SEARCH

SEARCH BY CITATION

Keywords:

  • hybrid capture;
  • human papillomavirus;
  • cervical intraepithelial neoplasia;
  • cytologic screening

Abstract

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

To test the reliability of the Hybrid Capture II (HC-II) assay detecting 13 high-risk human papillomavirus (HR-HPV) types for the screening of cervical lesions, we monitored by cytology, HR-HPV testing, colposcopy and biopsy, 3,091 women with normal smears at the first entry. Our primary endpoint was clinical progression defined as the presence of a high-grade lesion (HGSIL) at the biopsy. In our population of 659 HR-HPV-infected women, 241 (36.6%) had a positive HR-HPV test at 2 to 4 examinations with a final histological diagnosis of HGSIL in 51 cases (21.2%) within 4 to 36 months, while women with regressive HPV infection did not develop any lesion during the same period. In the cohort of 2,432 women testing negative for HR-HPV infection, only 2 women (0.08%) developed a HGSIL. Both were HR-HPV positive 18 and 24 months after the first entry, at the time of diagnosis of disease. The RR of incident HGSIL when a HR-HPV was detected at enrollment in women with normal smears was 96.7 (95% CI, 95.8–97.7). The RR increased to 237.3 (95% CI, 222.8–251.8) when the HR-HPV test remained positive at 2 controls, and to 314.3 (95% CI, 260.7–367.9) when the HR-HPV test was positive at 3 controls. The evaluation of the viral load of HR-HPV by the HC-II did not represent a sensitive approach to predict the recurrence of HR-HPV infection and/or the apparition of HGSIL. Nevertheless, a recurrent HR-HPV infection detected with HC-II represents a reliable tool to select populations at risk for the development of HGSIL. © 2002 Wiley-Liss, Inc.

High-risk human papillomaviruses (HR-HPV) have now been conclusively demonstrated to be the causative agents of cervical cancer.1, 2, 3 Indeed, it has been reported that 99.7% of all cervical squamous cell carcinomas contain HR-HPV.4 Infections with HR-HPV are associated with a relative risk of between 8 and 11 for the development of squamous intraepithelial lesions (SIL)5 and only low-grade SIL that contain HR-HPVs progress to high-grade disease.6 While most authors agree that moderate and severe dysplasias should be managed regardless of HPV type as most of them contain HR-HPV, HPV testing may be useful for triaging women with a cytological diagnosis of atypical squamous cells of undetermined significance (ASCUS) in their cervical smears and for the follow-up of women treated for CIN.7, 8, 9, 10, 11, 12, 13 In the same way, there is an increasing interest in using HR-HPV DNA detection either alone or in addition to the classic cytological examination as a method for primary screening for cervical preneoplastic and neoplastic lesions.

Six years ago, a commercial HPV detection test, Hybrid Capture-II (HC-II), was introduced.14 HC-II is a non radioactive, easy to perform, relatively rapid (5 hr of handling), liquid hybridization assay designed to detect 18 HPV types divided into high-risk (types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59 and 68) and low-risk (types 6, 11, 42, 43 and 44) groups. The first studies have been particularly promising, showing a good agreement with PCR results, with a detection of HR-HPV DNA in more than 90% of high grade SIL (HGSIL).15, 16, 17, 18, 19 This test has been proposed to be used routinely on large series of women to improve the sensitivity and negative predictive value of cytological screening for SIL. Although the sensitivity of this test is higher than that of cytology, its specificity remains low. This is largely explained by the numerous transient HR-HPV infections and the high prevalence of HR-HPV infection in women with normal smears. For example, in our study that explored the use of this assay for primary screening of 7,932 women explored in primary screening with this assay, we found that a total of 773 women (11.9%) out of 7,339 women with smears within normal limits presented with an HR-HPV infection.15 Numerous studies in the literature using Polymerase Chain Reaction (PCR) for the detection of HR-HPV have emphasized that the persistence of HR-HPV infection is significantly associated with progressive disease.20, 21, 22, 23, 24 Moreover a high viral load may be considered as a risk factor and preferentially observed in potentially progressive and high-grade lesions.25, 26, 27, 28 This parameter can be semi-quantitatively evaluated by the relative light unit values provided by the HC-II assay. Thus, the detection of a recurrent HR-HPV infection with a baseline high viral load may be a useful discriminating test to select the population of women at risk for the development of cervical SIL.

The aim of the present work was to evaluate the reliability of the HC-II assay for the screening of cervical preneoplastic and neoplastic lesions in a population of 3,091 women with no cervical abnormalities at the first cervical smear. These women were monitored by cytology, HR-HPV testing and colposcopy. Our primary endpoint was clinical progression defined as the presence of a HGSIL at the biopsy.

MATERIAL AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Study population

A total of 3,091 women with a median age of 39 years (range 17 to 77 years) was recruited for the study between August 1997 and June 2002. This population was restricted to women who underwent their biennal or triennial routine screening in the Department of Obstetrics and Gynecology of the C.H.U. of Reims. All of these women had a cervical smear within normal limits at baseline. We excluded subjects on the basis of a recent cytologic abnormality and/or an untreated cervical lesion in the past 2 years, as well as pregnant women and patients with AIDS. All women were informed of the aim of the study and gave their consent.

