In a study of human sarcomas maintained in culture for periods up to two years, the following observations were made.
The most prominent cell type in serially cultured osteosarcomas was fibroblastic in appearance. After 16–20 wk in culture some lines spontaneously developed foci of altered cells resembling the foci produced in monolayer cultures by oncogenic viruses. The presence of these foci in the sarcoma cultures was transient, and usually they did not reappear; but in one instance they recurred with a characteristic periodicity of several weeks. From one of the sarcoma lines, in which foci appeared after 5 months in culture, two subcultures were established from stored frozen cells and these both exhibited foci after approximately the same lapse of time. The same phenomenon has been seen with another line, suggesting that the time of appearance of foci is characteristic for particular sarcomas. Foci of similar type could sometimes be induced in monolayer cultures of human fibroblasts by filtered medium from cultured sarcomas; this bore no relation to the presence or absence of foci in the sarcoma cultures at the time the filtrate was prepared. Electron microscopy of the spontaneous and induced foci, and of the sarcoma cultures, revealed no demonstrable virus.
12 out of 15 sarcoma cultures contained an antigen (S) demonstrable by indirect immunofluorescence with human sera. It was not present in any of the original sarcoma specimens, nor in any culture lines other than sarcomas. At least 3–4 wk in culture appear to be required for its demonstration. The antigen was cytoplasmic, occurred in only a small proportion of the cells, and was unpredictably variable in its expression, even in the same culture line. It could be induced in monolayer cultures of human fibroblasts by filtrates of medium from sarcoma cultures. As with the foci, the induction of S antigen in indicator cultures was not dependent upon the expression of antigen in the sarcoma line from which the filtrates were obtained. There was no association between the presence of foci and of antigen, nor was there any apparent relation between the ability of filtrates to induce foci and their ability to induce antigen. 80% or more of the general population have S antibody, and the titer of antibody in patients with sarcoma is no higher than in normal subjects. Thus, as in the case of Burkitt's lymphoma antigen, it appears that most individuals have been exposed to S antigen. But unlike Burkitt's lymphoma, no relation has so far been established between any particular disease and a corresponding high titer or frequency of occurrence of S antibody.
The occurrence of foci of altered cells and of a common antigen, and the transmission of these two characters to indicator cells by filtrates, are all suggestive of a virus specifically associated with human sarcomas, one to which the general population is widely exposed, as indicated by the prevalence of antibody.