The association of p16(INK4A) and fragile histidine triad gene expression and cervical lesions.

OBJECTIVE: This cross-sectional study was intended to assess the association between immunohistochemical analysis of p16 and fragile histidine triad (FHIT) and the presence of precancerous cervical lesions. MATERIALS AND METHODS: Women seen at Pérola Byington Hospital, São Paulo, Brazil, with histologically confirmed cervicitis (n = 31), cervical intraepithelial neoplasia (CIN) 1 (n = 30), CIN 2,3 (n = 30), and cervical cancer (n = 7) had also cervical material collected for liquid-based cytology, human papillomavirus Hybrid Capture 2 (HC2) test, and p16 and FHIT immunohistochemical reactions. RESULTS: p16 and FHIT reactions were scored as the following: <1%, 1% to 5%, >5% to 25%, and >25%. Receiver operating curve analysis was used to select p16 and FHIT score cutoffs for further categorical analyses. All but one of the 37 CIN 2,3/cancer cases had a p16 score of greater than 1% to 5%. Among the 61 cervicitis/CIN 1 cases, 46 (75%) had a p16 score lower than 1% to 5%. In contrast, no association of FHIT expression and severity of cervical lesions could be demonstrated in this data set. Receiver operating curve analyses suggested the score of 1% to 5% for p16 as the cutoff that best discriminates CIN 2,3/cancer from cervicitis/CIN 1. No cutoff for FHIT scores could be suggested with data set. CONCLUSIONS: p16, but not FHIT expression, has the potential to be used as complementary diagnostic tool to investigate human papillomavirus-induced cervical lesions, if these results are confirmed in larger studies.

Screening for cervical cancer in high-risk populations: DNA pap test or Hybrid Capture II test alone?

This study was designed to evaluate whether Hybrid Capture II (HC2) test alone refer women to colposcopy as appropriately as DNA Papanicolaou (Pap) test, in the context of a high-risk group of women using the recently validated DNACitoliq LBC system. Women with suspected cervical disease were included in this cross-sectional study at a tertiary center in São Paulo, Brazil, for further workup. All women had cervical material collected for LBC and HC2 for high-risk human papillomavirus (hrHPV)-DNA test. Irrespective of cytology and HC2 results, colposcopy, and cervical biopsy when applicable, was systematically performed. All tests were performed blindly. Sensitivity, specificity, positive and negative predictive values, and overall accuracy of both methods were computed in relation to histology. A total of 1,080 women were included: 36.4% (393/1080) had ACUS+, 10.2% (110/1080) were high-grade squamous intraepithelial lesions (HSIL) or cancer. Mean age was 33.5 years. All women underwent colposcopy, and cervical biopsies were performed in 38.4% (415/1080): 33% (137/415) of the biopsies were negative, 14.4% (155/415) were low-grade squamous intraepithelial lesions (LSIL), 10.7% (116/415) were HSIL, and 0.6% (7/415) were cancer. HC2 sensitivity to diagnose biopsy-proven HSIL was 100%. Because all HSIL cases had a positive HC2 test, sensitivity could not be improved by adding LBC. Specificity and positive and negative predictive values of DNA Pap were not significantly different from HC2 test alone when considering LSIL+ histology as "gold standard" and HSIL+ histology. As a screening strategy for women with high-risk for cervical cancer, DNA Pap test does not seem to add substantially to HC2 alone in terms of appropriately referring to colposcopy.

Pap test in a high-risk population comparison of conventional and liquid-base cytology.

Liquid-based cytology (LBC) is believed to have better sensitivity than conventional smears (CSs) and offers the possibility to perform molecular assay. The goal of this work was to study the performance of CS and LBC in a high-risk population and to compare the results with the hybrid capture (HC) II for high-risk human papillomavirus (HPV). Samples were collected from selected women with clinical suspicions of low genital tract lesion at Pérola Biygnton Hospital (São Paulo, Brazil). After CS preparation, the brush was introduced in the endocervix and a new sample was collected and rinsed in the preservative medium of the system. The residual material was used to HC2. From 925 cases, LBC was unsatisfactory in 4 (1.51%) cases and CS was unsatisfactory in 100 cases (10.81%); among theses cases HC2(+) reactions was observed in 54 (54%) CSs and 3 (21.4%) LBCs. Considering cases with atypia of undetermined significance (squamous and glandular), 85 (77.27%) cases from LBC and 44 (86.4%) from CS were positive for HC2 assay for high-risk DNA-HPV. The difference among the methods was not significant (P < 0.38). The diagnosis improvement of LBC in comparison with CS was 86% in satisfactory samples, 92.76% in undetermined atypical lesions (including glandular), 83% in positive low-grade squamous intraepithelial lesions (LSIL(+)), and 86.84% in high-grade SIL(+) (HSIL(+)). HC2(+) reactions were observed in 144 CS cases and 266 LBC cases with abnormalities. Our results have showed that LBC was superior to CS in a high-risk population to detect lesions with high concordance with HC2(+) reactions; CSs also exhibit a high concordance with HC2 assay but with inferior performance to detect lesions.