Copyright ©

2011 Inform

a UK Limite

d

Not for S

ale or Commerc

ial Distri

bution

Unauthoriz

ed use prohibite

d. Auth

orised users

can download,

display, view and print a

single copy for p

ersonal u

se

Current Medical Research & Opinion Vol. 27, No. 3, 2011, 569–578

0300-7995 Article 5629/546730

doi:10.1185/03007995.2010.546730 All rights reserved: reproduction in whole or part not permitted

Original articlePopulation-based frequency assessment ofHPV-induced lesions in patients with borderlinePap tests in the Emilia-Romagna Region: thePATER study

S. CostaDepartment of Gynaecology and Obstetrics, S. Orsola-

Malpighi Hospital, University of Bologna, Bologna, Italy

S. VenturoliDepartment of Haematology and Oncological Sciences

L. e A. Seragnoli – Microbiology Section, S. Orsola-

Malpighi Hospital, University of Bologna, Bologna, Italy

F.S. MenniniA. MarcellusiCEIS Sanita (CHEM – Centre for Health Economics and

Management), Faculty of Economics, University of Tor

Vergata, Rome, Italy

M. PesaresiDepartment of Gynaecology and Obstetrics, S. Orsola-

Malpighi Hospital, University of Bologna, Bologna, Italy

E. LeoDepartment of Haematology and Oncological Sciences

L. e A. Seragnoli – Microbiology Section, S. Orsola-

Malpighi Hospital, University of Bologna, Bologna, Italy

A. FalascaE. MarraDepartment of Gynaecology and Obstetrics, S. Orsola-

Malpighi Hospital, University of Bologna, Bologna, Italy

M. CriccaDepartment of Haematology and Oncological Sciences

L. e A. Seragnoli – Microbiology Section, S. Orsola-

Malpighi Hospital, University of Bologna, Bologna, Italy

D. SantiniDepartment of Pathology, S. Orsola-Malpighi Hospital,

University of Bologna, Bologna, Italy

M. ZerbiniDepartment of Haematology and Oncological Sciences

L. e A. Seragnoli – Microbiology Section, S. Orsola-

Malpighi Hospital, University of Bologna, Bologna, Italy

G. PelusiDepartment of Gynaecology and Obstetrics, S. Orsola-

Malpighi Hospital, University of Bologna, Bologna, Italy

Abstract

Background:

The PATER study assessed the frequency of high-risk (HR) and low-risk (LR) human papillomavirus (HPV) in

HPV-induced lesions in patients with borderline cytology.

Methods:

This retrospective observational cohort study was designed to evaluate ASCUS patients detected through a

local cervical cancer screening programme and referred to the Department of Gynaecology and Obstetrics at

the S. Orsola-Malpighi University Hospital in Bologna, in the period between January 2000 and December

2007.

Results:

In 1047 patients aged 38.4� 9.6 years (range 23–65 years), 34.8% (n¼ 364) was positive for HR- or

LR-HPV DNA. The mean age of women with HPV infection was significantly lower compared with the

negative group (36.8� 9.4 versus 39.3� 9.6 years; p50.001). Overall, 357 (34.1%) women had

cervical lesions: 279 (26.6%) had CIN1, 18 (1.7%) CIN2, and 60 (5.7%) CIN3þ. HR-HPV genotype was

detected in 83.3%, and 91.5% of patients with CIN2 and CIN3þ respectively. Among the 124 CIN1

HPV-positive women, 8.9% harboured LR-HPV genotypes, 80.6% HR-HPV and 10.5% a combination of

HR- and LR-HPV. HPV-6 and 11 accounted for 19.4% of all HPV-positive CIN1 lesions.

Conclusion:

Our study suggest that: in ASCUS patients over 40 years there is a low risk of positivity for HPV infection; the

HPV DNA testing in patients with CIN3þ and a mean age close to 40 years is highly sensitive (98.3%) and

acceptably specific (75.5%); the frequency of LR-HPV (alone or in combination with HR) in ASCUS cytology is

not negligible. A tetravalent-based HPV vaccination alongside the screening programme would provide

considerable clinical, organizational, and economic benefits.

Introduction

Cervical cancer screening programmes have been implemented in Italy on aregional basis since 19961, and women aged between 25 and 64 years are rec-ommended to undergo screening for cervical lesions once in a 3-year interval ifthe Papanicolaou (Pap) smear is negative, according to the European guidelinesin cervical cancer screening2.

Pap test reporting classifications evolved during the last decades, especiallywith the current standard being represented by the Bethesda System3.

! 2011 Informa UK Ltd www.cmrojournal.com HR- and LR-HPV infections in borderline Pap tests Costa et al. 569

Cur

r M

ed R

es O

pin

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Fran

cis

A C

ount

way

Lib

rary

of

Med

icin

e on

01/

17/1

1Fo

r pe

rson

al u

se o

nly.

Atypical squamous cells of undetermined significance (ASCUS) is the mostcommon abnormal Pap test result accounting for 1–12% of all cytological diag-noses1,4–6. Nevertheless, the positive predictive value (PPV) for cervical intrae-pithelial neoplasia grade 2–3 or cancer (CIN2þ) is about 5% in population-based screening programmes, and 7–16% in controlled clinical trials4,7,8.

Although the PPV is quite low, ASCUS should always undergo an accurateevaluation. Several strategies are usually implemented: immediate colposcopy isreputed as the safest option; however, it is also associated with high healthcarecosts and possible overtreatment. Since the sensitivity of a single cytology isrelatively low, a program of repeated cytology has been using in most countries,with a potential risk of defaulting or clogging1,7–9.

Combined data from several large studies, performed in different settings andpopulations, have evaluated the performance of human papillomavirus (HPV)DNA testing in the management of ASCUS cytology, with a sensitivity indetecting CIN2 or 3 included between 83% and 100% using the HybridCapture 2 (HC2) procedure5,8–15. However, few trials assessed the frequencyof high-risk (HR) and low-risk (LR) HPV genotypes in patients with ASCUS onthe basis of population-based cervical cancer screening reports11–15.

