Histopathologic follow-up and human papillomavirus DNA test results in 290 patients with high–grade squamous intraepithelial lesion papanicolaou test results

Authors

  • Faye F. Gao MD, PhD,

    1. Department of Pathology, Magee-Womens Hospital of University of Pittsburgh medical center, Pittsburgh, Pennsylvania
    Search for more papers by this author
  • R. Marshall Austin MD, PhD,

    1. Department of Pathology, Magee-Womens Hospital of University of Pittsburgh medical center, Pittsburgh, Pennsylvania
    Search for more papers by this author
  • Chengquan Zhao MD

    Corresponding author
    1. Department of Pathology, Magee-Womens Hospital of University of Pittsburgh medical center, Pittsburgh, Pennsylvania
    • Magee-Womens Hospital of University of Pittsburgh medical center, 300 Halket Street, Pittsburgh, PA 15213
    Search for more papers by this author
    • Fax: (412) 641-1675


  • The authors acknowledge and thank Chunqiao Tian, PhD (Biostatistician, Precision Therapeutics, Inc), for his direction and review of the statistical analysis. The authors also acknowledge cytopathologists Jing Yu, MD, Olga Navolotskaia, MD, Tee Lang, MD, and Payam Arya, and cytotechnologists Joseph Beere, Amy Colaizzi, Rose Crane, Jonee Matsko, and Xin Zhao for participating in blinded slide reviews.

Abstract

BACKGROUND:

The study documents histopathologic outcomes and high-risk (hr) human papillomavirus (HPV) test results in a large cohort of patients with high-grade squamous intraepithelial lesion (HSIL) liquid-based cytology (LBC) Pap test results.

METHODS:

A total of 352 patients with HSIL results (338 cervical and 14 vaginal) who had hrHPV testing and 290 patients with biopsy follow-up were studied. hrHPV detection rates were compared at different ages, with or without an endocervical/transformation zone sample (EC/TZS), and for cervical and vaginal HSIL Pap smears. Histopathologic follow-up findings were also compared. hrHPV-negative HSIL slides were re-evaluated in a blinded manner.

RESULTS:

A total of 325 of 338 (96.2%) cervical HSIL and 12 of 14 (87.5%) vaginal HSIL tested hrHPV-positive. A total of 271 of 281 (96.4%) EC/TZS-positive cervical HSIL and 54 of 57 (94.7%) EC/TZS-negative cervical HSIL tested hrHPV-positive. The percentage of hrHPV-positive HSIL declined slightly with increasing age. 197 of 273 (72.3%) hrHPV-positive cervical HSIL had histopathologic cervical intraepithelial neoplasia (CIN) 2/3+ follow-up, including 8 squamous carcinomas, compared with 4 of 12 (33.3%) hrHPV-negative HSIL with CIN2/3 (no carcinomas). 167 of 241 (69.2%) EC/TZS-positive HSIL had CIN2/3+ follow-up, compared with 34 of 44 (77.3%) EC/TZS-negative HSIL. Equivocal HSIL morphology characterized some HPV-negative HSIL without CIN2/3+ follow-up.

CONCLUSIONS:

hrHPV was detected in LBC vials from 96.2% of 338 cervical HSIL and 85.7% of 14 vaginal HSIL. CIN2/3+ was significantly more likely with hrHPV-positive cervical HSIL than with hrHPV-negative cervical HSIL. Presence or absence of an EC/TZS did not significantly impact HSIL hrHPV or CIN2/3+ rates. Some hrHPV-negative HSIL cases may represent HSIL cytologic mimics. Cancer (Cancer Cytopathol) 2011;. © 2011 American Cancer Society.

