Self-sampling for human papillomavirus (HPV) testing as cervical cancer screening option. Experience from the LAMS study

PURPOSE

To compare Hybrid Capture II (HC2) in detecting high-risk (HR) HPV in patient-collected vaginal samples with those obtained using gynaecologist collected samples.

METHODS

Patients were submitted to Pap smears, visual inspection with acetic acid (VIA) and HC2 for hr-HPV.

RESULTS

A total of 1,081 HC2 tests for HR-HPV were performed: 770 (71.2%) samples were collected by a physician and 311 (28.8%) were self-collected by the patients. In detecting any cervical lesion, the sensitivity of HC2 collected by a physician was higher (92.86%) than that (37.5%) in the self-sampling group. Negative predictive value (NPV) was high for both, 99.69% and 93.75%, respectively. Using the CIN2 cutoff, performance of HC2 was significantly improved: 92.9% and 62.5%, respectively. HC2 specificity for any cervical lesion and for CIN2 or higher were close to 90% in both groups.

CONCLUSIONS

Self-sampled HPV testing is a powerful option to increase the detection of cervical lesions in women segregated from prevention programs.

Cell proliferation activity unrelated to COX-2 expression in ovarian tumors

The objective of this study was to assess the expression of Cyclooxygenase-2 (COX-2) and cell proliferation activity (Ki67 expression) in benign, borderline, and malignant serous and mucinous ovarian tumors. Expression of COX-2 and Ki67 proteins were evaluated by immunohistochemistry, in paraffin-embedded sections of ovarian epithelial tumors. The study included 113 serous (67 benign, 15 borderline, and 31 malignant) and 85 mucinous (48 benign, 28 borderline, and 9 malignant) tumors, removed from women who underwent laparotomy between January 1997 and December 2003. From benign to malignant tumors, there was a progressive positive trend in COX-2 expression in both serous and mucinous tumors, more evident in mucinous ones (P < 0.001). Comparing histologic types, COX-2 expression was more prominent in serous than in mucinous benign tumors (P < 0.01), but this difference was not significant in the borderline (P= 0.11) or malignant categories (P= 0.71). There was a progressive Ki67 positivity in line with the tumor histologic gradient for both serous (P < 0.01) and mucinous lesions (P < 0.01), but this increasing expression did not correlate with COX-2 expression in the present series (P= 0.78). There was a higher COX-2 expression in serous ovarian adenomas than in mucinous ones. COX-2 positivity increases in line with the morphologic gradient, from benign to malignant in both histologic types, but it was more prominent in mucinous lesions, pointing to different oncogenic pathways related to different histologic types. A correlation between the expression of COX-2 and Ki67 was not found, suggesting that COX-2 may be required for carcinogenesis, but this pathway is not responsible for cell proliferation in ovarian tumors.

Prediction of high-grade cervical disease with human papillomavirus detection in women with glandular and squamous cytologic abnormalities

The objective of this study was to assess whether human papillomavirus (HPV) detection with hybrid capture II (HC II) can help predict the presence and the nature, glandular or squamous, of histologic cervical lesions in women referred due to atypical glandular cells (AGC) or high-grade squamous intraepithelial lesion (HSIL). A total of 247 women were included. Referral Pap smears comprised AGC (51 cases), AGC plus HSIL (28 cases), adenocarcinoma in situ (10 cases), and HSIL (158 cases). All patients were tested for high-risk HPV with HC II and had a histologic assessment of their cervix. Histologic analysis showed 38 women with (15.3%) cervicitis, 194 with (75.5%) squamous lesions, and 15 with (9.2%) glandular neoplasia. The overall rate of high-risk HPV detection was 77%. Almost 70% of AGC-HPV-negative patients did not have a pathologically proven cervical neoplasia, whereas 76% of women with AGC-HPV-positive result were diagnosed with a squamous or glandular neoplasia. Most (95%) of the lesions in patients with AGC-HSIL were of squamous nature, and HPV detection did not contribute to their differentiation from glandular lesions. We conclude that in women with AGC, HPV positivity strongly correlated with the presence of glandular or squamous cervical lesion but did not help distinguishing women with squamous from those with glandular neoplasia.

