Assisted primary screening using the automated ThinPrep Imaging System

Area of Science:

• Gynecologic Pathology
• Cervical Cancer Screening
• Medical Device Technology

Background:

• Cervical cancer screening relies on cytological examination of Papanicolaou (Pap) smears.
• Manual screening is labor-intensive and subject to inter-observer variability.
• Advancements in imaging technology aim to improve the accuracy and efficiency of cervical cytology.

Purpose of the Study:

• To evaluate the clinical performance of the ThinPrep Imaging System (TIS) in cervical cancer screening.
• To compare the sensitivity and specificity of TIS-assisted screening with manual screening.
• To assess the impact of TIS on cytologist screening productivity.

Main Methods:

• A prospective clinical trial involving 10,742 ThinPrep specimens across 4 US clinical sites.
• Slides were screened manually, then TIS-assisted, followed by adjudication by 3 pathologists.
• Sensitivity and specificity analyses were performed using adjudicated diagnoses as the gold standard.

Main Results:

• TIS-assisted screening demonstrated statistically higher sensitivity for atypical squamous cells of undetermined significance (ASCUS) or higher.
• Sensitivity for low-grade squamous intraepithelial lesion (LSIL) and high-grade squamous intraepithelial lesion (HSIL) was equivalent or higher with TIS-assisted screening.
• Specificity was equivalent for ASCUS+ and LSIL+, and significantly higher for HSIL+.
• Cytologist daily screening rates doubled with TIS-assisted screening.

Conclusions:

• TIS-assisted screening equals or exceeds manual screening sensitivity without compromising specificity.
• The ThinPrep Imaging System can significantly improve cervical cancer screening productivity.
• Further studies are needed to evaluate the cost-effectiveness of TIS.