High-grade squamous intraepithelial lesion (HSIL) cytology results are associated with a high risk for histopathologically verifiable precancerous cervical disease1-3 and also a much smaller, age-dependent risk for presence of an underlying invasive cervical carcinoma.4-7 Accordingly, consensus follow-up guidelines recommend timely colposcopic examination or a diagnostic excisional procedure for adults with HSIL Pap results, acknowledging that colposcopy alone may fail to sample a significant number of small or inconspicuous precancerous lesions.8 Adding to the challenge of assessing noncorrelating histopathologic follow-up findings after HSIL Paps, many high-grade lesions spontaneously regress, especially in younger adults.9-14 With the advent of Food and Drug Administration (FDA) -cleared routine Pap and high-risk (hr) human papillomavirus (HPV) DNA cotesting for women 30 and older,15, 16 an increasing number of women with HSIL Pap test results now also have hrHPV DNA test results available to further assess histopathologic follow-up findings and to potentially help inform follow-up management decisions.17 Nonetheless, reports on the range of expected histopathologic follow-up outcomes for large numbers of patients with HSIL Pap results based on widely used FDA-cleared liquid-based cytology (LBC) and from-the-vial hrHPV DNA testing remain limited. Some researchers have speculated that HPV clearance precedes lesional regression18; however, others caution that it remains uncertain whether persistent infections associated with disease progression are characterized by continuously detectible HPV or by a state of latency during which the virus is undetectable due to quantitative or qualitative changes in the infectious state.19 By studying several hundreds of patients with HSIL, hrHPV test results, and histopathologic follow-up over a 5-year period in a large academic US women's hospital practice, we sought to further characterize correlations of widely used FDA-approved, hrHPV DNA tests with histopathologic follow-up outcomes in patients with LBC HSIL results.

MATERIALS AND METHODS

Case Selection and Cytologic Methods

After obtaining Institutional Review Board approval from the University of Pittsburgh Medical Center (UPMC), a retrospective study was initiated. A computer-based search was carried out on the Magee-Womens Hospital (MWH) of UPMC database over a 62-month span, between July 2005 and September 2010, to identify LBC Pap tests interpreted as HSIL with hrHPV DNA testing from the LBC vial. The presence or absence of a endocervical/transformation zone sample (EC/TZS) in cervical Paps was documented. All specimens were processed and evaluated in the pathology laboratory at MWH and reported using current Bethesda System (TBS) 2001 terminology.20 All LBC were ThinPrep Pap tests (TPPT) (Hologic Corp., Marlborough, Mass) prepared according to manufacturer's specifications from PreservCyt samples using an automated processor (ThinPrep 3000). Staining of slides was done on a Sakura Tissue Tek Automated Slide Stainer (Sakura Finetek USA, Torrance, Calif) according to an FDA-approved manufacturer's protocol. Location guided computer-assisted screening of TPPT slides was accomplished using the ThinPrep Imaging System (TIS).21 EC/TZS status in cervical Paps was determined according to the method described in TBS 2001, that is, at least 10 well-preserved endocervical or squamous metaplastic cells, singly or in clusters. In all cases in which the 22 image-guided fields did not contain enough endocervical or squamous metaplastic cells, the cytotechnologists manually rescreened the entire TPPT slide to determine EC/TZS status using the same criteria. The MWH cytopathology laboratory is a large subspecialized academic hospital laboratory that usually reports more than 100,000 Pap tests per year from a large, integrated hospital health system that serves a metropolitan area with a significantly older age population profile than the national average.22 UPMC is a large, integrated private health system in which Pap tests are collected by a highly diverse group of clinical providers that includes gynecologists, family physicians, internists, nurse practitioners, physician assistants, and house-staff trainees. The reporting profile of the laboratory is documented in numerous recent publications.23-31

HPV Test Methods

For the study group of patients with both HSIL LBC and from-the-vial HPV test results, hrHPV testing was ordered by clinicians using 1 of 2 available ordering protocols: routine hrHPV cotesting with Pap testing in women 30 and older, or hrHPV DNA cotesting regardless of age. hrHPV DNA testing on residual ThinPrep vial fluid used the commercially available FDA-approved Hybrid Capture II (HC2) method32 (Qiagen Corp., Hinden, Germany). Results were categorized as either HC2-positive or HC2-negative based on a threshold of 1 pg/mL HPV DNA.

Histopathologic Follow-Up

Histopathologic follow-up included endocervical curettage, cervical biopsy, vaginal biopsy for vaginal HSIL Paps, and cervical conization by loop electrosurgical excision procedure (LEEP) or cold-knife conization. The time elapsed from Pap testing until colposcopic examination and cervical biopsy, follow-up procedures, and histopathologic findings were abstracted from CoPath files. For women undergoing 2 or more procedures during the follow-up period, only the most severe histologic diagnosis was recorded. The most severe histologic diagnosis was recorded as “endpoint” for the follow-up duration count. In this study, cervical intraepithelial neoplasia (CIN)/vaginal intraepithelial neoplasia (VAIN) lesion terminology is used exclusively for histopathologic findings, whereas Bethesda low-grade squamous intraepithelial lesion (LSIL)/HSIL terminology is used exclusively for Pap test results. At MWH, all histopathologic diagnoses of CIN/VAIN 2/3 are confirmed by a documented review by a second pathologist.33 Liberal elective use of immunohistochemical staining with P16 and Ki-67 is also used by staff pathologists to increase the reliability of CIN/VAIN 2/3 diagnoses.34

