University Hospital for Infectious Diseases, Zagreb,Croatia1
University Hospital for Gynecology and Obstetrics, Zagreb2
Private Clinic for Gynecology and Obstetrics Harni, Zagreb3

Authors Affiliations: Diagnostic Department (Drs.Adriana Vince and Tatjana Jeren)
STD Outpatient Department (Dr.Davorin Skalko)
Department for Gynecology (Drs.Marina Ivanišević, Dr.Vesna Harni)

This study was not supported by any grant.

Corresponding author: Dr.Adriana Vince, MD, PhD, Diagnostic Department, University Hospital for Infectious Disease, 10 000 Zagreb, Mirogojska 8, CROATIA

Fax ++385 1 4603 131
avince@fran.bfm.hr

SUMMARY

Background: Human papillomavirus (HPV) has been shown to be the major risk factor for the development of cervical carcinoma, the second most common cancer among woman worldwide. Cervical cytology has been the main screening tool for detection of premalignant lesions in last 50 years.

Objective: to evaluate the accuracy of cervical cytology regarding HPV detection in reactive and mild to moderate dysplastic changes that include cellular pattern suggestive of HPV presence.

Patients and Methods: 466 woman aged 17 - 58 years whose cervical smears indicated HPV presence with or without signs of mild and moderate dysplasia were included.

Group I (44 patiens) had only reactive changes + HPV,

group II (250 patiens) CIN 1 + HPV,

group III (172 patiens) had CIN 2 + HPV cytological findings.

In all of the patients additional cervical swabs were obtained and proceeded for HPV molecular detection using the Digene Hybrid capture II.

Results: HPV was detected in 289/466 (62%) patients. The high-risk HPV was present in 263 (56,4%) and the low-risk type in 26 (5,5%) of patients. The detection rate was 25% in patients with reactive changes, 55% in patients with CIN 1, and 81% in CIN 2 group.

Conclusion: the study shows the limitations of cytology regarding HPV detection on a rather large patient´s group. Molecular detection of HPV should obligatory be used as an adjunct to cytology in patients whose PAP smears show only mild dysplastic changes in order to identify the patiens who are really at risk for the development of high-grade cervical premalignant lesions.

Key words: human papillomavirus, cervical intraepithelial neoplasia, cytological screening

INTRODUCTION

Human papillomavirus (HPV) is the major risk factor for the development of cervical carcinoma worldwide (zur Hausen, 1994). It has been shown in a recent study on the large histology series that oncogenic HPV types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 66, 68) are present in up to 99,7% of cervical carcinomas (Walboomers et al, 1999).

Incidence of cervical carcinoma in the Western European countries as well in USA is below 10/100 000 women, but in most Middle and Eastern european countries is still above 20/1000 000 (Parkin et al, 1992). Papanicolau (PAP) cervical screening has helped reduce cervical cancer rate dramatically since its implementation in 1950s. Pap test reporting classifications have evolved and been refined, different countries using their own classification, but most of the cytologists report their findings according to the current Bethesda system (Revised Bethesda System, 1991). this system allows the cytological differentiation between benign lesions, low squamous intraepithelial lesion (LSIL), high squamous intraepithelial lesions (HSIL) and squamous carcinoma (Kurman et al, 1994).

In developed countries however, cytology screening is failing to further reduce the incidence of cervical carcinoma, because the method has its limitations regarding sensitivity (Koss LG, 1989). On the other hand there is a trend towards overdiagnosing of cellular changes indicative of HPV infection which leads to the overtreatment of those patiens (Richart et al, 1993; Flanelly et al, 1994).

The availability of HPV commercial molecular tests (DIGENE; Hybrid capture II test) seems to be a promising adjunct to cytology in follow up of patients with squamous intraepitehelial lesion: CIN. Its sensitivity and specificity has already been tested in several studies with encouraging results for routine use in the diagnostic molecular laboratories (Clavel et al, 1999; Poljak et al, 1999).

