HPV detection
HPV testing
• HPVs can not be cultured in vitro
• No serological or protein tests available that
Peter J.F. Snijders
Dept. of Pathology
VU University Medical center
Amsterdam
are sufficiently sensitive for reliable detection
of HPV presence
• Currently used HPV detection assays rely on
detection of viral nucleic acids (DNA, RNA)
– Principle: hybridisation with complementary
sequences
Some commercial HPV tests
Target amplification:PCR
1st cycle
2nd cycle
3rd cycle
4th cycle
Test comparison on cervical scrapes of women with
CIN3
%
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
Hybrid Capture 2 Qiagen
Digene HPV genotyping RH kit Qiagen
Digene HPV genotyping LX kit, Qiagen
Amplicor HPV Test Roche
Linear array Roche
SPF10/LiPa DDL
NucliSens EasyQ HPV Biomerieux
Aptima Gen-Probe
Human Papilloma Virus reagents Third Wave
BIOPAP QTS HPV Kit Loxo
Reveal HPV Real-Time HPV Detection Kit GenoID
AID STD assay GenID
AID HPV screening kit GenID
AID HPV typing kit GenID
Linear ArrayExtra HPV Genotyping Kit Innogenetics
PCR Human Papillomavirus Detection Set Takara Mirus Bio
HPV DNA Chip Biomedlab
Array Papillomavirus Genomica
ProDect Chip HPV typing Bcs Biotech S.P.A
PapType Genera Biosystems
LCD Array HPV 3.5 Chipron
Seeplex HPV Genotyping Seegene
Viroactiv Virofem
HPV OncoTest Invirion Diagnostics
Genpoint Tm HPV test Dako-Oxoid
Abbott RealTime High Risk HPV Abbott
Luminex HPV Genotyping, Multimetrix/Progen
Papillocheck, Greiner BioOne
PreTect HPV Proofer Norchip
Test comparison on cervical scrapes of women with
normal cytology without high-grade CIN in follow-up
%
p<0.001
N=45
N=264
Principles most commonly used
(hr)HPV DNA detection assays
HrHPV E6/E7 mRNA detection assays
• Hybridisation followed by signal amplification
– Hybrid capture 2 (HC2; Qiagen): RNA probe
cocktail (commercial)
– Invader technology: Third Wave Invader HPV
test (commercial)
• Isothermal RNA amplification methods
– PreTect HPV-Proofer (Norchip), NucliSense EasyQ
HPV (Biomerieux): 5 hrHPV types
– APTIMA (GenProbe): 13 hrHPV types
• In situ hybridisation (ISH): in situ hybridisation of
DNA with probe cocktail (commercial)
• Broad spectrum PCR: DNA amplification with
consensus primers or multiplex format (some
commercially available)
Commonly used HPV broad spectrum
PCR assays target L1 sequences
E6
E2
E7
E5
E1
Enzyme immuno assays
L1
– Cocktail of oligoprobes per well (no genotyping)
(Eg. GP5+/6+-PCR, SPF10, Roche Amplicor)
L2
E4
0
1000
2000
3000
4000
5000
PGMY, MY11/09
6000
7000
7900
450 bp
GP5+/6+
150 bp
SPF10
65 bp
Roche HPV Amplicor™
~165 bp
Example Enzyme Immuno Assay (EIA)
PCR with one biotinlabeled primer
Capture products on
streptavidin-coated micro
well
NaOH denaturation
pNPP
AP
Incubation with antiDIG-AP conjugate
Commonly used read-out systems for PCR
assays
Hybridisation with
DIG labeled probe
Reverse hybridization assays (genotyping); oligoprobes
immobilized on:
• Strips/filters
–
–
–
–
Linear Array (PGMY; Gravitt et al. 1998)
LiPA (SPF10; Kleter et al., 1999)
RLB (GP5+/6+-PCR; van de Brule et al., 2002)
Consensus HR HPV genotyping strip (GP5+/6+-PCR, PGMY,
Amplicor; de Koning et al., 2006)
Enzyme immuno assay
Commonly used read-out systems for PCR
assays
Principle of line probe assay
Enzyme immuno assays
– Cocktail of oligoprobes per well (no genotyping)
