Intended for healthcare professionals

Feature Banner

Why have UK doctors been deterred from prescribing Avastin?

BMJ 2015; 350 doi: (Published 01 April 2015) Cite this as: BMJ 2015;350:h1654
  1. Deborah Cohen, investigations editor, The BMJ
  1. dcohen{at}

Doctors in England’s NHS have been left seemingly unable to prescribe a cheap, safe, and effective drug despite its flourishing use elsewhere in Europe and the US. In the first of a two part investigation Deborah Cohen examines which agencies are responsible, whether commissioners and prescribers actually have more freedom to use intravitreal bevacizumab than might at first seem to be the case, and whether the UK’s drug marketing authorisation system really acts in the best interests of public health

Despite widespread critical coverage in The BMJ and elsewhere, the dominance of expensive drug ranibizumab (Lucentis) over cheaper bevacizumab (Avastin) to treat wet age related macular degeneration (AMD) continues unabated in the UK.

New evidence uncovered by The BMJ raises questions about the legal and regulatory positions that have skewed clinical practice, fuelled drug costs for the NHS, and left doctors confused about what they can and can’t prescribe. But in recent weeks clinical leaders have begun to fight back. Over 100 clinical commissioning groups (CCGs) in England have written to the health secretary, NHS leaders, and the General Medical Council, urging a resolution. The current situation is “untenable,” they say. Allowing CCGs to use bevacizumab instead of ranibizumab could release £102m (€138m; $152m) a year for patient services.

The NHS spends £244m a year on ranibizumab, the second highest amount for any drug.1 However, research and development costs do not explain why ranibizumab is priced 10-20 times higher than bevacizumab, Philip Rosenfeld, professor of ophthalmology at the Bascom Palmer Eye Institute in Florida told The BMJ. He was involved in the early phase trials of ranibizumab and was one of those who pioneered the use of bevacizumab for wet AMD. He said bevacizumab is also more expensive to make than ranibizumab.

Ranibizumab is a monoclonal antibody fragment derived from the same parent monoclonal antibody as bevacizumab. Both drugs act by inhibiting vascular endothelial growth factor (VEGF), preventing blood vessel growth. Roche holds the intellectual property rights for both, although Novartis has the rights to market ranibizumab in Europe. Bevacizumab is licensed for use only in cancer conditions, and Roche has never applied for a marketing authorisation for ophthalmic conditions—despite repeated calls from politicians to do so. This means only the more expensive ranibizumab has such a licence, and prescriptions of bevacizumab— even in its repackaged form for ophthalmic use—are “off label,” both the European Medicines Agency and the Food and Drug Administration have confirmed to The BMJ.

Bevacizumab came on to the UK market first and was widely prescribed off-label for wet AMD before ranibizumab’s approval by the National Institute for Health and Care Excellence in 2007—and it was openly considered a wonder drug. Even after ranibizumab was approved by NICE, former health secretary, Patricia Hewitt, reassured the Commons that bevacizumab would still be available for wet AMD on the NHS.2 Hewitt also urged the manufacturers to undertake clinical trials to allow the drug to be licensed for ophthalmic use.

But eight years on, bevacizumab is scarcely used for AMD in the NHS, and the companies have not done the necessary trials. NICE has not conducted a technology appraisal of bevacizumab for wet AMD because it doesn’t have a marketing authorisation for the relevant condition—although it has signalled in the past it would be willing to do so. Minutes from a NICE workshop show that Roche said decisions not to develop bevacizumab for ocular use have been “due to corporate considerations.”

The company’s refusal to conduct trials of bevacizumab forced ophthalmologists to ask, in the New England Journal of Medicine: “Who pays for a drug in a clinical trial in which there is no pharmaceutical company or partner?”3

The answer was clear: the state would step in. At least six trials comparing bevacizumab and ranibizumab for wet AMD have now reported their outcomes. And a meta-analysis by the Cochrane Collaboration last year found no difference in efficacy and no difference in deaths or serious adverse events or specific subsets of adverse events with the exception of gastrointestinal disorders.4

An accompanying article shows how the manufacturers didn’t want to do the trials themselves and when it was agreed that the public should fund both a comparative and a dosing trial, they did all they could to scupper them—even turning to the Royal National Institute of Blind People (RNIB) for help. Then once these trials were published, they embarked on a campaign to undermine and divert attention from the results, raising safety concerns themselves and via their key opinion leaders and charities.

