James Raftery: Should the NHS use the new meningitis B vaccine?

The argument over whether the NHS should fund Bexsero, the new meningitis B vaccine from Novartis, raises a global issue about the price of the new vaccine, as well as questions about the role of cost effectiveness analysis in setting prices for vaccines, and the processes of the Joint Committee on Vaccinations and Immunisation (JCVI), which acts as the NICE of vaccines and immunisations.

The good news is that Bexsero is a new, apparently effective, vaccine for meningitis B. The bad news is its price: £75 per dose. Modelling of cost effectiveness estimated that it cost over £100k per quality adjusted life year (QALY.) The price at which it would meet NICE’s threshold for cost effectiveness was put at £9 per vaccine dose.

The total cost to the NHS depends on who is vaccinated and how often. Treating infants, toddlers, and adolescents would cost well over £1billion, or 1% of NHS spend. In March 2014, the JCVI finally recommended three doses for infants only. The cost of the drug at list price would be £157m.

Although Bexsero is licensed for use in the European Union, Canada, and Australia, Novartis has not yet proposed it for license in the US. This is partly due to the meningitis B strain being less common there than elsewhere, (in turn due to earlier use of meningitis vaccine C outside the US.) Prices for drugs and vaccines are usually set for the US market which accounts for over 50% of the global pharmaceutical market. Cost effectiveness plays little, if any, role in the US market.

The UK is the first to assess the cost effectiveness of Bexsero and decide on its use. A committee originally set up to advise on polio, the JCVI has evolved to be an independent advisory committee within Public Health England. Its brief under the 2009 Health Protection Vaccination Act includes advising ministers on vaccination, but with a twist: for topics referred to it by the minister, the minister must accept its advice if the vaccine is shown to be cost effective. The JCVI can only recommend a vaccine if it is cost effective.

This is unusual. The JCVI brief differs from NICE in being so explicit. Ministers neither would, nor have, rejected advice from NICE. What matters here is that the relevant minister referred Bexsero to the JCVI in 2013, thus obliging the department to fund it if it was found cost effective, but apparently without considering what would happen if it was not found cost effective.

Assessing the efficacy of a vaccine such as Bexsero poses problems due to the rarity of the disease. Surrogate markers have to be used, specifically the capacity of the sera of immunised individuals to kill the relevant strains of meningococcus B in a laboratory assay.  These were the primary outcomes of the licensing trials. This and other factors (efficacy, duration, carriage and side effects) complicated the modelling of both efficacy and cost effectiveness.

The results of the modelling for the JCVI have varied. A published version showed that the vaccine could be cost effective at the NICE threshold of £30k per QALY only if the price was reduced to £9 per vaccine dose. The JCVI minutes for July 2013 report that when the model was rerun with what it considered more plausible assumptions, then there was no price at which the vaccine could meet this cost effectiveness threshold.  Yet minutes of its meeting of February 2014, and the JCVI position statement revert to the original position that Bexsero could be cost effective at a lower price.

Part of these confusions result from how the JCVI processes differ from those of NICE. It can consult, but is not obliged to. It relies on publication of minutes rather than draft guidance. It appears to conduct much of its cost effectiveness modelling itself. It has no appeal system. These factors have made it more difficult for the JCVI to defend its advice on Bexsero.

There has been a chorus in favour of Bexsero, including a letter from 100 experts to secretary of state Jeremy Hunt, organised by the charity Meningitis Now, but also by shadow health ministers Darren Miller, Andy Burnham, and shadow childrens’ minister Steven McCabe. Given that it was the Labour government in 2009 that effectively made the JCVI the NICE of immunisation and vaccine this is predictably but deplorably ironic.

These developments have landed Jeremy Hunt and the Department of Health in a mess. The department has stated that the vaccine will be made available on the NHS subject to a suitable price being negotiated with Novartis. This is an unusual way to negotiate and hardly one designed to gain a large price reduction. If the economic evaluation had been carried out by NICE, discussion about a patient access scheme would have been on the agenda. However the price reduction from £75 to £9 per dose that is necessary for the vaccine to meet NICE’s cost per QALY threshold of £30k seems too large for any such scheme

If the NHS were to refuse to fund Bexsero, it would have been likely that at least some parents would have purchased it privately. The price, while high, might not be considered unaffordable to some. The presence of side effects (mild fever) may contribute to “vaccine hesitancy” particularly if Bexsero was given with other routine infant vaccinations. Giving Bexsero separately with concomitant prophylactic paracetamol, as recommended by JCVI, will increase the cost to the NHS.

The controversy revealed differences in methods for assessing cost effectiveness. NICE was shown to use two different discount rates, one for health technology assessment (3.5% both costs and benefits except exceptionally) and another for public health interventions (1.5%). This led the JCVI modelling to consider both in its most recent modelling. Although use of the latter lower discount rate led to lower cost per QALY estimates, it made no difference to the finding that Bexsero was not cost effective.

What will happen next remains unclear. As the modelling relies on many assumptions to do with efficacy, duration of protection, and carriage, which can only be resolved in a population based evaluation, such an evaluation should be planned. This could probably be done most efficiently as post implementation surveillance following introduction into the routine immunisation programme. However, given how the story has evolved, a climb down via some form of patient access scheme with a confidential price discount seems more likely. Such a deal may save political face, but will leave the rest of the world to negotiate the price of Bexsero alone with Novartis.

Finally, Pfizer also has a vaccination for meningitis B, focused mainly on adolescents. The price set for Bexsero will set a higher benchmark for any Pfizer vaccine that might otherwise have been the case. Once again, the rest of the NHS will bear the opportunity cost of having to introduce a technology which is less cost effective than many common treatments.

See also:

James Raftery is a health economist with several decades experience of the NHS. He is professor of Health Technology Assessment at Southampton University. A keen “NICE-watcher,” he has provided economic input to technical assessment reports for NICE but has never been a member of any of its committees. The opinions expressed here are his personal views.