Cytologic diagnosis

At the first gynecologic examination, 2 samples were taken from 1,337 women: first, a cytologic smear with an Ayre's spatula, for classical cytology, and then a sample for the HC-II test using a brush provided in the HC-II kit. These samples were suspended in 1 ml of specimen transport medium (STM) for HPV testing (Digene, Gaithersburg, MD). A second group of 1,754 women had only one cervical scrape with a Cervexbrush (Rovers Medical Devices, Oss, Netherlands) at the first entry. Samples were prepared in PreservCyt medium for liquid-based cytology with the ThinPrep technique (Cytyc Corporation, Boxborough, Mass) and 4 ml of the sample were used for HPV testing. In total, including the first samples and the follow-up of women, 7,388 cervical smears and scrapes were collected. Smears were classified according to the Bethesda system for reporting cervical or vaginal cytological diagnosis. Women were selected with adequate smears including metaplastic and/or endocervical cells, according to the criteria of Bethesda, which represented 91% of our total smears. However, even if the smear was not considered to be adequate, women were also included with smears evocative of lesions. Altogether, 4.4% of the samples treated for liquid-based cytology could not be used for HPV testing and have been excluded. The cytotechnicians and pathologists involved in the study were not informed about the results of the HPV testing. All smears that exhibited cytological abnormalities and biopsy specimens were examined by the same 2 independent pathologists; when examining the biopsy specimens, the pathologists had no knowledge of cytology results. The results were compared and if the first 2 diagnoses disagreed, a third pathologist reviewed the case with no knowledge of preceding diagnoses. Consensus diagnoses were determined by a two-thirds majority when possible and all remaining discrepancies were resolved by conference review. Patients with HGSIL were systematically treated by loop electrosurgical excision procedure (LEEP). Data from these LEEP specimens were included in the disease definitions.

Colposcopic referral

In our protocol, all the women with smears within normal limits but presenting with an HR-HPV infection were systematically recalled 3 to 12 months later (mean interval = 6 months) for a new cytological examination and HR-HPV testing, followed by colposcopy if a lesion and/or a HR-HPV infection was still detected. Punch biopsy specimens were taken from the areas colposcopically suspicious for CIN. The women with a second positive HR-HPV test without any detectable lesions were recalled 6 to 12 months later for a third visit with cytological examination and HR-HPV testing using the same indications as above for colposcopy and biopsy. Women with regressive HR-HPV infection were also recalled for colposcopy 12 months later. By contrast, women with initial normal smears and no detectable HR-HPV infection were followed with a classical biennal or triennal cervical screening where a new cytological examination and HR-HPV testing were performed as a follow-up control. Some of these women were randomly selected for a colposcopic examination as a control for verification bias. There were not particular selection criteria for these colposcopy controls. The median age was 38 years (range 22 to 66 years). Colposcopy was performed by the gynecologist informed of an initial HR-HPV test negative, when the women went for new routine screening. The primary endpoint of our study was the detection of a histologically proven HGSIL at the biopsy and/or in the LEEP specimen.

HPV testing

When conventional cytology was performed, specimens for HPV DNA testing were suspended in 1 ml of STM transport medium (Digene, Gaithersburg, MD) and stored at −20°C until further processing. When samples were used for liquid based cytology, 4 ml of the sample was centrifuged and the cell pellet was resuspended in 200 μl of phosphate-buffered saline for HPV testing. HPV DNA detection was performed by the commercially available HC-II System (Digene). All scrapes were analyzed for the presence of HR-HPV types 16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59 and 68. This enzyme-linked immunosorbent assay is based on a sandwich hybridization followed by a nonradioactive alkaline phosphatase reaction with chemoluminescence in microplates. The chosen positive threshold of this test was 1.0 pg/ml of HPV DNA as recommended by the manufacturer.

Samples were classified as positive for HR-HPV DNA if the relative light unit (RLU) reading obtained from the luminometer was equal to or greater than the mean of the 3 positive control values supplied by the HC-II kit. Since some authors have reported that increasing HPV DNA levels of HR-HPV types were the principal predictors of CIN 8, we used as proposed, the ratio RLU/positive controls values to quantify HR-HPV DNA in our samples. Moreover, we added other positive controls such as SiHa cell lines (1 to 2 copies of HPV type 16 per cell) to check the reproducibility of the HC-II sensitivity.

Statistical methods

The statistical methods used were mostly descriptive. Sensitivity, specificity, positive and negative predictive values were determined by comparing the results of each test to the gold standard of histology. A few high-grade lesions may have been missed if they were negative on both tests and thus women were not referred for colposcopy. Therefore, there was a colposcopy control group representing 10% of these double negative women. Ninety-five percent confidence interval for these values were assessed using either binomial or normal distribution, according to the data. We analyzed the risk of occurrence of HGSIL as an incident finding in relation to HR-HPV infection status at enrollment and during the 2 following visits and for that purpose, the longitudinal relative risks (RRs) and respective 95% confidence intervals (CIs) of incident cervical lesions over time have been calculated. The homogeneity test was performed to compare the respective efficiencies of conventional and liquid-based cytology for predicting recurrent HR-HPV infection and the incidence of HGSIL.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Recurrent HR-HPV infection and detection of cervical lesions

The median follow-up was 12 months (4 to 39 months) for the 659 women with initial HR-HPV infection and 27 months (9 to 59 months) for the 2,432 women without any HR-HPV detectable at the first smear. The results of the follow-up of women with HR-HPV infection are summarized in Table I.