How various organisations view bevacizumab for wet AMD

The World Health Organization has also backed bevacizumab for ophthalmic use, adding the drug to its essential medicines list. When faced with a controversial decision whether to include the cheaper drug off-label or to allow patients around the world to go blind unnecessarily, Cees de Joncheere, director of essential medicines and health products at WHO, told The BMJ: “WHO’s interest is in protecting patients rather than protecting companies.”

Last year the French and Italian governments passed laws to allow the reimbursement of off-label medicines, specifically referring to bevacizumab.

However, three trade groups representing the drug industry have filed a complaint with the European Commission against Italy and France for expanding the off-label use of medicines. They maintain that Italy is undermining the EU regulatory system, patent protections, and incentives for drug development.

Last year, an investigation by Italy’s competition authority led to a £150m fine for Roche and Novartis for allegedly colluding to prevent the use of bevacizumab in wet AMD. It alleged that despite little data on harms of bevacizumab both drug companies colluded in frightening doctors in a series of scientific articles and conference publications. In December, an Italian court dismissed an appeal by industry, and Italy’s health ministry has announced that it will seek damages of €1.2bn (£890 000; $1.6bn).

In October 2014, the French medicines regulator also wrote to Roche to gather more information about the safety and efficacy of bevacizumab. It is seeking to issue temporary authorisation to use the drug in wet AMD.

Given this favourable evidence base, and the money being spent on ranibizumab, what is stopping the use of a cheaper and equally effective alternative?

GMC guidance

The simplest answer is the GMC. Its guidance on prescribing and managing medicines discourages off-label prescribing when there is a licensed alternative, as in the case of bevacizumab and ranibizumab.

But the GMC has not always been so hostile to off-label prescribing. Questions about the ophthalmic use of bevacizumab led the GMC to launch a consultation into the use of off label and unlicensed drugs in 2011. The initial proposal included a new clause that would encourage more doctors to prescribe off label for reasons of cost. The clause said that doctors could prescribe off label if “there is no appropriately licensed alternative available or they are satisfied, on the basis of authoritative clinical guidance that it is as safe and effective as an appropriately licensed alternative.”

Speaking about the proposed new clause at a meeting held by the RNIB in August 2011, GMC policy adviser Michael Keegan, said: “Doctors have an ethical obligation not only to provide the best care, but also to make the best use of available resources.” He added, “The consultation draft includes a clause that could encourage more doctors to prescribe off-label for reasons of cost.”

But the GMC appears to have done a U turn. The final guidance, issued in early 2013, says that doctors “should usually prescribe licensed medicines in accordance with the terms of their licence.” It concedes that “prescribing unlicensed medicines may be necessary,” but only “where there is no suitably licensed medicine that will meet the patient’s need.”

Why this change when GMC documents seen by The BMJ suggest that doctors’ organisations and NHS bodies were “overwhelmingly supportive” and there was “strong statistical support” for relaxation of the guidance to allow for “the consideration of cost effectiveness alongside patients’ clinical interests?”

How important was lobbying by the Medicines and Healthcare Products Regulatory Agency and industry in the GMC’s change of heart? Documents obtained under freedom of information request show that the GMC consulted the Association of the British Pharmaceutical Industry (ABPI) and the MHRA on the matter, and continued to negotiate with both organisations after the consultation period had closed in May 2011.

The BMJ has learnt that both the ABPI and the MHRA, acting on behalf of the government, lobbied against the proposed new clause. The ABPI did not want cost to be a factor in prescribing decisions, arguing that it should be made clear that “budgetary responsibility never compromises clinical responsibility” and that financial gain cannot take priority over the health of the public.

“It would be unfortunate if GMC guidance was used, by NHS commissioners for example, to exert pressure on doctors to put financial considerations ahead of the best interests of patients,” an email from the ABPI to the GMC said. Indeed, the guidance issued in 2013 is about prescribing and not commissioning.

The ABPI argued the changes “will fundamentally undermine the licensing process as well as have serious implications for patients’ safety and the health of the industry, not least in being a direct threat to innovation”—a point also made to government ministers by Novartis. It was a view the MHRA also shared.

As the ABPI and the MHRA continued to press for the removal of the clause, the GMC tried to reach a compromise, saying that “authoritative clinical guidance should be limited to NICE and parallel bodies in UK countries.”

