Intended for healthcare professionals


What is stopping the NHS from using bevacizumab for macular degeneration and other retinal disorders?

BMJ 2014; 349 doi: (Published 19 November 2014) Cite this as: BMJ 2014;349:g6887
  1. Andrew Lotery, professor1,
  2. Carrie MacEwen, president2
  1. 1Clinical and Experimental Sciences, Faculty of Medicine, University Hospital Southampton, Southampton SO16 6YD, UK
  2. 2Royal College of Ophthalmologists, London NW1 4QW, UK
  1. Correspondence to: A Lotery A.J.Lotery{at}

Government must act to remove the hurdles

Age related macular degeneration (AMD) is the commonest cause of blindness among elderly people in the developed world.1 The treatment of neovascular AMD was revolutionised by clinical trials of ranibizumab (Lucentis), a monoclonal antibody against vascular endothelial growth factor. These showed that vision was improved with repeated monthly intravitreal injections.1 The National Institute for Health and Care Excellence (NICE) approved its use in 2008 and subsequently approved aflibercept, which also inhibits vascular endothelial growth factor, in 2013.

Since then, carefully conducted clinical trials have shown that a similar drug, bevacizumab (Avastin), is as effective as ranibizumab at improving or stabilising vision in patients with neovascular AMD. The two largest studies are CATT (Comparison of AMD Treatments Trials) in the US2 and the IVAN (Inhibition of VEGF in Age-related Choroidal Neovascularisation) trial in the UK.3 Bevacizumab, however, is not licensed for use in eyes.

There are concerns that bevacizumab is not as safe as ranibizumab. However, a recent Cochrane review found no difference between the two drugs for deaths, all serious systemic adverse events, or specific subsets of adverse events with the exception of gastrointestinal disorders.4 It concluded that, from a safety point of view, there was no significant evidence to support the preferential use of either bevacizumab or ranibizumab in the treatment of neovascular AMD.4

Financial implications

The IVAN study found that using bevacizumab instead of ranibizumab to treat neovascular AMD would save NHS England £102m (€128m; $160m) a year.5 In the US switching to bevacizumab would save nearly $29bn over 10 years without substantially altering patient outcomes.6 Therefore switching to bevacizumab seems an obvious way to minimise costs in this time of austerity. Furthermore, there are other retinal conditions for which the best treatment is vascular endothelial growth factor inhibitors. These include retinal vein occlusions and diabetic macular oedema. NICE has also approved ranibizumab for these conditions. The evidence base is, as yet, not so strong for bevacizumab in these diseases, but it is likely that it will be shown to be effective, again at a much cheaper cost.

All vascular endothelial growth factor inhibitors need to be given repeatedly over many years. In the context of an ageing population, the drug costs to the NHS can be expected to escalate exponentially.

In 2011, after the CATT study was published, the Royal College of Ophthalmologists stated: “The college believes that the NHS executive should urgently instruct NICE and the Medicines and Healthcare Products Regulatory Agency (MHRA) to evaluate the use of Avastin in the treatment of AMD and produce national guidelines for the use of anti-VEGF agents in AMD.”7 That call was ignored.

Obstacles to using bevacizumab

Commissioners are expected to enact NICE guidance, and NICE has not considered bevacizumab. This makes it difficult to commission the use of bevacizumab and to deprive patients of NICE sanctioned treatment. Furthermore, the General Medical Council (GMC) states that doctors should prescribe unlicensed drugs only if “there is no suitably licensed medicine that will meet the patient’s need.”8 Without unequivocal GMC and NICE support, ophthalmologists are understandably concerned that they may be assuming unacceptable personal liability by using an unlicensed drug when a licensed alternative exists. Where these concerns don’t exist, ophthalmologists are happy to prescribe bevacizumab. For example, bevacizumab is the market leading drug for neovascular AMD in the US.9 Closer to home, in Guernsey, where NICE guidance does not apply, bevacizumab is the only therapy commissioned for the treatment of neovascular AMD.

To change treatment in the UK NICE must appraise bevacizumab along with licensed drugs for the treatment of neovascular AMD. Other vascular endothelial growth factor inhibitors do not necessarily need to be excluded because, for example, some patients may benefit from a switch from one to another and aflibercept has a practical advantage in that it can be given at two monthly instead of monthly intervals.10 Therefore there should still be the option of using alternatives to bevacizumab if they are in the patient’s best interest. Until bevacizumab has been appraised the GMC must also be unambiguous in supporting doctors who use the off-label drug instead of licensed alternatives.

There are reasons why NICE has not considered bevacizumab. The same company Roche makes ranibizumab and bevacizumab (Novartis has marketing rights outside the US for ranibizumab) so there is little financial incentive for Roche to pursue NICE appraisal for bevacizumab or to have it licensed for use in neovascular AMD.11 12 Therefore in this unprecedented situation, either the regulators must find a way to licence a drug without the sponsorship of the company that owns it or NICE must find a way to consider an off-label drug that is not being submitted for appraisal by its owners. Bevacizumab could then be used routinely in the UK, saving the NHS millions of pounds a year.

The hospital eye service is facing a serious and ever increasing capacity problem because of the demand for frequent intravitreal injections. Consequently, patients may not be getting treatment when they need it and not getting the best results. The money saved by switching to bevacizumab could facilitate investment in these services. Given the overwhelming evidence for the effectiveness and safety of bevacizumab in the treatment of neovascular AMD, central government should act to overcome the bureaucratic hurdles that prevent its use.


Cite this as: BMJ 2014;349:g6887


  • Competing interests: We have read and understood the BMJ Group policy on declaration of interests and declare the following interests: AL has served on advisory boards for Novartis, Bayer, and Roche and acted as an investigator in clinical trials sponsored by these three companies. He has received travel grants to attend educational meetings and honorariums for giving lectures from Novartis and Bayer. His institution has received educational grants from Novartis. He was a co-investigator on the IVAN study and provided a bevacizumab AMD treatment service for patients in Guernsey. He co-wrote the Royal College of Ophthalmology AMD treatment guidelines. CM is president of the Royal College of Ophthalmologists, which has received educational and grant support from Novartis and Bayer and research funds from Bayer.

  • Provenance and peer review: Commissioned; not externally peer reviewed.


View Abstract