MULTIPLE MYELOMA 



tion, can settle down in an appropriate physiological environ- 

 ment and proliferate. 



The fact that each myeloma patient produces his own 

 characteristic and individual serum protein, with its sharp 

 spike evidence of homogeneity, provides support for what 

 many workers might consider a weak point of the clonal 

 selection theory, that each clone produces a specific antibody 

 globulin whose pattern is genetically determined. The 

 globulins produced in multiple myelomatosis are abnormal, 

 but they remain consistently of the one physical and chemical 

 pattern which has been, as it were, chosen from a probably 

 unlimited number of possibilities. This constancy of pattern 

 can hardly be interpreted in any other fashion than as a 

 genetically controlled quality of the clone. 



On general grounds one could not expect to find a 

 myeloma globulin functioning as a normal antibody even 

 if it sprang from a clone which was antibody-producing. The 

 distortion of structure is not likely to leave antibody specificity 

 unchanged and, even if it did, the chance of finding the right 

 antigenic determinant among the hundreds (or thousands) 

 which might be involved would be very slight. We have 

 already referred to the rather rare instances in which 

 Gajdusek's AIGF test is given by macroglobulinaemia sera. 

 A considerable number of myeloma sera have been tested 

 with completely negative results. In the macroglobulin 

 serum most closely studied by Mackay (1956) there was 

 indirect evidence of heterogeneity similar to that found with 

 other high-titre sera from other clinical conditions. In the 

 light of experience with myeloma sera we should consider 

 that we are deaUng with two distinct abnormaUties of 

 globulin production which will only very rarely affect a 

 clone of mesenchymal cells simultaneously. The first is the 

 production of an abnormal but genetically controlled type 

 of globulin, the second the production of protein molecules 

 which, in regard to one aspect not yet clearly characterized, 

 are of randomly distributed patterns. 



183 



