Alternating high-risk human papillomavirus infection

Consequences of Progression to Cervical Intraepithelial Neoplasia

Abstract

BACKGROUND.

Nearly every Dutch woman will be exposed to genital human papillomavirus (HPV) at least once during her lifetime, and most likely several times. In the current study, the authors investigated the prevalence of high-risk–HPV (HR-HPV) infection and the likelihood of progression to cervical intraepithelial neoplasia (CIN).

METHODS.

In this study, the course of HR-HPV infection in 703 women was observed. From a database of 720,016 negative cytology smears, the authors selected 703 women based on the availability of at least 2 HR-HPV polymerase chain reaction tests. The authors database stores not only the HPV data but also all other cytologic and histologic data, allowing the detection of women who progressed from negative cytology to CIN within a period of 10 years.

RESULTS.

Of the 703 selected women, 159 were found to have alternating HR-HPV infection (change from a negative HR-HPV test to a positive test or vice versa), 40 had a persistently positive HR-HPV test, and 504 women had a persistently negative HR-HPV test. The percentage of alternating HPV infection declined over time from 37% to 7%. Of the women age older than 40 years, 17% had an alternating HR-HPV infection, 2 of whom developed CIN. These findings led the authors to conclude that all the women in the current study with an increased risk of developing type 2 or 3 CIN were identified using 2 HPV tests. Women age older than 40 years still have a significant risk of acquiring a HR-HPV.

CONCLUSIONS.

In light of the current study findings, the authors believe it is worth considering the inclusion of women age 40 years and older who have negative cytology for HPV testing as part of the Dutch national screening program. Cancer (Cancer Cytopathol) 2006. © 2006 American Cancer Society.

In the Netherlands, human papillomavirus (HPV) infection is common among women with normal cytology.1 The prevalence of HPV is reported to be highest in sexually active young women and decreases with age.2, 3

Persistent infection with ≥1 high-risk types of human papillomavirus (HR-HPV) is reported to be the unifying risk factor for the development of cervical intraepithelial neoplasia (CIN).4, 5

The prevalence of both HPV infection and cervical cancer are closely related to age, but the age profiles differ. In the Netherlands, the peak prevalence of HPV infection precedes the peak prevalence of high-grade CIN and cervical cancer.6

A positive HPV test can be followed by a negative one, and vice versa. Such alternating infections are frequently encountered in our screening practice. The transient HPV infections are frequently studied but, to the best of our knowledge, there are no studies published to date that include a large number of sexually active women of all ages to obtain insight into the consequences of alternating HR-HPV infection in women with negative cytology.

Most HPV studies rely on the analysis of referral populations (ie, on women with cytologic abnormalities).2, 4, 5 For the purposes of healthcare planning decisions, studies using nonreferral data and cytology-negative smears would appear to be preferable. Therefore, we decided to readdress an existing database with HR-HPV and the histology results of patients participating in primary screening for cervical cancer, thereby allowing us to obtain insight in the consequences of alternating HR-HPV infection on progression to CIN.

MATERIALS AND METHODS

Study Population

Between 1993 and 2003, the Leiden Cytology and Pathology Laboratory classified 720,016 conventional smears as having negative cytology. Of these smears, 80% originated from women who participated in the national screening program. In the Netherlands, all women between the ages of 30 and 60 years are invited once every 5 years to be tested for cervical cancer by the national population screening program authority, based on community-centerd population registries. The remaining 20% of the smears were indicated because of individual or clinical reasons.

During this period, 21,335 HPV polymerase chain reaction (PCR) tests were performed. The majority of the HR-HPV tests were performed as part of an ongoing quality control and assurance program; cytologically negative smears were randomly selected and tested for HR-HPV. A total of 19,793 smears were randomly selected from a total of 720,016 smears with negative cytology. The remaining 1542 tests were performed because the cytologically negative smear contained koilocytes.

From this set of smears, we identified 703 women with at least 2 HR-HPV tests. There were 640 women (91%) with 2 tests, 53 women (8%) with 3 tests, and 10 women (1%) with ≥4 tests for a total of 1482 tests. Based on the outcomes, the women from our study population were classified as being HPV-positive or HPV-negative (ie, all HR-HPV tests were either positive or negative); the others were classified as having alternating HR-HPV infection. In our study population, all age groups were well represented (Table 1). In the database, not only the HR-HPV data but also other cytologic and histologic data are stored, thereby allowing us to find women who progressed from a negative cytology to CIN within 10 years.