– Eg. GP5+/6+-PCR, SPF10, Roche Amplicor
Reverse hybridization assays (genotyping)
Type-specific oligoprobes immobilized on:
• Strips/filters
–
–
–
–
Linear Array (PGMY; Gravitt et al. 1998)
LiPA (SPF10; Kleter et al., 1999)
RLB (GP5+/6+-PCR; van de Brule et al., 2002)
RH genotyping strip (GP5+/6+-PCR, PGMY, Amplicor; de
Koning et al., 2006)
Reverse Line Blot (GP5+/6+ amplimers)
digene RH genotyping strip (Qiagen): genotyping of
18 HPV types (GP5+/6+ amplimers)
16
16
52
16
81
33
42
72
6
81
11
56
42
42
82/mm4
39,43
16
18
16
31
26
40
33
42
16,42
31
16
42
42
16,70
31
56
16
16
81
51
56,50
HPV types in samples
HPV type 16 18 26 31 33 35 39 45 51 52 53 56 58 59 66 68a 68 73 82i 82m
Reverse hybridization assays (genotyping)
Type-specific oligoprobes immobilized on:
•Microarrays
– HPV oligo array (Kim et al. 2003)
– DNA array primer extention (APEX; Gheit et al. 2006)
– PapilloCheck (Greiner Bio-One)
•Microsphere beads
– Multiplex HPV genotyping (MPG; Schmitt et al. 2006)
– HPV DNA suspension array (HPV-SA; Jiang et al. 2006)
– Templex assay (Han et al. 2006)
OLIGO lanes
Conjugate control HPV16 HPV18 HPV26 HPV31 HPV33 HPV35 HPV39 HPV45 HPV51 HPV52 HPV53 HPV56 HPV58 HPV59 HPV66 HPV68 HPV73 HPV82 β globin -
16
18
31
33
35
39
45
51
52
56
58
59
66
68
6
11
26
34
40
42
43
44
53
54
55
57
61
70
72
73
CP6108
71
81
82/is39
82/mm4
83
84
Commonly used read-out systems for PCR
assays
Enzyme immuno assays
– Cocktail of oligoprobes per well (no genotyping)
– Eg. GP5+/6+-PCR, SPF10, Roche Amplicor
Reverse hybridization assays (genotyping)
Type-specific oligoprobes immobilized on:
• Strips/filters
–
–
–
–
Linear Array (PGMY; Gravitt et al. 1998)
LiPA (SPF10; Kleter et al., 1999)
RLB (GP5+/6+-PCR; van de Brule et al., 2002)
Consensus HR HPV genotyping strip (GP5+/6+-PCR, PGMY,
Amplicor; de Koning et al., 2006)
Novel assays: a.o. real-time multiplex PCR (eg. GenoID, Abbott)
Potential outcomes of hrHPV exposure cervix
HPV DNA tests can be used for
several distinct purposes
• Epidemiological: assessing burden of HPV
infections
• Vaccine monitoring: determining protection against
incident HPV infections
• Cervical screening and triage: detecting/predicting
disease
– hrHPV testing should be considered for the detection of
CIN2/3 or cancer, not simply viral infections
hrHPV testing for cervical screening and triage
of women with Pap2/3a1
• Not necessarily hrHPV DNA presence per sé but
elevated levels of viral DNA confer an increased
risk of ≥CIN2/3 (Joseffson et al., Ylitalo et al.,
2000)
hrHPV testing for cervical screening
• For screening hrHPV testing is not inherently useful
unless it is considered in the context of CIN2/3 or cancer
• Key issue: optimal balance between clinical sensitivity/
specificity
– to minimize redundant or excessive follow-up
procedures
• Current clinically validated tests (i.e. HC2 and GP5+/6+PCR): clinical sensitivity of 90-95% for ≥ CIN2/3
• Cautious against misguided attempts to achieve a 100%
clinical sensitivity by increasing the analytical sensitivity
– such small gains will result in a dramatic decrease in
clinical specificity (ie. more false positives)
2-3 years
10-12 years
Productive
infection
hrHPV
infection
CIN 1/2
Transforming infection
20%
lesions
Normal
CIN 1/2
80%
Transient
infection: HPV
clearance
CIN 2/3
Carcinoma
CIN 2/3
Carcinoma
Latent
infection?
