SEARCH

SEARCH BY CITATION

Keywords:

  • SurePath;
  • ThinPrep;
  • human papillomavirus;
  • Hybrid Capture 2

Abstract

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. Conflict of Interest Disclosures
  8. References

BACKGROUND:

Human papillomavirus (HPV) DNA testing using Hybrid Capture 2 assay with ThinPrep Papanicolaou (Pap) collection is the only US Food and Drug Administration-approved method for the triage of women with atypical squamous cells of undetermined significance (ASCUS). Although SurePath Pap collection has been used for Hybrid Capture 2 HPV DNA testing, clinical validation of this method has been scarce.

METHODS:

From a cervical cancer-screening program in Mississippi, we analyzed data from screenings of 8380 women with ASCUS Pap results who underwent reflex Hybrid Capture 2 HPV DNA tests during a course of 4 years. Of these, 4145 were screened with the ThinPrep collection system, and 4235 were screened with SurePath. Results of follow-up biopsies within 3 months of Pap tests were available for the ThinPrep group (229 cases) and the SurePath group (455 cases). Hybrid Capture 2 positive rates and the follow-up biopsy results from both groups were compared.

RESULTS:

Hybrid Capture 2 detected high-risk HPV DNA in 68.8% of ThinPrep and 66.7% of SurePath-collected specimens (P = .37). Detection rates for CIN2+ and CIN3+ were also comparable between ThinPrep (21.4%, 3.1%) and SurePath (15.4%, 4.2%) using Hybrid Capture 2 (P = .06, P = .45). In ThinPrep-collected specimens, 4.4% were quantitatively insufficient for Hybrid Capture 2 testing. Significantly more equivocal Hybrid Capture 2 results were observed in SurePath (11.4%) than in ThinPrep specimens (3.2%). However, 67.4% of women with equivocal Hybrid Capture 2 results had negative 1-year Pap cytology follow-up in the SurePath group.

CONCLUSIONS:

Hybrid Capture 2 positive rates and CIN2-3 detection rates were comparable for the SurePath and ThinPrep Pap collection systems, thus supporting the use of SurePath for Hybrid Capture 2 testing. Cancer (Cancer Cytopathol) 2009. © 2009 American Cancer Society.

Human papillomavirus (HPV) infection with high-risk types is responsible for the development of >99% of cervical carcinomas and high-grade cervical precancerous lesions, that is, cervical intraepithelial neoplasia grade 2 or 3 (CIN 2/3).1-3 This well-established etiological link between HPV and cervical carcinomas/precancerous lesions has prompted clinical utilization of HPV DNA testing in cervical dysplastic clinics and cervical cancer prevention programs to predict cervical carcinoma/precancerous lesions. In the United States, HPV DNA testing has been integrated into cervical cancer screening programs and has become the standard of care in the triage of women with Papanicolaou (Pap) cytology test results of atypical squamous cells of undetermined significance (ASCUS).4 In addition, HPV DNA testing has been more frequently used along with Pap cytology testing for coscreening women aged ≥30 years.5

To date, Hybrid Capture 2 (Qiagen, Valencia, Calif) HPV DNA testing assay using ThinPrep Pap collection devices (Hologic, Marlborough, Mass) is the only US Food and Drug Administration-approved HPV DNA testing method in the triage of women with ASCUS Pap results and coscreening of women aged ≥30 years by Pap cytology testing. The Hybrid Capture 2 assay uses DNA:RNA hybridization with RNA probes that are complementary to 13 high-risk HPV genotypes (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68). The specific HPV DNA:RNA hybrids formed in solution can be captured by alkaline phosphatase-conjugated antibodies bound to the solid-surface of a microtiter plate. After chemiluminescent reagent is added, the captured HPV DNA:RNA hybrids emit luminescent signals that can be detected by a luminometer. The intensity of the luminescent signals is proportional to the quantity of the targeted DNA in the specimen, and the results are reported as relative light units (RLUs). Because Hybrid Capture 2 is not a target amplification assay and uses DNA:RNA hybridization, cross-contamination can be minimized. Hybrid Capture 2 testing is also fully automated, with high interlaboratory reproducibility.6 The Hybrid Capture 2 HPV DNA testing assay is by far the most extensively studied and documented HPV DNA testing assay. More importantly, it was clinically validated in the ASCUS low-grade squamous intraepithelial lesion (LSIL) triage study, a strictly controlled, randomized clinical validation.7 The ASCUS LSIL triage study demonstrated a higher clinical sensitivity in detecting CIN2-3 and a lower referral rate for colposcopy using reflex Hybrid Capture 2 HPV DNA testing compared with repeat Pap testing.7 In the United States, a few commercial liquid-based Pap collection devices have been used for Pap cytology testing as well as reflex HPV DNA testing. In addition to ThinPrep, the SurePath (BD Diagnostics, TriPath, Burlington, NC) Pap collection device is also used for Pap cytology testing and for Hybrid Capture 2 HPV DNA testing. Although the US Food and Drug Administration granted approval to use the SurePath Pap collection device for Pap cytology testing, and recently also approved the automated screening system (BD FocalPoint GS Imaging System), the Hybrid Capture 2 HPV DNA testing with SurePath Pap specimen is still under Food and Drug Administration evaluation. One major impediment is the lack of a randomized clinical validation study, such as the ASCUS LSIL triage trial. Such research requires a considerable budget for colposcopy and biopsy confirmation in a study population that is large enough to generate the statistical power to reach a conclusion. To use a non-Food and Drug Administration-approved SurePath Pap collection device, most institutions conducted their own validation testing.

