The cheaper drug, bevacizumab, should be referred to NICEBMJ 2007; 334 doi: https://doi.org/10.1136/bmj.39128.708565.3A (Published 22 February 2007) Cite this as: BMJ 2007;334:381
- James P Raftery, professor of health technology assessment ()1,
- Andrew Lotery, professor of ophthalmology2
- 1University of Southampton, Southampton SO17 1SG
- 2Southampton Eye Unit, Southampton General Hospital, Southampton SO16 6YD
We think that Chakravarthy and Lim could have said more about the pricing of ranibizumab and bevacizumab.1 Both drugs are owned by a single company, Roche/Genentech, which has no intention of licensing the cheaper. The US price of ranibizumab is $1950 or roughly £1000 per injection. Monthly injections would cost £12 000 per patient. Bevacizumab, which is licensed for cancer treatment, could cost as little as £17 per injection, as the dosages used for eyes are minute compared with cancer. In the US, off-licence bevacizumab is estimated to cost $17-50 (£8-25) including the costs of splitting up the larger cancer doses. By refusing to license bevacizumab for macular degeneration, Roche/Genentech is raising the price by an unprecedented factor of over 50.
Given the lack of data directly comparing these two drugs, we support the call for a head to head trial (indeed we are part of a team bidding to do such a trial). We wish to make three further points.
Firstly, we have modelled how much more effective bevacizumab would have to be relative to ranibizumab in order to meet the National Institute of Health and Clinical Excellence (NICE) threshold of £30 000/QALY. Using best estimates of current US prices of $1950 and (a high) $50, ranibizumab would need to be 2.5 times more efficacious to meet NICE's threshold. This seems highly unlikely, given the similarity of the two drugs and the observational data that exist on the effectiveness of avastin. Even if ranibizumab's price was reduced to $500, it would have to be more than 5% better than bevacizumab to be cost effective.
Secondly, the review by NICE of ranibizumab versus standard care for patients with the predominantly classic form of macular degeneration, due to report by October 2007, could imperil any head to head trial in the UK. Should NICE find in favour of ranibizumab, then those patients may well prefer to be treated with ranibizumab rather than being randomised. Any UK trial must recruit quickly.
Thirdly, bevacizumab has been excluded from the NICE review because it is unlicensed. Exclusion of unlicensed drugs is normally sensible owing to lack of data. However, given that Roche/Genentech, which owns both drugs, has no plans to license the cheaper, an exception should be considered. Even if a trial were to show bevacizumab to be equivalent to ranibizumab, it would require Department of Health authorisation before bevacizumab could be widely used. The department should urgently consider referring bevacizumab for NICE appraisal.
Competing interests: Both JR and AL are part of a bid for funding of a head to head comparison of bevacizumab and ranibizumab.