Statistical Analysis

hrHPV detection rates were compared between different groups, including cervical vs. vaginal HSIL Paps, EC/TZS present versus EC/TZS absent HSIL Paps, and according to age group, using the Pearson chi-squared or Fisher exact test. The Mantel-Haenszel chi-squared test was applied to assess the association of hrHPV positive/negative, EC/TZS presence/absence, and age grouping with follow-up results since the histopathologic finding was an ordinal outcome. All analyses were conducted on the SAS 9.1 system (SAS Institute Inc., Cary, NC). A P value <.05 was considered statistically significant.

RESULTS

Prevalence of hrHPV DNA

A total of 352 patients with HSIL LBC results (338 cervical and 14 vaginal) who also had hrHPV testing were included in this study. The median age of women with cervical HSIL was 33 years (range, 16-75 years) and 55 years (range, 41-93 years) for vaginal HSIL. The overall hrHPV detection rate was 95.7% (95% CI, 93.1%-97.6%; Table 1). hrHPV was detected slightly more often in cervical samples compared with vaginal samples; however, this difference was not statistically significant (96.2% vs 85.7%; P = .115). Furthermore, for cervical HSIL, there was no difference in hrHPV detection rate among women with a EC/TZS compared with patient samples without a EC/TZS (96.4% vs 94.7%; P = .466). For women with cervical HSIL, the age-stratified hrHPV detection rates are shown in Table 2. Although hrHPV DNA detection rates declined slightly with increasing age, there was no statistically significant variation in the hrHPV detection rates among the different age groups (P = .501).

Table 1. hrHPV Detection Rates With Cervical HSIL Paps With or Without a Endocervical/Transformation Zone/Sample and With HSIL Vaginal Paps (n = 352)
HSIL Pap SitesEC/TZSNo. of PatientsNo. of hrHPV+% hrHPV+
  1. Abbreviations: EC/TZS, endocervical/transformation zone sample; hrHPV, high-risk human papillomavirus; HSIL, high-grade squamous intraepithelial lesion; N/A, not applicable; P = .115 for cervical HSIL vs vaginal HSIL (Fisher exact test); P = .466 for EC/TZS + vs EC/TZS- (Fisher exact test).

CervicalPresent28127196.4%
Absent575494.7%
All33832596.2%
VaginalN/A141285.7%
Table 2. hrHPV Detection Rates by Age Group With Cervical HSIL Paps (n=338)a
Age Group (years)No. of PatientsNo. of hrHPV+% hrHPV+
  • Abbreviations: hrHPV, high-risk human papillomavirus; HSIL, high-grade squamous intraepithelial lesion.

  • a

    P = .501 for comparing hrHPV detection rates across age groups.

<201212100%
20-29969497.9%
30-3914113696.5%
40-49464393.5%
≥50434093.0%

Histopathologic Follow-Up Results

Of 338 women with cervical HSIL Pap tests, 285 (84.3%) had at least 1 follow-up cervical biopsy result during the study period. The follow-up histopathologic results are shown in Table 3. Overall, CIN2/3+, CIN1 and absence of CIN (negative) diagnoses were rendered respectively in 201 (70.5%), 57 (20.0%) and 27 (9.5%) of the 285 patients. CIN2/3+ histopathologic follow-up diagnoses were significantly more common among women with hrHPV-positive HSIL results when compared with women with hrHPV-negative HSIL results (P = .009). In the hrHPV-positive HSIL group, 197 of 273 (72.2%) women had a subsequent histopathologic diagnosis of CIN2/3+, including 8 cases of invasive cervical squamous carcinoma, whereas only 4 of 12 (33.3%) patients in the hrHPV-negative HSIL group had a follow-up histopathologic diagnosis of CIN2/3. No cervical carcinomas were documented among the 12 patients with hrHPV-negative HSIL. Follow-up outcomes did not appear to be significantly affected by age. When patients were stratified into 2 age groups (<30 years vs ≥ 30 years), 57 of 85 (67.1%) women younger than 30 years had CIN2/3+ follow-up diagnoses compared with 144 of 200 (72.0%) women 30 years and older (P = .403). When patients were stratified into <50 years and ≥ 50 years 2 age groups, 175 of 241 (72.6%) women younger than 50 years had CIN2/3+ follow-up diagnoses compared with 26 of 44 (59.1%) women 50 years and older. Although the CIN2/3 detection rate was higher in women younger than 50 years than in older group, the difference was not statistically significant (P = .070) due to the small number of cases for 50 years and older. Histopathologic follow-up outcomes were also not statistically significantly different when comparing HSIL Paps with or without a EC/TZS (Table 4). Of 241 women with HSIL results and a EC/TZS, 167 (69.3%) had CIN2/3+ diagnoses. Among 44 women with HSIL results and absence of a EC/TZS, 34 (77.3%) had CIN2/3+ diagnoses (P = .259).