Important distinctions
• Analytical sensitivity and specificity
hrHPV infections
• Clinical sensitivity and specificity
≥CIN2/3 (clinically relevant hrHPV infections)
Comparison study between GP5+/6+-PCR and
SPF10-PCR
• 45 cases and 264 controls selected from women
with normal cytology at baseline
• cases = women who developed lesions ≥CIN3
• median age 33 years; median follow-up: 2.7 years
• controls = women with repeat normal cytology
at follow-up or histology ≤CIN1
• median age 41 years; median follow-up: 5.8 years
Hesselink et al., JCM 2008
Nested case-control study women with normal
cytology
Overall
Screening
cohort
Viral load analysis in concordant vs
discordant SPF10/GP5+/6+-PCR samples
Hospital
cohort
Type-specific real time PCR
90
9
Cases: CIN3
80
Controls: ≤CIN1
70
60
p<0.001
50
GP5+/6+-PCR
SPF10
p<0.001
40
p<0.001
HPV 16
HPV 18
HPV 52
p<0.001
p<0.001
p=0.006
8
log [HPV copies/scrape]
%
30
7
6
5
4
20
10
3
GP-/
SPF+
(n=13)
0
Cases
N=45
Controls
N=264
Cases
N=25
Controls
N=193
Cases
N=20
Controls
N=71
GP+/
SPF+
(n=6)
GP-/
SPF+
(n=14)
GP+/
SPF+
(n=4)
GP-/
SPF+
(n=15)
GP+/
SPF+
(n=4)
HPV type
SPF10 positivity rate was significantly higher than that of GP5+/6+PCR for control women, not for case women
Summary GP5+/6+-PCR and SPF10-PCR
comparison
• Analytical sensitivity SPF10 too high for screening
Dramatic increase in the number of false positives (low
clinical specificity)
Samples negative by GP5+/6+-PCR but positive with SPF10 had significantly
lower viral loads
Comparison ISH (Ventana) and
HC2
• Hospital population: women referred to
gynaecologist because of abnormal smear in
screening program or post-treatment monitoring
• LBC specimens (PreservCyt; n=76)
• Conversely, insufficient analytical sensitivity will translate
to unacceptably low clinical sensitivity for ≥CIN2/3
• Prescreened for positivity by hrHPV GP5+/6+PCR
Hesselink et al. Cancer Cytopathol. 2004
ISH (Ventana) vs HC2
CaSki
Severe
dyskaryosis
Moderate
dyskaryosis
Severe
dyskaryosis
100
90
80
%
70
60
50
40
30
20
10
0
ISH
HC2
<=CIN 1
CIN2/3
N=46
N=30
Clinical sensitivity of ISH too low: too many false negatives
HC2 ≈ GP5+/6+-PCR
Hesselink et al., 2004
ROC curve for HC2 values using
cytology threshold >BMD
Conclusions
1,0
Cut-off: 1.0 RLU/CO
GP5+/6+ (3.2 RLU/CO)
0,9
0,8
• hrHPV tests display variable clinical proporties
• For cervical screening: high clinical sensitivity
and at the same time high clinical specificity (limit
false positives)
0,7
Positivity rate
0,6
0,5
0,4
• At the moment only HC2 and GP5+/6+-PCR are
known to fulfill this on the basis of data from
large prospective trials
0,3
0,2
0,1
0,0
0,0
0,1
0,2
0,3
0,4
1 – negativity rate
HC2 and GP5+/6+-PCR are clinically compatible
Hesselink et al.
