In 1890 it was shown that resistance to diphtheria toxin could be transferred from one animal to another by transfer of serum (1). From this discovery, passive antibody therapy was developed as an effective treatment for infectious diseases and for neutralization of toxins, and continues to be used to this day. Meanwhile, there have been continued efforts to use antibodies for cancer therapy, starting with the pioneering work of Hericourt and Richet in 1895 (2), which was the forerunner of the "magic bullet" concept. However, all of the early work on tumor therapy led ultimately to disappointment (3). The problems were readily acknowledged, i.e., lack of specificity and reproducibility, lack of purity, and the xenogeneic immune response. Developments over the past 20 years, as described throughout this book, have effectively overcome all of these technical problems, often in very ingenious ways. The difficulty we have now is different. There are just too many potential new antibody-based treatments for them all to be properly evaluated in the clinic. Many will still fail because of factors that are hard to predict from experiments: unexpected toxicity, biological heterogeneity of the target disease, or lack of access to the appropriate tissue.