Page 19 - Screening for Cervical Cancer: Systematic Evidence Review

Chapter I. Introduction

Algorithm-based rescreening identifies slides that exceed a selected probability for

containing abnormal cells. The AutoPap 300QC system (Neopath, Inc.—when reviewed, now

TriPath), FDA approved in May 1998, can be set for different thresholds that result in 10%, 15%,

or 20% review rates. Because a classification algorithm is applied, the slides selected for review

are more likely to contain abnormalities than are negative Pap slides randomly selected for

review. The system is reported to identify 70% to 80% of those slides misdiagnosed as normal

(i.e., false negatives) at the time of manual screening. AutoPap is also employed for both routine

screening and rescreening purposes outside the United States, and publications evaluate both

types of use.

We excluded review of the performance of AutoSCREEN, a product for interactive,

computer-assisted screening and rescreening. Its manufacturers no longer produce it after a

corporate merger between the AutoPap parent company and the original producer. We also have

not reviewed approaches that require additional time, equipment, expertise, or materials at the

time of the clinical pelvic examination, such as speculoscopy, screening colposcopy, or

cervicography to photograph the cervix. These are not suitable for widespread implementation

in the United States.

Tools for HPV Testing

Cellular changes associated with HPV infection can be seen on visual examination of

cervical cytologic or biopsy specimens and, to a lesser degree, at colposcopy. However, the

presence of these changes is neither sensitive nor specific.

Direct HPV testing methods for assessing the presence and type of HPV are more

promising; they rely on identification of HPV viral DNA. Some versions of testing are

qualitative, detecting only the presence or absence of HPV DNA; some are quantitative,

9