Adjunctive human papillomavirus testing in the 2-year follow-up of women with low-grade cervical cytologic abnormalities: A randomized trial and economic evaluation

Archives of Pathology & Laboratory Medicine, Sep 2003 by Lytwyn, Alice, Sellors, John W, Mahony, James B, Daya, Dean, Et al

* Context.-Although human papillomavirus (HPV) testing may aid in managing low-grade abnormality on screening cervical cytology, patient compliance with repeat testing programs requires consideration.

Objectives.-To determine effectiveness and costs of repeated Papanicolaou (Pap) test and oncogenic HPV testing for detecting cervical intraepithelial neoplasia 2 or 3.

Design.-We conducted a randomized controlled trial of combined Pap test and cervical HPV testing by Hybrid Capture 1 test compared with Pap test alone; tests were performed every 6 months for up to 2 years. The study end point was colposcopic examination performed on all women at 2 years, or earlier if an HPV test was positive or if a Pap test showed high-grade squamous intraepithelial lesion.

Setting.-Sixty-six community family practices.

Participants.-Two hundred fifty-seven women with atypical squamous cells of undetermined significance or low-grade squamous intraepithelial lesion on screening cervical cytology.

Main Outcome Measures.-Detection of histologically confirmed cervical intraepithelial neoplasia 2 or 3, fully allocated costs, and loss to follow-up.

Results.-Combined Pap test and HPV testing detected 11 (100%) of 11 cases of cervical intraepithelial neoplasia 2/3, whereas Pap test alone detected 7 (63.6%) of these 11 cases (P = .14); corresponding specificities were 39 (46.4%) of 84 and 45 (71.4%) of 63 (P = .005). The cost-effectiveness ratio was Can $4456 per additional case of high-grade cervical intraepithelial neoplasia. Sixty-nine (26.8%) of the 257 women (24.6% combined group vs 29.1% Pap test only group, P = .41) defaulted from testing or from colposcopy when referred with an abnormal result.

Conclusions.-Combined testing was more costly but may detect more cases of cervical intraepithelial neoplasia 2/3 than Pap test alone. However, poor adherence limits usefulness of a management strategy that requires repeated follow-up.

Up to 10% of screening Papanicolaou (Pap) tests show low-grade squamous intraepithelial lesions (LSILs) or atypical squamous cells of undetermined significance (ASCUS).1-3 Approximately 10% of women with ASCUS and possibly up to 20% of women with LSIL are reported to have histologic cervical intraepithelial neoplasia (CIN) 2 or 3.4-10 The management of ASCUS and LSIL ranges from repeat Pap test to immediate colposcopy.2,11 However, colposcopy services are costly2 and may be overwhelmed if all women with ASCUS or LSIL are referred for examination.

A Canadian national workshop recommended that women with mild squamous dysplasia undergo Pap testing every 6 months for up to 2 years, and colposcopic examination if a Pap test shows cytologic progression or if the 2-year smear shows any degree of dysplasia.12 It is not known whether repeat cytology, as proposed in this recommendation, will be successful in identifying women with CIN 2/3. The sensitivity of a single repeat Pap test with a high-grade squamous intraepithelial lesion (HSIL) as threshold for colposcopy referral is low, ranging from 11.1% to 44.1%,4,7,8 although repeated testing may result in increased sensitivity. An additional concern is that studies suggest many women default from long-term surveillance.4,13,14

Oncogenic types of human papillomavirus (HPV) are present in more than 90% of cervical cancers15 and highgrade CIN.16 In cross-sectional analyses of women with low-grade abnormalities on screening Pap test, the addition of HPV testing identified more CIN than repeat Pap test and seemed to decrease unnecessary colposcopic referrals.5-9 Our results from a randomized trial indicate that immediate HPV testing may detect a greater number of CIN 2/3 cases than delayed Pap test.4 However, 1-time HPV testing by itself was insufficient to identify all women with CIN 2/3.

We report the results, according to randomization group, of a trial comparing combined Pap test and Hybrid Capture 1 (HC1) (Digene Corporation, Gaithersburg, Md) oncogenic HPV testing with Pap test alone for the detection of histologically confirmed CIN 2/3 in women with ASCUS and LSIL on screening Pap test. Women in community family practices were tested every 6 months for up to 2 years and then referred to a colposcopy clinic for exit colposcopic examination. To maximize generalizability and retention, the women returned to their own primary care doctors for testing rather than attending the colposcopy clinic for these procedures.

METHODS

Participants

From May 1995 to April 1998, consecutive women aged 16 to 50 years who were members of 66 community-based family practices and who had ASCUS or LSIL on screening cervical cytology were invited to participate in the study (Table 1). The women's age, entry smear findings, and reasons for study refusal were recorded for women who were eligible but chose not to enter the trial. The Research Ethics Board, Hamilton Health Sciences Corporation, Hamilton, Ontario, approved this study.

Protocol

Family physicians asked eligible women to return in 6 months. At the return visit, women gave written informed consent, completed a self-administered questionnaire on sociodemographic characteristics, reproductive history, and sexual behavior, and were randomly assigned to 1 of the 2 management strategies.