Riddle: what do aplastic anemia, paroxysmal nocturnal hemoglobinuria (PNH) and "hypoplastic" leukemia have in common?

I N 1961, WE REPORTED 20 cases of severe aplastic anemia in which infusions of allogenic ( homologous ) bone marrow had been used as one of the therapeutic methods.1 Seven of these patients made apparently complete recoveries; whether coincidentally or in rlationship to the marrow infusions is not clear. Since then, the use of allogeneic bone marrow infusions has been wellnigh discarded for the induction of transplantation, chiefly because of the difficulties involved with suppression of the rejection phenomenon, as well as for the possibility of development of the graft-vs.-host reaction. Of the recovered cases referred to above, three patients have subsequently ( as of June 1967 ) developed the characteristic features of PNH. Originally it occurred to us that this unusually high incidence of PNH might have some obscure relationship to the infused allogenic marrow, but since PNH may follow aplastic anemia without the mediation of introduced marrow, this idea did not appear very likely. During a recent trip to the Far East where aplastic anemia appears to be unduly prevalent ( perhaps because the use of chloramphenicol is relatively uninhibited ) , it was evident that the incidence of PNH was also unduly high. Thus, in Manila, the Philippines, Dr. Allen Caviles of the Philippines General Hospital informed me that he had observed 71 cases of aplastic anemia in three years, 53 of which had been subject to follow-up; one of these had developed PNH. In the same period, nine cases of PNH had also been observed, five of them having been previously diagnosed as aplastic anemia. Dr. Tien-tse Hwang in Taipei, Taiwan, reported that he had observed 10-14 new cases of aplastic anemia annually, as well as seven cases of PNH at the two hospitals where he worked, one of them the large National Defense Hospital. Among the first 10 cases of hypoplastic anemia he had seen in 1966, one of them subsequently developed PNH. From these several observations, the factor of coincidence for the two apparently disparate conditions of aplastic anemia and PNH seems unlikely. Dacie and Gilpin2 were the first to broach the possibility that PNH and aplastic anemia might be related. This was subsequently further emphasized by Dacie3’4 and particularly in Lewis and Dacie’s recent paper. Of 46 cases of aplastic anemia, seven had a positive Ham test for PNH and two actually developed clinical evidence of the disease. Conversely, of 60 patients with PNH, 15 showed aplastic anemia sometime during their course. In two stmch