Dr Stephen Inglis (Chair)
Dr Simon Kroll
Dr Anthony Harnden
Dr Syed Ahmed
Prof Andy Hall
Dr Douglas Fleming
Dr Paul Heath
Dr Mike McKendrick
Dr Andrew Pollard
Prof Judith Breuer
Prof Elizabeth Miller
Dr John Edmunds
Dr David Brown
Dr Katy Sinka
Scottish Executive
Dr Elizabeth Stewart
MHRA
Dr Philip Bryan
DH
Prof David Salisbury
Dr Peter Grove
Bela Vatsa (minutes)
Pamela Gardiner
Apologies: Dr Ray Borrow, Dr Richard Roberts, Prof Paul Griffiths,
Prof David Goldblatt, Dr Mair Powell, Dr Alessia Melegaro, Dr Lorraine
Doherty, Dr Dorian Kennedy, Dr Karen Noakes, Heather Mellows.
1. Task and method of working
1.1 Objective
The terms of reference for the group are to provide advice to the main
committee on:
- The current burden of disease, by age distribution, resulting from
varicella and herpes zoster infection in the UK and any limitations
to the data
- The efficacy, duration of protection, and safety of varicella and
herpes zoster vaccines and any limitations to the available evidence.
- The benefits of introducing varicella and/or herpes zoster vaccine
to the UK immunisation programme, any impact of vaccination on older
unvaccinated age groups and surveillance requirements, and the most
effective vaccination schedule.
- The cost-benefits of varicella/shingles vaccination
- And to review existing varicella policies
1.2 Timescale for work to be completed
Various models will be developed to predict the likely impact and cost-effectiveness
of vaccination. The subgroup will be tasked with evaluating the models
and assumptions made. The group's aim will be to provide overall advice
on universal varicella/zoster vaccination to the main committee in 2009.
Since modelling work on use of vaccine in relation to antenatal varicella
screening is almost complete, it should be possible to provide advice
specifically on this topic to JCVI in time for its June 2008 meeting.
1.3 Procedures for declaration of interest
Declarations (for GSK and Sanofi-Pasteur) had been requested in advance
of the meeting. No-one has declared any personal specific interests.
Members were reminded that all relevant interests needed to be declared
before participation. It was possible that during the course of the
discussion other potential conflicts might be recognised, for example
relating to antiviral drugs or other therapeutic products. These would
need to be borne in mind and declared as appropriate.
1.4 Invitations for submissions of information
It was noted that submissions of relevant information could be requested
from interested parties including manufacturers.
2. The burden of disease
The group was provided with a summary of published data, an update from
the latest data from RCGP, and a presentation on deaths in children during
the 2006/07 season.
A number of concerns were raised about the burden data:
a) As varicella/zoster are not notifiable, the data must be interpreted
with some caution. It was noted, however, that most zoster cases are
likely to be seen by GPs and hence registered. It was also noted that
in modelling studies, various ways of estimating disease burden had
been used, without affecting the overall outcome significantly.
b) There are limited data available on population subgroups, including
incidence of zoster in ethnic minorities.
c) Estimates of varicella mortality were probably too low. Varicella
is not necessarily recorded where hospitalisation has been due to bacterial
infection. There are probably also many deaths that begin with zoster
but develop into secondary complications and zoster is not recorded.
Finally, a large number of cases are treated by GPs and so are not reflected
in hospital data.
The group received a brief presentation on a recent study that revealed
higher than expected mortality data in children. Among the questions
raised were whether this reflected circulation of a more virulent varicella,
if it might be indicative of an epidemic year, or whether it was the
result of paediatricians ascertaining varicella deaths better. It was
noted that a more virulent strain would have been expected to result
in more adult deaths but this has not been the case. An epidemic was
also considered unlikely; the deaths were recorded in National Statistics
death data prior to active collection for the study, and neither data
from the Royal College of General Practitioners (RCGP) nor data from
Scotland were suggestive of an epidemic year.
The group also discussed the issue of recurrent infection - a study
by Seward (USA) indicates that 13% of varicella diagnosed by paediatricians
is recurrent.
The notion that serology can be used to estimate disease burden reliably
was challenged. It was suggested that only a small proportion of seroconverters
may get symptomatic illness. It was agreed that a comparison of serology
and verbal history was required to address this point. Although some
relevant data from a USA study are available, UK-specific data should
be collected
ACTION: Secretariat to compare methods (i.e. using serology and
verbal screening) to estimate disease burden in the UK and circulate
to the group as soon as possible.
It was suggested that data on varicella infection in those 18 months
and under would be informative as these children are more likely to
develop shingles at a young age as a result of interaction with maternal
antibodies.
ACTION: Douglas Fleming to provide data on age distribution of varicella
in those aged <=18months (for update at next meeting).
In general it was felt that the weakest elements of the data supporting
current modelling were the effects of boosting on zoster and on the
overall quality of life impact from post-herpetic neuralgia (PHN). It
was recognised that it would be difficult to generate more robust data
on these issues. However, it would be valuable to obtain further advice
on the latter from a pain expert.
ACTION: Anthony Harnden to invite Andrew Rice (pain specialist,
Imperial) to join the sub-group.
It was also noted that the last seroepidemiological survey had been
carried out in 1996. It was agreed that this should be updated; this
could be done relatively quickly.
