IMMUNITY IN TUMOR TRANSPLANTATION 431 



cifically on the transplanted leucocytes, might be responsible for the peculiari- 

 ties noted. 



There exist, then, tumors in which an extraneous growth stimulus (virus) 

 determines the reaction of the host against the transplant and the specificity of 

 the immune sera. In other cases the reaction may be determined by growth 

 stimuli intrinsic in the cancerous cells (Gi). These stimuli may perhaps be 

 similar to factors active in embryonal tissue, although in some very essential 

 respects conditions prevailing in tumors and embryonal tissues differ. In still 

 other instances, perhaps, substances related to tissue and organ differentials, 

 such as those present in pus and necrotic tissue, may give rise to immune sub- 

 stances. However, definite data as to the nature of some of the antigenic sub- 

 stances found in these cancers, and also in leukemia, which would differentiate 

 them from normal tissues, are as yet lacking. Moreover, some characteristics 

 of tumors may change in the course of serial transplantations ; we have dis- 

 cussed the probable nature of these changes in a preceding chapter. Dmochow- 

 ski, in the case of some mammalian tumors, found indications that also the 

 antigens may change as the result of long-continued serial transplantations. 

 Similar are the recent observations of MacDowell and his associates in mouse 

 leukemia. Mice belonging to strain C58 clevelop leukemia in a very large per- 

 centage of cases. Leukemic cells from a C58 animal with spontaneous leukemia 

 can readily be transplanted into other C58 mice, where they proliferate and 

 so transfer their disease to the hosts. When several lines of leukemic cells 

 were propagated through a large number of generations of C58 mice, in the 

 course of these passages the cells gained in proliferative power and transmitted 

 the disease more readily to other C58 mice. MacDowell found that, through 

 graded inoculation with increasing doses of such leukemic cells, C58 mice can 

 be immunized against these various propagated lines, so that in the end the 

 transfer of such cells no longer calls forth leukemia in mice thus treated. But, 

 if a C58 mouse, immunized against these special lines of C58 leukemic cells, 

 is inoculated with cells taken directly from a case of leukemia arising spon- 

 taneously in a C58 mouse, then the inoculated mouse succumbs to the disease. 

 The immunization with the serially propagated leukemic cells from strain C58 

 protects only against these special propagated lines, but not against new 

 leukemic cells which have not yet been propagated in passages. It appears 

 therefore, as if, as a result of the serial propagation, not only did the leukemic 

 cells acquire a greater growth energy and become therefore more virulent, 

 but there must also have taken place in all the serially propagated lines the 

 same type of modification of the antigen, which made this antigen different 

 from that present in the leukemic cells from primary spontaneous cases. As 

 to the nature of this antigen, it may represent an intrinsic stimulus or an 

 extrinsic virus, or something akin to a tissue or organ differential ; but for 

 the reasons stated, it is not probable that this antigen or any of the special 

 tumor antigens originate as a result of somatic mutations occurring in the 

 cells. There occurs then, in cancer cells, in addition to the organismal differen- 

 tials, various other kinds of antigens, but there remains some doubt at the 

 present time as to the nature of some of these antigens of the second type. 



