Oral contraceptive pills are associated with artifacts in ThinPrep Pap smears that mimic low-grade squamous intraepithelial lesions

A broad-based approach to differentiate CIN from its mimics: The utility of in situ hybridization and immunohistochemistry

The hematoxylin and eosin slides of 100 consecutive cases diagnosed as CIN 1-2 were combined with 25 CIN 1 and 25 negative for CIN as documented by in situ HPV testing. The 150 cases were then reviewed blinded and scored as "CIN" or "negative for CIN". Each of the 50 controls was correctly scored. Of the 100 cases, 62 were diagnosed as CIN and the other 38 were scored as negative for CIN on re-review. Each of the CIN cases was positive for HPV as proven by the in situ detection of either HPV DNA or the L1 capsid protein. The 38 cases diagnosed as negative for CIN and 38 of the CIN cases were tested for HPV DNA by in situ hybridization and for a panel of biomarkers that included p16, Ki67, importin-β, exportin-5, and Mcl1 plus the L1 HPV capsid protein. Each of the 38 CIN cases was positive for HPV as well as each biomarker that localized towards the basal aspect of the lesion. Two of the 38 negative for CIN cases were positive for HPV DNA/L1 capsid protein and each of the biomarkers. The other 36 cases were negative for HPV DNA/L1 protein and each of the biomarkers showed baseline expression. Thus, 36% of the diagnoses of CIN 1-2 were incorrect and this could have been prevented with either in situ detection of the viral DNA/capsid protein or the immunohistochemistry detection of a panel of biomarkers that included p16, Ki67, importin-β, exportin-5, and Mcl1.

Gynecologic cytology on conventional and liquid-based preparations: a comprehensive review of similarities and differences

Liquid-based preparations (LBPs) have largely replaced conventional Papanicolaou smears (CPS) for cervical samples in the United States and in many other industrialized countries. The two FDA-approved LBP currently in use include ThinPrep (TP), (Hologic Inc., Bedford, MA) and SurePath (SP), (BD Diagnostic, Burlington, NC). Split-sample and direct-to-vial studies have shown that LBPs show an overall improvement in sample collection and processing, reduce artifacts that interfere in diagnosis, are more sensitive, can be utilized for ancillary tests and are a cost-effective replacement for CPS. Comparative analyses of diagnostic accuracy indicate that LBPs perform at least as well as CPS. However, the added advantages of standardized, automated preparations and screening, reduced unsatisfactory rate, improved specimen adequacy and ability to perform human papillomavirus (HPV) test, are enough to continue use of LBP. The cytologic features in LBP are similar to CPS with subtle differences, particularly in background information. There are also subtle differences between the two LBPs, SP and TP, which are reflective of different sampling devices, collection media, and processing techniques. Architecturally, LBP shows smaller cell clusters and sheets and more dyscohesion. Cytologically, enhanced nuclear features and smaller cell size are more prominent. Advances in liquid-based Papanicolaou's (Pap) test have lead to well-defined patient management guidelines by the American Society for Colposcopy and Cervical Pathology. Herein, we review these aspects of Pap test including, morphology, automation, ancillary tests (HPV and immunochemistry), pertinent QA/QC monitors, patient management guidelines, and review of pertinent literature.

The utility of HPV in situ hybridization and the PAS test in improving the specificity of the diagnosis of CIN 1

The histologic features of cervical intraepithelial neoplasia (CIN 1), caused by infection by the human papillomavirus (HPV), can overlap with those of its mimics that can lead to an over diagnosis of this sexually transmitted disease. In this study, 67 consecutive cervical biopsies that were diagnosed as CIN 1 from the surgical files of Ohio State University Medical Center were analyzed. Twenty controls (10 CIN 1 cervical biopsies and 10 normal cervical tissues) were also studied. The 87 biopsies were reevaluated blinded to the original diagnosis and the results were correlated with detection of HPV DNA by in situ hybridization and glycogen by the periodic acid solution (PAS)/PAS-D stain, respectively. HPV was detected by in situ hybridization in 55/67 cases (82%); no virus was evident in the negative controls whereas each of the 10 CIN 1 controls was virus positive. A PAS test demonstrated in the mature squamous component of the negative controls a strong signal in cells with prominent and uniform halos, which was lost with diastase treatment, indicative of abundant glycogen. The PAS/PAS-D tests in the CIN 1 lesions showed rare variable sized glycogen deposits in the dysplastic cells. Nine (15%) cases initially diagnosed as CIN 1 were HPV negative by in situ hybridization and had halolike cells that were strongly and uniformly positive for glycogen. This data underscores the value of glycogen and HPV analyses in improving the specificity of the diagnosis of CIN 1.

