Biomorphometric analysis of human prostatic carcinoma by using three-dimensional computer models.

Abstract

Advances in the detection of carcinoma of the prostate during the last 15 years have accounted for a sharp increase and then an abrupt decrease in the incidence of the disease. A more recent decline in its mortality rates has been variously interpreted as either the success of early detection and improved treatment or lead-time bias. The recently reported Prostate Cancer Prevention Trial had an overall detection rate that approached the 30%-40% prevalence rates reported in autopsy series in which men died of other causes. However, the prognostic information that can be obtained from prostate cancer found on biopsy is limited. Three-dimensional computer modeling is one technique that allows multiple studies on "immortal" prostates to test methods of biopsy sampling accuracy and to assist in the determination of the disease's severity. Computer modeling can assess detection rates and assesses tumor multifocality and heterogeneity. It can provide a more accurate representation of tumor volume, aiding in therapeutic decision making, and can assess sampling errors of various biopsy methods. It has been shown to be superior to wire-frame technique by immortalizing the original shape and dimensions of the surgically excised prostate gland. Moreover, our 3-dimensional computer modeling system improves upon other systems: It is more than a simple extension of the planimetric technique, and it is able to demarcate clearly the boundaries of Gleason grades just 1 grade apart.