But the then chief executive of the MHRA, Kent Woods, disagreed, as did the ABPI—the MHRA is the regulatory authority and not NICE. “We have concerns about the suggestion that NICE (or any of its counterparts in other UK countries) is in a position to provide authoritative guidance that an off-label medicine is as safe and effective as an appropriately licensed alternative,” he wrote, adding that it “could confuse clinicians as to the respective roles of MHRA and NICE.”

Novartis had also made the same point to government ministers: that the MHRA was the only authority with the “competence to assess the safety, efficacy and quality of medicines. It is not the role of NICE.”

The GMC pointed out that off-label prescribing when there is a licensed alternative routinely occurs in the NHS. NICE guidelines recommend off-label prescribing alongside licensed alternatives for cost effectiveness reasons—for example, amitriptyline for neuropathic pain and sertraline for generalised anxiety disorder. These guidelines still stand and cost is taken into account.

However, internal GMC discussions obtained under freedom of information show the MHRA was “uncompromising,” saying that it was never appropriate to prescribe off-label when a licensed drug was available. The MHRA—along with the RNIB and the ABPI—was against the proposed clause that would have allowed the NHS to make judicious use of its resources and presented “ongoing difficulties” for the guidance, GMC discussions show.

Confusion over terminology

The new guidance has left doctors and commissioners confused and frustrated.

Last year, the Vale of York Clinical Commissioning Group noted what other services it could fund with the £4m it was spending on ranibizumab: eight neonatal care beds for a year, a full emergency department for over six months, or 5000 cataract operations.

In December the group’s chief clinical officer, Mark Hayes, wrote to the GMC for advice.

“We have the situation whereby the guidance of the GMC prevents doctors from using a product outside of its license in preference to a licensed product,” he wrote.

But the GMC is standing by its guidance. Fionnula Flannery, a GMC policy manager, replied that “it would be unlawful to prescribe unlicensed medicines—including for an indication not covered in the licence—on the grounds of cost.”

Her letter explains that the GMC’s position is based on European law, which she says was clarified by a European Court of Justice ruling in March 2012 against the Polish government. However, the relevance of this ruling to the prescribing of bevacizumab is not immediately clear.

Firstly, European law relates to the manufacture and marketing of drugs and says nothing about prescribing of drugs once they are licensed.

The GMC recognises this. “We are unaware of any domestic or European legislation that restricts the use of licensed drugs to their licensed purposes, or which establishes criteria restricting the circumstance in which licensed drugs may be prescribed for purposes outside the terms of their licence,” a GMC policy official said in internal discussions.

The lack of European legislation around prescribing is a point that the MHRA has also subsequently confirmed to The BMJ.

Secondly, the ruling was against the importing of unlicensed generic drugs. Bevacizumab is not a generic. Novartis has repeatedly made clear that the two drugs are different, with their own molecular structures and dosages. The French medicines regulator has also said that they are two “radically different products.”

Another source of confusion is the GMC’s decision to merge off-label and unlicensed prescribing in its final guidance.

Throughout its deliberations—including those with the ABPI— the GMC had made a clear distinction between drugs that did not have a UK licence (unlicensed) and drugs that were prescribed and used outside the terms of their UK licence—off-label prescribing. The ABPI seemed to agree with this distinction.

However, this differentiation was not reflected in the final guidance. Advice from “leading counsel” in June 2012—after the verdict of the Polish case—led the GMC to restructure the guidance “to cover a single category of unlicensed medicines,” which encompassed medicines that did not have a UK licence and using medicines outside their licence (off-label).

Nine months after The BMJ’s initial freedom of information request, the GMC eventually released its internal deliberations. However, the legal advice, which was so critical to the final guidance, was withheld.

Even though it may have a direct impact on patients, parts of the discussion were heavily redacted because the GMC said the public interest was best maintained by not releasing them and they attracted “legal professional privilege.”

The BMJ asked the GMC why it combined off-label and unlicensed drugs into one category. A spokesperson for the GMC said that the MHRA agreed that the “plain English terminology used in our guidance is consistent with the legal framework surrounding the supply of off-label and unlicensed medicines.”

Reports funded by Bayer (which makes another VEGF inhibitor, aflibercept) and Novartis have also made much of this blurring between medicines use arguing that that off-label prescribing is a “subset of unlicensed prescribing” in legal terms.