To date, only 2 validation studies of Hybrid Capture 2 DNA testing with SurePath Pap specimens have been published.8, 9 The ASCUS LSIL triage trial and other well-documented studies using ThinPrep Pap collection have laid out a solid basis for Hybrid Capture 2 clinical application. Hybrid Capture 2 results using the SurePath collection system are usually compared with those in the ASCUS LSIL triage trial to evaluate the testing efficacy of Hybrid Capture 2 assay.8 In Mississippi, Hybrid Capture 2 HPV DNA testing for the triage of women with ASCUS has been implemented in the cervical cancer-screening program since 2003. From 2003 to 2005, the Mississippi State cervical cancer-screening program used the SurePath Pap cytology collection device for Hybrid Capture 2 HPV DNA testing. In 2005 the state shifted Pap cytology collection from SurePath to ThinPrep. This provided an opportunity to compare the 2 most commonly used Pap cytology collection systems. The large screening population was predominantly a cohort of young African American women. In this study, we evaluated Hybrid Capture 2 HPV DNA testing in SurePath and ThinPrep specimens from women with ASCUS results by comparing Hybrid Capture 2 positive rates and follow-up biopsies.

MATERIALS AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. Conflict of Interest Disclosures
  8. References

The institutional review board (IRB) at the University of Mississippi Medical Center approved this study (IRB 2008-0004). The IRB waived the requirement for written consent.

Study Population

From the Tri-Regional Cervical Cancer Screening Management Project sponsored by the Mississippi State Health Department at the University of Mississippi Medical Center, we retrospectively collected data from records of women who had Pap results of ASCUS and reflex HPV DNA testing using the Hybrid Capture 2 assay. From November 2003 to August 2005 (22 months), 124,300 women were screened using specimens collected by the SurePath collection system (Becton Dickinson, Franklin Lakes, NJ). Of these, 4235 women had Pap results of ASCUS and reflex Hybrid Capture 2 HPV DNA tests. After the switch from SurePath to the ThinPrep collection system, 114,979 women were screened from October 2005 to December 2007 (27 months). Of these, 4145 women who had ASCUS Pap results were tested for HPV DNA using Hybrid Capture 2 assay. These 8380 women were included in this study.

Liquid-Based Pap Cytology Testing

The SurePath Pap specimens were collected and processed according to the manufacturer's guidelines. The specimens were processed by centrifugal sedimentation through a density gradient to partially remove nondiagnostic debris and inflammatory cells. The concentrated cells were resuspended in 1000 μL of buffered deionized water. An aliquot of 200 μL of cell-rich suspension was then used to prepare a SurePath slide using a Prepstain slide processor (Becton Dickinson). The residual cell suspension (800 μL) was used for Hybrid Capture 2 HPV DNA testing.

The ThinPrep specimens were processed according to the manufacturer's guidelines using a ThinPrep 3000 processor (Hologic). In this process, part of the specimen passes through a 20-mm-diameter polycarbonate filter to remove nondiagnostic debris and inflammatory cells. A variable fraction of a specimen was used to achieve an optimal cell density for ThinPrep slides. Once the optimal cell density was obtained, the filter stamped these cells on a ThinPrep glass slide. The remaining specimen was used for Hybrid Capture 2 DNA testing.

