l84 DISCUSSION 



used either non-specific tumours, which means a genetically incom- 

 patible system, or else, if they used isologous systems they worked with 

 long-transplanted tumours, like Baserga and his co-workers, for in- 

 stance. In our laboratory, B. Tornberg carried out experiments of this 

 type but with spontaneous tumours. He removed spontaneous mam- 

 mary cancers operatively from C3H mice and divided the material into 

 two groups. One group was treated with cortisone, while the other 

 was left untreated. The frequency of lung metastases was compared in 

 these two groups. Not only did cortisone not show any promoting 

 effect upon metastasis in this system, but it had some slight inhibitory 

 effect. In contrast, it had a promoting effect on a long-transplanted 

 mammary cancer. I think it still remains to be shown whether the 

 entire cortisone effect on metastasis is not due to an inhibition of homo- 

 graft reaction. We do not perhaps yet have to postulate that the latent 

 tumour cells that he around in the lymph nodes and other tissues and 

 seem to be promoted by cortisone treatment are necessarily latent 

 because they are inhibited by an immune mechanism. The mechanism 

 may be something other than an immunological one. 



Feldman: It is certainly true that cortisone might do a number of 

 things. For one thing, it might even activate the formation of metastases 

 — not by means of suppressing any resistance — but due to the effect of 

 cortisone on blood capillaries, which results in capillary arrest of tumour 

 cells. This type of effect was recently reported by E. Zeidman (1961. 

 Proc. Amer. Ass. Cancer Res., 3, 281). We have been studying tumours 

 which apparently metastasize via the lymphatics. Whether the promo- 

 tion of metastasis in chimeras, where the immune reactivity is 

 decreased, is indeed due to suppression of some sort of resistance 

 phenomenon, is still an open question. 



