A comparison of liquid-based cytology and Pap smear as a screening method for cervical cancer

Abstract

The implementation of population-based screening for cervical cancer with Pap smear in the early sixties was set to detect and treat precancerous lesions, hopefully preventing a subsequent invasive cervical cancer. Epidemiological data indicate that organized screening has a major impact on morbidity and mortality from cervical cancer. The limited sensitivity of a single smear necessitates repeated smears in organized program. It is suggested that liquid-based cytology improves the sensitivity. The aim of this split-sample study was to compare ThinPrep liquid-based cytology with conventional Pap smear, relying on a laboratory with long-term experience of the latter. In total, 137 women with atypical Pap smear in population-based cervical screening were enrolled for the split-sample study. The performance of both techniques (ThinPrep liquid-based cytology and conventional Pap smear) were compared and validated by a histological follow-up. Women without representative histological biopsy were excluded from the study. Pap smear had sensitivity for detection of CIN2-3 of 47% compared to 66% for liquid-based material. The concordance of the two sampling techniques with the histological diagnosis was 37% and 53%, respectively, this difference being statistically significant. The proportion of reports on atypical squamous cells of undetermined significance (ASCUS) was significantly less in the liquid-based material, 4.3% compared to 8% of the conventional smears. This improved sensitivity in combination with the possibility to perform reflex testing such as HPV DNA or p16 immunocytochemistry without renewed sampling gives ThinPrep a substantial advantage and makes the liquid-based technique interesting.