Both SurePath and ThinPrep specimens were processed at the Department of Pathology, the University of Mississippi Medical Center. Following the Bethesda System, the Pap slides were screened by cytotechnologists and verified by cytopathologists at the Department of Pathology, the University of Mississippi Medical Center.

Hybrid Capture 2 HPV DNA Testing and Validation

The residual specimens from either SurePath or ThinPrep liquid-based Pap specimens were processed for Hybrid Capture 2 HPV DNA testing at the Department of Pathology, the University of Mississippi Medical Center. Hybrid Capture 2 testing was performed using high-risk HPV kit #5199-1220 (Qiagen) according to the manufacturer's recommendations. The high-risk HPV panel consisted of 13 types of HPV (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59, and 68). The RLU readings were obtained using a DML 2000 luminometer (Qiagen). The positive cutoff value was determined by the mean of the positive control samples. The ratio of the RLUs from the patient sample to the positive cutoff value was determined. For a ratio <1, the result was classified as negative. In the specimens with a ratio >1 but with an RLU <1000, the result was considered equivocal. For a ratio >1 with an RLU >1000, the result was considered positive.

In-house validation was conducted at the Molecular Pathology Lab, the University of Mississippi Medical Center before the implementation of Hybrid Capture 2 HPV DNA testing for SurePath specimens. A total of 111 cervical specimens were collected using a split-sample collection method for SurePath (Becton Dickinson) and PreservCyt (Hologic). Pap results were classified as normal, ASCUS, and squamous intraepithelial lesion. Hybrid Capture 2 testing was performed in Pap specimens collected in both Pap collection systems. By using PreservCyt as a standard, Hybrid Capture 2 results from SurePath specimens were evaluated for each patient. The SurePath yielded 1.8% false-positive results, 2.7% false-negative results, and 9.9% equivocal results. The cutoff limits for positive, negative, and equivocal categories were determined by taking all of this information into account. The women with equivocal Hybrid Capture 2 results were followed up for 1 year. The majority of these women tested negative for high-risk HPV DNA.

Follow-Up Biopsy and Pap Cytology

Follow-up biopsy specimens were processed and interpreted at the Department of Pathology, the University of Mississippi Medical Center. We retrieved and compared the results of the follow-up biopsies performed within 90 days (3 months) after Pap testing from 446 women in the SurePath group and 229 women in the ThinPrep group. The follow-up biopsies were classified as benign, CIN1, CIN2, and CIN3. No carcinoma case was observed in the study group. One-year follow-up Pap cytology results were collected and compared from 301 women from the SurePath group and 77 women from the ThinPrep group with equivocal Hybrid Capture 2 results.

Statistics and Data Analysis

Descriptive statistics were calculated. The Wilcoxon rank sum test was used to assess the difference in continuous variables between women using ThinPrep and SurePath Pap testing. The chi-square test was performed to assess the association between the tests and disease status. P values (2-sided test) of <.05 were considered significant. All computations were carried out using SAS version 8.0 (SAS Institute, Cary, NC).

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. Conflict of Interest Disclosures
  8. References

The Study Population

Ages ranged from 9 to 65 years in the SurePath group, with a mean of 23.0 years, and 12 to 62 years in the ThinPrep group, with a mean of 25.4 years. In the SurePath group of 4235 women, 2870 (67.8%) women were African American, 1174 (27.7%) were Caucasian, and 191 (4.5%) were from other ethnic populations, such as Hispanic American, Native American, or Asian American. In the ThinPrep group of 4145 women, 2635 women (63.6%) were Africa American, 1274 (30.7%) were Caucasian, and 236 (5.7%) were from other ethnic groups (Table 1). The age distribution is illustrated in Table 2. In the ThinPrep group, there were significantly fewer women ≤19 years of age. Slightly more women in the ThinPrep group were age ≥20 years.