The PATER study (Population-based frequency Assessment of HPV-inducedlesions in patients with borderline Pap Tests in the Emilia-Romagna Region)was designed to investigate the frequency of HR and LR-HPV genotypes inASCUS women who have undergone cervical cancer screening in the districtof Bologna, Emilia-Romagna Region, correlating the virological data with his-tological diagnosis.

Patients and methods

Study design

A retrospective and observational cohort study was planned to assess the fre-quency of HR- and LR-HPV in patients living in the city of Bologna and neigh-bouring areas with a confirmed cytological diagnosis of ASCUS. Women werereferred to the dysplasia clinic at the Gynaecology and Obstetrics Department ofS. Orsola-Malpighi University Hospital in Bologna, Italy. All ASCUS weredetected in an observation timeframe included between January 2000 andDecember 2007. The assessment of frequencies was drawn from a specific data-base developed by cross-linking clinical and microbiological archives which arestored in different departments of the University Hospital. The conjoining pro-cedure (based on the cross-linking of different data) was previously implementedin Italy, providing results methodologically and statistically accurate and reli-able16–18. The study was approved by the local institutional review board.

Data source

Data analysed in this study were retrieved from two different archives. Clinicalcharacteristics of patients such as age, personal health code, diagnostic andtreatment procedures (i.e. colposcopy and histology), and duration of follow-up were historically recorded in a population-based clinical archive. The viro-logical archive includes data related to patients (i.e. name, age, personal healthcode), and all molecular tests (i.e. genotyping by Polymerase Chain Reaction –PCR) carried out to identify genotype HR- and LR-HPV from each specimen.All personal data (name or personal health code) were replaced by a univocalnumerical code, making both databases anonymous at source (in full compliancewith the Italian Privacy Law – Decree 196, 30/06/2003). The univocal

Address for correspondence:Prof. Francesco Saverio Mennini, CEIS Sanita – Centre

for Health Economics and Management (CHEM),

Faculty of Economics – University of Rome ‘Tor

Vergata’, Via Columbia, 2, 00133 Rome, Italy.

Tel.: þ39 06 72595642; Mob: þ39 333 4991647;

Fax: þ39 06 233 245 53;

f.mennini@uniroma2.it

Key words:Borderline cytology – Cervical cancer screening – CIN

lesions – HR-HPV – LR-HPV

Accepted: 7 December 2010; published online: 12 January 2011

Citation: Curr Med Res Opin 2011; 27:569–78

Current Medical Research & Opinion Volume 27, Number 3 March 2011

570 HR- and LR-HPV infections in borderline Pap tests Costa et al. www.cmrojournal.com ! 2011 Informa UK Ltd

Cur

r M

ed R

es O

pin

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Fran

cis

A C

ount

way

Lib

rary

of

Med

icin

e on

01/

17/1

1Fo

r pe

rson

al u

se o

nly.

numerical code, attributed to all subjects included in theanalysis, made it possible to retrospectively match clinicalpatients’ characteristics with virological data and, finally,to perform a frequency assessment of HPV types amongpatients with borderline cytology. The observational andretrospective study design did not require the collection ofinformed consent from the subjects included (Law Decree196/2003, art. 110).

Patient population

All patients with a confirmed cytological diagnosis ofASCUS have been considered eligible for inclusion.Patients with any other cytological diagnosis wereexcluded from the study. All eligible patients were triagedto diagnostic procedures and/or periodic follow-up visitsaccording to the algorithm reported in Figure 1. Womenunderwent colposcopy within one month after the Papsmear report, according to the regional screening protocol.In each patient a cytological specimen was collected inorder to perform a specific genotyping test to identifyHR- and LR-HPV at the time of colposcopy.

Colposcopic procedures

Conventional colposcopy was performed systematically onall women enrolled by a Zeiss OPM 1F colposcope after a5% acetic acid solution application. Acetowhite abnormalareas were graded as abnormal transformation zone grade 1(AnTZ-G1: minor colposcopic abnormalities), AnTZ-G2(major colposcopic abnormalities) or cancer and assessedby punch biopsy19. Endocervical curettage (ECC) was also

performed when the squamo-columnar junction (SCJ)was not entirely visible both in normal and in abnormalcolposcopies, as elsewhere described20,21.

Molecular genotyping procedures

In the Microbiology Laboratory of the S. Orsola-MalpighiUniversity Hospital in Bologna all cervical samples weremarked only with the study number. One millilitre of sus-pension from the eso/endocervical sample was centrifugedand the cell pellet was extracted according to the instruc-tions of the Qia Amp DNA Mini kit (QIAGEN).Extracted samples were genotyped using the PCR-ELISAassay, according to a methodology previously described22.Briefly, a consensus primer pair MY09/MY11 was used forPCR amplification. The typing of HPV DNA wasmade by different hybridization reactions, using tenbiotinylated oligonucleotide-specific probes able to recog-nize eight high-risk mucosal genotypes (HPV-16, 18, 31,33, 35, 45, 52, and 58) and two low-risk genotypes (HPV-6and 11).

Histological evaluation

The histologic examination was performed according toroutine practice. Histologic interpretation of biopsy speci-mens was conducted at the Pathology Department andclassified as negative, CIN1, 2 or 3, or invasive cancer.The slides were sent to the Pathology Quality ControlGroup for re-evaluation. All histologically confirmedhigh-grade lesions diagnosed as CIN2þ were treated by

ASC-US

Negative

Visible SCJ

Not visibleSCJ

Positive

Punch Biopsy

Screening ECC

Colposcopy

HPV DNA Test

IF

Suspectedinvasive KG1-G2

IF

IF

VisibleSCJ

Not visibleSCJ

Punch Biopsy & ECC

Punch Biopsy

SCJ – squamo-columnar junction. G1 – minor colposcopic abnormalities. G2 – major colposcopic abnormalities. ASCUS patients underwent HPV DNA test and colposcopy. Punch biopsy was performed in any acetowhite areas, and endocervical curettage (ECC) was performed both in negative and positive colposcopic patients when the SCJ was not visible.

Figure 1. Flow diagram of the study for the management of ASCUS.

