Despite molecular biological advances in understanding human cancers, translation into therapy has been less forthcoming; targeting neoplastic cells still requires that tumor-specific markers, preferably those on the cell surface, be identified. The epidermal growth factor receptor (EGFR) exists in a deletion-mutant form, EGFRvIII, which has been identified by genetic and immunological means in a subset of gliomas and non-small cell lung carcinomas. Specific polyvalent antisera to the extra-cellular portion of the variant were readily induced, but immunization using a synthetic linear peptide representing the unique EGFRvIII primary sequence has been unsuccessful in mice or macaques. We report here five specific monoclonal antibodies (mAbs) developed through long-term immunization protocols using the EGFRvIII-specific synthetic peptide and the intact variant in different formats that maintained secondary and tertiary conformation. These mAbs identify the EGFRvIII on the cell surface with relatively high affinity (KA range, 0.13 to 2.5 × 109 m-1) by live cell Scatchard analysis. These mAbs are specific for EGFRvIII as determined by RIA, ELISA, Western blot, analytical flow cytometry, autophosphorylation, and immunohistochemistry. Isolating specific mAbs enabled us to analyze normal and neoplastic human tissue and establish that EGFRvIII is truly tumor specific for subsets of breast carcinomas and for previously reported non-small cell lung carcinomas and gliomas. Also, this receptor is not expressed by any normal human tissues thus far examined, including elements of the peripheral, central nervous, and lymphoid systems. With mAbs, we identified a higher incidence of EGFRvIII positivity in gliomas than previously described and identified an EGFRvIII-positive subset of breast tumors; also, we observed that the EGFRvIII epitope is not expressed in normal tissues, and we demonstrated the localizing and therapeutic potential of the mAbs for tumors expressing this epitope. Our observations strongly warrant development of this mAb-antigen system as therapy for breast, lung, and central nervous system tumors.
This work was supported in part by NIH Grants CA 11898, CA 56115, CA 32672, NS 29955, and NS 20023 and by American Cancer Society Grant 89-171.