The European Federation of Pharmaceutical Industries and Associations has also referred to the GMC’s decision when lobbying the European Commission against the use of off-label medicines when there is a licensed one available.

Niall Dickson, chief executive of the GMC, told The BMJ, “The main problem is the law. The European Court has in effect ruled out the adoption of blanket policies that permit the off- label/unlicensed prescribing of medicines on the grounds of cost.”

But the EMA is clear about the distinction. After consultation with EMA’s legal team, a spokesperson said: “The European Commission and the agency are of the view that any use of medicines outside the terms of the licence—ie, not in accordance with the authorised product information—is considered ‘off-label.’”

This definition would apply to a wide range of drugs used in the NHS. It includes variation of dose, frequency, a condition outside the defined indications, routes of administration, or contrary to listed warnings.

Question over licensing status

The GMC guidance raises a new anxiety about whether bevacizumab for ophthalmic use is unlicensed. Those who argue that ophthalmic use of bevacizumab is off-label do so on the grounds that it is the same drug being used in a different condition. But is intravitreal bevacizumab in fact a different product which is currently without a licence at all in the UK? The MHRA thinks so.

In the UK, bevacizumab has to be repackaged in compounding pharmacies for intravitreal use because it is not manufactured in small enough doses. The MHRA told The BMJ that aseptically transferring medicines into smaller units “constitutes a manufacturing step and the resultant product is unlicensed.” However, its website is unclear. Up until 2009 the UK’s regulator described intravitreal bevacizumab as “off-label.” Following a phone conversation with Novartis in 2011, the agency added information saying it was “unlicensed,” documents obtained through freedom of information requests show.

In the confusion a new term has been coined— “unlicensed use”— to describe bevacizumab’s use in the eye, and Novartis has repeatedly used it in letters to government. The EMA is also clear that this new term “unlicensed use” has no meaning in law.

Alex Foss, consultant ophthalmologist and lead investigator of the ongoing TANDEM trial examining bevacizumab dosing in wet AMD points to the lack of logic in this reasoning—that when a doctor draws a drug up in a clinic and injects it for an indication not mentioned in the drug’s label, this is considered off-label, but to draw up a drug under aseptic conditions in a pharmacy and then inject it is “unlicensed.”

“I think in the eyes of many this will be considered not only to be nonsense but dangerous nonsense as clearly the latter is safer for the patient,” he says.

When The BMJ asked EMA and the US Food and Drug Administration about repackaging bevacizumab into smaller vials and syringes for use in the eye, both agencies were clear that this still constitutes off-label use because the drug is licensed for other indications.

The law

The BMJ asked David Lock, QC, to provide a guide through the legal maze.5 He says there is a theoretical possibility that a doctor could be reported to the GMC for prescribing bevacizumab instead of ranibizumab for wet AMD. However, there is no record that any doctor has been investigated by the GMC for doing so.

The legal confusion surrounding the use of bevacizumab could have been resolved if the courts had ruled on the case of the Southampton, Hampshire, Isle of Wight, and Portsmouth (SHIP) Primary Care Trust (PCT). However, The BMJ has learnt that the government was reluctant for the case to go ahead.

A group of commissioners at SHIP had agreed to fund bevacizumab to treat wet AMD to make better use of NHS resources in 2011. Patients were also given the option of treatment with ranibizumab. But Novartis forced a judicial review of the policy. And by the end of July 2012, SHIP agreed to withdraw the policy, bringing the judicial review to a close. At the time, the chair of SHIP, Jonathan Montgomery, said: “We remain of the view that the policy was lawful, sensible, and safe for patients.” Those involved in the case had to sign confidentiality agreements. However, The BMJ has learnt that some SHIP staff received phone calls from senior NHS officials urging them not to take the case forward.

SHIP executives met with Novartis several times, and sources have said that the company seemed desperate to do a deal and avoid a public hearing. Emails and letters obtained under freedom of information seem to support this assertion. A letter to former health secretary Andrew Lansley said that Novartis wanted “to resolve this issue without recourse to the courts.”

Correspondence between Lansley and Novartis about the SHIP judicial review is missing from the documents sent to The BMJ. A spokesperson for the Department of Health said that an initial search “did not reveal the letter.”

A letter from Novartis to Lansley in July 2012, sent just after SHIP decided to revoke its new policy, said that the company had “welcomed the opportunity to work together with SHIP to explore potential options [that would] allow more patients to have access to Lucentis,” adding: “We have made a confidential commercial proposal to the SHIP Cluster Board.”

