Tissue culture lines have become widely utilized in the study of human cancer. Legitimate questions arise as to the degree of similarity between the in vitro cells and their in vivo counterparts. The author has used eleven monoclonal antibodies (mAbs) of defined specificities to compare antigen (Ag) expression of human renal cancers in vivo and in vitro. Nine of the eleven Ags defined by these mAbs have been previously immunochemically characterized. These mAbs were generated by immunizing mice with human cancer cell lines. Nephrectomy specimens from six patients with adenocarcinoma were available for study as both frozen sections (in vivo) and tissue culture lines (in vitro). Frozen sections were typed by an indirect immunoperoxidase assay while established cell lines were typed by an anti-mouse immunoglobulin rosette assay. Antigens were scored either positive (+) or negative (-). Results revealed three patterns of antigenic expression: (1) consistent expression in vivo and in vitro (gp160, gp115, gp120r, gp120nr, GD3, B2.6, HLA-DR); (2) in vivo (-):in vitro (+) (AJ2, gp40, gp130); and (3) in vivo (+):in vitro (-) (S25). These results suggest that while most antigens (Ags) are consistently expressed or repressed when cells adapt to tissue culture, some Ags, e.g., AJ2, gp40 and gp130, appear to be induced and others, e.g., S25, are suppressed in their new milieu. Moreover, expression of these 11 Ags did not change with up to 40 passages in tissue culture. The demonstrated similarities between in vivo and in vitro cells coupled with the logistic advantages offered by cell lines makes an in vitro system a worthwhile approach to the study of human cancer. However, the fact that differences exist between the two systems underscores the need to extend interesting in vitro findings to the in vivo system prior to making á priori extrapolations or assumptions.