A large variety of monoclonal antibodies (MAb) have been identified that bind to antigens on cancer cells and there are many mechanisms by which antibodies can kill cancer cells. However, very few antibodies are able to kill sufficient numbers of cells to cause tumor regression in experimental animals and even fewer are useful in the treatment of cancers in humans. Therefore it is often necessary to arm the antibody with a cytotoxic agent such as a cytotoxic drug, a radioisotope or a protein toxin. My laboratory has been interested in using protein toxins for this purpose. In this paper I will review how we became interested in using protein toxins to make immunotoxins, and the various stages in the development of immunotoxins culminating in our current clinical trials. In 1977, Mark Willingham and I were trying to apply new optical techniques to uncover the organelles involved in the endocytosis of growth factors, hormones and viruses. In a series of experiments we were able to show that diverse agents such as insulin, α 2 macroglobulin, epidermal growth factor and adenovirus all entered cells via coated pits and endocytic vesicles. We called these vesicles receptosomes to emphasize their role in receptor-mediated endocytosis [1]. They are now referred to as endosomes. We also showed that several different ligands of this type could occupy the same endocytic vesicle. After several years of research, we were joined in these studies by a postdoctoral fellow, David FitzGerald. He was an expert on protein toxins, particularly Pseudomonas exotoxin A (PE). Most of the ligands we were studying were directed to lysosomes or were returned to the cell surface, but adenovirus was different. It ended up in the cytosol. We hypothesized that adenovirus disrupted the membrane of the endocytic vesicle, allowing all its contents including the virus to reach the cytosol. To test this hypothesis we exposed cells to adenovirus and PE together, and found that adenovirus increased the cell killing activity of PE by 100 to 1,000 fold. It did this by enabling PE to directly reach the cytosol, where it could inactivate elongation factor 2 instead of reaching the cytosol by a more circuitous route [2]. In addition to this very interesting result, this experiment opened our eyes to the possibility of targeting PE to cancer cells using antibodies. At this time, the immunotoxin field was in its infancy and many investigators were attaching their favorite toxins to various antibodies.