Correlation of Pap smear, cervical biopsy, and clinical follow-up with an HPV typing microarray system

A human papillomavirus (HPV) microarray system allows the determination of HPV type in clinical samples. The purpose of this study was to determine the presence of HPV in liquid-based Pap smears with the MyGene MyHPV Chip Kit HPV genotyping microarray test (MyGene Assay), and to correlate this with the cytology and biopsy diagnoses, clinical follow-up, HPV Hybrid Capture data, and HPV sequence analyses. Four hundred and two Pap smears (93 ThinPrep, 309 SurePath) were available for study. Correlation of HPV DNA detection by the MyGene Assay with the Pap smear diagnosis showed a detection rate of 19/97 (19%) for normal Pap smears, 181/242 (74%) for atypical squamous cells of undetermined significance (ASCUS), and 61/63 (97%) for squamous intraepithelial lesions (SILs). Biopsy data on 248 women were available. HPV was noted by the MyGene Assay in the Pap smear in 98/100 (98%) of the cases, for which the corresponding biopsy had been diagnosed as SIL, compared with 103/148 (69%) of the cases for which the biopsy had been negative for SIL. Clinical follow-ups were available for 200 women with ASCUS Pap smears. A significant increase was observed in the rate of biopsy-proven SILs in women with ASCUS Pap smears that were HPV-positive (63/66=95%) as compared with those that were HPV-negative (96/134=71%, P<0.05). The MyGene Assay and Hybrid Capture system gave equivalent results for all the categories studied, except for the presence of multiple infections, as determined by viral sequence analysis. Specifically, the Hybrid Capture system overestimated the presence of dual infection (low-risk and high-risk positive) by 48% and missed many cases of multiple infections, especially when 2 or more high-risk types were present. It is concluded that the MyGene Assay allows for the routine typing of HPV in liquid-based Pap smears, and that the presence of HPV DNA in ASCUS Pap smears is strongly correlated with a biopsy-proven SIL.

Determinants of human papillomavirus-negative, low-grade squamous intraepithelial lesions in the atypical squamous cells of undetermined significance/low-grade squamous intraepithelial lesions triage study (ALTS)

BACKGROUND

Although low-grade squamous intraepithelial lesions (LSIL) most often are the result of infection by human papillomaviruses (HPV), a small proportion of women with LSIL have negative HPV tests. Using the Atypical Squamous Cells of Undetermined Significance/LSIL Triage Study (ALTS) population, the authors evaluated the significance of HPV-negative LSIL.

METHODS

Women with cytologic interpretations of LSIL by referral Papanicolaou (Pap) tests or enrollment ThinPrep tests (range, 1195-1476 women, depending on the specimen type and the reviewer) had HPV testing performed by both Hybrid Capture 2 and polymerase chain reaction (PCR)-based linear array for 27 HPV types.

RESULTS

Using 4 independent cytologic definitions of LSIL, only 3-11% of women with LSIL were found to have HPV-negative results on both HPV tests. The demographic characteristics of women with HPV-negative LSIL were consistent with those of a low-risk population; many were age > 35 years, and many reported no or only 1 recent sexual partner. The absolute risk of a histologic diagnosis of cervical intraepithelial neoplasia (CIN) Grade 3/carcinoma during the 2-year trial was lower for women with HPV-negative LSIL (range, 2-4%) compared with the absolute risks for oncogenic HPV-positive women with LSIL (range, 13-19%). However, at the next 6-month follow-up visit, 12%-32% of the women with HPV-negative LSIL had a positive HPV test. Finally, visual inspection of cervigrams demonstrated a clear association between a larger os and negative HPV test results compared with women who had HPV-positive LSIL. This may have influenced HPV sample adequacy.

CONCLUSIONS

Based on the ALTS data, the authors found no evidence to support the existence of HPV-negative LSIL as a distinct biologic entity related to the risk of cervical carcinoma. Such results appear to represent cytologic misinterpretations or falsely negative HPV tests.