Table 1. Distribution of Women With Atypical Squamous Cells of Undetermined Significance Papanicolaou Results and Reflex Hybrid Capture 2 Human Papillomavirus DNA Testing Using SurePath and ThinPrep Pap Collection Systems
 SurePath (%)ThinPrep (%)
Africa American2870 (67.8)2635 (63.6)
Caucasian1174 (27.7)1274 (30.7)
Other191 (4.5)236 (5.7)
Total4235 (100.0)4145 (100.0)
Table 2. Hybrid Capture 2 Human Papillomavirus DNA Testing Results in Women With Atypical Squamous Cells of Undetermined Significance Papanicolaou Results
Age, yNegativeEquivocal (%)Positive (%)Total (%)P
SPTPSPTPSPTPSPTP
  • SP indicates SurePath; TP, ThinPrep.

  • *

    This number does not include 185 TP specimens, which were quantitatively insufficient for HC2 testing.

<1919683153191051 (75.1)499 (83.0)1400 (33.0)601 (15.2)<.001
20-24254352191521227 (73.4)1226 (75.2)1672 (39.5)1630 (41.2).23
25-291462937834403 (64.3)549 (62.7)627 (14.8)876 (22.1).52
30-3494154358124 (49.0)201 (55.4)253 (6.0)363 (9.2).12
35-396512915859 (42.5)78 (36.3)139 (3.3)215 (5.4).24
>4081175121351 (35.4)87 (33.0)144 (3.4)275 (6.9).43
Total8361186484 (11.4)134 (3.2)2915 (68.8)2640 (66.7)4235 (100.0)3960 (100.0)*.37

HPV DNA Testing Using Hybrid Capture 2

HPV DNA was detected in 2915 (68.8%) SurePath specimens and 2640 (66.7%) ThinPrep specimens (Table 2). The Hybrid Capture 2 positive rates were not significantly different between the SurePath and ThinPrep groups (P = .37). When Hybrid Capture 2 results were stratified by age, Hybrid Capture 2 positive rates in the 2 groups were comparable in women aged ≥20 years (P = .12-.52). The Hybrid Capture 2 positive rate was significantly lower in women aged ≤19 years in the SurePath group (75.1%) than in the ThinPrep group (83.0%) (P<.001) (Table 2). In both the SurePath and ThinPrep groups, Hybrid Capture 2 positive rates decreased significantly with age, from 75.1% in women aged ≤19 years to 35.4% in women age ≥40 years in the SurePath group (P-trend <.001) and from 83.0% in women aged ≤19 years to 33.0% in women aged ≥40 years in the ThinPrep group (P-trend <.001).

Significantly more equivocal Hybrid Capture 2 results were observed in the SurePath cases (484, 11.4%) than in the ThinPrep group (134, 3.2%) (P<.001).

Specimens were quantitatively insufficient for Hybrid Capture 2 testing in 185 (4.4%) ThinPrep specimens, versus 0 SurePath specimens (P<.001).

Comparison of Follow-up Biopsy Results in Women With Positive Hybrid Capture 2 Results

Follow-up biopsies were available in 684 women who had Pap results of ASCUS and positive Hybrid Capture 2 HPV DNA testing results within 3 months (90 days) of Pap testing. Of these, 455 biopsies were from women from the SurePath group and 229 were from women in the ThinPrep group.

A comparison of biopsy diagnoses in the SurePath and ThinPrep groups is listed in Table 3. No significant difference was observed between the 2 groups whether CIN2+ (P = .06) or CIN3+ (P = .46) was used as a cutoff threshold.

Table 3. Comparison of Follow-Up Biopsies in Women With Atypical Squamous Cells of Undetermined Significance/Hybrid Capture 2 Positive Results
Biopsy ResultsSurePath (%)ThinPrep (%)P
  1. CIN indicates cervical intraepithelial neoplasia.

≥CIN270 (15.4)49 (21.4).064
≥CIN319 (4.2)7 (3.1).45
Total455 (100.0)229 (100.0) 

Comparison of Follow-Up Pap Cytology and Biopsy Results in Women With Equivocal Hybrid Capture 2 Results

Of 484 cases of SurePath Pap specimens with equivocal Hybrid Capture 2 results, 62.2% (301 of 484) had follow-up Pap tests, and 9.7% (47 of 484) had follow-up biopsy in 1 year. In ThinPrep specimens with equivocal Hybrid Capture 2 results, 1 year follow-up Pap tests and biopsies were performed in 57.5% (77 of 134) and 14.9% (20 of 134), respectively (Table 4). Of the 301 SurePath specimens with follow-up Pap tests, 203 (67.4%) were classified as negative for intraepithelial lesion or malignancy, whereas 60% (46 of 77) of ThinPrep specimens with equivocal Hybrid Capture 2 results were classified as negative for intraepithelial lesion or malignancy. The rates of abnormal Pap results including ASCUS, LSIL, and high-grade squamous intraepithelial lesion (HSIL) were comparable (P = .689) between the SurePath and ThinPrep groups.