Table 3. Histopathologic Follow-Up Diagnoses After hrHPV-Positive Cervical HSIL and After hrHPV-Negative Cervical HSIL (n = 285)a
hrHPV ResultsNo. of PatientsFollow-Up Histopathalogic Diagnoses No. of Patients (%)
CIN2/3+CIN1Negative
  • Abbreviations: CIN, cervical intraepithelial neoplasia; hrHPV, high-risk human papillomavirus; HSIL, high-grade squamous intraepithelial lesion.

  • a

    Includes 8 invasive squamous cell carcinomas.

  • a

    P = .009 for comparing histopathologic findings with hrHPV+HSIL vs. hrHPV-HSIL (exact Mantel-Haenszel chi-squared test).

Positive273197* (72.2%)52 (19.1%)24 (8.8%)
Negative124 (33.3%)5 (41.7%)3 (25.0%)
All285201 (70.5%)57 (20.0%)27 (9.5%)
Table 4. Histopathologic Diagnoses after Cervical HSIL Paps with or without a Endocervical/Transformation Zone Sample (n=285)a
EC/TZSNo. of PatientsFollow-Up Histologic Result No. of Cases (%)
SCCCIN3CIN 2CIN1Negative
  • Abbreviations: CIN, cervical intraepithelial neoplasia; EC/TZS, endocervical/transformation zone sample; HSIL, high-grade squamous intraepithelial lesion; SCC, squamous cell carcinoma.

  • a

    P = .259 for comparing histologic finding between TZ/ECS+ vs. TZ/ECS- (Mantel-Haenszel chi-squared test).

EC/TZS +2415 (2.1)68 (28.2)94 (39.0)52 (21.6)22 (9.1)
EC/TZS -443 (6.8)14 (31.8)17 (38.6)5 (11.4)5 (11.4)

The average follow-up duration for these 285 biopsied patients (the time period between the HSIL Pap test and the last biopsy) was 5.0 months, ranging from 0.2 to 50 months. In addition to cervical biopsy, other histopathologic procedures included 82 LEEP/cone diagnostic excisional procedures and 26 hysterectomies. The average time period between the HSIL Pap test and initial diagnosis of CIN 2/3+ was 3.4 months, ranging from 0.2 month to 48 months. A total of 186 of 201 (92.5%) of histopathologic CIN 2/3+ diagnoses were rendered within 6 months, and 171 (85.1%) were diagnosed within 3 months.

Five of 14 women with vaginal HSIL Pap tests had histopathologic follow-up results, with 4 VAIN 2/3+ diagnoses in the subsequent vaginal biopsies. Due to the limited number of vaginal samples and the different site of origin, this group of patients was analyzed separately from the follow-up analyses of patients with cervical samples.

Blinded secondary review interpretations of HPV-negative cervical HSIL Paps Slides from 11 of 12 women with HPV-negative cervical HSIL Paps and histologic follow-up results were available for review. Three cytopathologists and 6 cytotechnologists each reviewed these 11 cases. All reviewers were blinded to patient identification (except for patient age), history, original cytologic interpretation, biopsy interpretation, and other reviewers' interpretations. Reviewers were asked to evaluate and interpret these slides as they did in daily practice and reviewers did not know the purpose of the study. Table 5 lists the age, final gynecologic procedure, histopathologic diagnosis, and the secondary review results by 9 reviewers. Figure 1 shows photomicrographs for 4 of these cases. Some slides, originally interpreted as HSIL with no CIN detected on histopathologic follow-up, in retrospect showed high nuclear/cytoplasmic ratios but had less pronounced chromatin and nuclear membrane abnormalities and were judged to represent possible false-positive HSIL mimics.

Figure 1.