Table I. FOLLOW-UP OF WOMEN WITH HR-HPV INFECTION WITH NORMAL SMEARS1
Number of women at the 1st entryFirst control 4 to 10 monthsHGSIL at histologySecond control 10 to 18 monthsHGSIL at histologyThird control 18 to 39 monthsHGSIL at histology
  • 1

    NL, normal smear; LGSIL, low grade lesion; HGSIL, high grade lesion.

659 NL HPV+4 ASCUS HPV− 2 NL HPV−   
 30 ASCUS HPV+24 ASCUS HPV+11 NL HPV+ 
     1 NL HPV− 
   3 LGSIL HPV+ 1 NL HPV+ 
     1 NL HPV− 
   3 HGSIL HPV+3  
   2 NL HPV+   
   2 NL HPV− 2 NL HPV− 
 7 LGSILHPV− 4 NL HPV−   
 28 LGSILHPV+11 ASCUS HPV+ 1 NL HPV+ 
   4 LGSIL HPV+ 1 LGSIL HPV+ 
     1 NL HPV− 
   2 HGSIL HPV+2  
   2 NL HPV+ 2 LGSIL HPV+ 
   2 NL HPV−   
 38 HGSIL HPV+371 LGSIL HPV+ 1 NL HPV+ 
 203 NL HPV+13 ASCUS HPV+11 NL HPV− 
   7 LGSIL HPV+ 2 LGSIL HPV+ 
     1 HGSIL HPV+1
     2 NL HPV− 
   2 LGSIL HPV−   
   1 HGSIL HPV+   
   53 NL HPV+ 3 ASCUS HPV+1
     3 LGSIL HPV+ 
     15 NL HPV+1
     4 NL HPV− 
   42 NL HPV− 9 NL HPV− 
 349 NL HPV− 1 ASCUS HPV+   
   4 LGSIL HPV+ 1 LGSIL HPV− 
     1 NL HPV− 
   7 NL HPV+ 2 NL HPV− 
   84 NL HPV− 1 NL HPV+ 
     28 NL HPV− 

HR-HPV testing was found positive at the second examination in 299 cases. In this cohort, 30 women had ASCUS, 28 had smears evocative of a low grade lesion, 38 had smears evocative of a HGSIL and 203 smears remained within normal limits with a HR-HPV infection. In this cohort of women with a second positive HR-HPV testing, 41 HGSIL have been discovered at the histological examination after colposcopy. One HGSIL corresponded to low grade lesions at the cytological examination, 2 to ASCUS, 37 to cytological HGSIL and 1 case remained cytologically normal. In this population with recurrent HR-HPV infection, 134 women had a third examination, 86 cases remained HR-HPV positive, while 48 HR-HPV infections regressed. The colposcopic controls revealed 7 HGSIL, with ASCUS in 2 cases and a smear evocative of HGSIL in 5 cases. Forty-one women with a third positive HR-HPV testing had a fourth control. At this control, HR-HPV testing was still positive in 31 cases, while regression was observed in 10 cases. Three HGSIL were detected at the histological examination, with, in one case, a smear within normal limits. Thus, in total, recurrent HR-HPV infection was found in 241 women (36.6%) and was associated with a histological diagnosis of HGSIL in 51 patients (21.2%). These HGSIL were detected within 16 months after the first examination in 41 cases. In these 41 HGSIL, the initial smear was examined by conventional cytology in 13 cases and by liquid-based cytology in 28 cases. If we considered the population of 426 women > 30 years aged, a total of 140 recurrent HR-HPV infections (32.9%) was observed associated with 31 HGSIL (22.1%). The incidence of recurrent HR-HPV infections was significantly lower (p<0.02) than in the population of women < 30 years aged (43.3%).

In the population of 417 women with regressive or transient HR-HPV infection (63.3%), most cervical smears remained normal and colposcopic and histological examinations when cell abnormalities were observed, did not reveal any HGSIL.

The results of the follow-up women tested negative for HR-HPV are summarized in Table II. Out of 2,432 women with initial smear within normal limits and without no detectable HR-HPV infection, 190 presented a HR-HPV infection in their follow-up, transient in 66 cases, associated with ASCUS in 11 cases, with smears evocative of low grade lesion in 22 cases. No HGSIL was detected in these cases. However in 2 women, a HGSIL associated with a HR-HPV infection was diagnosed at the cytology and histologically confirmed 18 and 24 months after the initial smear. The 225 colposcopic controls performed randomly in the 2,214 women without any HR-HPV infection and with smears within normal limits did not reveal any HGSIL. The cytology remained within normal limits in these women. In our series, the median duration of HR-HPV infections was 12 months (5 to 59 months).

Table II. FOLLOW-UP OF WOMEN WITH NORMAL SMEARS WITH NO HR-HPV INFECTION AT ENROLLMENT1
Number of women at the 1st entryFirst control 9 to 36 monthsHGSIL at histologySecond control 24 to 59 monthsHGSIL histologyCytology at 3rd control
  • 1

    NL, normal smear; LGSIL, low grade lesion; HGSIL, high grade lesion.