Current Medical Research & Opinion Volume 27, Number 3 March 2011

! 2011 Informa UK Ltd www.cmrojournal.com HR- and LR-HPV infections in borderline Pap tests Costa et al. 571

Cur

r M

ed R

es O

pin

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Fran

cis

A C

ount

way

Lib

rary

of

Med

icin

e on

01/

17/1

1Fo

r pe

rson

al u

se o

nly.

surgical procedures, while CIN1 were followed up accord-ing to regional screening guidelines.

Statistical analysis

The value of continuous parameters has been reported asmean� standard deviation (SD). In order to reveal signif-icant differences in baseline characteristics among groups,one-way analysis of variance (ANOVA) for continuousvariables was performed. The Pearson’s �2 test was usedfor categorical variables and the Bonferroni post hoc testwas carried out as appropriate. The statistical significanceof differences between averages was calculated using thet-test for independent samples. Missing data were replacedby individual means of valid data. All tests were two-tailed,and a level of a lower than 0.05 was considered to desig-nate the statistical significance. The proportional hazardassumption was examined to investigate the relationshipbetween the rate of HPV positivity and the age variable.Thereafter, the odds ratios and 95% confidence intervalswere calculated, taking an age greater than 40 years asreference category23. Statistical analyses were performedwith SPSS Windows, version 16 (Chicago, IL, USA,2002).

Results

The study population numbered 1047 patients with a con-firmed cytological diagnosis of ASCUS. Patients’ meanage corresponded to 38.4� 9.6 years (range 23–65 years),and the HPV positivity rate was equal to 34.8% (n¼ 364),with 2.1% (227 1047) of patients harbouring LR-HPVtypes, 30.8% (3227 1047) HR-HPV types, whileboth HR- and LR-HPV types were detected in 1.9%(207 1047) of patients.

In Figure 2 is depicted the distribution of evaluatedpatients, whereas in Table 1 demographic characteristicsof both HPV-positive and negative patient groups arereported. The mean age of women with HR/LR-HPVinfection was lower compared with the HPV negativegroup (36.8� 9.4 versus 39.3� 9.7 years) and this differ-ence reached statistical significance (p50.001). Inpatients aged less than 40 years, the HPV positivity ratewas equal to 67.9% in comparison to a rate of 32.1%observed in those over 40 (p50.001). Specifically, inyounger women the risk of being HPV-positive was 1.68-fold greater than older patients (CI 95%: 1.28–2.19;p¼ 0.0001).

As also depicted in Figure 3, the peak of distributioncurve by age of HPV-positive group is substantially higher(line a�) and shifted (line ag) to the left toward the lowerages in comparison to the group of HPV-negative patients.A histologic examination was performed in 765 women,357 (46.7%) with abnormal colposcopy and 408 (53.3%)with normal colposcopic findings and SCJ not entirelyvisible (Table 2). In 282 women with normal colposcopyand SCJ visible, a biopsy was not carried out and thosesubjects were considered as negative for pathology.Overall, 357 (34.1%) women were positive for cervicallesions: 279 (26.6%) with CIN1 and 78 (7.5 %) withCIN2þ (18 CIN2, 59 CIN3 and one invasive cancer).

HPV NegativeN = 683

HPV PositiveN = 364

ASCUSN = 1,047

CIN1N = 155

CIN2N = 3

CIN3N = 1

CIN2N = 15

CIN1N = 124

CIN3N = 58

InvasiveK

N = 1

Nolesion

N = 166

Nolesion

N = 524

Figure 2. Patient distribution according to the study design.

Table 1. Distribution of ASCUS patients by HPV positivity and age.

HPV positive HPV negative p

Number of pts 364 683Mean age in yrs (SD) 36.8 (9.4) 39.3 (9.7) 50.001a

Distribution by age (%)540 yrs 247 (67.9) 383 (55.8)40–65 yrs 117 (32.1) 300 (44.2)Total 364 (100.0) 683 (100.0) 50.001b

at-test for independent samples.b�2 test.

Current Medical Research & Opinion Volume 27, Number 3 March 2011

572 HR- and LR-HPV infections in borderline Pap tests Costa et al. www.cmrojournal.com ! 2011 Informa UK Ltd

Cur

r M

ed R

es O

pin

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Fran

cis

A C

ount

way

Lib

rary

of

Med

icin

e on

01/

17/1

1Fo

r pe

rson

al u

se o

nly.

The mean age of HPV-positive patients with a histologicalabnormality was considerably lesser in comparison to thatof HPV-negative ones (36.2� 9.4 versus 39.5� 9.3 yearsin CIN1, p¼ 0.004; 31.9� 3.4 versus 35.0� 3.5 years inCIN2, p¼ 0.177; and 37.5� 9.0 versus 44.0� 0.0 years inCIN3, p¼ 0.479, respectively; in CIN2þ statistical signif-icance was not achieved due to the reduced sample ofpatients in HPV-negative groups). On the contrary, theage of patients with negative histology was significantlyhigher in HPV-negative women (43.9� 8.8 versus40.8� 9.6 years, p¼ 0.003).

The sensitivity and specificity of HPV DNA testingwere 94.9% and 75.5% for CIN2þ and 98.3% and75.5% for CIN3þ, respectively. Moreover, the PPV andthe negative predictive value (NPV) were 42.6% (CI 95%:35.2–49.9%) and 98.7% (IC 95%: 97.5–99.9%) forCIN2þ and 37.1% (IC 95%: 29.6–44.6%) and 99.7%(IC 95%: 99.0–100.0%) for CIN3þ, respectively.

Within the 683 women of the HPV-negative group,155 CIN1 (22.7%), 3 CIN2 (0.4%) and 1 CIN3 (0.1%)were detected, while in the 364 women of the HPV-posi-tive group 124 CIN1 (34.1%), 15 CIN2 (4.1%) and 59CIN3þ (16.2%) were diagnosed.

A single or a multiple HR-HPV genotype was detectedin 35.8%, 83.3%, and 91.5% of patients with CIN1, CIN2,and CIN3þ lesions respectively, while a single LR-HPVtype was found in 3.9% of patients with CIN1 (Table 3).Whenever combined with HR-HPV genotypes, the overall

frequency of LR-HPV was more than double, reaching avalue of 8.6%.