ACTION: HPA to carry out a new seroepidemiological study.
It was agreed that though there were some areas where data could and
should be strengthened, overall the available data on burden of disease
were robust enough to support modelling of vaccine cost effectiveness.
3. The efficacy and adverse events from varicella and herpes zoster
vaccines.
With regard to safety, the chair requested that the group should at
this early stage in its deliberation consider only major issues that
might render further discussion on possible vaccine implementation irrelevant.
No such issues were identified.
It was noted that there is currently insufficient evidence to be able
to distinguish between the safety profile of single varicella vaccines
and those combined with MMR. A ProQuad safety and efficacy study is
in progress, and Priorix-Tetra is used in Germany.
Studies have shown that varicella vaccines were highly efficacious
against severe varicella disease (97% (95% CI 93-99%) against moderate
to severe disease). Efficacy against zoster in adults was approximately
50%.
It was noted that although varicella vaccination campaigns had so far
been based on a single vaccine dose, administration as part of an MMRV
combination would require two doses. In addition, recent data from the
US showing waning responses over time might result in recommendations
from manufacturers to move to two doses for single vaccines also.
ACTION: It was agreed that a two-dose schedule would need to be
included in the HPA cost-effectiveness modelling.
The possibility that vaccine virus could be transmitted from vaccinated
to unvaccinated individuals was raised. Though small, this risk would
need to be considered at a later stage - it was not relevant to the
cost-effectiveness modelling.
4. The predicted impact of childhood vaccination on the burden of
varicella and shingles.
The group discussed published data from the USA. These studies indicate
that mass childhood vaccination has reduced the incidence of varicella.
The data are also consistent with modelling studies that have predicted
an increase in zoster as a result of reduced virus prevalence and hence
decreased opportunities for boosting of immunity to zoster by natural
infection. In two studies where varicella rates were shown to decrease,
there was also a significant increase in zoster post varicella vaccination.
The group noted, however, that the available USA data are not sufficiently
robust to be able to estimate the full impact of vaccination.
ACTION: Secretariat to obtain updated USA data on the impact of
varicella vaccination on zoster (for update at next meeting).
The group also discussed the probable need for a catch-up campaign,
and also studies on serological tests (FAMA) and correlates of protection.
These issues would be addressed in the modelling work.
5. Planned modelling work to estimate the impact and cost-effectiveness
of varicella and zoster vaccination programmes in the UK.
The HPA is currently developing cost-effectiveness models for both
zoster and varicella that should be ready early in 2008. These analyses
will not be based on new serology data and so will need to be updated
when the data are available.
The HPA's published work on varicella/zoster has been peer-reviewed
so the group accepted that a further review process is not necessary.
The two main uncertainties are the estimates for the treatment costs
and QALY losses with (postherpetic neuralgia) PHN and the average duration
of protection resulting from either natural exposure to VZV or zoster
vaccination. Data on costs have come from several sources, but these
are mostly from the USA. The available UK information is out of date
and treatment patterns are likely to have changed. It was noted that
there may be additional available sources of useful information on this.
ACTION: Judy Breuer and John Edmunds to follow up on PHN treatment
costs (for update at next meeting).
A European Collaborative Project underway, led by John Edmunds (HPA),
will include up-to-date data on relevant contact patterns. It will also
consider the impact of boosting immunity to zoster by exposure to natural
infection. This could result in changes to the current HPA model.
6. Antenatal varicella screening and vaccination
It was agreed that the possibility of post partum vaccination of susceptibles,
detected as a result of ante-natal screening (either verbal or through
serological analysis), was worth considering as an interim strategy
in the absence of a decision on universal vaccination. Such a programme
could be introduced quickly at relatively low cost, and could be reviewed
subsequent to a decision on universal vaccination.
A cost-effectiveness model for such a strategy had been developed and
reviewed by the subgroup. Specific points raised by the review were:
- whether verbal screening could really be done without cost
- how representative Bangladeshi women are of other women from tropical
countries
- whether the duration of protection was sufficient to cover a 10+year
period of parity
Any decision on introduction of screening would need to be consistent
with National Screening Committee policy.
ACTION: The National Screening Committee to be consulted by the
Secretariat.
Overall, however, the model was considered robust enough for a decision
to be taken. A recommendation will be made at the next meeting in April
in time for a final recommendation to JCVI in June.
7. Current advice on varicella vaccination recommendations for healthcare
workers.
The group was asked whether the current green book recommendations
were appropriate. Judy Breuer noted that although serology can be used
to determine if someone is immune post-vaccination, many people lose
antibody fairly quickly (30% of cohort after 1 year) and there is evidence
of protection in the absence of detectable antibody. Therefore the value
of post-immunisation serological testing as currently recommended was
questionable.
ACTION: Judy Breuer to provide evidence based statement in support
of this change to be considered at the next subgroup meeting.
ACTION: Secretariat to consider how to amend the Green Book to reflect
this.
8. Next steps:
8.1 Agenda for next meeting
Cost-effectiveness analysis
Antenatal screening
Green Book change on HCW section.
8.2 Date of next meeting
April 2008
8.3 Further input /expertise/actions required
Social research on acceptability of varicella vaccination is needed.
A survey has been conducted previously, but this is now dated.
ACTION: Secretariat to make the previously conducted study available
for the next meeting and consider including varicella vaccine within
new surveys.
You
are here: 