Method for selection of primate tumor-associated antigens suitable as in vivo targets for antibodies

Abstract

The present invention is a process for the selection of antigens which are suitable targets for in vivo antibody localization in human tumors or other altered (or diseased) tissue. The process provides a simplified and rapid technique for discovering useful in vivo targets for antibodies and is useful in cancer detection and therapy in humans or other primates, whether or not the antigens are specific to tumors. More specifically, the invention relates to a process for the selection of antigens suitable as targets for antibodies which localize in a tumor in vivo in which antibodies are first prepared distinguishable from those present in the animal in which biofiltration is to occur and that bind to antigens present in the tumor to be targeted. These antibodies are then injected into a non-tumor-bearing primate, into a tumor-bearing animal, and into a non-tumor bearing animal of the same species as the tumor-bearing animal to permit biofiltration of the antibodies. The biofiltered antibodies are next recovered from each of the non-tumor-bearing primate, the tumor-bearing animal, and the non-tumor-bearing animal and are employed to identify antigens whose antibodies are not retained in vivo in the primate and the non-tumor-bearing animal. The antibodies that are not retained in vivo by the non-tumor-bearing primate and the non-tumor bearing animal are then compared with those antibodies that are actually retained in vivo in the tumor-bearing animal to identify the antigens corresponding to those antibodies that are selectively retained in the tumor-bearing animal. In an alternative process, the tumor-bearing animal may be replaced by a perfusable surgically removed human organ bearing a tumor, and the non-tumor-bearing animal may be replaced by a surgically removed normal organ. Passage through a primate, however, remains the method of choice for selecting antigens whose antibodies are unabsorbed in vivo.

Dependencies are inferred from the claim text.