Table 4. Comparison of Follow-Up Pap Tests and Biopsies in Women With Equivocal Hybrid Capture 2 Results
 SurePath (%)ThinPrep (%)P
  1. NILM indicates negative for intraepithelial lesion or malignancy; ASCUS, atypical squamous cells of undetermined significance; LSIL, low-grade squamous intraepithelial lesion; HSIL, high-grade squamous intraepithelial lesion; CIN, cervical intraepithelial neoplasia.

Pap cytology   
 NILM203 (67.4)46 (60).689
 ASC-US46 (15.3)15 (19) 
 LSIL46 (15.3)13 (17) 
 HSIL6 (2.0)3 (4) 
 Total301 (100.0)77 (100) 
Biopsy   
 Negative18 (38)3 (15) 
 CIN124 (51)12 (60).133
 CIN2-35 (11)5 (25).204
 Total47 (100)20 (100) 

Follow-up biopsies showed 24 cases of CIN1 and 5 cases of CIN2-3 in the SurePath group and 12 cases of CIN1 and 5 cases of CIN2-3 in the ThinPrep group. Similarly, no significant differences were observed between SurePath and ThinPrep in follow-up biopsies with abnormal results (P = .204) (Table 4).

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. Conflict of Interest Disclosures
  8. References

In this retrospective study, we compared the ThinPrep and SurePath collection system results for Hybrid Capture 2 HPV DNA testing. We collected data encompassing 4 years of records (totaling 8380 women) who had Pap results of ASCUS and reflex Hybrid Capture 2 HPV DNA testing in a state-sponsored cervical cancer-screening program. Of these women, 4235 had Pap tests using specimens collected by the SurePath system, and 4145 had Pap tests using specimens collected by the ThinPrep system. We observed no statistically significant differences in Hybrid Capture 2 positive rates in the 2 groups as a whole or by age except for women aged ≤19 years. In the 684 follow-up biopsies performed within 3 months of Pap cytology testing, no significant difference in detection rates for CIN2+ or CIN3+ were observed between the SurePath and ThinPrep groups. On the basis of these observations, we conclude that SurePath is comparable to ThinPrep Pap collection using Hybrid Capture 2 HPV DNA testing assay for predicting CIN2-3 in women with ASCUS results.

Hybrid Capture 2 HPV DNA testing in the triage of women with ASCUS Pap results has proven to have higher sensitivity in predicting CIN2+ than repeat Pap cytology testing. In the ASCUS LSIL triage trial, the ThinPrep Pap collection device was used to collect cervical specimens. From 4 institutions in the United States, 3488 women with ASCUS Pap results were enrolled in the study. Hybrid Capture 2 HPV DNA testing had a sensitivity of 95.9% to detect CIN2+ and a positive predictive value of 19.6% when CIN2 was used as the cutoff threshold.7 In a pooled meta-analysis including 20 studies on the triage of ASCUS, Arbyn et al demonstrated an overall Hybrid Capture 2 sensitivity of 92.5% (95% confidence interval [CI], 90.1%-94.9%) and 95.6% (95% CI, 92.8%-98.4%) for detecting CIN2+ or CIN3+, respectively.10

Because the ASCUS LSIL triage trial is the most comprehensive randomized clinical validation study in the triage of women with ASCUS using the Hybrid Capture 2/ThinPrep method, it has frequently been used as a gold standard to evaluate Hybrid Capture 2 clinical validation in other liquid-based cytology specimens, such as SurePath. Recently, Ko et al, at Massachusetts General Hospital, compared Hybrid Capture 2 DNA testing in 2319 women with ASCUS Pap results using SurePath Pap collection with the ASCUS LSIL triage trial.8 The study showed a similar sensitivity (93.4%), negative predictive value (95.7%), and positive predictive value (17.5%) to predict CIN2+ compared with the ASCUS LSIL triage trial. The strength of their study is that biopsy data were available for both Hybrid Capture 2 positive (485) and negative (140) cases to generate calculated clinical sensitivity, specificity, positive predictive value, and negative predictive value.