2432 NL HR-HPV−19 ASCUS HR-HPV− 1 NL HR-HPV+  
   6 NL HR-HPV−  
 10 ASCUS HR-HPV+ 1 NL HR-HPV+  
 9 LGSIL HR-HPV− 7 NL HR-HPV−  
 20 LGSIL HR-HPV+ 1 ASCUS HR-HPV−  
   1 NL HR-HPV+  
   4 NL HR-HPV−  
 2 HGSIL HR-HPV+2   
 158 NL HR-HPV+ 1 ASCUS HR-HPV+  
   2 LGSIL HR-HPV+  
   1 LGSIL HR-HPV−  
   60 NL HR-HPV− 2 NL HR-HPV+
 2214 NL HR-HPV−    

Thus, the RR of incident HGSIL when a HR-HPV was detected at enrollment in women with normal smears was 96.7 (95% CI, 95.8–97.7). The RR increased to 237.3 (95% CI, 222.8–251.8) when the HR-HPV test remained positive at the second control, and to 314.3 (95% CI, 260.7–367.9) when the HR-HPV test was positive at 3 controls. When we considered the method of cytology at enrollment, the RR was 124.9 (95% CI, 104.2–144.5) for conventional cytology, and 81.6 (95% CI, 64.5–98.7) for liquid-based cytology.

Viral load, recurrent HR-HPV infection and development of HGSIL

Considering the viral load of high-risk HPV detected by the HC-II test, the median viral load at first entry was 3.9 for regressive and transient infections, while it was 19.7 for recurrent HR-HPV infections. For an initial cervical sample/positive control ratio > 10, the RRs of a recurrent HR-HPV infection and of the development of HGSIL were respectively 2.1 (95% CI, 0.8–3.3) and 2.7 (95% CI, 1–4.4).

With a view to clinical application, we chose to test the sensitivity, specificity and positive predictive value of the HR-HPV DNA level for the recurrence of HR-HPV infection and the detection of histologically proven high-grade lesions in our total population and in women aged > 30 years old who could represent a cohort at higher risk for development of high-grade lesions.29 The results obtained are summarized in Tables III and V. We also tested the different efficiencies of conventional cytology and liquid-based cytology for predicting recurrent HR-HPV infection and the incidence of HGSIL. These results are reported in Tables IV and VI. When we used the homogeneity test, we obtained a better efficiency of the liquid-based cytology for predicting recurrent HR-HPV infection (p< 0.01) and the incidence of HGSIL (p<0.001).

Table III. EVALUATION OF THE VIRAL LOAD (VL IN RLU; ORDINARY CUT-OFF = 1 RLU) ESTIMATED BY THE HC-II ASSAY FOR PREDICTING A RECURRENT HR-HPV INFECTION IN WOMEN WITH INITIALLY CYTOLOGICALLY NORMAL SMEARS1
PopulationVLSensitivity 95% C.I.Specificity 95% C.I.PPV 95% C.I.NPV 95% C.I.
  • 1

    PPV, Positive Predictive Value; NPV, Negative Predictive Value; C.I., Confidence Interval.

All women>3207/250 (82.8%)183/409 (44.7%)207/433 (47.8%)183/226 (81%)
  78.1–87.5%39.9–49.5%43.1–52.5%75.9–86.1%
 >10159/250 (63.6%)259/409 (63.3%)159/309 (51.4%)259/350 (74%)
  57.6–69.5%58.6–67.9%45.8–56.9%69.4–78.6%
Women >30 years old>3136/140 (97.1%)2/286 (0.7%)136/420 (32.3%)2/6 (33.3%)
  94.3–99.8%0–1.7%27.8–36.7%0–71%
 >1081/140 (57.8%)120/286 (41.9%)81/247 (328%)120/179 (67.0%)
  49.6–65.9%36.2–47.6%26.9–38.6%60.1–73.9%
Table V. EVALUATION OF THE VIRAL LOAD (VL IN RLU; ORDINARY CUT-OFF = 1 RLU) ESTIMATED BY THE HC-II ASSAY FOR THE DETECTION OF A HISTOLOGICALLY PROVEN HGSIL IN WOMEN WITH INITIALLY CYTOLOGICALLY NORMAL SMEARS1
PopulationVLSensitivity 95% C.I.Specificity 95% C.I.PPV 95% C.I.NPV 95% C.I.
  • 1

    PPV, Positive Predictive Value; NPV, Negative Predictive Value; C.I., Confidence Interval.

All women>345/51 (88.2%)220/608 (36.2%)45/433 (10.4%)220/226 (97.3%)
  79.4–97.1%32.4–40%7.3–13.3%95.2–99.4%
 >1036/51 (70.6%)335/608 (5.1%)36/309 (11.6%)35/350 (11.6%)
  58.1–83.1%49.1–61.1%8.0–15.2%8.2–14.9%
Women >30 years old>328/31 (90.3%)3/395 (0.7%)28/420 (6.7%)3/6 (50%)
  84.6–95.6%0–1.5%4.3–9.1%10–90%
 >1013/31 (41.9%)161/395 (40.7%)13/247 (5.3%)161/179 (89.9%)
  24.5–59.3%35.8–45.5%2.5–8.1%87.6–92.1%
Table IV. EVALUATION OF THE VIRAL LOAD (VL IN RLU; ORDINARY CUT-OFF = 1 RLU) ESTIMATED BY THE HC-II ASSAY FOR PREDICTING A RECURRENT HR-HPV INFECTION IN WOMEN WITH CYTOLOGICALLY NORMAL SMEARS EXAMINED BY CONVENTIONAL CYTOLOGY AND LIQUID-BASED CYTOLOGY
PopulationVLSensitivity 95% C.I.Specificity 95% C.I.PPV 95% C.I.NPV 95% C.I.
  1. 1PPV, Positive Predictive Value; NPV, Negative Predictive value; C.I., Confidence Interval.

