Validation of AutoPap® primary screening system sensitivity and high-risk performance

OBJECTIVE: To confirm or deny the reported sensitivity of the AutoPap(R) Primary Screening System (AutoPap) (TriPath Imaging Inc., Burlington, North Carolina, U.S.A.) in a moderate-sized laboratory and to determine performance characteristics for the "clinically high risk" (CHR) patient population. STUDY DESIGN: Archives were searched for low and high grade squamous intraepithelial lesion (LSIL, HSIL), adenocarcinoma in situ (AIS) and cancer (Ca) with follow-tip biopsies demonstrating a lesion of at least the reported Pap smear's severity. Smears fulfilling these criteria and a matched normal, control slide front the same day of preparation were subjected to evaluation on the AutoPap. Two hundred eighty-three smears from 254 patients were enrolled in the study, including 80 LSIL, 178 HSIL, 5 AIS and 20 Ca. Specific criteria established CHR status. Fisher's exact test was applied to determine AutoPap performance differences for non-CHR and CHR populations. RESULTS: AutoPap successfully classified as "Review" all cases as follows: 91.2% LSIL (73180), 96.6% HSIL (1721178), 100% AIS (515), and 100% Ca (20120). Fisher's exact P values, P-LSIL = 1.00 and PHSIL+ =.411, confirmed statistically equivalent performance. CONCLUSION: The results confirmed the sensitivity data reported in the Food and Drug Administration approved labeling of the AutoPap and indicated no statistically significant differences in performance characteristics in a CHR population when compared to patients without CHR status for all grades of abnormality examined.