MISCLASSIFICATION OF MICROINVASIVE CERVICAL-CANCER AND CARCINOMA-IN-SITU OF THE CERVIX

We estimated the extent of misclassification of microinvasive cervical cancer and carcinoma-in-situ by reviewing the histology of all cases of squamous cell microinvasive disease (N = 61) and a random sample of cases of squamous cell carcinoma-in-situ (N = 35) reported to the Vic torian Cancer Registry in 1992. Five cases (3 microinvasive and 2 carc inoma-in-situ) were excluded for technical reasons or because we were unable to obtain the slides for review. Four pathologists reviewed the slides and came to a consensus diagnosis. Eleven cases (19.0%, 95% CI 8.9-29.1) of squamous cell microinvasive cancer had squamous cell car cinoma-in-situ on review and three cases (5.2%, 95% CI 0.0-10.9) were invasive cancer. Two cases (6.1%, 95% CI 0.0-14.1) of carcinoma-in-sit u had microinvasive disease on review. Adjusting squamous cell cervica l cancer incidence rates for overcalling of squamous cell carcinoma-in -situ as squamous cell microinvasive cancer results in a 5.7% reductio n in the age-standardized incidence rate of squamous cell cervical can cer; however, adjusting for both the overcalling of squamous cell carc inoma-in-situ and the undercalling of squamous cell microinvasive canc er as squamous cell carcinoma-in-situ results in a 49% increase in the age-standardized incidence rates of squamous cell cervical cancer. Er rors in cervical cancer incidence rates hinder the evaluation of cervi cal cancer screening programs.