JCM, 2006
• Adherence of candidate tests to HPV test
guidelines is necessary to avoid adverse affects
of HPV testing
Proposal international guidelines for HPV
test requirements for primary cervical
screening (formulated relative to HC2)
Candidate test should:
Validation guidelines of candidate HPV
assays (comparative analysis with HC2)
1. Clinical sensitivity:
1. Have a clinical sensitivity for ≥CIN2 not less than 90% of
that of HC2 in women >30 years
• Sensitivity HC2: 97.9% (95%CI:95.9-99.9)
• Translates into very high NPV (reassurance) allowing for
extending screening intervals test neg. women
Compared to HC2, relative sensitivity of at least 90% assessed by
non-inferiority score test
≥60 smears of a representative set of women from a screening
population with ≥CIN2 detected by HC2, either or not combined with
cytology, and the candidate test (power of 80%)
2. Clinical specificity:
≥800 smears of women ≥30 years without ≥CIN2 randomly taken
from a screening population tested by HC2, either or not combined
with cytology, and the candidate test (power of 80%)
When 2.500 samples tested: power >99%
2. Have a clinical specificity for ≥CIN2 not less than 98% of
that of HC2 in women >30 years
• Specificity HC2: 91.3% (95%CI: 89.5-93.1)
3. Should display intra-laboratory reproducibility and interlaboratory agreement with a lower confidence bound ≥87%
3. Intra-laboratory reproducibility in time and inter-laboratory
agreement:
≥500 smears (incl. 30% positive in reference lab by a clinically
validated test). Kappa at least 0.5
Meijer et al. IJC, in press
Meijer et al. IJC, in press
Acknowledgements
Laboratory requirements for HPV testing
VU University medical center
1. Infrastructure for molecular testing in case nucleic acid
amplification technology is used
2. Accreditation for clinical molecular testing
Standard operation procedures (SOP) and good
laboratory practice
3. Quality control HPV test performance and sample
processing
Proficiency testing including regular intra-laboratory
evaluation
Inter-laboratory performance evaluated by sending
proficiency panels
Dept of Pathology
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
•
B. Hesselink
D. Heideman
A. Brink
M. Bleeker
N. Bulkmans
F. van Kemenade
A. van den Brule
L. Rozendaal
A. Groothuismink
N. Fransen-Daalmeijer
M. Lettink
M. Verkuijten
D. Boon
R. Pol
R. van Andel
P. v. Diemen
C. Meijer
Dept. Epidemiology and
Biostatistics
•
•
J. Berkhof
F. Voorhorst
Dept. Obstetrics and
Gynaecology
•
R. Verheijen
Albert Schweitzer Hospital
Dept. Obstetrics and
Gynaecology
•
C. Hogewoning
St. Radboud University
medical center
•
•
W. Melchers
L. Massuger
International collaborators
(guidelines)
DDL, Voorburg
• D. Geraets
• M. de Koning
• W. Quint
•
•
•
•
•
•
•
P. Castle (USA)
J. Cuzick (UK)
M. Arbyn (Belgium)
J. Dillner (Sweden)
G. Ronco (Italy)
E. Franco (Canada)
X. Bosch (Spain)
Bead-based multiplex genotyping
HPV type-specific probes coupled to distinct coloured bead sets
(one set per probe)
GP5+
biotinylated GP6+
mix beads with
PCR products
L1 ORF
HPV genome
PCR
heat denaturation
type-specific polymorphic region
reporter dye
Luminex analyser
+
wash +
analysis
Schmitt et al., JCM 2006
hybridisation
wash
wash plate
biotinylated hybridised complexes