Among the 124 women with a diagnosis of CIN1 andbelonging to the HPV-positive group, 8.9% harboured LR-HPV genotypes exclusively, 80.6% HR-HPV types and10.5% a combination of HR and LR-HPV. Within the18 women with CIN2 lesions, 16.7% were HPV-negativewhile in the remaining 83.3% only HR-HPV genotypeswere detected. Finally, in the group of 60 women withCIN3þ lesions, only one was HPV-negative, whilst inthe 59 HPV-positive patients (58 CIN3 and one invasivecarcinoma) 93.1% had HR-HPV and in 6.9% both HR-and LR-HPV genotypes were detected. In 690 patientswithout lesions (408 with a negative histologic reportand 282 without biopsy), 524 (75.9%) were HPV-negativeand in the remaining 166, 11 (6.6%), 152 (91.6%), and 3(1.8%) LR-HPV, HR-HPV, and LRþHR genotypes weredetected respectively. Interestingly, the mean age ofpatients (n¼ 13) with CIN1 lesions induced by a com-bined HRþ LR-HPV infection was significantly lower(33.7� 10.4 years) compared with all other HPV-positivegroups (36.4� 9.6 years in HR-HPV group [n¼ 100], and36.6� 5.9 years in LR-HPV group [n¼ 11]; p¼ 0.024).A similar result was observed in patients with CIN3lesions, although statistical significance was not reacheddue to the reduced sample size.

A comprehensive description of the frequency ofHPV genotypes by specific histological diagnosis is

4.0%b

ag

3.0%

2.0%

1.0%

HPV-positive group

HPV-negative group

0.0%

23 25 27 29 31 33 35 37 39 41 43 45 47 49 51 53 55 57 59 61 63 65

Line ab gives the difference in percentage at peak between distribution curves by age forHPV positive and negative patient groups. Line ag provides the difference of agebetween groups in years.

Figure 3. Distribution of HPV positive and negative patient groups by age.

Current Medical Research & Opinion Volume 27, Number 3 March 2011

! 2011 Informa UK Ltd www.cmrojournal.com HR- and LR-HPV infections in borderline Pap tests Costa et al. 573

Cur

r M

ed R

es O

pin

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Fran

cis

A C

ount

way

Lib

rary

of

Med

icin

e on

01/

17/1

1Fo

r pe

rson

al u

se o

nly.

reported in Table 4. Regardless of histological diagnosis,HPV-16 was the most frequent genotype detected in HPV-positive patients with cervical lesions (29.0% in CIN1,53.3% in CIN2, and 61.7% in CIN3þ) and in those with-out lesions (27.1%). In women with CIN2 and CIN3lesions, the other more frequent HR-HPV genotypeswere HPV-31 (which accounted for 20.0% and 17.2% ofCIN2 and CIN3), followed by HPV-33 (detected in 20.0%and 12.1% of CIN2 and CIN3 HPV-positive lesions,respectively). The percentage of women with HPV-18increased from 6.7% in CIN2 to 12.1% in CIN3.Multiple HR-HPV infections were detected in 27.5% ofHR-HPV positive samples.

Within CIN1 patients, HPV-52, HPV-31, HPV-18,and HPV-58 were the most frequent genotypes recorded

beyond HPV-16 with rates corresponding to 21.0%,19.4%, 11.3%, and 11.3% respectively. HPV-6 and 11accounted for 19.4% of all HPV-positive CIN1 lesions,though 10.4% was detected in a combination with HRgenotype. Finally, HPV-6 and 11 were also detected inpatients with CIN3 lesions, invariably in a combinedinfection with HR-HPV.

Discussion

Cervical screening programmes prevent invasive cancer,thus women with abnormal cytology can be identified andreferred for further examinations and therapy when ahistologically precancerous cervical lesion is confirmed.

Table 2. ASCUS patient population distributed according to their histologic pattern and HPV positivity.

HPV� HPVþ Total p

Positive HistologyCIN1

Number of pts 155 124 279Mean age in yrs (SD) 39.5 (9.3) 36.2 (9.4) 38.2 (9.5) 0.004a

Distribution by age (%)540 yrs 81 (52.3) 88 (71.0) 169 (60.6)40–65 yrs 74 (47.7) 36 (29.0) 110 (39.4)Total 155 (100.0) 124 (100.0) 279 (100.0) 0.002b

CIN2Number of pts 3 15 18Mean age in yrs (SD) 35.0 (3.5) 31.9 (3.4) 32.4 (3.5) 0.177a

Distribution by age (%)540 yrs 3 (100.0) 15 (100.0) 18 (100.0)40–65 yrs 0 (0.0) 0 (0.0) 0 (0.0)Total 3 (100.0) 15 (100.0) 18 (100.0) NP

CIN3Number of pts 1 58 59Mean age in yrs (SD) 44.0 (0.0) 37.5 (9.0) 37.6 (9.0) 0.479a

Distribution by age (%)540 yrs 0 (0.0) 38 (65.5) 38 (64.4)40–65 yrs 1 (100.0) 20 (34.5) 21 (35.6)Total 1 (100.0) 58 (100.0) 59 (100.0) 0.356b

ICCNumber of pts – 1 1Mean age in yrs (SD) – 33.0 (0.0) 33.0 (0.0) NP

Negative HistologyNegative Report

Number of pts 308 100 408Mean age in yrs (SD) 43.9 (8.8) 40.8 (9.6) 43.2 (9.1) 0.003a

Distribution by age (%)540 yrs 110 (35.7) 52 (52.0) 162 (39.7)40–65 yrs 198 (64.3) 48 (48.0) 246 (60.3)Total 308 (100.0) 100 (100.0) 408 (100.0) 0.005b

Test Not PerformedNumber of pts 216 66 282Mean age in yrs (SD) 32.5 (6.7) 32.6 (8.1) 32.5 (7.1) 0.926a

Distribution by age (%)540 yrs 189 (87.5) 53 (80.3) 242 (85.8)40–65 yrs 27 (12.5) 13 (19.7) 40 (14.2)Total 216 (100.0) 66 (100.0) 282 (100.0) 0.105b

at-test for independent samples.b�2 test.ICC – suspected invasive cervical cancer; NP – statistical test not performed.