All women>3207/250 (82.8%)183/409 (44.7%)207/433 (47.8%)183/226 (81%)
  78.1–87.5%39.9–49.5%43.1–52.5%75.9–86.1%
 >10159/250 (63.6%)259/409 (63.3%)159/309 (51.4%)259/350 (74%)
  57.6–69.6%58.6–68%45.8–57%69.4–78.6%
Women with 1st smear with conventional cytology>350/61 (82%)24/108 (22.2%)50/134 (37.3%)24/35 (68.6%)
72.4–91.6%14.4–30.0%29.1–45.5%52.9–83.7%
 >1042/61 (68.8%)36/108 (33.3%)42/114 (36.8%)36/55 (65.4%)
  57.2–80.4%24.4–42.1%27.9–45.6%52.8–78%
Women with 1st smear with liquid-based Cytology>3157/189 (83.1%)159/301 (52.8%)157/299 (52.5%)159/191 (83.2%)
76–88.4%47.2–58.4%46.8–58.2%77.9–88.5%
 >10117/189 (61.9%)223/301 (74.1%)117/195 (60%)223/295 (75.6%)
  55–68.8%69.2–79.0%53.1–66.9%70.7–80.5%
Table VI. EVALUATION OF THE VIRAL LOAD (VL IN RLU; ORDINARY CUT-OFF = RLU 1) ESTIMATED BY THE HC-II ASSAY FOR THE DETECTION OF A HISTOLOGICALLY PROVEN HGSIL IN CYTOLOGICALLY NORMAL SMEARS EXAMINED BY CONVENTIONAL CYTOLOGY AND LIQUID-BASED CYTOLOGY1
PopulationVLSensitivity 95% C.I.Specificity 95% C.I.PPV 95% C.I.NPV 95% C.I.
  • 1

    PPV, Positive Predictive Value; NPV, Negative Predictive value; C.I., Confidence Interval.

All women>345/51 (88.2%)220/608 (36.2%)45/433 (10.4%)220/226 (97.3%)
  79.4–97.1%32.4–40%7.3–13.3%95.2–99.4%
 >1036/51 (70.6%)335/608 (55.1%)36/309 (11.6%)335/350 (11.6%)
  58.1–83.1%49.1–61.1%8.0–15.2%8.2–14.9%
Women with 1st smear with conventional cytology>318/19 (94.7%)34/150 (22.7%)18/134 (13.4%)34/35 (97.1%)
84.6–100%16.0–29.4%7.6–19.2%91.5–100%
 >1017/19 (89.5%)53/150 (35.3%)17/114 (14.9%)53/55 (96.4%)
  75.7–100%27.6–42.9%8.4–21.4%91.5–100%
Women with 1st smear with liquid-based cytology>327/32 (84.4%)186/458 (40.6%)27/299 (90.3%)186/191 (97.4%)
71.8–97%36.1–45.1%86.9–93.6%95.1–99.6%
 >1019/32 (59.4%)282/458 (61.6%)19/195 (9.7%)282/295 (95.6%)
  42.4–76.4%57.1–66.0%5.5–13.8%93.3–97.9%

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

Our longitudinal study on the follow-up of 3,091 women that attended cervical screening with baseline cervical smears within normal limits clearly demonstrates the highly significant association of recurrent HR-HPV infection detected with the HC-II assay, with the apparition of HGSIL. Indeed, in our population of 659 HPV infected women with a median follow-up of 12 months, a total of 241 (36.6%) had a recurrent HR-HPV infection, with a final diagnosis of HGSIL in 51 of them (21.2%). In a practical point of view, 41 of these HGSIL were diagnosed within 16 months after a smear within normal limits. In contrast, women with regressive HPV infection did not develop any lesion and only 2 women out of 2,432 (0.08%) who were HR-HPV negative at baseline developed a HGSIL at follow-up. The RRs of incident HGSIL were 96.7 when a HR-HPV was detected at enrollment in women with normal smears, 237.3 when the HR-HPV testing remained positive at the second control and 314.3 when the HR-HPV testing was positive at 3 controls. Although a more intensive follow-up consecutive to the detection of a HR-HPV infection necessarily introduces a bias of selection resulting in more and early diagnoses of HGSIL, our results are in agreement with the published data from studies using HPV detection by PCR, where it was reported that only women with initial and persistent HR-HPV infections either in cytologically abnormal or normal smears exhibit progressive CIN.5, 21, 22, 23, 24, 25, 30, 31, 32 However, the positive predictive value of a recurrent HR-HPV infection for detection of a HGSIL in HR-HPV infected women with normal smears does not vary by age when we separate the women aged > 30 years old (21.2% for the total population vs. 22.1% for women aged > 30 years old). One of the weaknesses of the HC-II assay is that it cannot detect specific HPV types and it has been well demonstrated that persistently detected molecular HPV types are preferentially associated with cervical neoplasias.33, 34 Indeed, we have no evidence that the recurrent HR-HPV infections detected by the HC-II assay are truly persistent infections since in some cases a recurrent infection may correspond to the clearance of a HR-HPV type and a new infection with a new type. A secondary HR-HPV typing of the HR-HPV positive samples that demonstrate a real persistent infection with the same HPV genotype would likely increase the positive predictive value of HPV detection for screening HGSIL. Nevertheless, even if the HC-II assay gives a crude information on the presence of HR-HPV of various and eventually different types, recurrent HR-HPV infection detected with this assay selects a population at higher risk for the development of HGSIL among women with normal smears.

