The impact of LSIL-H terminology on patient follow-up patterns: a comparison with LSIL and ASC-H.

The Papanicolaou (Pap) test category of "low grade squamous intraepithelial lesion, cannot exclude high grade squamous intraepithelial lesion" (LSIL-H) is not recognized by The Bethesda System but is commonly used. It is essentially an amalgamation of the official LSIL and ASC-H categories. Since these two categories have similar follow-up algorithms, the clinical utility of the combined LSIL-H category is unclear. We have therefore studied follow-up patterns for these three entities in our laboratory to determine the real-world impact of each in our patient population. We searched our pathology database over an 18-month period to find Pap tests (predominantly ThinPrep) interpreted as LSIL-H (137), LSIL (2,189), and ASC-H (101). Like other studies, we found that the discovery rate of high grade dysplasia in biopsies after LSIL-H (31.9%) was similar to ASC-H (35.3%) and was higher than LSIL (7.6%; P < 0.0001). In women with no previous history of dysplasia, the frequency of biopsy follow-up after the initial Pap test was significantly higher for LSIL-H (68.3%) than for LSIL (49.6%; P = 0.0038) and similar to ASC-H (62.3%). We also found that women with an initial negative biopsy or a biopsy positive for low grade dysplasia were more likely to undergo an additional biopsy if the initial Pap test was LSIL-H (36.2%) than if it was LSIL (18.2%; P = 0.0023). ASC-H (26.9%) had an intermediate rate. In our patient population, the use of the terminology LSIL-H is associated with follow-up biopsy patterns much more similar to ASC-H than to LSIL.

Only a small fraction of high-grade cervical lesions are discovered after an interpretation of atypical squamous cells of undetermined significance when using imager-assisted, liquid-based papanicolaou tests and the Bethesda 2001 system.

CONTEXT: Previous work has reported that most high-grade cervical neoplasia is seen in patients with preceding Papanicolaou test results of atypical squamous cells of undetermined significance. This information was based on conventional test results and the Bethesda 1991 reporting system and was determined before the current treatment guidelines. OBJECTIVE: Our objective was to perform a retrospective review of all histologically confirmed, high-grade cervical neoplasia to determine the diagnosis of the preceding liquid-based Papanicolaou test. DESIGN: A total of 189 histologically confirmed, high-grade cervical intraepithelial neoplasia (CIN) cases grade 2 and greater were identified for a 1-year period. RESULTS: Of the 189 cases, 10 (5.3%) had a previous diagnosis of atypical squamous cells of undetermined significance; 55 (29.1%) had low-grade squamous intraepithelial lesions; 31 (16.4%) had low-grade squamous intraepithelial lesions, unable to rule out a high-grade squamous intraepithelial lesion; 21 (11.1%) had atypical squamous cells, unable to rule out a high-grade squamous intraepithelial lesion; 68 (36%) had high-grade squamous intraepithelial lesions; 1 (0.5%) had atypical glandular cells; 1 (0.5%) had adenocarcinoma in situ; and 2 (1%) had invasive carcinoma. Combined "low grade" Papanicolaou test results (atypical squamous cells of undetermined significance and low-grade squamous intraepithelial lesion) preceded 51 of 103 cases of CIN 2 (49.5%) and 14 of 103 cases (13.6%) of CIN 3/cancer, whereas "high grade" Papanicolaou test results (atypical squamous cells, unable to rule out a high-grade squamous intraepithelial lesion; low-grade squamous intraepithelial lesions, unable to rule out a high-grade squamous intraepithelial lesion; high-grade squamous intraepithelial lesions; atypical glandular cells; adenocarcinoma in situ; and invasive carcinoma) preceded 52 of 103 CIN 2 cases (50.5%) and 72 of 103 CIN 3/cancer cases (69.9%). CONCLUSIONS: Our data show that we can now more-reliably predict high-grade dysplasia on routine Papanicolaou tests. Only a small fraction of histologically confirmed CIN 2/3 cases are found following a Papanicolaou test diagnosis of atypical squamous cells of undetermined significance.