Current Medical Research & Opinion Volume 27, Number 3 March 2011

574 HR- and LR-HPV infections in borderline Pap tests Costa et al. www.cmrojournal.com ! 2011 Informa UK Ltd

Cur

r M

ed R

es O

pin

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Fran

cis

A C

ount

way

Lib

rary

of

Med

icin

e on

01/

17/1

1Fo

r pe

rson

al u

se o

nly.

An ASCUS report represents a condition where thepathologic significance of the abnormal cells in thesmear cannot be properly defined. This heterogeneouscategory including benign, reactive and premalignant cel-lular changes, accounts for 5–10% of all cytologicaldiagnoses1.

The PATER study was conducted to analyse the pres-ence of cervical lesions (CIN1þ) and HPV positivity(both HR- and LR-HPV) in ASCUS Pap smears derivedfrom a population-based screening. In our study, the over-all rate of HR-HPV corresponded to 32.7% (3427 1047),a value lower compared with that reported in the currentliterature24. Actually, in the ALTS study 50.6% (CI 95%:47.6–53.6%) of ASCUS was positive for HR-HPV geno-types5; however, this study was conducted on a younger

population (aged 29 on average), in which the prevalenceof HR-HPV infection is generally higher and the sponta-neous regression has already been well documented25.Moreover, comparing the histological diagnoses of ourseries (1047 patients) with the ‘colposcopy arm’ of theALTS study (1163 patients)26, the rate of CIN3þ caseswas similar (5.6% for both studies), although the propor-tion of CIN1 and CIN2 in PATER study was respectivelyhigher and lower: 26.6% vs. 14.5% and 1.7% vs. 6.0%.These variations probably reflect different characteristicsof populations included in the two studies. In the ALTSstudy, HPV DNA testing detected 96.3% of the womenwith CIN3þ 5,27. In the PATER study we observed thatthe HPV test identified 98.3% and 94.9% of women diag-nosed with CIN3þ and CIN2þ respectively.

Table 3. Frequency of HR- and LR-HPV genotypes by histologic pattern.

HR-HPV LR-HPV LR-HPVþ HR-HPV No HPV Total p

Positive HistologyCIN1

Number of pts 100 11 13 155 279Mean age in yrs (SD) 36.4 (9.6) 36.6 (5.9) 33.7 (10.4) 39.5 (9.2) 38.0 (9.4) 0.024a

Distribution by age (%)540 yrs 70 (70.0) 8 (72.7) 10 (76.9) 81 (52.3) 169 (60.6)40–65 yrs 30 (30.0) 3 (27.3) 3 (23.1) 74 (47.7) 110 (39.4)Total 100 (100.0) 11 (100.0) 13 (100.0) 155 (100.0) 279 (100.0) 0.016b

CIN2Number of pts 15 – – 3 18Mean age in yrs (SD) 31.9 (3.4) – – 35.0 (3.5) 32.4 (3.5) 0.177a

Distribution by age (%)540 yrs 15 (100.0) – – 3 (100.0) 18 (100.0)40–65 yrs – – – – –Total 15 (100.0) – – 3 (100.0) 18 (100.0) NP

CIN3Number of pts 54 – 4 1 59Mean age in yrs (SD) 37.9 (9.0) – 32.5 (8.6) 44.0 (0.0) 37.6 (9.0) 0.406a

Distribution by age (%)540 yrs 35 (64.8) – 3 (75.0) – 38 (64.4)40–65 yrs 19 (35.2) – 1 (25.0) 1 (100.0) 21 (35.6)Total 54 (100.0) – 4 (100.0) 1 (100.0) 59 (100.0) 0.366b

ICCNumber of pts 1 – – – 1Mean age in yrs (SD) 33.0 (0.0) 33.0 (0.0) NP

Negative HistologyNegative Report

Number of pts 91 7 2 308 408Mean age in yrs (SD) 40.8 (9.2) 38.4 (14.7) 51.5 (3.5) 43.9 (8.8) 43.2 (9.1) 0.008a

Distribution by age (%)540 yrs 48 (52.7) 4 (57.1) 0 (0.0) 110 (35.7) 162 (39.7)40–65 yrs 43 (47.3) 3 (42.9) 2 (100.0) 198 (64.3) 246 (60.3)Total 91 (100.0) 7 (100.0) 2 (100.0) 308 (100.0) 408 (100.0) 0.013b

Test Not PerformedNumber of pts 61 4 1 216 282Mean age in yrs (SD) 32.9 (8.3) 28.0 (4.8) 31.0 (0.0) 32.5 (6.7) 32.5 (7.1) 0.599a

Distribution by age (%)540 yrs 48 (78.7) 4 (100.0) 1 (100.0) 189 (87.5) 242 (85.8)40–65 yrs 13 (21.3) – – 27 (12.5) 40 (14.2)Total 61 (100.0) 4 (100.0) 1 (100.0) 216 (100.0) 282 (100.0) 0.275b

aANOVA (one way) and Bonferroni post hoc test.b�2 test.ICC – suspected invasive cervical cancer; NP – statistical test not performed.

Current Medical Research & Opinion Volume 27, Number 3 March 2011

! 2011 Informa UK Ltd www.cmrojournal.com HR- and LR-HPV infections in borderline Pap tests Costa et al. 575

Cur

r M

ed R

es O

pin

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Fran

cis

A C

ount

way

Lib

rary

of

Med

icin

e on

01/

17/1

1Fo

r pe

rson

al u

se o

nly.

We found an overall HPV positivity rate (HR and/orLR) equivalent to 34.8% (HR-HPV genotypes accountingfor 30.8%, LR-HPV for 2.1%, and both LR- and HR-HPVtypes for 1.9%). This HPV positivity rate was lower thanthat recently reported by Kjaer et al.28, who found that82.0% of patients were positive for LR- and/or HR-HPVgenotypes (3% with LR-HPV types, 56.9% with HR-HPV,and 22.1% with both LR- and HR-HPV). The Kjaer et al.study, however, was performed considering a cluster ofASCUSþ LSIL (low-grade squamous intraepitheliallesions) cytological diagnosis. In agreement with datafrom Kjaer et al., in our study women with CIN1 had evi-dence of infection with HR- as well as LR-HPV genotypes,which is consistent with the fact that low-grade neoplasiacan be caused both by LR and HR types and, as expected,none of the women with CIN2 or CIN3þ had an infectionwith LR-HPV genotypes alone. The frequency ofHPV infection increased with the severity of the histolog-ical diagnosis (reaching 98.3% in CIN3þ). This is consis-tent with the large body of evidence that infection withHPV is a necessary cause of invasive cervical cancerworldwide29.