Table 1. HR-HPV Data of 703 Women with Multiple HPV Tests Stratified by Age

	No. of women		Alternating HR-HPV*		HR-HPV–Positive test†		HR-HPV–Negative test‡
Age, years	No.	%	No.	%	No.	%	No.	%
HR-HPV indicates high-risk human papillomarvirus. * Defined as HR-HPV–positive and HR-HPV–negative in any order or number. † Defined as all tests were HR-HPV positive. ‡ Defined as all tests were HR-HPV negative.
<30	59	100%	21	36%	9	15%	29	49%
30–39	233	100%	68	29%	19	8%	146	63%
40–49	245	100%	41	17%	6	2%	198	81%
>50	166	100%	29	17%	6	4%	131	79%
Total	703		159		40		504

Molecular Biology

To obtain enough cervical cells, we scraped the lower half of the smear after removing the cover slip and the excess glue. Half of the smear, containing ectocervical and endocervical cells, was scraped off with a sterile disposable blade under conditions that ensured a minimal risk of contamination. The scraped off cells were suspended in 1000 μl of TRIS-HCL buffer (10 mM [pH 8.3]), and centrifuged for 5 minutes at 13,000 revolutions per minute. This cell pellet was used for PCR as described by Van den Brule et al.7 In this procedure, only the HR-HPV RNA probe mix (B) was used, which detected 11 types of HPV (HPV-16, HPV-18, HPV-31, HPV-33, HPV-35, HPV-39, HPV-45, HPV-51, HPV-52, HPV-56, and HPV-58).8,9

RESULTS

HR-HPV Tests

Of the 703 women with multiple HR-HPV tests, there were 40 women in the HR-HPV–positive group (those with ≥2 subsequent HR-HPV–positive tests without any negative tests) and 504 women in the HR-HPV–negative group (those with ≥2 HR-HPV–negative tests without any positive tests). Finally, there were 159 women in the HR-HPV alternating group (with both negative and positive tests in any order or number). The HPV data for these 703 women are presented in Table 1.

Of the women age younger than 40 years, 30.5% had an alternating HR-HPV infection. The older women demonstrated a lower but still remarkable rate of alternating HR-HPV infection of 17%.

Of the 640 women with 2 tests, 128 (20%) were found to have alternating HR-HPV infection, 29 (5%) were found to have HR-HPV–positive tests, and 483 (75%) were found to have HR-HPV–negative tests. Of the 53 women with 3 tests, 22 (42%) were found to have alternating HR-HPV infection, 11 (21%) were found to have HR-HPV–positive tests, and 20 (38%) were found to have HR-HPV–negative tests. Of the 10 women with ≥ 4 tests, 9 (90%) were found to have alternating HR-HPV infection and 1 woman (10%) was found to have a HR-HPV–negative test; none of the women had an HR-HPV–positive test.

Time Window between 2 HPV Tests

In Table 2, the time between the 2 HPV tests was stratified into 3 intervals: <1 year, 1 to 5 years, and >5 years (for those women with 3 and 4 tests, the average time was used.) The HPV-positive group percentages demonstrated a strong decline in time (from 21% to 0%), indicating the transience of HPV infections. The same finding was present in the group of patients with alternating HR-HPV infection, although in this group, a mixture of HPV acquisition and clearance was present. The HR-HPV–negative group was a complement of the 2 other groups.

Table 2. Time Window between 2 HPV Tests

	No. of women		Alternating HR-HPV*		HR-HPV–Positive test†		HR-HPV‒Negative test‡
Time between 2 tests	No.	%	No.	%	No.	%	No.	%
HPV indicates human papillomavirus; HR, high-risk. * Defined as HR-HPV–positive and HR-HPV–negative in any order or number. † Defined as all tests were HR-HPV positive. ‡ Defined as all tests were HR-HPV negative.
<1 y	43	100%	16	37%	9	21%	18	42%
1–5 y	566	100%	136	24%	31	56%	399	70%
>5 y	94	100%	7	7%	0	0%	87	93%
Total	703		159		40		504

Development of CIN

During the follow-up period, 6 of the 703 women (0.9%) developed CIN. The data from these 6 women are shown in Table 3. All 6 women had at least 1 positive HR-HPV test. In 1 woman age 41 years in whom cervical cancer was diagnosed, the interval time between the negative cytology finding and the histologic diagnosis of cancer was 6 years. The second woman with alternating HR-HPV tests was found to have type 2 CIN, 8 years after her negative HPV test. In the group of 504 women with a HR-HPV–negative test, none developed CIN. Therefore, in all the women with incident CIN, HR-HPV infection appears to have been present before the diagnosis of CIN.

Table 3. Data From Six Patients Who Developed CIN

Age at negative cytology, years	Test result from first HR-HPV test	Test result from second HR-HPV test	Age at biopsy, years	Histology diagnoses
CIN indicates cervical intraepithelial neoplasia; HR-HPV, high-risk human papillomavirus; +, positive; −, negative.
24	+	+	31	CIN, type 3
29	+	+	31	CIN, type 2
31	+	+	42	CIN, type 3
35	+	+	36	CIN, type 3
35	−	+	41	Invasive carcinoma
41	−	+	49	CIN, type 2

In the study group of 292 women age younger than 40 years, the prevalence of HR-HPV was 40% (117 women). The 411 women age 40 years and older were found to have a lower prevalence of HR-HPV of 20% (82 women). Three women in both groups developed histologic abnormalities.