Human papillomavirus testing as an optional screening tool in low-resource settings of Latin America: experience from the Latin American Screening study

Hybrid capture II (HC II) test for oncogenic human papillomaviruses (HPV) was carried out in a cohort of 4284 women at their first clinical visit. Overall prevalence of HPV was 17.1%, decreasing with age from 33.9% among women below 20 years to only 11.0% among those older than 41 years. HPV prevalence was significantly higher among current smokers (odds ratio [OR] = 1.31; 95% CI 1.1-1.6), in women with two or more lifetime sexual partners (OR = 1.9; 95% CI 1.6-2.4), and those women with two or more sexual partners during the past 12 months prior to examination (OR = 1.6; 95% CI 1.2-2.2). HPV detection increased in parallel with increasing cytologic abnormality, being highest in women with high-grade squamous intraepithelial lesion (P= 0.001). Specificity of the HPV test in detecting histologically confirmed cervical disease was 85% (95% CI 83.9-86.1). Sensitivity of the HPV test in detecting histologic abnormalities increased in parallel with disease severity, ranging from 51.5% for cervical intraepithelial neoplasia (CIN) 1 to 96.5% for CIN 3 and 100.0% for cancer, with respective decline of positive predictive value. These data suggest that HPV testing with HC II assay might be a viable screening tool among this population with relatively high prevalence of cervical disease.

Atypical glandular cells: criteria to discriminate benign from neoplastic lesions and squamous from glandular neoplasia

OBJECTIVE

To evaluate the presence of some criteria in cervical smears with atypical glandular cells and their correlation with histological patterns to identify pre-neoplastic and neoplastic lesions.

METHODS

Seventy-three women referred with an atypical glandular cell smear, who had undergone conization or hysterectomy, were included in this study. Referral Pap smears were reviewed using the set of 27 cyto-morphological criteria that was correlated with the histological diagnosis.

RESULTS

Histological results showed intraepithelial or invasive neoplasia in 35 (48%) cases and benign lesions in 38 (52%) cases. After logistic regression and decision tree analysis an increased nuclear/cytoplasmic ratio and the presence of dyskeratotic cells were strongly associated with intraepithelial or invasive neoplasia and the differential cyto-morphological criteria for glandular lesions were decreased cytoplasm, irregular nuclear membranes and the presence of nucleoli.

CONCLUSION

The analysis of individual cyto-morphological criteria can better predict intraepithelial or invasive neoplasia and differentiate glandular from squamous lesions.

P53 and Bcl-2 as prognostic predictors in epithelial ovarian cancer

The study was designed to evaluate the prognostic importance of clinical and pathologic variables with p53 and Bcl-2 in epithelial ovarian cancer using multivariate analysis. Tumor tissues from 90 patients were analyzed immunohistochemically for p53 and Bcl-2 expression. Hazard ratios were calculated in univariate and multivariate survival analyses. Forty-two (47%) were considered positive for p53 expression and 18 (20%) were positive for Bcl-2. Positive expression for p53 was less frequent in patients in FIGO stage I (22%). Positive staining for Bcl-2 correlated significantly with the histologic type (P < 0.01). No direct correlations could be demonstrated between p53 and Bcl-2 expression and age or histologic grade. In univariate analysis, p53 and Bcl-2 expression were not significantly correlated with overall survival, disease-free survival, or progression time. FIGO stage III and IV and residual disease > or =2 cm3 after first surgery were significantly correlated with poor outcome in univariate analysis. FIGO stage retained their independent prognostic value in multivariate analysis. Neither p53 nor Bcl-2 had any significant influence on outcome in multivariate survival analysis. FIGO stage proved to be the only significant independent prognostic factor in epithelial ovarian cancer, although residual disease remains correlated with disease-free survival.

P53 and Bcl-2 as prognostic predictors in epithelial ovarian cancer

The study was designed to evaluate the prognostic importance of clinical and pathologic variables with p53 and Bcl-2 in epithelial ovarian cancer using multivariate analysis. Tumor tissues from 90 patients were analyzed immunohistochemically for p53 and Bcl-2 expression. Hazard ratios were calculated in univariate and multivariate survival analyses. Forty-two (47%) were considered positive for p53 expression and 18 (20%) were positive for Bcl-2. Positive expression for p53 was less frequent in patients in FIGO stage I (22%). Positive staining for Bcl-2 correlated significantly with the histologic type (P < 0.01). No direct correlations could be demonstrated between p53 and Bcl-2 expression and age or histologic grade. In univariate analysis, p53 and Bcl-2 expression were not significantly correlated with overall survival, disease-free survival, or progression time. FIGO stage III and IV and residual disease ≥2 cm3 after first surgery were significantly correlated with poor outcome in univariate analysis. FIGO stage retained their independent prognostic value in multivariate analysis. Neither p53 nor Bcl-2 had any significant influence on outcome in multivariate survival analysis. FIGO stage proved to be the only significant independent prognostic factor in epithelial ovarian cancer, although residual disease remains correlated with disease-free survival.