Abstract 5771: Performance of uterine cervix liquid based cytology, in patients with ASCUS by conventional pap smear

Abstract

Abstract Background: The diagnosis of uterine cervix premalignant lesions is ​​the goal of cervical cancer prevention programs; Conventional Pap smear (CPS) with its limitations remains the screening method used in these programs in many Latin-American countries. In Colombia, after the implementation of CPS as a primary screening test, a less than expected decrease in mortality from cervical cancer has been observed. Actions aimed to improving cytological diagnosis could reduce the number of patients followed and treated unnecessarily. The demonstration of high specificity for liquid-based cytology (LBC) to adequately categorize cases with ASCUS in CPS, will serve as support for public health authorities, in their decisions about the best method to the screening programs. Design: Were included all patients with ASCUS results in CPS, referred to a Cervical Pathology Unit in Cartagena, Colombia. Material for LBC was obtained and LBCs were interpreted by two pathologists, blind to previous CPS or colposcopy diagnosis. Colposcopy were classified as negative/normal and positive/abnormal, colposcopic abnormalities were biopsied. Histological analyzes were performed by two pathologists blind to CBL outcome. Pathology diagnosis used were: CIN1/HPV infection: low-grade squamous intraepithelial lesion (LSIL), CIN2/CIN3: high-grade squamous intraepithelial lesion (HSIL). Diagnostic gold standards used to calculate LBC performance were: normal/negative for SIL: if colposcopy did not reveal lesion or colposcopy was abnormal but biopsy revealed no lesion; abnormal/positive: biopsy diagnosis ≥CIN2. Sensitivity, specificity, positive (PPV), negative predictive values (NPV), ​​ and Kappa concordance and correlation test were performed. Results: A total of 114 cases taken from patients with ASCUS in CPS were included, 77 of them corresponded to patients ≥30 years old. LBC sensitivity and specificity to detect HSIL/≥CIN2 were 76.5% and 66.0% respectively; PPV and NPV were 28.3% and 94.1%, when the cut-off point was a cytological result of HSIL. The concordance with the histological study was k = 0.2492 (p= 0.01). LBC is possible to reduce doubtful results and colposcopy remissions. Using LBC there were fewer results of ASCUS which is the most common cytological abnormality in screening populations that imposes heavy burden for follow-up. Concordance in HSIL cases with histological study was 100%, but in general the accuracy measured by the kappa statistic is low. Conclusions: These results support the evidence that LBC is no more sensitive than CPS for detection of HSIL/≥CIN2. The sensitivity and predictive values for HSIL/≥CIN2 were not significantly different between CPS and LBC, but, taking into account the benefits of unsatisfactory minors, and material availability for molecular biomarkers and HPV detection in the same sample, it could be used in cervical cancer program management guidelines. Citation Format: Ines Benedetti, Yoled Vizcaino, Lia Barrios. Performance of uterine cervix liquid based cytology, in patients with ASCUS by conventional pap smear [abstract]. In: Proceedings of the Annual Meeting of the American Association for Cancer Research 2020; 2020 Apr 27-28 and Jun 22-24. Philadelphia (PA): AACR; Cancer Res 2020;80(16 Suppl):Abstract nr 5771.