In women with ASCUS the youngest age played a crit-ical and significant role, suggesting that an age less than 40years is associated with a higher probability of having notonly a HPV infection but also a histological abnormality.Statistical significance was not always achieved because ofthe reduced size or missing samples in the comparisongroup. Consequently, the mean age of patients with nega-tive histology is likely to be higher in general population.HPV-16 was the most frequent genotype in all histologicaldiagnoses and the percentage of women with HPV-16increased from CIN1 (29%) to CIN3þ (61.7%) as reportedby Kjaer et al. previously28. Significantly less frequent thanHPV-16 were HPV-31 (16.2%), 58 (10.4%), 33 (9.5%),52 (9.5%) and 18 (7.9%) as reported by others28,30. Thedifferences in the frequency of individual HPV typesobserved in our study compared to others may be causedby the different sensitivity for HPV types in utilized tests31.

Although representing a reference, the ALTS study leftseveral relevant issues in screening unaddressed. Actually,studies focused on the frequency of possibly (LR-HPV) andcertainly oncogenic viral genotypes (HR-HPV) in border-line cytology are few and contradictory32. On the contrary,

Table 4. Frequency of HPV genotypes by histological diagnosis.

HPV Genotypes CIN1 CIN2 CIN3 ICC NR TNP Total

HPV-6Number of ptsa 18 – 3 – 7 4 32Rate on histology (%) 14.5 – 5.2 – 7.0 6.1 7.4

HPV-11Number of ptsa 6 – 1 – 2 1 10Rate on histology (%) 4.8 – 1.7 – 2.0 1.5 2.3

HPV-16Number of ptsa 36 8 36 1 32 13 126Rate on histology (%) 29.0 53.3 62.1 100.0 32.0 19.7 29.2

HPV-18Number of ptsa 14 1 7 – 6 6 34Rate on histology (%) 11.3 6.7 12.1 – 6.0 9.1 7.9

HPV-31Number of ptsa 24 3 10 – 18 15 70Rate on histology (%) 19.4 20.0 17.2 – 18.0 22.7 16.2

HPV-33Number of ptsa 11 3 7 – 13 7 41Rate on histology (%) 8.9 20.0 12.1 – 13.0 10.6 9.5

HPV-35Number of ptsa 4 1 3 – 6 4 18Rate on histology (%) 3.2 6.7 5.2 – 6.0 6.1 4.2

HPV-45Number of ptsa 7 1 2 – 1 3 14Rate on histology (%) 5.6 6.7 3.4 – 1.0 4.5 3.2

HPV-52Number of ptsa 26 1 4 – 5 5 41Rate on histology (%) 21.0 6.7 6.9 – 5.0 7.6 9.5

HPV-58Number of ptsa 14 2 3 – 14 12 45Rate on histology (%) 11.3 13.3 5.2 – 14.0 18.2 10.4

aThe total number of patients may exceed 364 because some of them (n¼ 60) could have more than one genotype at the same time.ICC – invasive cervical cancer; NR – negative report; TNP – test not performed.

Current Medical Research & Opinion Volume 27, Number 3 March 2011

576 HR- and LR-HPV infections in borderline Pap tests Costa et al. www.cmrojournal.com ! 2011 Informa UK Ltd

Cur

r M

ed R

es O

pin

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Fran

cis

A C

ount

way

Lib

rary

of

Med

icin

e on

01/

17/1

1Fo

r pe

rson

al u

se o

nly.

the PATER study provided probably one of the first signif-icant observations concerning the following topics:(1) the assessment of frequencies of HR-HPV types in

women who have been diagnosed as ASCUS with(52.4%, 1877 357) and without (22.5%, 1557690) histological abnormalities aged 23–65, repre-senting the target population of cervical cancerscreening programmes;

(2) the high percentage of borderline cytology which isnegative at the subsequent histological examination(65.9%, 6907 1047);

(3) data on the frequency of LR-HPV (HPV types 6 and11) infections producing mild cervical precancerouslesions (19.4% of CIN1), which are likely to regressor cause the development of benign epithelial prolif-erations such as genital warts.

The appearance of koilocytosis (which is used bypathologists as a hallmark of HPV-infected epithelialcells in Pap smears and in histology) is observed in bothLR- and HR-HPV infections. Recently many authorsobserved that the E5 and E6 proteins from both LR- andHR-HPV cooperate to induce koilocyte formation inhuman cervical cells33,34. Koilocytosis is included amongCIN1 diagnostic parameters, therefore an overestimationof CIN1, induced by LR-HPV infections, can be translateinto unnecessary costs, actually burdening the screeningprogramme.

Population-based data from our study should be deemedas relevant not only to accurately assess the frequency ofHR- and LR-HPV genotypes in women with ASCUS, butalso to channel the best possible use of resources in screen-ing. The high number of women with borderline cytologycan lead to a significant increase in total expenditure asso-ciated either with healthcare provided (direct costs) orwith its socioeconomic impact (indirect costs induced byillness-related absenteeism, workplace productivity loss orpresenteeism, out-of-pocket costs, and so on) for eachhigh-grade lesion detected35–36. In a recent publication itwas reported that with the tetravalent vaccination a reduc-tion of direct costs associated with the prevention ofabnormal pap smears, colposcopies, LSIL, and ASCUSinduced by HPV in Italy would be approximately E43 mil-lion37. Depending on the percentage of LR-HPV-inducedASCUS inputted, roughly 16–21% of this amount is gen-erated by HPV-6 and 11, excluding any expense caused bygenital condylomas37.