DISCUSSION

The results of the current study, based on the analysis of 703 women with multiple tests who were extracted from the dataset of 21,480 cytologically negative smears with HPV tests, suggest that alternating HR-HPV infection is common among all women throughout the Netherlands. In the literature, the term “alternating HPV infection” is not used. We propose to add this term to the medical vocabulary because this phenomenon might be significant in our understanding of cervical carcinogenesis. The transience of HR-HPV infection is well known from the literature and describes the loss of HPV infection. This phenomenon occurs more frequently in younger women,4, 10 and appears to decline with age. In addition, the acquisition of HPV was found to predominate at a younger age.11 We found a nevertheless remarkably high prevalence of 17% for alternating HR-HPV infection in women age older than 40 years (Table 1). The probability of finding a woman in the HR-HPV–positive group must be even smaller than that found in the current study, because some women could have cleared a HR-HPV infection between 2 positive tests.12, 13 This would have to had gone undetected because of our coarse sampling regime. These women should have been included in the group of patients with alternating HR-HPV infection. Therefore, it appears likely that alternating HR-HPV infection is even more common than was found in the current study.

The declining age effect for persistent and alternating HR-HPV infection in Table 1 can be attributed to various independent causes (eg, fewer sexual partners and an acquired HPV immunity).14 Another possible explanation is the latent presence of the HPV virus during the majority of the woman's life, at levels below the detection threshold of the HPV tests.15, 16 This infection is reactivated at irregular instances (ie, a recurrent self-infection). Another explanation for monogamous couples is an oscillation of infection and clearing between sexual partners.17, 18 In instances in which this was the case, the infection was always present in either 1 or both of the partners, and was never cleared completely.

During the follow-up period in the current study, 6 of the 703 women (0.9%) developed CIN. These 6 women had at least 1 positive HR-HPV test. Other studies have indicated that the continuous presence of HR-HPV is necessary for the development, maintenance, and progression of CIN lesions.4, 19–23 In 3 studies performed in women with negative cytology, a substantial proportion of the women had developed either type 2 or type 3 CIN at the time of last follow-up.19, 23, 24 In the study by Rozendaal et al.,23 0.4% of the women with negative cytology developed type 3 CIN, and 6 had a HR-HPV–positive test whereas only 1 woman had a HR-HPV–negative test. In the study by Bulkmans et al.,24 0.6% of the women developed type 3 CIN; 12 of these women had type 3 CIN with a positive HR-HPV test and 1 woman had a negative HR-HPV test. These percentages appear to be in agreement with the findings of the current study. In the study by Koutsky et al.,19 the incidence of CIN was found to be 12%. We suspect that the higher incidence is because of a selection bias in the studied population; the women had presented themselves for the evaluation of sexually transmitted diseases. Unfortunately, in both studies by Rozendaal et al.,23, 24 the authors were unable to examine alternating HR-HPV infection because patients were tested only once for HPV. This also might be the reason why in both studies, a woman developed type 3 CIN without a positive HR-HPV test. By testing women more often, these women might be identified. In the Netherlands, a large study had focused on the question of the possible consequences of a positive HR-HPV test for progression to CIN in Dutch women. This study might provide more insight into the efficiency of the use of a HR-HPV test in the national screening program.25

The results of the current study demonstrate that a HR-HPV test can alternate from positive to negative and vice versa. In the HART study (a multicenter screening study of 11,085 British women) by Cuzick et al.,16 there were 189 women with normal or atypical cells of undetermined significance (AS-CUS) cytology who were selected for a second HPV test within 12 months. Of these women, 80 had 2 HR-HPV–negative tests, 117 women had 2 HR-HPV–positive tests, and the remaining 92 women had a positive test followed by a negative one and therefore were considered to have alternating HR-HPV infection.16

Clinical and experimental evidence suggest that, similar to other mechanisms of carcinogenesis, HPV infection alone is not sufficient to result in cervical cancer but requires other factors that may influence local immune responses and/or result in other genetic alterations.26, 27 In an overview study, Cuzick et al. supported the use of HPV testing as the primary screening test, with cytology reserved for women whose test was positive for HPV.28

As is clear from the current study, the frequency of HR-HPV infection is high even for women age older than 40 years. Therefore, we believe that it might be wise to include women age 40 years and older who have negative cytology for HPV testing. The results of the current study also demonstrate that of those women who were tested at least twice and found to be HR-HPV negative, none had developed CIN at the time of last follow-up. For these women with a low risk of developing cervical cancer, the interval between testing might be prolonged.