In our study, the sensitivity, specificity, and positive predictive value of Hybrid Capture 2 testing to predict CIN2-3 cannot be estimated because of the lack of biopsy data from women with negative Hybrid Capture 2 results. Instead, we present a back-to-back comparison of the 2 liquid-based Pap collection methods in the same cervical cancer-screening program. In addition to showing similar Hybrid Capture 2 positive rates stratified by age between the 2 Pap collecting devices, we observed similar positive predictive values of Hybrid Capture 2 testing between SurePath (15.4%) and ThinPrep (21.4%) for detecting CIN2+ (P = .06). For detecting CIN3, which is more clinically relevant, our study also showed similar rates, with 4.2% in the SurePath group and 3.1% in the ThinPrep group (P = .45). By comparison, Ko et al observed 17.5% positive predictive value for CIN2+ in the Massachusetts General Hospital cohort, which was similar to that in our SurePath group. However, their positive predictive value for CIN3+ (7.8%) was higher than that of our SurePath group.8 We speculate that the lower rate of CIN3 in our study might be partially attributable to the younger age in our screening population, with a mean of 23.0 years in SurePath and 25.4 years in ThinPrep; CIN3 diagnosis usually peaks in women aged 25 to 30 years.11 By using CIN3+ as a cutoff threshold, our data in both SurePath (4.2%) and ThinPrep (3.1%) were comparable to those in the ASCUS LSIL triage trial (5.1%) and the pooled meta-analysis, including 20 studies by Arbyn et al (4.3%; 95% CI, 2.7%-5.9%).10

The Hybrid Capture 2 positivities in our SurePath and ThinPrep cohorts were at the high end compared with the range in published studies, with 68.8% in the SurePath group and 66.7% in the ThinPrep group. In the ASCUS LSIL triage trial with combined data from 4 institutions in the United States, the Hybrid Capture 2 positive rate was 56.1%.7 Ko et al, at Massachusetts General Hospital, observed a 40.1% Hybrid Capture 2 positive rate.8 The positive rate from the pooled 20 Hybrid Capture 2 studies was 42.3% (95% CI, 38.1%-46.3%). Our high Hybrid Capture 2 positivity may also be associated with the younger age of women in the screening program: 23.0 years old in the SurePath group and 25.4 years old in the ThinPrep group, compared with the average age of 29 years in the ASCUS LSIL triage trial.7 Age-dependent Hybrid Capture 2 HPV positivity has been well documented.12 The prevalence of HPV infection is higher among young women and decreases with age. It has been recognized that HPV infections in young women are more likely to be transient than persistent.12, 13 Therefore, HPV infection in younger women has significantly less clinical implication.

We also observed significantly decreased Hybrid Capture 2 positivities with age in both the SurePath and ThinPrep groups in women with ASCUS Pap results in our study. To reduce unnecessary colposcopy and biopsy, the protocol of the cervical cancer-screening program in Mississippi was modified in August 2006 in accordance with the 2006 Consensus Guidelines, with a recommendation that women aged ≤20 years with ASCUS Pap results no longer be screened with a reflex Hybrid Capture 2 HPV DNA test. As a consequence, we observed significantly fewer young women aged ≤ 20 years in the ThinPrep (15.2%) than in the SurePath group (33.0%). This change might also affect Hybrid Capture 2 positivity in this age group, which showed a significantly higher Hybrid Capture 2 positivity in the ThinPrep group (83.0%) than in the SurePath group (75.1%). We therefore believe that Hybrid Capture 2 data collected from this age group may not be comparable between the 2 groups because of the biased specimen selection for Hybrid Capture 2 testing in the ThinPrep group. The main limitation in our study, the inability to randomize because of its retrospective nature, means that the younger mean age in the SurePath group might result in a higher Hybrid Capture 2 positive rate in SurePath specimens of all ages, although it must be said that the difference we found was not statistically significant.

In our study, we observed a significantly higher insufficiency rate in ThinPrep (4.5%) than in SurePath (0) collection for Hybrid Capture 2 testing and a significantly higher rate of equivocal Hybrid Capture 2 results in the SurePath group (11.4%) than the ThinPrep group (3.2%). We speculate that differences in specimen processing contributed to the high insufficiency rate for Hybrid Capture 2 testing with ThinPrep and the high equivocal rate of Hybrid Capture 2 results with SurePath.