Some limitations of this study have to be examined.First, in spite of significant benefits derived from the cer-vical cancer screening, some Italian regions have not yetfully implemented their secondary prevention programmesof cervical cancer. As a consequence, an excess of varia-tion concerning compliance with cervical cancer screen-ing can be found across Italy and approximately 30% ofItalian women has never been screened in their life38.Nevertheless, the Emilia-Romagna is one of those Italian

regions where cervical screening is acknowledged as a ref-erence of a well-implemented organizational model of pre-ventive medicine. Second, since the clinical archive wasnot structured to track each patient longitudinally afterthe completion of healthcare interventions performed inour centre, a complete patient’s follow-up was not avail-able. However, given the algorithm used in this study, it isvery likely that patients may have returned to our centre atthe appearance of any new cervical lesion or in case ofanother positive Pap test.

In conclusion, the findings of our study can drive theidentification of new solutions improving the managementof the borderline Pap test. In this context, a tetravalent-based HPV vaccination alongside the cervical cancerscreening programme would provide substantial clinical,organizational, and economic benefits specifically associ-ated with the significant reduction of management costsof borderline cytology induced by HR-HPV as well asLR-HPV (6 and 11) genotypes. The accurate evaluationof the frequency of HR- and LR-HPV infections in a pop-ulation with ASCUS aged 23–65 is useful both for cervicalcancer screening and a HPV vaccination strategy whichmust be implemented in a synergistic combination withscreening programmes. Since the prevalence of ASCUSand low-grade squamous intraepithelial lesions largelyexceed that of high-grade squamous intraepithelial lesions(from 5–10 and 2–5 fold for ASCUS and LSIL, respec-tively), vaccination should permit in the short tomedium term a significant reduction of borderline cytol-ogy, in addition to high-grade lesions decline. This couldallow a reduction of healthcare and social costs and a sig-nificant optimization of diagnostic and treatment proce-dures32,39. Moreover, this considerable economic reservesaved could be used for the management of other publichealth priorities.

TransparencyDeclaration of fundingThe study was supported with unrestricted funding from SanofiPasteur MSD, Italy.

Declaration of financial/other relationshipsS.C., S.V., F.S.M., A.M., M.P., E.L., A.F., E.M., M.C., D.S., M.Z.and G.P. have disclosed that they have no significant relation-ships with or financial interests in any commercial companiesrelated to this study or article.

The CMRO peer reviewers have disclosed that they have nosignificant relationships with or financial interests in any com-mercial companies related to this study or article.

AcknowledgementsThe authors wish to acknowledge the support of the OrganizedScreening Programme Service of Bologna Health Unit whichprovided screening data for carrying out the assessment.

Current Medical Research & Opinion Volume 27, Number 3 March 2011

! 2011 Informa UK Ltd www.cmrojournal.com HR- and LR-HPV infections in borderline Pap tests Costa et al. 577

Cur

r M

ed R

es O

pin

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Fran

cis

A C

ount

way

Lib

rary

of

Med

icin

e on

01/

17/1

1Fo

r pe

rson

al u

se o

nly.

These date were previously presented as a poster at the 16thInternational meeting of the European Society of GynecologicalOncology (ESGO) Belgrade, Serbia, October 11–14, 2009. CostaS., Marcellusi A., Mennini FS., Venturoli S. ‘Population-basedfrequency Assessment of HPV infections in borderline Pap Testsin the Emilia-Romagna Region – the PATER study’.

References1. Giorgi Rossi P, Ricciardi A, Cohet C, et al. Epidemiology and costs of cervical

cancer screening and cervical dysplasia in Italy. BMC Public Health 2009 9:71

2. Coleman D, Day N, Douglas G, et al. European guidelines for quality assur-

ance in cervical cancer screening. Eur J Cancer 1993;29A(Suppl. 4):S1-38

3. Kurman R, Solomon D, Nayar R, et al. The Bethesda System for Reporting

Cervical Cytology: Definitions, Criteria, and Explanatory Notes. New York:

Springer, 2003

4. Cocchi V, Sintoni C, Carretti D, et al. External quality assurance in cervical/

vaginal cytology: interlaboratory agreement in the Emilia-Romagna region of

Italy. Acta Cytol 1996;40:480-8

5. Solomon D, Schiffman M, Tarone R (for the ALTS study group). Comparison of

three management strategies for patients with atypical squamous cells of

undetermined significance: baseline results from a randomized trial. J Natl

Cancer Inst 2001;93:293-9

6. Sherman ME, Schiffman MH, Lorincz AT, et al. Toward objective quality

assurance in cervical cytopathology. Correlation of cytopathologic diagnoses

with detection of high-risk human papillomavirus types. Am J Clin Pathol

1994;102:182-7

7. Ronco G, Giorgi Rossi P. New paradigms in cervical cancer prevention: oppor-

tunities and risks. BMC Women’s Health 2008;8:23

8. Ronco G, Giorgi-Rossi P, Carozzi F, et al. Efficacy of human papillomavirus

testing for the detection of invasive cervical cancers and cervical intraepithe-

lial neoplasia: a randomised controlled trial. The Lancet Oncology 2010;

11:249-57

9. Arbyn M, Martin-Hirsch P, Buntinx F, et al. Triage of women with equivocal or

low-grade cervical cytology results: a meta-analysis of the HPV test positivity

rate. J Cell Mol Med 2009;13:648-59

10. Lynge E, Antilla A, Arbyn M, et al. What’s next? Perspectives and future needs

of cervical screening in Europe in the era of molecular testing and vaccination.

Eur J Cancer 2009;45:2714-21

11. Arbyn M, Buntinx F, van Ranst M, et al. Virologic versus cytologic triage of

women with equivocal Pap smears: a meta-analysis of the accuracy to detect

high-grade intraepithelial neoplasm. J Natl Cancer Inst 2004;96:280-93

12. Wright Jr TC, Lorincz A, Ferris DG, et al. Reflex human papillomavirus deoxy-

ribonucleic acid testing in women with abnormal Papanicolaou smears. Am J

Obstet Gynecol 1998;178:962-6

13. Szarewski A, Ambroisine L, Cadman L, et al. Comparison of predictors for

high-grade cervical intraepithelial neoplasia in women with abnormal smears.

