A method for quantitative autoradiography over stained sections of tumors exposed in vivo to radiolabeled antibodies.

Abstract

Quantitative autoradiography can be an important element in calculating dose patterns delivered to tumors and normal tissues of animals injected with radiolabeled antibodies. In this article we develop a computer algorithm that permits quantitative evaluation of grain density over large areas of stained tissue with minimal grain counting. This algorithm, using spectral classification of brightfield and darkfield images, image ratios, and manual counting of small sections of an autoradiograph, yields images defined in terms of grain density, independent of staining artifacts. These quantitative images give a precise indication of intratumor radioimmunoglobulin targeting, and can serve as source functions for dosimetric calculations.