This interpretive review attempts to dovetail advanced work by different groups of investigators on blood group and carcinoma (CA) glycoconjugates that have terminal, immunoreactive Tn epitopes (GalNAc alpha-O-Ser/Thr), and on the interaction of those structures with complementary antibodies and lectins. Fenlon et al. (1987) and Leathem and Brooks (1987) found a positive correlation between primary breast CA aggressiveness and its affinity for Helix pomatia (HPA) lectin. This phenomenon was used successfully to accurately predict, in studies on 305 breast CA patients, early or late CA recurrence and patient survival time. The innate specificity of the large HPA combining groove (aside from its avid reactivity with appropriately spaced GalNAc alpha-O-) remains obscure, despite careful investigation for more than a decade (Baker et al., 1983). Leathem and Brooks presumed that HPA recognizes a hitherto "undefined biological marker" that indicates a breast CA's aggressiveness. Our own work has shown that the chemically fully defined Tn epitope, as measured with human polyclonal and murine monoclonal anti-Tn antibodies, occurs in immunoreactive form in approximately 90% of all breast and lung adenoCAs studied. Tn is occluded and non-reactive in healthy and non-CA-diseased tissues. We found that CA-associated Tn is an adhesion molecule in attachment to healthy cells; an increase in its density on breast CA cell membranes parallels greater aggressiveness of breast tumors in both humans and mice (the only species studied). Thus, Tn may be all or a major part of the postulated "as yet undefined biological marker" associated with high breast CA aggressiveness. Besides being helpful in the elucidation of some aspects of breast CA pathogenesis, these findings on primary breast CA have clinical implications in that they should facilitate stratification of breast CA patients for adjuvant treatment.