The
Natural History of HPV
Infection
and Cervical
Carcinogenesis
Mark Schiffman,
MD, MPH
National
Cancer Institute
Lessons
Drawn From 3 NCI Cohorts
And
Colleagues
Portland Kaiser Permanente (20,000+
Women receiving Pap smears)
Guanacaste,
Costa Rica (9,000+women in
a high-risk general population)
ASCUS-LSIL triage study (5,000+women
with mildly abnormal Pap smears)
Reconsider
the CNI Continuum
Surface of
Epithelium
Basal
Epithelium
Normal LOW GRADE S.I.L. HIGH GRADE
S.I.L.
C.I.N. 1 C.I.N.
2 C.I.N. 3
CONDY- VERY MODER- SEVERE
CARCI- INVASIVE
LOMA- MILD MILD ATE DYS NOMA CANCER
TOUS DYS- DYS- DYS- PLASIA IN SITU
ATYPIA PLASIA
PLASIA
PLASIA
NATURAL History of Cervical Carcinogenesis
Persistence
(>1 year)
Norm Infection--------HPV Progression
Pre Invasion CANCER
clearance Infected Cancer
Cervix Cervix Regression
Mild
Cytologic
Abnormalities
HPV and Cervical Cancer
Over
100 types of HPV, over 40 genital types
Approximately
15 types associated with cervical cancer; most common 16,18,31
and 45 (focus of this talk)
Condyloma acuminatum most commonly types 6
& 11 “Benign” types include tupes 53 and 61.
HPV
16,18,31 And 45 Cause
Most
Cervical Ca Worldwide
16 18 31 45 Ng/0th
% of Cancers
Positive for Type
Sexual
Transmission of HPV
Established
% 
Lifetime # Sex
Partners
HPV 16 DNA Seropos 16
Out
of Thousands of women typed, 61 women were found at enrollment
to have HPV 16 DNA by PCR, with normal cytology and no history of previous CIN.
HPV
16 Persistence Among 61 Initially Cytologically
Months
Viral Clearance
HPV 16 DNA detection drops to
about ane-half
Within a year, and
somewhat more by two years.
The two instances of HPV 16
persistence found uncesored at 4 years both were
associated with high-grade CIN.
Clearance apparently type specific
and type independent. Non-oncogenic clear faster.
Our conclusions agree with
prospective data summarized by Burk recently in Papillomavirus Reports (2001;12:119-123).
Guanacaste, Costa Rica
We designed the Guanacaste study to examine the origins of high-grade CIN
in a population without prior effective screening and treatment.
HPV testing of all specimens and histopathology reviews
of possible incident high grade endpoints are underway.
Guanacaste: Study Population and
Baseline Screening Tests
Census of population: 9, 175 women
underwent gynecologic screening at baseline (> 90 % participation).
PCR testing of all enrollment specimens generated a cohort of 219 women
infected with HPV 16, but with only low-grade or no cytologic abnormalities
(prevalent CIN 2 + and cancers were censored).
Follow-up
v Women were followed by
repeat examinations for new cervical abnormalities. For safety, we treated and
censored any evidence of CIN 2 +
v
For this analysis, we were able to complete repeat viral
testing of the 5 – 6 year (or latest available) specimens.
Outcome of 219 HPV 16 infections
followed over 5 – 6 years
Oucome of
Infection
Regress
Pers, NL Pers, Abn CIN 2 CIN
3 ICC
Persistence and
Progression
Persistent HPV 16 is highly
associated with progression to precancer and cancer
We now define persistence as
repeat detection at 1 –2 years, recognizing that the longer HPV 16 persists,
the higher the absolute risk for CIN 3 and cancer.
Co – factors include smoking, multiparity, but viral load is a complicated issue.
Risk for Cervical
Cancer and Precancer Associated with an Abnormal
Baseline Pap smear and/or a Positive HPV test over 45 months.
Pap and/or # # Absolute Relative
HPV Women Cases Risk Risk
Positive (95
% CI)
No 17573 16 0.09
Yes 3213 102 3.2 35 (21- 59)
What About Viral Latency?
Natural history studies through 10
years of follow-up suggest that latency is not a big factor, but these studies
are still too short
Latency is a theorical
concern, but if cancers arise from reactivated virus, will affect screening and
vaccination plans
Evidence to date suggest very low viral load is good, not a high risk.
Cervical Cancer Risk Factors
Age onset sexual activity (HPV)
Number of sexual partners (HPV)
Smoking (co-factor)
Very high parity (co-factor)
Compromised immune system (HIV and
other immunosuppression), risk to CIN 3
LACK OF SCREENING.
Cervical Neoplasia: Pyramid of Diagnoses in the
CA 15,000
Problem of 2,000,000 HPV
Infections Annually in the
We can now screen very sensitively
so the issue
Of reassurance (negative
predictive value) is technically “solved”, but the issue of positive predictive value is not
We need to know what to do for
women who screen positive, to prevent morbidity from excessive referrals,
over-treatment, etc.
Histologic confirmation of CIN 1
is very problematic (see Cox and Guido presentations)
Thanks again to my many
collaborators.
Regresar
a www.cealgin.com