Women >or=30 years of age with low grade squamous intraepithelial lesion (LSIL) have low positivity rates when cotested for high-risk human papillomavirus: should we reconsider HPV triage for LSIL in older women?

High-risk human papillomavirus (HR-HPV) testing for colposcopy triage of low grade squamous intraepithelial lesion (LSIL) is not recommended because of high positive rates in young women. It remains unclear whether HR-HPV testing may be useful for triage of older women. We compiled HR-HPV data for women aged >or=30 years with LSIL for the period March 1, 2006 to February 28, 2008. Follow-up cervical biopsy information was collected for the period March 1, 2006 to August 15, 2008. We used the Hybrid Capture II test performed on residual material from liquid-based Pap tests. Of 735 women, 254 had HR-HPV testing, and of these 144 had positive HR-HPV results. Among women with positive HR-HPV results 79 underwent biopsy (54.9%) and 11 had cervical intraepithelial neoplasia (CIN) 2 or 3 (13.9% of women with biopsy follow-up). A total of 481 women did not undergo HR-HPV testing, of whom 192 underwent biopsy (39.9%) and 11 had CIN 2 or 3 (5.7% of biopsied women [P = 0.04]). Among women who tested negative for HR-HPV and had follow-up biopsies, only one had a high grade lesion found (CIN 2). The overall HR-HPV positive rate in tested women >or=30 years old with LSIL was 56.7% if women who had reflex HR-HPV testing for ASC-US are included. The HR-HPV positive rate in residual material from Pap tests interpreted as LSIL was 63.8%. Among women >or=30 years of age with LSIL, CIN 2-3 is significantly more likely in HR-HPV positive women. Relatively few older women with LSIL test positive for HR-HPV. Colposcopy triage using HR-HPV may be justified in this population.

The utility of p16INK4a and Ki-67 staining on cell blocks prepared from residual thin-layer cervicovaginal material.

BACKGROUND: Cell blocks can be prepared from residual thin-layer cervicovaginal (ThinPrep) material and can be used in immunohistochemical staining assays for p16INK4a and Ki-67, which are surrogate markers related to human papillomavirus infection and cell proliferation, respectively. The objectives of the current study were 1) to investigate the feasibility and the role of cell block preparations in identifying significant neoplastic and preneoplastic lesions of the uterine cervix and 2) to assess the feasibility of using p16INK4a and Ki-67 immunohistochemical staining patterns on cell blocks to identify significant preneoplastic cervical lesions. METHODS: Cervicovaginal cytology specimens from 85 patients were analyzed. Cytologic diagnoses based on ThinPrep Papanicolaou test results were as follows: squamous cell carcinoma was diagnosed in 3 specimens, high-grade squamous intraepithelial lesions (HSIL) were diagnosed in 27 specimens, low-grade squamous intraepithelial lesions (LSIL) were diagnosed in 20 specimens, and atypical squamous cells of uncertain significance (ASCUS) were diagnosed in 11 specimens. Diagnoses of negativity for intraepithelial lesions or malignancy (NILM) were made in 24 specimens. Cell block sections were stained with hematoxylin and eosin and were immunostained with antibodies against p16INK4a protein and Ki-67 antigen. RESULTS: The cytomorphologic diagnoses made using cell block preparations were as follows: SCC in 2 specimens, HSIL in 20 specimens, LSIL in 30 specimens, NILM in 32 specimens, and no diagnosis in 1 specimen. In 62 cases (73%), the diagnoses made using cell block preparations were in agreement with the ThinPrep diagnoses. Immunostaining of cell blocks for p16INK4a and Ki-67 exhibited a statistically significant association (P < 0.05) with the presence of significant lesions on either cell block or ThinPrep analysis. CONCLUSIONS: To the authors' knowledge, p16INK4a has not been analyzed previously in ThinPrep cell blocks, and the correlation between Ki-67 expression and cell block diagnoses also has not been reported previously. The current results indicate that cell blocks prepared from residual ThinPrep material represent an additional reliable diagnostic tool in the evaluation of cervical samples. Furthermore, immunohistochemical studies may be helpful in differentiating significant preneoplastic changes from other cervical lesions, such as atrophy.