Cellular morphology in 4 cases reported as high-risk (hr) human papillomavirus (HPV) -negative high-grade squamous intraepithelial lesion (HSIL). (A) Case 6. Original magnification, ×600. Cluster of small cells with marked nuclear hyperchromasia and nuclear membrane irregularity and unevenly distributed coarse chromation. Cervical biopsy confirmed cervical intraepithelial neoplasia (CIN) 3. (B) Case 3. Original magnification, ×600. Cluster of small cells with increased normal/cancerous (N/C) ratio. The cells showed dense metaplastic squamous cytoplasm. No CIN lesion was identified based on either a cervical biopsy or a follow-up loop electrosurgical excision procedure (LEEP) biopsy. (C) Case 5. Original magnification, ×600. Cohesive cluster of small cells with increased N/C ratio. Irregularity of nuclear membranes and hyperchromasia of nuclei are minimal (see case A). Follow-up by LEEP biopsy was negative for CIN. (D) Case 7. Original magnification, ×400. Cluster of large cells with nuclear enlargement and increased N/C ratio. Nuclear hyperchromasia and nuclear membrane irregularity may in retrospect be insufficient for an interpretation of HSIL. Follow-up histopathologic diagnosis was CIN1.

Table 5. Blinded Secondary Reviews of hrHPV-Negative 11 HSIL Paps
Case NoAge (years)Diagnostic ProcedureHistologic diagnosisHSILASC-HLSILASC-US
  1. Abbreviations: ASC-H, atypical squamous cells, cannot exclude high-grade squamous intraepithelial lesion; ASC-US, atypical squamous cells of undetermined significance; BX, biopsy; CIN, cervical intraepithelial neoplasia; ECC, endocervical curettage; HSIL, high-grade squamous intraepithelial lesion; LEEP, loop electrosurgical excision procedure; LSIL, low-grade squamous intraepithelial lesion.

134BX, ECCCIN381
236BX, ECC, LEEP, TAHCIN345
323BX, ECC, LEEPNegative72
454BX, ECC, LEEPCIN1333
550ECC, LEEPnegative3231
669BX, ECC, LEEP, TAHCIN39
749BX, ECC, TAHCIN1171
823BX, ECCCIN1315
948BX, ECCNegative2313
1030LEEP, ECCCIN1522
1131BX, ECC, LEEPCIN1351

DISCUSSION

Timely follow-up of patients with HSIL Pap results is regarded as mandatory because of the high risk of treatable potentially progressive high-grade squamous intraepithelial lesions35 and because of the risk of an underlying asymptomatic cervical carcinoma.4, 5

Accordingly, the Clinical Laboratory Improvement Amendments of 1988 (CLIA 88) requires laboratories to monitor patients with HSIL Pap results with no recorded follow-up.36 In this study, 8 of 338 patients (2.4%) with HSIL LBC findings were ultimately diagnosed with invasive cervical squamous carcinoma, similar to the roughly 2% rate reported in 1996 in a large College of American Pathologists' Q-Probe study, a nationwide survey reflecting experience with conventional Pap smears.5 With the introduction of LBC and computer-assisted screening, now widespread in the United States, detection of high-grade squamous intraepithelial lesions has in many studies been reported to have increased,37-39 even while a significant number of the detected high-grade CIN lesions appear to be small lesions which may be difficult to identify on colposcopy and which may have more uncertain progressive potential.40, 41 Both multiple biopsies42, 43 and diagnostic excisional procedures also appear to increase the likelihood of establishing a histopathologic diagnosis of CIN2/3 after abnormal cervical screening test results.3 This study presents data that hrHPV-positive HSIL results are associated with increased positive predictive value for presence of an underlying histopathologically verifiable CIN2/3+ lesion and suggests logically that efforts to document such lesions in hrHPV-positive HSIL cases should be particularly thorough. Our data also document for the first time that both hrHPV test results and histopathologic follow-up correlations in large numbers of patients with cervical HSIL are independent of cytologic sampling of the transformation zone or endocervical canal.

Although novel correlation data on vaginal HSIL presented in this report are limited, the similar correlation findings we report with HPV-tested vaginal HSIL cases and cervical HSIL cases mirror results we reported earlier in comparing patients with vaginal LSIL results, HPV results, and histopathologic follow-up with older patients with cervical LSIL.44 In both vaginal and cervical samples, a positive hrHPV test is associated with increased positive predictive value for follow-up histopathologic verification of a high-grade intraepithelial lesion. To the best of our knowledge, no similar data on the correlation of hrHPV testing with HSIL vaginal cytology and histopathologic follow-up findings have been previously reported.

