tmmunologyt~Jdr{y, De+ember 1981 239 Use of radiolabelled monoclonal anti-CEA antibodies for the detection of human carcinomas by external photoscanning " " hy and tomosclntlgrap Jean-Pierre Mach, Franz Buchegger, Michel Forni, Jurg Ritschard, Christian Berche, Jean-Denis Lumbroso, Magali Schreyer, Christophe Girardet, ,Roberto S. Accolla and Stefan Carrel Ludwig Institute f~)rCancer Research, Lausanne Branch, and Department of Biochemistry University of Lausanne, 1066 Epalinges, Switzerland; Clinque M~dicale and Division de Madecine Nuclaaire from the Department of Medicine University of Geneva, 1211 Geneva, Switzerland; lnstitut Gustave Roussy and Institut de Radiobiologie Clinique, 4800 Villejuif, France. Paul Ehrlich 's inspired concept of 'magic bullets 'for the cure (~fdiseases has beer~revitalized by recent advances in immunology j . In parlicular, the development of ceil fusion technology allowing the production r![ monoelonal antibodies ( Mabs) with exquisite specifcities 2 triggered new hopes that we may now have the perfect carrier molecules with which to deliver cytotoxic drugs 3 or loxins 4 to the hidden cancer cells. This article reviews data on one aspect of the magic bullet concept, the use of radiolabelled antibodies as tracers for turnout localization. It will also discuss the very recent clinical use of 13lI-labelled Mabs against carcinoembryonic antigen ( CEA) ~ lo detect carcinoma either by conventional external photosearmir~g or by single photon emission compulerized lomogra- phy ( SPECT). This alliance (the most modern tools from immunology ( Mabs) and nuclear medicine ( SPECT) appears promising as a way to improve the sensitivity of 'immunoscintigraphy '. However, this approach is not yet tea+for wi&spread clinical use. Experimental models Research on tumour localization of radiolabelled antibodies was initiated almost 30 years ago by Pressman C' and Bale 7, who showed that labelled anti- bodies against Wagner osteosarcoma or Walker carci- noma cells were concentrated in vivo by these turnouts. One of Pressman's major contributions in this field was to introduce the paired labelling method 8, in which both antibodies and control normal IgG, each labelled with a different isotope, are injected simul- taneously into the same turnout-bearing animal. The measurement of radioactivity from each isotope in a dual channel scintillation counter allows one to dist- inguish the specific localization of antibodies in a turnout ft'om the non-specific accumulation of normal IgG, which is known to occur in the inflammatory and necrotic region of the tumour '~. In 1971, using this method in golden hamsters bearing a human chorio- carcinoma grafted into their cheek pouches, Q uinones et al. m showed that 12~I-labelled rabbit IgG anti- human chorionic gonadotrophin (hCG) localized in the turnout. The control normal lgG, however, was also concentrated in the tumour and only at day 4 after the injection was there a significant, although moderate, increase of antibody over normal IgG in the tumour. In 1974, we introduced into this field the model of nude mice bearing grafts of human colon carcinoma and the use of affinity purified antibodies against a well characterized antigen, such as CEA 11. We showed that purified 131I-labelled goat anti-CEA antibodies could reach up to a 9 times higher concen- tration in the tumour than in the liver, while the con- centration of control normal IgG in the turnout was never higher than 2.3 times that in the liver. However, c l'~Ne~Wt/N,+~t h I h~ll;~nd l~im,;c(li~ ,t] l+:css 1981 C)lO ~ I,;19/gl/<louo ~3~3~qP/Su2 ~11