The Fate of Intravenously Injected Tumor Cells

In spite of a large volume of information from clinical and experimental observations on tumor metastasis, the subject is still largely a matter of speculation. Blood-borne metastases, often dramatic and bizarre, have attracted much attention, particularly in the study of “silent” or “healing” emboli seen in human material and carefully described by Schmidt (9), Teacher (12), and others. Iwasaki (3) and Takahashi (11) have shown strikingly similar changes in animals with intravenously injected tumor. Probably the closest approximation to spontaneous vascular metastases in man was obtained in the experimental massage of tumor by Tyzzer (13), Knox (4), and Marsh (8). This method, however, is complicated by the possibility of naturally occurring metastases by other routes. There is no doubt that tumor emboli may become organized and disappear or remain latent over long periods without establishing true metastasis. Whether this is due to death of some of the cells which initiates thrombus formation, or whether it is due to mechanical factors or to resistance of the host is not certain. The relative significance of the biological character of the tumor and of the local reaction to implantation in metastasis is still not clear, although Slye (10) has given very interesting evidence as to the latter. Since many animal tumors metastasize readily by the blood stream, they are well adapted for this type of experimentation. As the simplest experimental approach to the study of metastasis, we have attempted to follow the way in which tumor nodules are established in the lung after the intravenous injection of a known number of uninjured tumor cells, by killing the animals at short intervals. Similar experiments carried out by others are clouded by the use of mixtures of crushed, autolyzed, and viable cells. Since no process of dispersing and suspending tumor cells for injection has heretofore been carried out without cell injury, the experimental results have been confused by the effects of products of cellular degeneration. The toxicity of such suspensions has long been known and is confirmed in our own work. Consequently when we improved and standardized a procedure for the intravenous injection of uninjured tumor cells, the advisability of reinvestigating the subject was obvious both as a means for establishing a basis for further work on metastases with this tumor, and for comparison with previous work. The method is based on the observation of Hesse (2) that peritoneal fluid resulting from intraperitoneal inoculation of Flexner-Jobling carcinoma contained tumor cells capable of transmitting the tumor (Fig. 1).