ULTRAVIOLET RADIATION AND CANCER 551 



vitamin D suggest that the activation of the provitamin takes place in the 

 corneum, or at least very superficially (see Blum, 1950a), and it appears 

 that most of the steroids of the skin are found in the corneum rather than 

 in the deeper tissues. Yet the site of carcinogenesis in the case of the 

 mice is deeper than this. There is other evidence against this hypothesis. 

 The polycyclic hydrocarbon carcinogens, which are the type we should 

 expect to find formed under these circumstances, are quite diffusible. If 

 strain A mice are treated with carcinogens either by injection or by paint- 

 ing on the skin, in which case the carcinogen is of course licked off and gets 

 into the alimentary canal, there is a great increase in the incidence of 

 cancers of the lung (Shimkin, 1940). Yet the incidence of lung tumors in 

 this same strain was not found to be greater among mice in which cancer 

 had been induced by ultraviolet radiation from that in untreated con- 

 trols.'^ Thus it appears that no diffusible carcinogen is formed by the 

 ultraviolet radiation, and this is strong evidence against the participation 

 of a polycyclic hydrocarbon carcinogen in this type of carcinogenesis.'^ 



THE ROLE OF SUNLIGHT IN CANCER OF THE SKIN OF MAN 



After a discussion of the results of animal experiments, the clinical 

 findings bearing on the question of the etiological role of the ultraviolet 

 radiation of sunlight in cancer of the skin of man may be better evaluated. 

 In our introduction four general lines of evidence were mentioned, and 

 these may now be considered. 



Topographical Distribution. All authorities agree that cutaneous 

 cancer occurs in the white race predominantly on the face ; for example, in 

 91 per cent of the 1626 cases reviewed by Lacassagne (1933) the tumors 

 were on that area. Basal-cell and squamous-cell cancers predominate 

 among skin cancers, and it is these types which display the strong predi- 

 lection for the face. The other principal site of these tumors is the back of 

 the hand. This evidence alone strongly supports the idea that sunlight, 

 which reaches other areas of the body to a very small extent, has a role in 

 the etiology of basal-cell and sciuamous-cell cancers. The occasional 

 appearance of one of these tumors on an unexposed area of the body would 

 not negate this conclusion, since no doubt there are other agents which 



^^ Actually there was a slight decrease in the treated animals, the probable causes of 

 which have been discussed by Blum (1944). The experiments also indicated a differ- 

 ence in general susceptibility to the development of cancer within tlic genetically 

 homogeneous strain. 



^* There have been various experiments showing that exposure to ultraviolet radia- 

 tion decreases the incidence of cancers in mice painted with carcinogenic hydrocarbons. 

 Recently Engelbreth-Holm and Iversen (1947) offered evidence that this is due to 

 photooxidation of the hydrocarbon. This explanation was offered by the writer as 

 early as 1940. Other possible factors may be involved (e.g., Blum, 1943b), but in any 

 event there seems no reason to connect photochemical reactions of these carcinogens 

 with the induction of cancer by ultraviolet radiation. 



