Morphologic analysis of false negative SurePath® slides using Focalpoint™ GS computer-assisted cervical screening technology: An Australian experience.

BACKGROUND: The trial results of BD FocalPoint™ GS computer assisted screening(FP) of BD SurePath(®) liquid based cervical cytology slides (SP) were published in Diagnostic Cytopathology in 2012.(1) METHOD: The FP-reviewed SP slides were compared to conventional cervical Pap smears (CON) in a split sample of 2198 routine specimens. In all 47 confirmed high grade (HG) cases, FP either selected fields of view (FOV) containing cells suspicious or diagnostic of HG (46/47), or prompted full screening of the slide (1/47) leading to detection of the HG lesion. In this study for the HG cases which were reported initially as negative (NEG) or low grade (LG), the original slides were reviewed to determine the reasons for the false negative reporting and to identify features characteristic of a HG pattern. RESULTS: Indicators of a high risk pattern for HG Squamous Intraepithelial lesions (HSIL) included the recognition of pale abnormal immature metaplastic cells. Indicators of a high risk pattern for HG Glandular Intraepithelial lesion (AIS) were single columnar cells with the nucleus occupying the widest area of the cell. Pavemented sheets of glandular cells and isolated mitoses were additional clues to the diagnosis. In endocervical and endometrial adenocarcinomas tumour diathesis tends to be absent in SurePath slides. Single cells with ingested neutrophils and obscured eccentric nuclei were another clue to the detection of endometrial adenocarcinoma. CONCLUSION: These morphological features are illustrated to help identify HSIL and HG glandular lesions when viewing the FOV presented by the FocalPoint(TM) GS technology.

Results of an Australian trial using SurePath liquid-based cervical cytology with FocalPoint computer-assisted screening technology.

BD FocalPoint GS™ computer-assisted screening of BD SurePath® liquid-based cervical cytology slides (SP + FP) was compared with screening an accompanying conventional cervical Papanicolaou (Pap) smear (CON) in a split sample trial of 2,198 routine specimens. The rate of unsatisfactory specimens in the SP + FP arm was 0.2% compared with 4.1% in the conventional Pap smear, a significant reduction. There was no statistically significant difference between SP + FP and CON for the detection of histologically confirmed high-grade (HG) lesions in the routine split sample specimens (n = 9). To further test the sensitivity of SP + FP for HG lesions, 38 SurePath slides from confirmed HG cases, without an accompanying CON, were interpolated among the routine smears. In every one of the 47 confirmed HG cases, either HG cells were present in the microscope fields selected by FocalPointGS™ for review by the screening cytologist (46 of 47), or full screening of the slide was indicated by the FocalPointGS™ (1 of 47), confirming the effectiveness of SP + FP technology for primary screening. In a small number of cases, the screening cytologist did not recognize the abnormality even though on review HG cells were present in fields selected by FocalPointGS™. The overall detection rate was 93% for HG squamous lesions; 89% for known HG endocervical glandular lesions; and 91% for known endometrial carcinoma. In conclusion, the SP + FP detected 100% of HG abnormalities in the trial set; significantly reduced the rate of unsatisfactory specimens; and improved the overall screening rate of detection of HG abnormalities particularly of glandular lesions when compared with other screening technologies.