Routine Pap and HPV cotesting in women 30 and older, FDA-approved in 2003, now provides additional information in a significant subset of women undergoing cytologic screening.17, 45 Several clinical trials have found that hrHPV testing can help detect some additional cases of histopathologic CIN2/3+ not detected with either conventional Pap smears46, 47 or LBC.48, 49 The extent of this increased disease detection with hrHPV testing appears to reflect both the methodology and quality of the cytologic screening used.50 In this study, an hrHPV-positive result significantly increased the positive predictive value of HSIL cytology for histopathologic verification of CIN2/3 or cervical squamous carcinoma, when compared with findings in the small subset of women with hrHPV-negative HSIL results. Although adjunctive HPV test results are not included as a consideration in current initial HSIL follow-up management guidelines,8 hrHPV-positive HSIL findings appear to provide additional incentive in efforts to locate an HPV- associated epithelial lesion when initial biopsy findings show no diagnostic CIN changes to account for the positive hrHPV result.

On the other hand, an hrHPV-negative HSIL result presents several special challenges. The largest reported follow-up experience in women with hrHPV-negative HSIL results is from cotesting experience in women screened with conventional Pap smears at Kaiser Permanente in Northern California between 2003 and 2008 and tested for hrHPV by HC2 hrHPV tests separately collected in Standard Transport Medium (STM) tubes.50 Histopathologic CIN2/3+ diagnoses were established in HC2-negative cases in 46 of 996 (4.6%) patients with HSIL conventional Pap smear results, including 515 diagnoses of CIN3 and 1 diagnosis of cervical cancer. In the same report, histopathologic CIN2/3+ diagnoses were established in HC2-negative cases in 44 of 510 (8.6%) hrHPV-negative atypical squamous cells, cannot exclude high-grade squamous intraepithelial lesion (ASC-H) (equivocal HSIL) smear cases, including 14 diagnoses of CIN3 and 4 diagnosed cases of cervical cancer. In the Kaiser report, a negative hrHPV result had sufficiently limited negative predictive value in this conventional smear practice setting to lead the authors to conclude that hrHPV testing should not be used for triage after ASC-H (equivocal HSIL) conventional smear results in any age group.51 At least 1 study on hrHPV testing with ASC-H LBC results has come to similar conclusions.52 Other studies have reported different findings.26, 53

The full significance of negative hrHPV test results in the face of significantly abnormal cytology findings, such as HSIL, remains controversial. Our blinded retrospective cytology review findings support the possibility that some HSIL interpretations in hrHPV-negative cases may represent false-positive HSIL mimics. Others have reported that HSIL interpretations independently confirmed in secondary reviews are more likely to be histopathologically verified.54 Still others have speculated that a negative hrHPV test precedes regression of intraepithelial lesions.18 False-negative HC2 hrHPV test results in patients with high-grade lesions have also been associated with small high-grade lesions shedding few abnormal cells.55 In Kaiser's reported cotesting experience, 64 of 1275 (5%) of CIN3 diagnoses following abnormal (excluding atypical squamous cells of undetermined significance [ASC-US]) cytology findings were in patients with negative HC2 hrHPV test results.51 These findings are similar to the ALTS report that carcinogenic HPV was detected in 95.4% of 608 women diagnosed with histopathologic CIN3.56 By combining data from 3 Kaiser Northern California reports,51, 57, 58 it can be further estimated that 35 of 232 (15%) invasive cervical carcinomas in Pap and HPV cotested Kaiser patients 30 and older were diagnosed in patients with HC2-negative test results and that 53 of 232 (23%) carcinomas were diagnosed in patients with negative Kaiser conventional Pap smear results. Whether or not a negative hrHPV result has sufficient negative predictive value to serve as a stand-alone screening test, irrespective of the underlying degree of potentially detectible cytologic abnormality, remains a matter of active ongoing discussion.59, 60 Cytology and hrHPV cotesting do provide some additional margin of safety to minimize the possibility of underlying undetected significant disease, including some cases of hrHPV-negative but cytologically detectible CIN3 lesions, cervical cancers, and endometrial cancers.46, 60 Preliminary findings at MWH on a subset of previously screened cervical cancer patients are also consistent with these observations, documenting negative HC2 hrHPV results in 4 of 28 (14%) invasive cervical squamous carcinomas, all with detectible HPV DNA in tumor tissues.61 Postintegration viral copy number reduction and low viral loads in some cervical cancers and their precursors may in part account for the larger than expected numbers of hrHPV-negative cervical cancers and precancers reported now in several large series.62-65 Accordingly, no professional group currently advocates returning women with hrHPV-negative HSIL to routine screening.

CONFLICT OF INTEREST DISCLOSURES

The authors made no disclosures.

Ancillary