Transcript Poster

The UK HIV Drug Resistance Database: Development and use for national surveillance
Deenan Pillay1 and Hannah Green2 on behalf of the UK Collaborative Group on HIV Drug Resistance and the UK Register of HIV Seroconverters
1. UCL, London, UK 2. MRC Clinical Trials Unit, London, UK
BACKGROUND
Determining the clinical correlates of HIV drug resistance, and
estimation of resistance surveillance parameters, require large
databases. A UK-wide non-proprietary collaboration of all
virology laboratories and major clinical centres was established to
pool resistance data and interpretation patterns, together with
pre- and post- test virological, clinical and therapeutic information.
Resistance data is generated by “in house”, VGI (Bayer), and ABI
(Abbott) sequencing methodologies as well as by VIRCO.
We describe the first outputs from the UK database; the
prevalence of resistance in treatment naïve patients, patients with
acute/recent infections (seroconverters) and treatment
experienced patients.
RESULTS
Treatment naïve vs Seroconverters
As of May 2003, 9798 test results, obtained since 1996, from
6684 patients have been entered into the database. 60% of
these patients have clinical data in the UK CHIC database. 2279
(23%) tests were performed whilst patients were naive to
treatment and 5815 (59%) on treatment experienced patients.
The treatment status of the remaining 1704 (17%) tests is
unknown. Resistance test failures and tests with missing results
are excluded from further analyses.
Of 152 seroconverters fulfilling the UK Register case definition,
26 (17%) had evidence of transmitted resistant HIV. Resistance
to any ART class was 15% between 1996-1998 before stabilising
at 18% from 1999 onwards. The prevalence of resistance over
time was similar to resistance in treatment naïve patients.
% of patients with  1 primary mutation (any class)
30
18
20
16
10
14
0
1996-7
10
1500
8
Treatment status
The ART treatment status of patients (naïve to treatment or
experienced) at the time of each resistance test was determined
either by using clinical data or the reason for the test given on the
resistance test request form.
Various checks were made on a sample of patients thought to be
naïve to confirm treatment history prior to the resistance test:
• Resistance test request forms were manually checked for
additional ART information
• Patient notes were manually checked to assess the accuracy of
treatment history contained in the UK CHIC database.
Out of a sample of 72 patients checked, 7 were found to be
treatment experienced at the time of the resistance test.
Resistance
Resistance was defined as at least one key (primary) mutation
according to IAS-USA (2003). The analysis was based on the
first test for each patient whilst naïve to ART treatment and the
first test per year when treatment experienced.
Seroconverters
Tests reported to have been undertaken in ‘seroconverters’ are
followed up through the UK Register of Seroconverters to
establish whether they fulfil a common case definition; that the
resistance test was performed within 18 months of a documented
negative HIV antibody test or laboratory confirmed acute infection
in an ART-naïve individual. Data from the UK register was used
to measure the prevalence of resistance amongst patients with
acute/recent infections.
50
40
12
Clinical data
Without related clinical information, resistance data are of limited
value. Clinical data were obtained from electronic databases
wherever possible including the UK Collaborative HIV Cohort
study (UK CHIC). If no accurate clinical database was available,
clinics were asked to complete a standard form.
80
60
20
2000
METHODS
90
70
Number of tests by year and treatment status
2500
Treatment experienced:
% of patients with  1 primary mutation
1998
NRTI
1999
PI
2000
2001
NNRTI
2002-3
Any Class
6
1000
4
2
500
0
1996-1998
0
1996-7
1998
Treatment naive
1999
2000
2001
Treatment experienced
2002-3
1999-2000
Naive
2001-2003
CONCLUSIONS
Seroconverters
Status unknown
Treatment naïve
Treatment experienced
1966 patients had a resistance test whilst naïve to ART treatment;
270 (14%) with evidence of resistant HIV. For the time periods
1996-1998, 1999-2000 and 2001-2003, resistance to any ART
class was 9%, 15% and 16% respectively. NRTI resistance
followed a similar trend. After an initial rise in PI and NNRTI
resistance, there is no clear trend after 1999.
For treated patients, resistance to any ART class remained stable
over time (71%, 78 %, 78% for time periods 1996-1998, 19992000 and 2001-2003 respectively). NNRTI resistance
significantly increased over time with the rate of increase
declining in recent years. PI resistance peaked in 1999 before
stabilising at 27% in the 2001-2003 period. There was no
evidence of a trend in NRTI resistance.
Prevalence of resistance was not associated with risk group,
ethnic origin or the closest CD4 measurement to the test date.
Treatment naive:
% of patients with  1 primary mutation
25
20
1996-1998
1999-2000
2001-2003
Any class
71%
78%
78%
NRTI
PI
NNRTI
67%
26%
20%
71%
32%
39%
70%
27%
49%
• A large HIV-1 drug resistance clinical database containing nearly
10,000 results has been successfully established in the UK. The
clinical data related to these results have been and will continue to
be audited for accuracy.
• Approximately 15% of drug naïve patients who have had a
resistance test are infected with virus with at least one key
resistance associated mutation.
• Little difference is observed in prevalence of resistance when
comparing nearly 2000 drug naïve individuals with a smaller group
of seroconverters. This suggests that surveillance of chronic drug
naïve patients may be useful for monitoring transmitted drug
resistance. However, because they are chronically infected (in
contrast with seroconverters), the prevalence observed in a given
calendar period does not necessarily reflect the level of
transmitted resistance at that point in time.
• Future uses of the database will focus on virological predictors of
treatment response, in addition to continual surveillance of drug
resistance in drug naïve and experienced patients.
15
10
5
0
1996-7
1998
1999
NRTI
PI
2000
NNRTI
2001
2002-3
Any Class
UK Collaborative Group on HIV Drug Resistance Steering Committee:
Judy Breuer, Sheila Burns, Sheila Cameron, Ian Chrystie, Duncan Churchill, John Clarke, David Dunn, Philippa Easterbrook, Barry Evans, David Goldberg, Mark Gompels, Teresa
Hill, Paul Kellam, Andrew Leigh-Brown, Clive Loveday, Ryanne Matthias, Anna Maria Murdin-Geretti, Andrew Phillips, Deenan Pillay, Kholoud Porter, Anton Pozniak, Caroline
Sabin, Peter Tilston, Ian Williams
UK Register Steering Committee:
Abdel Babiker, Valerie Beral (Chair), Ray Brettle, Chris Carne, Janet Darbyshire, Philippa Easterbrook, Noël Gill, Richard Gilson, David Goldberg, David Hawkins, Don Jeffries,
Anne Johnson, Margaret Johnson, George Kinghorn, Linda Lazarus, Ken McLean, Philip Mortimer, Andrew Phillips, Deenan Pillay, Sarah Rowland-Jones, Jonathan Weber, Ed
Wilkins; Coordinator: Kholoud Porter
Contact: [email protected]