SurePath, which uses a density gradient to remove nondiagnostic debris, has the advantage of preserving cells proportionally for Hybrid Capture 2 testing. Because only a small aliquot of resuspended cells (⅕) is used for producing a cytology slide, a significant portion of the material is available for Hybrid Capture 2 testing. Therefore, specimen insufficiency for Hybrid Capture 2 testing is rare. Recently, Alsharif et al reported a low insufficiency rate (0.23%) for the SurePath Pap test.14 In contrast, ThinPrep, using a filter to remove nondiagnostic debris, was designed to optimize cell density for cytology slide preparation during specimen processing. Therefore, specimens with low cellularity may lead to sample insufficiency for a subsequent Hybrid Capture 2 testing.

In women with equivocal Hybrid Capture 2 results, Pap and biopsy follow-up results were comparable between the SurePath and ThinPrep collection methods. We speculate that the higher Hybrid Capture 2 equivocal rate in SurePath specimens (11.4%) compared with that in ThinPrep specimens (3.2%) might also be associated with differences in specimen processing. When nondiagnostic debris is removed using density gradient, a fraction of specimens containing cervical epithelial cells may also be removed, resulting in more cell loss than is the case for the filter-based method of the ThinPrep system. Knoepp et al, at Massachusetts General Hospital, observed 4% equivocal Hybrid Capture 2 results in SurePath, which is substantially lower than our equivocal Hybrid Capture 2 results.9 We cannot make a direct comparison between the 2 studies, because different cutoff values were used to define equivocal Hybrid Capture 2 results at the 2 institutions. In 191 cases of SurePath specimens with Hybrid Capture 2 equivocal results, Knoepp et al, retested Hybrid Capture 2 and observed 93% (178 of 191) positive rates in the first retest. In 1-year Pap cytology follow-up, Knoepp et al observed abnormal Pap results in 83% of ASCUS+ samples. Therefore, the authors conclude that the equivocal Hybrid Capture 2 results should be considered as positive Hybrid Capture 2 results based on their Hybrid Capture 2 retesting and follow-up results. In the cervical cancer-screening program in Mississippi, women with equivocal Hybrid Capture 2 results were followed up by annual Pap screening tests and not by retesting with Hybrid Capture 2. Although we do not have repeat Hybrid Capture 2 testing results for a comparison, the annual Pap cytology follow-up results in the 2 studies can be compared. From 301 women who had equivocal Hybrid Capture 2 results and the annual Pap cytology follow-up results, we observed predominantly negative for intraepithelial lesion or malignancy (67.4%) Pap results, whereas Knoepp's study showed predominantly ASCUS (79%) Pap results. Although we observed higher LSIL (15.3% vs 4.5%) and HSIL (2.0% vs 0.5%) than did Knoepp's study, collectively, we observed 32.6% with ASCUS compared with 84% with ASCUS in Knoepp's study. The clinical significance of equivocal Hybrid Capture 2 results in our study appears different from that in Knoepp's study. Because women with equivocal Hybrid Capture 2 results in our cervical cancer screening were managed with annual Pap tests without repeating Hybrid Capture 2 test, the higher equivocal Hybrid Capture 2 rate in SurePath specimens in our study did not result in a substantial increase in colposcopic follow-up. Predominantly negative Pap cytology follow-up in our study lends further support to the current clinical management for women who had ASCUS and equivocal Hybrid Capture 2 results, especially in a state cervical cancer screening program with a limited budget. Our observation with regard to equivocal Hybrid Capture 2 results from women with ASCUS suggests that in practice, the clinical significance of equivocal Hybrid Capture 2 results should be based on an in-house evaluation to determine the best clinical management for women in a given practice setting, either in dysplasia clinics or in cervical cancer screening programs.

Acknowledgements

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. Conflict of Interest Disclosures
  8. References

We thank Mr. Walter Pagel, Scientific Publications, The University of Texas M. D. Anderson Cancer Center, for editing the manuscript.

Conflict of Interest Disclosures

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. Conflict of Interest Disclosures
  8. References

The authors have no financial interests with any commercial products mentioned in this article.

References

  1. Top of page
  2. Abstract
  3. MATERIALS AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. Acknowledgements
  7. Conflict of Interest Disclosures
  8. References