In our experience, the absence of detectable lesions at the first smear could be considered as a false-negative diagnosis of the cytology, especially when a HGSIL is detected within a few months after the first entry. Nevertheless, the reexamination of these smears of our series confirmed the initial diagnosis except for the presence of rare groups of ASCUS in only 6 cases (4 with conventional cytology and 2 with liquid-based cytology) and only 2 out of 51 cases remained cytologically normal at baseline with a histologically proven HGSIL in their follow-up. The different modalities of collection of the samples (Ayre's spatula and cytobrush) may explain these false-negative results when conventional cytology was used. Nevertheless, if we consider the 41 cases of HGSIL discovered within 16 months after the first entry, they corresponded to 13 (7.7%) out of 169 women with initial conventional cytology and to 28 (5.7%) out of 490 women with liquid-based cytology at enrollment. The use of the same cervical specimens collected for liquid-based cytology and HPV detection certainly reduces the discrepancies between the results of cytology and HPV testing for the detection of HGSIL. However, the higher sensitivity of liquid-based cytology for the detection of HGSIL observed in our study and reported in the literature15, 35 remains lower than the sensitivity of HPV testing.

Cuzick et al.25 showed first that in women with cytological abnormalities, HPV positivity at a high-level detected by a semi-quantitative PCR was strongly related to high-grade CIN. Moreover, Ho et al.21 have also suggested that SIL with a high viral load are more likely to persist than those with a low level of HPV DNA. Ylitalo et al.27 and Josefsson et al.26 using a sensitive real time quantitative PCR assay have clearly demonstrated that cervical carcinoma in situ associated with HPV type 16 occurs mainly in HPV type 16 positive women who had consistently high viral loads long term. Recently, Van Duin et al.28 have suggested that in women with normal cytology, an increased HPV type 16 load conferred an increased risk of developing a cervical lesion. Thus, a high viral load may be considered to be a risk factor and preferentially observed in potentially progressive lesions and in HGSIL. This parameter could be semi-quantitatively evaluated by the relative light unit values (RLU) provided by the HC-II assay. Indeed, for an initial cervical sample/positive control ratio > 10, the RRs of a recurrent HR-HPV infection and of the development of HGSIL are, respectively, 2.1 (95% CI, 0.8–3.3) and 2.7 (95% CI, 1–4.4). However, when this parameter is used with a clinical application to differentiate women with recurrent HR-HPV infection from women with regressive infection, it does not provide reliable sensitivity and specificity. For example, for a cervical sample/positive control ratio > 10, the sensitivity is only 70.6% and the specificity 55.1% for predicting a HGSIL. If we consider women > 30 years old who may represent a population at higher risk for the development of HGSIL, for the same values of viral load, the sensitivity for predicting a HGSIL falls to 41.9% and the specificity to 40.7%. Similarly, in our previous study, the semi-quantitative approach provided by the HC-II assay for the assessment of the viral load could not clearly distinguish among cases with or without HGSIL.15 One major bias of this parameter evaluation remains the cell number in the cervical specimen which is not quantitated and can vary substantially from one sample to another. A standardization of the sample with the use of cervical specimens collected with a broom device for thin-layer cytology may improve the semi-quantitative approach by HC-II in the future. Indeed, when we used an homogeneity test, we found that liquid-based cytology was more efficient than conventional cytology for predicting the incidence of a HGSIL and recurrent HR-HPV infection. Moreover, the global high viral load detected by HC-II is rather confusing, since it may represent one or multiple HPV types among the 13 high-risk types detected by the kit. Thus, in our experience at the present time, contrary to quantitative real time PCR, HR-HPV DNA quantitation by HC-II cannot be considered as a relevant sensitive test to be used in practice to predict the recurrence of HPV infection and to screen HGSIL.

In conclusion, our study using the sensitive, easy to perform HC-II assay confirms the usefulness and reliability of this test to select from a population of women with normal cervical smears those who are HR-HPV positive and therefore at significantly increased risk for the development of SIL. The evaluation of the viral load by this assay needs a standardization of the cervical sample with an initial identical cell amount that may give more reproducible results. At the present time, however, it cannot be used as a discriminating predictive parameter. It has been well established that the mean infection duration for HR-HPV infection is of 13.5 months36 (12 months in our experience). Consequently, we propose that women with normal smears and HR-HPV infection have to be controlled every 6 months with cytological examination and HPV testing until HPV infection regresses. Women with recurrent HR-HPV infection and/or with cytological abnormalities appearing in their follow-up should be systematically referred for colposcopy. Liquid-based cytology with the possibility of HPV testing on the same sample allows for easy implementation of this policy that will significantly increase the detection of HGSIL in a screening program. At the same time, women with a normal smear and no HR-HPV infection have a very low risk for developing a HGSIL. In consequence, as previously proposed by Meijer et al.,10 in such conditions, the use of HPV testing would allow the screening interval to be safety lengthened to 5 years with a cost-effective benefit.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES

We thank I. Putaud and C. Mangeonjean for their technical help, Dr. M.-F. Poncelet-Deletoile and Dr. C. PIA for their collaboration and all the women for their participation in our study.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. REFERENCES
  • 1
    Bosch F, Lorincz A., Munoz N, Meijer CJLM, Shah KV. The causal relation between Human Papillomavirus and cervical cancer. J Clin Pathol 2002; 55: 24465.
  • 2
    Lorincz AT, Reid R, Jenson AB, Greenberg MD, Lancaster W, Kurman RJ. Human papillomavirus infection of the cervix: relative risk association of 15 common anogenital types. Obstet Gynecol 1992; 79: 32837.
  • 3
    Zur Hausen H. Molecular pathogenesis of cancer of the cervix and its causation by specific human papillomavirus types. Curr Top Microbiol Immunol 1994; 186: 13156.
  • 4
    Walboomers JMM, Jacobs MV, Manos MM, Bosch FX, Kummer JA, Shah KV, Snijders PJ, Peto J, Meijer CJ, Munoz N. Human Papillomavirus is a necessary cause of invasive cervical cancer worldwide. J Pathol 1999; 189: 129.
  • 5
    Koutsky LA, Holmes KK, Critchlow CW, Stevens CE, Peevonen J, Beckmann AM, De Rouen TA, Galloway DA, Vernon D, Kiviat NB. A cohort study of the risk of cervical intraepithelial neoplasia grade 2 or 3 in relation to papillomavirus infection. N Engl J Med 1992; 327: 12728.
  • 6
    Gaarenstroom KN, Melkert P, Walboomers JMM, Van den Brule AJ, Van Bommel PF, Meijer CJ, Voorhorst FJ, Kenemans P, Helmerhorst TJ. Human papillomavirus DNA genotypes: prognostic factors for progression of cervical intraepithelial neoplasia. Int J Gynecol Cancer 1994; 4: 738.
  • 7
    Cole HM. Human papillomavirus DNA testing in the management of cervical neoplasia. JAMA 1993; 270: 297581.
  • 8
    Cox JT, Lorincz AT, Schiffman MH, Sherman ME, Cullen A, Kurman RJ. Human papillomavirus testing by hybrid capture appears to be useful in triaging women with a cytologic diagnosis of atypical squamous cells of undetermined significance. Am J Obstet Gynecol 1995; 172: 94654.
  • 9
    Manos MM, Kinney WK, Hurley LB, Shieh-Ngai J, Kurman RJ, Ransley JE, Fetterman BJ, Hartinger J., McIntosh KM, Pawlick GF, Hiatt RA. Identifying women with cervical neoplasia using Human Papillomavirus DNA testing for equivocal Papanicolaou results. JAMA 1999; 28: 160510.
  • 10
    Meijer CJLM, Helmerhorst TJM, Rozendaal L, Van Der Linden JC., Voorhorst FJ, Walboomers JMM. HPV typing and testing in gynaegological pathology: has the time come? Histopathology 1998; 33: 836.
  • 11
    Richart RM, Wright TC. Controversies in the management of low grade cervical intraepithelial neoplasia. Cancer 1993; 71: 141321.
  • 12
    Solomon D, Schiffman M, Tarone R. 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: 293299.
  • 13
    The Atypical Squamous Cells of Undetermined Significance/ Low Grage Squamous Intraepithelial Lesions Study (ALTS) Group. Human papillomavirus testing for triage of women with cytological evidence of low grade squamous intraepithelial lesions: baseline data from a randomized trial. J Natl Cancer Inst 2000; 92: 397402.
  • 14
    Lorincz AT. Hybrid Capture method for detection of human papillomavirus DNA in clinical specimens. Papillomavirus Rep 1996; 7: 15.
  • 15
    Clavel C, Masure M, Bory JP, Putaud I, Mangeonjean C, Lorenzato M, Nazeyrollas P, Gabriel R, Quereux C, Birembaut P. Human papillomavirus testing in primary screening for the detection of high-grade cervical lesions: a study of 7932 women. Br J Cancer 2001; 89: 16161623.
  • 16
    Cuzick J, Beverly E, Ho L, Terry G, Sapper H, Mielzynska I, Lorincz A, Chan WK, Krausz T, Soutter P. HPV testing in primary screening of older women. Br J Cancer 1999; 81: 554558.
  • 17
    Ratnam S, Franco EL, Ferenczy A. Human Papillomavirus testing for primary screening of cervical cancer precursors. Cancer Epidemiol Biomarkers Prev 2000; 9: 94551.
  • 18
    Riethmuller D, Gay C, Bertrand X, Bettinger D, Schaal JP, Carbillet JP, Lassabe C, Arveux P, Seilles E, Mougin C. Genital Human Papillomavirus infection among women recruited for routine cervical cancer screening or for colposcopy determined by Hybrid Capture II and polymerase chain reaction. Diagn Mol Pathol 1999; 8: 15764.
  • 19
    Schiffman M, Herrero R, Hildesheim A, Sherman ME, Bratti M, Wacholder S, Alfao M, Hutchinson M, Morales J, Greenberg MD, Lorincz AT. HPV DNA testing in cervical cancer screening: results from women in a high-risk province of Costa Rica. JAMA 2000; 283: 8793.