Cancer Epidemiol Biomarkers Prev 2008;17:3033-42

14. Ronco G, Cuzick J, Segnan N, et al. HPV triage for low grade (L-SIL) cytology is

appropriate for women over 35 in mass cervical cancer screening using liquid

based cytology. Eur J Cancer 2007;43:476-80

15. Ronco G, Giubilato P, Naldoni C, et al. Extension of organised cervical cancer

screening programmes in Italy and their process indicators. Epidemiol Prev

2008; 32(2 Suppl. 1):37-54

16. Russo P, Capone A, Attanasio E, et al. Pharmacoutilization and costs of

osteoarthritis: changes induced by the introduction of a cyclooxygenase-2

inhibitor into clinical practice. Rheumatology 2003;42:879-87

17. Degli Esposti L, Di Martino M, Saragoni S, et al. Pharmacoutilization of statin

therapy after acute myocardial infarction. A real practice analysis based on

administrative data. Ital Heart J 2004;5:120-6

18. Favato G, Mariani P, Mills RW, et al. ASSET (Age/Sex Standardised Estimates

of Treatment): A research model to improve the governance of prescribing

funds in Italy. PLoS ONE 2007;2:e592. Available at: http://www.plosone.org/

article/info%3Adoi%2F10.1371%2Fjournal.pone.0000592 [Last accessed

March 2010]

19. Massad LS, Jeronimo J, Katki HA, et al. The accuracy of colposcopic grading

for detection of high-grade cervical intraepithelial neoplasia. J Low Genit Tract

Dis 2009;13:137-44

20. Costa S, Nuzzo MD, Rubino A, et al. Independent determinants of inaccuracy

of colposcopically directed punch biopsy of the cervix. Gynecol Oncol

2003;90:57-63

21. Ferris DG, Litaker MS; for the ALTS Group. Prediction of cervical histologic

results using an abbreviated Reid Colposcopic Index during ALTS. Am J

Obstet Gynecol 2006;194:704-10

22. Musiani M, Venturoli S, Gallinella G, et al. Qualitative PCR-ELISA protocol for

the detection and typing of viral genomes. Nat Protoc 2007;2:2502-10

23. Scott DR, Hagmar B, Maddox P, et al. Use of human papillomavirus DNA

testing to compare equivocal cervical cytologic interpretations in the United

States, Scandinavia, and the United Kingdom. Cancer 2002;96:14-20

24. The Atypical Squamous Cells of Undetermined Significance/Low-Grade

Squamous Intraepithelial Lesions Triage Study (ALTS) Group. Human papil-

lomavirus testing for triage of women with cytologic evidence of low-grade

squamous intraepithelial lesions: data from a randomized trial. JNCI 2000;

92:397-402

25. Adelstein AM, Husain OAN, Spriggs AI. Cancer of the cervix and screening.

BMJ 1981;282:564

26. Solomon D, Schiffman M, Tarone R. ASCUS LSIL Triage Study (ALTS) con-

clusions reaffirmed: response to a November 2001 commentary. Obstet

Gynecol 2002;99:671-4

27. ASCUS-LSIL Triage Study (ALTS) Group. A randomized trial on the manage-

ment of low-grade squamous intraepithelial lesion cytology interpretations.

Am J Obstet Gynecol 2003;188:1393-400

28. Kjaer SK, Breugelmans G, Munk C, et al. Population-based prevalence,

type- and age-specific distribution of HPV in women before introduction of

an HPV-vaccination program in Denmark. Int J Cancer 2008;123:1864-70

29. Walboomers JM, Jacobs MV, Manos MM, et al. Human papillomavirus is

a necessary cause of invasive cervical cancer worldwide. J Pathol 1999;

1:12-19

30. Clifford GM, Smith JS, Aguado T, et al. Comparison of HPV type distribution in

high-grade cervical lesions and cervical cancer: a meta-analysis. Br J Cancer

2003;89:101-5

31. Klug S, Molijn A, Schopp B, et al. Comparison of the performance of different

HPV genotyping methods for detection of genital HPV types. J Med Virol 2008;

80:1264-74

32. Evans M, Adamson C, Papillo JL. Distribution of human papillomavirus types

in Thin prep Papanicolaou tests classified according to the Bethesda 2001

terminology and correlations with patient age and biopsy outcomes. Cancer

2006;106:1054-64

33. Gupta S, Takhar PP, Degenkolbe R, et al. The human papillomavirus type 11

and 16 E6 proteins modulate the cell-cycle regulator and transcription cofac-

tor TRIP-Br1. Virology 2003;317:155-64

34. Krawczyk E, Suprynowicz FA, Liu X, et al. Koilocytosis: a cooperative inter-

action between the human papillomavirus E5 and E6 oncoproteins. Am J

Pathol 2008;173:682-8

35. Confortini M, Carozzi F, Dalla Palma P, et al. Interlaboratory reproducibility of

atypical squamous cells of undetermined significance report: a national

survey. Cytopathology 2003;14:263-8

36. Confortini M, Montanari G, Prandi S. GISCi. Raccomandazioni per il controllo

di qualita in citologia cervico vaginale. Epidemiol Prev 2004;28:1-16

37. Mennini FS, Costa S, Favato G, et al. Anti-HPV vaccination: A review of recent

economic data for Italy. Vaccine 2009;27:A54-A61

38. Istituto nazionale di statistica (ISTAT) – Fattori di rischio e tutela della salute –

Indagine multiscopo sulle famiglie italiane ‘Prevenzione dei tumori

femminili: ricorso a Pap test e mammografia’ anni 2004/2005. Available

at: http://www.istat.it/salastampa/comunicati/non_calendario/20061204_

00/2006;2006 [Last accessed March 2010]

39. Gage JC, Hanson VW, Abbey K, et al. Number of cervical biopsies and sen-

sitivity of colposcopy. Obstet Gynecol 2006;108:264-72

Current Medical Research & Opinion Volume 27, Number 3 March 2011

578 HR- and LR-HPV infections in borderline Pap tests Costa et al. www.cmrojournal.com ! 2011 Informa UK Ltd

Cur

r M

ed R

es O

pin

Dow

nloa

ded

from

info

rmah

ealth

care

.com

by

Fran

cis

A C

ount

way

Lib

rary

of

Med

icin

e on

01/

17/1

1Fo

r pe

rson

al u

se o

nly.