The objective of our work was to shown on a large series of cytologically screened women the utility of HPV molecular detection in minor and moderate cervical abnormalities.

PATIENS and METHODS

466 women aged 17 - 58 years (means 31) who attented the STD Clinic of the University Hospital of infectious Diseases in the period from the beginning of 1998 till the end of 1999 for the routine cervical examination and whose PAP test results indicated HPV presence followed by reactive changes or mild to moderate dysplasia reported according to Bethesda as CIN 1 or CIN 2 (cervical intraepithelial neoplasia) were enrolled in the study.

The cytology smears were examined by two senior cytologists. The PAP test findings are summarized in table I.

Table I. Distribution of patients according to the cytology findings

Table II shows the distribution of the patiens according to the age and PAP findings. Most of our patients were in the third and fourth decade of their lives.

Table II. Distribution of patiens according to age and various cytology findings

To all of the patients the possibility of molecular HPV detection was explained, and they agreed for the additional cervical swabs to be taken.

In the period from 2 - 4 weeks from the PAP test examnitaion the second cervical swabs were obtained from each patient and proceeded for the Hybrid capture II HPV testing according to the instructions of the manufacturer (Digene, Silver Spring, MD, USA). The test is based on DNA/RNA hybridization, followed by signal amplification achieved by numerous monoclonal antibodies and alkine phosphatase molecules. It detects 18 types of HPV and can differentiate between two HPV groups: low-risk HPV types 6,11, 42, 43, 44 and high-risk HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, 68 (Poljak et al, 1999).

RESULTS

Overall HPV was detected in 289 (62%) of patients. It was the high risk type in 263 (56,4%) and low risk type in 26 (5,5%) patients (Table III).

Table III. HPV detection in various cytology groups

The detection rate was highest in cytology with moderate dysplasia (81%), while the group with reactive changes and HPV suspicion showed a low rate of molecular detection (25%).

HPV detection in various age group is shown in table IV. The highet rate of HPV detection was noticed in the age group from 30 - 39 years, while interestingly most cytologically false positives were seen in the age group from 40 - 49 years, where even 51% of patiens werw HPV negative, despite the cytological suspicion.

Table IV. Detection of HPV in according to the age

DISCUSSION

Cytological screening has been very helpful in reducing the cervical carcinoma incidence for the last 50 years (Koss, 1989). Yet clinicians as well as cytologists have to be aware of its limitations. Cytology is a subjective method and cellular chages usually associated to HPV infection (koilocytosis, dyskeratosis, hyperkeratosis) are not specific enough so they can easily be mistaken with nonspecific reactive or inflammatory chages (Kurman et al, 1994). Also similar 'artifacts' can occur during inadequate processing steps. Our results have shown on a rather large number of patiens that the accuracy of cytology regarding HPV detection in the abscense of dysplasia is very low, and also CIN 1 changes harbor the HPV only in 55% of patiens. Similar observations were already reported in the studies of Matsurua et al, 1996; Ferris et al, 1998).

Of course, HPV is known to cause a reversible infection, so the discordance between cytology and molecular detection can partly be explained by self-clearence of the virus during the interval between Pap test and molecular testing. It is also interesting to notice that the rate of HPV molecular detection was lowest in the patients in the patiens in the fifth decade of life, maybe it could be connected to the beginning of hormonal misbalance and its influence on cell morphology that can sometimes mimic the HPV indicating cellular changes.

However the accuracy of cytology was rather high when the signs of moderate dysplasia (CIN 2) were present.

We have to conclude that HPV molecular testing should obligatory be used in patiens whose PAP test indicate HPV presence followed only by reactive or mild dysplastic changes (CIN 1) in order to identify patients who are really at risk for the development of high-grade premalignant lesions. By the combining of both methods, at least in 50% of patients unnecessary stress and intensive follow-up can be avoided.