  • 20
    Ho GYF, Bierman R., Beardsley L, Chang CJ, Burk RD. Natural history of cervicovaginal papillomavirus infection in young women. N Engl J Med 1998; 338: 423428.
  • 21
    Ho GYF, Burk RD, Klein S, Kadish AS, Chang CJ, Palan P, Basu J, Tachezy R, Lewis R, Romney S. Persistent genital human papillomavirus infection as a risk factor for persistent cervical dysplasia. J Natl Cancer Inst. 1995; 187: 136571.
  • 22
    Nobbenhuis MAE, Walboomers JMM, Helmerhorst TJ, Rozendaal L, Remmink AJ, Rise E, Van Der Linden HC, Voorhorst FJ, Kenemans P, Meijer CJ. Relation of human papillomavirus status to cervical lesions and consequences for cervical-cancer screening: a prospective study. Lancet 1999; 354: 205.
  • 23
    Remmink AJ, Walboomers JMM, Helmerhorst TJM, Voorhorst FJ, Rozendaal L, Risse EK, Meijer CJ, Kenemans P. 1995. The presence of persistent high-risk HPV genotypes in dysplastic cervical lesions is associated with progressive disease: natural history up to 36 months. Int J Cancer 1995; 61: 30611.
  • 24
    Schlecht NF, Kulaga S, Robitaille J, Ferreira S, Santos M, Miyamura RA, Duarte-Franco E, Rohan TE, Ferenczy A, Villa LL, Franco EL. Persistent human papillomavirus infection as a predictor of cervical intraepithelial neoplasia. JAMA 2001; 286: 31063114.
  • 25
    Cuzick J, Terry G, Ho L, Hollingworth T, Anderson M. Human papillomavirus DNA in abnormal smears as a predictor of high-grade cervical intraepithelial neoplasia. Br J Cancer 1994; 69: 16717.
  • 26
    Josefsson AM, Magnusson P, Ylitalo N, Sorensen P, Qwarforth-Tubbin P, Andersen PK, Melbye M, Adami HO, Gyllenstein UB. 2000. Viral load of human papillomavirus 16 as a determinant for development of cervical carcinoma in situ: a nested case-control study. Lancet 2000; 355: 218993.
  • 27
    Ylitalo N, Sorensen P, Josefsson AM, Magnusson PK, Andersen PK, Ponten J, Adami HO, Gyllenstein UB, Melbye M. Consistent high viral load of human papillomavirus 16 and risk of cervical carcinoma in situ: a nested case-control study. Lancet 2000; 355: 21948.
  • 28
    Van Duin M, Snujders PJ, Schrijnamakers HF, Voorhorst FJ, Rozendaal L, Nobbenhuis MA, Van Den Brule AJ, Verheijen RH, Helerhorst TJ, Meijer CJ. Human Papillomavirus 16 load in normal and abnormal cervical scrapes: an indicator of CIN II/III and viral clearance. Int J Cancer 2002; 98: 5905.
  • 29
    Wright TC, Jr, Lorincz A, Farris DG, Richart RM, Ferenczy A, Mielzynska I, Borgatta I. Reflex human papillomavirus deoxyribonucleic acid testing in women with abnormal Papanicolaou smears. Am J Obstet Gynecol 1998; 178: 8626.
  • 30
    Liaw KL, Glass AG, Manos MM, Greer CE, Scott DR, Sherman M, Burk RD, Kurman RJ, Wacholder S, Rush BB, Cadell DM, Lawler P, et al. Detection of human papillomavirus DNA in cytologically normal women and subsequent cervical squamous epithelial lesions. J Natl Cancer Inst 1999; 91: 95460.
  • 31
    Rozendaal L, Walboomers JMM, Van der Linden JC, Voorhorst FJ, Kenemans P, Helmerhorst TJ, Van Vallegoijen M, Meijer CJ. PCR-based high-risk HPV test in cervical cancer screening gives objective risk assessment of women with cytomorphologically normal cervical smears. Int J Cancer 1996; 68: 7669.
  • 32
    Zielinski GD, Snijders PJ, Rozendaal L, Voorhorst FJ, Van der Linden JC, Runsinkl AP, De Schipper FA, Meijer CJ. HPV presence precedes abnormal cytology in women developing cervical cancer and signals false negative smears. Br J Cancer 2001; 85: 398404.
  • 33
    Villa LL, Sichero L, Rahal P, Ruiz A. Molecular variants analysis of human papillomavirus types 16 and 18 preferentially associated with cervical neoplasia. J Gen Virol 2000; 81: 295968.
  • 34
    Wallin KL, Wiklund F, Angstrom T, Bergman F, Stendahl U, Wadell G, Hallmans G, Dillner J. Type-specific persistence of human papillomavirus DNA before the development of invasive cervical cancer. N Engl J Med 1999; 341: 16338.
  • 35
    Monsonego J, Autillo-Touati A, Bergeron C, Dachez R, Liaras J, Saurel J, Zerat L, Chatelain P, Mottot C. Liquid-based cytology for primary cervical cancer screening: a multi-centre study. Br J Cancer 2001; 84: 360366.
  • 36
    Franco EL, Villa LL, Sobrinho JP, Prado JM, Rousseau MC, Desy M, Rohan TE. Epidemiology of acquisition and clearance of cervical human papillomavirus infection in women from a high-risk area for cervical cancer. J Infect Dis 1999; 180: 14151423.