete to become head of the influential CHMP. Pradaxa received its authorisation in 2008 – without the ‘double-blind’ test or significantly, without an antidote to combat its side effects that include uncontrollable bleeding.
The antidote arrived in 2015, too late for Nathalie’s mother and others. Boehringer has since reached a $650 million settlement with affected patients and families in the US.
An EMA spokesperson said: “EMA has a robust system for ensuring the independence of its work.” Boehringer Ingelheim said it was “common practice” to suggest experts to the EMA, but added: “Meanwhile practice has changed, and companies no longer make such recommendations.”
The firm said Pradaxa had important health benefits and added: “Given the severe impact of stroke events, regulators underline that the benefits of stroke prevention treatment outweigh the risk of adverse effects from bleeding.”
“They used scientific advice to obtain a weakening of the procedures,” argues Anne Chailleu, vice-president of Formindep, a non-profit which has spent years campaigning for victims. “These decisions are tainted by conflicts of interest.”
German firm Boehringer Ingelheim produces the Pradaxa drug.Shutterstock
Tomas Salmonson, the Swedish doctor who checked into the Amsterdam hotel, was chairman of the CHMP for six years until September 2018. The partnership he set up with his friend, Robert Hemmings (11 years a CHMP member) – Consilium Salmonson & Hemmings – has one main goal: to help pharmaceutical clients obtain EMA marketing authorisation.
Business has been brisk. Salmonson’s Swedish business amassed net income of more than €4.4 million since it started, analysis of corporate accounts reveal. Hemmings’ company declared retained earnings of €1.8 million, as of March 2023.
Guido Rasi, EMA director at the time of the hotel meetings, said he raised concerns: “When I heard that Salmonson and Hemmings were contacting the experts, I walked into the meeting, asked for a postponement and gave a clear signal that I was prepared to do anything to stop them, including referring the matter to Olaf [the EU anti-fraud office].”
Salmonson is also on the board of Pharmetheus, a Swedish consultancy, which pursues approvals for big pharma clients. In 2021 he joined the scientific advisory board of Winhealth, a Chinese company whose “partners” include the likes of Roche, Pfizer and Daiichi Sankyo. Salmonson said he has not done any work with Winhealth for a long time.
The reguator has faced scrutiny over conflict of interest concerns involving its experts.Shutterstock
Salmonson and Hemmings both denied any conflict of interest in their work.
Salmonson admitted he had “worked for companies that CHMP formed opinions (positive and negative) when I was the Chair”, since leaving the regulator. He added that all CHMP decisions were taken by the wider group. Hemmings said that he delivers “high quality drug development programmes that can lead to high quality dossiers being presented to regulatory authorities.”
The EMA said that its experts are “subject to the rules in place at the organisation that employs them.” Sweden’s Medical Products Agency, where Salmonson was employed at the time, admits there is a regulatory problem. “In Sweden, there is a lack of regulations on so-called transition restrictions in connection with an employee ending his employment with an authority and moving to, for example, the private sector.”
The British Medicines and Healthcare Regulatory Agency (MHRA) did not respond to the question of whether Hemmings had complied with the rules related to the cooling period after leaving the agency: “We are in the pre-election period in the UK and cannot comment on enquiries that do not relate to interests in the public health.”
“EMA has a very strict policy in place on handling competing interests of scientific committee members and experts.”
— European Medicines Agency
Recently, a small hope of transparency came from the European Court of Justice. On 14 March, the EMA was condemned for conflicts of interest in a case against a French laboratory. D & A Pharma had discovered that two experts working on its marketing authorisation dossier for Hopveus, a drug against alcohol dependency, were employed by a rival. The experts had been called in as part of an “ad hoc” committee, a rare arrangement used when there is no dedicated committee on the subject. Except, as the court noted, a psychiatry committee did exist.
At the same time, Spanish firm Pharma Mar, was waging a similar legal battle. The independent company, which had submitted an authorisation application for a drug to treat multiple myeloma, challenged the decision to reject it on the grounds that an expert used by the EMA worked for a competitor. “EMA has a very strict policy in place on handling competing interests of scientific committee members and experts,” the regulator said.
Hareth Nahi, who was added as an “additional expert” to the Scientific Advisory Group for Oncology, is in fact the co-owner of the patent for Cell Protect, a drug against multiple myeloma. He declined to comment for this article. Pharma Mar won their first case in 2020, but then three EU countries – Germany, Estonia and the Netherlands – appealed the verdict. They won. The court has now asked for a review, and a new decision is expected soon.
“It’s hard not to fall into the conspiracy theory, when you see that now the big laboratories are protected by countries”, says a source inside Pharma Mar, who asked to remain anonymous. “It’s a battle of David against Goliath.”
Editor: Chris Matthews