Nuclear roundness factor measurement for assessment of prognosis of patients with prostatic carcinoma. I. Testing of a digitization system.

Standard pathological grading systems for prostatic carcinoma based upon glandular architectural pattern or nuclear anaplasia have failed to predict the prognosis of individual patients. Quantitative morphometric analysis of nuclear shape has predicted the outcome of patients with prostatic carcinoma when evaluated by some but not all reported studies. Calculation of nuclear roundness factor by different investigators was complicated by equipment differences and lack of standardized methods for acquiring and reporting data. We have improved our system for nuclear contour digitization and determined its theoretical limitations by digitizing standardized objects. Measurement errors were independent of orientation or location of an object within a microscopic or digitizer tablet field, speed of digitization and introduction of a microscope extension tube and light emitting diode cursor. Perimeters and areas of circles and squares of actual or digitizer-projected diameter or sidelength greater than five mm. were measured with a reproducibility and accuracy of greater than or equal to 90%. When a microscopic circle of diameter similar to prostatic carcinoma nuclei was digitized at a magnification of 2580 X, perimeter and area measurements differed within or between observers by less than 5% and were more than 95% accurate. In order to calculate accurately and evaluate NRF for use in assessing the prognosis of patients with prostatic carcinoma investigators must precisely describe their digitization system, standardization method and observer reproducibility and accuracy when measuring circles that approximate the projected size of prostatic carcinoma nuclei.