molecular epidemiology of human papillomavirus (hpv) and ... papillomavirus and... · epidemiology...
Post on 20-Aug-2019
227 Views
Preview:
TRANSCRIPT
Epidemiology of HPV
and Cervical Neoplasia
Implications for Optimal
Vaccination and Screening
Mark Schiffman, MD, MPH
September 2009
TISSUE: Cervix
• Best understood site of HPV infection
• For all sites where HPV causes cancer:
HPV +
Transformation Zone =
Increased Risk of Carcinogenesis
Transformation
Zone -- Visually
Transformation Zone –
Histologically
10
8
1
9
11
7
5
6
4
2
15
3
13
EXPOSURE: HPVHPV6
HPV11
HPV13
HPV74
HPV44
HPV55
HPV91
HPV7
HPV40
HPV32
HPV42
HPV54
HPV52
HPV67
HPV33
HPV58
HPV16
HPV31
HPV35
HPV34
HPV73
HPV59
HPV18
HPV45
HPV70
HPV39
HPV68
HPV85
HPV26
HPV69
HPV51
HPV82
HPV30
HPV53
HPV56
HPV66
HPV57
HPV2a
HPV27
HPV71
HPV90
HPV61
HPV72
HPV62
HPV81
HPV83
HPV89
HPV84
HPV86
HPV87
HPV28
HPV3
HPV10
HPV29
HPV77
Burk,
Virology 2005
α1, 8, 10, 13
Genital Warts
α2, 3, 4, 15
Commensal
Infections
α5, 6, 7, 9, 11
Carcinoma &
Precursors*Evolutionary
Tree
(millions of
years)
18
16
Peak Ages: 15-25 25-35 45-50
DISEASE: Cervical Precancer and Cancer
Schiffman et al., Lancet, 2007
Epidemiologists Define Critical Steps
From Normal to Cancer
Normal
PrecancerAcute
HPV
Cancer Death
Epidemiologists need
reliable categories
to define risksGenital Warts
The First Step is HPV Infection
• Easily transmitted
• Each infection is
independent
• A woman can have
several, at the same
or different times
• The peak incidence in
a population is usually
at young ages
Normal
Acute
HPV
Rapid Clearance is the Rule
Normal
Acute
HPV
Persistence is Highly Associated
with Risk of CIN3
• Overt Persistence is key risk factor for precancer
• HPV type very important
• Co-factors like smoking or parity less important
PrecancerAcute
HPV
Clearance
90% of New Carcinogenic HPV
Infections Clear. Many but Not All
that Remain Indicate Precancer
Guanacaste Cohort (Rodriguez, JNCI 2007)
0.0%
0.5%
1.0%
1.5%
2.0%
2.5%
3.0%
3.5%
4.0%
4.5%
0-9 mo
10 mo - 4 yr
4 - 8 yr
8 - 12 yr
12 - 18 yr
HPV negative
Other ca-assoctypes
HPV18+
HPV16+
Kaiser Portland HPV Study (23,000 Women)
Cumulative Risk of Cancer, by HPV Type
Just
one
Test!
Prospective Study of Cancer Death
Cancer Death
Sankaranarayanan,
NEJM, 2009
Cancer Death
HPV
neg
8 / 24,000 0
HPV
pos
87 / 2,800 12
8 year follow-up in
randomized trial
(130,000 women)
HPV arm of trial
Clinical Viewpoint: Adapting
to New Knowledge of HPV
Parts of “Old” Prevention Strategy
• Pap smear
• Colposcopic impression if Pap abnormal
• Biopsy if needed
• Treatment of CIN
• Follow-up
• Cytology, colposcopy, and biopsy have
been great successes. We can now do
even better.
HPV Testing: Major Randomized
Trials
• All published in 2007-9– New England Journal, Lancet, J Natl Cancer Institute
• Show screening with HPV tests is more sensitive for early detection of CIN3 than cytology
• HPV testing with or followed by cytology might be useful in some places, or other triage tests might be used
• In low-resource regions, screen-and-treat?
• New IARC Study in India by Sankaranarayanan et al. is a landmark study
Schiffman and Castle, NEJM, 2006
Cervical Cancer Prevention Efforts Should Fit
Age Patterns in Natural History
If We Can Afford to Vaccinate…
Vaccinate before the peak of incidence, because the vaccines are
preventive, and they do not work after infection occurs
If We Screen with Limited Resources…
Screen after the peak of incidence, to improve specificity and positive predictive value
Defining a New Clinical View
of Cervical Carcinogenesis
There are many tests which,
alone or combined, predict
similar levels of risk
Cyto-
ASC-US
HPV-/ASC-US
HPV-/Cyto-All ♀
HPV+/ASC-USLSIL
HSIL
HPV+/Cyto-
~0%
Incre
asin
g 2
-Yr.
Ris
k o
f P
recancer
(CIN
3)
2%
10%
40%
100%Cytology HPV
HPV16+
HPV16-
HPV16/18
HPV16+
or
HPV18+
HPV16- &
HPV18-
“Risks”
Are
Simpler
To Use
Than
Algorithms
Don’t Screen Too Often with HPV ( 5-year cumulative incidence of CIN2+
following first HPV detection)
0
2
4
6
8
10
12
14
16
18
Enrollment First Follow-up 2nd Follow-up Never Positive
Year of First Positive HPV Test
% D
iag
no
sed
wit
h C
IN2+
N 1324 6626
Good risk stratification with enrollment testing
Don’t Screen Too Often with HPV ( 5-year cumulative incidence of CIN2+
following first HPV detection)
0
2
4
6
8
10
12
14
16
18
Enrollment First Follow-up 2nd Follow-up Never Positive
Year of First Positive HPV Test
% D
iag
no
sed
wit
h C
IN2+
N 1324 288 6626
Infections found a year later predict lower risk of CIN2+
Don’t Screen Too Often with HPV ( 5-year cumulative incidence of CIN2+
following first HPV detection)
0
2
4
6
8
10
12
14
16
18
Enrollment First Follow-up 2nd Follow-up Never Positive
Year of First Positive HPV Test
% D
iag
no
sed
wit
h C
IN2+
N 1324 288 203 6626
By year 2, any infections found are new and low-risk
Concluding Predictions
• Even better vaccines
• Decreased role for algorithms based on
cytology, colposcopy and targeted biopsy
• More reliance on HPV-related tests and
risk stratification to define management
• We need good epidemiology to inform new
strategies
top related