It is unclear what this offer from Novartis entailed, but Montgomery is quoted as saying the offer would “reduce significantly the £7.5m spent annually on Lucentis in the SHIP area.”

As a commissioner, public health consultant Daphne Austin, who was in West Midlands at the time, took a particular interest in the case. “It was clear the only way to settle the issue would be for a case to be put before the courts. Commissioners and the public health community were therefore hoping that the SHIP case would go ahead.

“One cannot be certain of the reasons for the case not proceeding, but it cannot be because the PCTs were convinced they were acting unlawfully, this I know at least to be true, and it seems to me, and many others, there were many interested parties not wanting the courts to settle the question,” she told The BMJ.

But though the SHIP judicial review was brought to a close, Novartis has not sought to stop other PCTs that have used a similar policy.

One PCT developed a pathway that allowed doctors to prescribe bevacizumab to patients with wet AMD without falling foul of judicial review. The approach cleverly exploited the fact that bevacizumab is still prescribed in private medicine.

Stockport PCT offered patients a choice of being treated at a local private hospital with bevacizumab, paid for by the NHS, or at the local NHS hospital with ranibizumab. They continued to do so despite letters from Novartis and the RNIB. The eye charity complained that the PCT was not offering patients a choice but treating patients with bevacizumab automatically.

After complaints Stockport commissioners changed the wording on their referral letter, but they did not face a lawsuit.

The RNIB has campaigned hard to promote ranibizumab in the UK. Before the drug was even licensed in 2007, the charity cohosted an advocacy service, part funded by Novartis, for patients seeking access to ranibizumab on the NHS.

Is the tide turning?

What prospects are there for a better solution for doctors and their patients? On the plus side, NICE has announced it will produce new clinical guidelines on the treatment of AMD, although it is not yet clear whether they will include bevacizumab. The RNIB and the Royal College of Ophthalmologists have called for NICE and the MHRA to evaluate bevacizumab for use in the eye. “Central government and the UK regulatory bodies must come together to find a solution,” an RNIB spokesperson said. Meanwhile pressure is building from patients. A letter to Jeremy Hunt and Andy Burnham from a patients’ participation network in London in February urged them to take action—including the compulsory licensing of bevacizumab.

There is no sign that the government plans to intervene. A reply on behalf of Hunt said that if he were to do that “he might be open to accusations of interfering in pharmaceutical market.”

But the health department has indicated that NICE is free to include bevacizumab for wet AMD. A spokesperson told The BMJ that it is for NICE to decide what falls within the scope of its guideline. “We recognise that NICE should consider the full breadth of current NHS practice in developing the scope for this guideline.”

So what will NICE do? Documents obtained under freedom of information show that rather than sanction NICE to do a technology appraisal for bevacizumab, the Department of Health has opted to pursue confidential patient access schemes to reduce the price of ranibizumab. A spokesperson says this is a straight price reduction.

But for a NICE clinical guideline, the institute may need to review these patient access schemes because these were taken into account in its appraisal of ranibizumab. And here’s the rub. To be able to review the technology appraisal NICE will have to get the agreement of stakeholders, which include Novartis and Bayer.

What’s more, Novartis has previously argued against including bevacizumab as a comparator in NICE appraisals of other VEGF inhibitors on the grounds that the drug is not routinely used in the NHS for ophthalmic conditions. But this is a circular argument. If doctors are worried about being sanctioned by the GMC for using “unlicensed” intravtireal bevacizumab, it is little surprise that uptake is low.

Doctors do use ophthalmic bevacizumab in the private sector. After ranibizumab was licensed, a RNIB official—whose name was redacted from emails obtained under FOI— asked the Royal College of Ophthalmology to completely reject bevacizumab-only services in the NHS. The RNIB wanted the inclusion of a statement that did not “close the door to the use of Avastin in private practice” but made it clear that bevacizumab “is not the cheap way out for the NHS.”

Austin points to the double standard in the whole debate. “Everyone knows that these drugs are offered as a matter of course in the private sector,” she says adding: “The proposition that it is ethical to provide the cheaper drug to a private individual, but unethical when the payer is the public purse is a demonstration of contempt for both the taxpayer and the patients whose care is sacrificed as a result,” she says.


Cite this as: BMJ 2014;349:h1654



View Abstract