Haemopoietic reconstitution (HR) using autologous peripheral blood stem cells (PBSC) was attempted after intensive chemotherapy or chemoradiotherapy in two patients with relapsed acute non-lymphoblastic leukaemia (ANLL). The PBSC were collected by leukapheresis very early in first remission and cryopreserved in liquid nitrogen. Both patients demonstrated early evidence of trilineage engraftment. The first patient received melphalan 200 mg/m2 followed by rescue with 1.3 X 10(8) mononuclear cells/kg body weight containing 29 X 10(4) granulocyte-macrophage progenitor cells (CFU-GM)/kg, and HR was evident by Day 14. The second patient was treated with supralethal chemoradiotherapy followed by rescue with 3.0 X 10(8) mononuclear cells/kg containing 23 X 10(4) CFU-GM/kg. He demonstrated early engraftment with near normal peripheral blood counts by Day 16. There was a subsequent fall in both bone marrow cellularity and peripheral blood counts to a level of low but persistent activity. There was a further phase of haematological recovery from 8 weeks following transplantation with an increase in peripheral blood counts and bone marrow cellularity until final relapse at 13 weeks. This study demonstrates that circulating stem cells have haemopoietic reconstitutive capacity, previously only shown with buffy coat cells from chronic granulocytic leukaemia. The minimum number of PBSC required for satisfactory engraftment remains unknown, although it seems probable that the ratio of pluripotent stem cells to committed progenitor cells is lower in very early remission peripheral blood than in either allogeneic normal bone marrow or autologous bone marrow collected later in stable remission. The question of leukaemic contamination of the PBSC remains to be answered.

Circulating myeloid progenitor cells (PB CFU-GM) were measured in the peripheral blood of 13 patients with acute non-lymphoblastic leukaemia (ANLL) as they entered first remission. The mean of the recorded peak levels was 2796 X 10(3) CFU-GM/l, representing a 25-fold increase above the mean value in normal subjects. These elevated levels of PB CFU-GM occurred regularly during the very early remission phase when platelet counts rose rapidly. Five of the patients had PB mononuclear cells collected by continuous-flow leukapheresis during this early recovery phase. CFU-GM were assayed as a measure of the number of haemopoietic stem cells in each collection. The cells were concentrated and then cryopreserved in liquid nitrogen. Leukapheresis was also performed on five normal subjects for comparison. Low numbers of CFU-GM were harvested from normal subjects, mean 0.33 +/- 0.06 X 10(4) CFU-GM/kg body weight for each leukapheresis. In ANLL patients entering remission, however, very large numbers of CFU-GM were regularly harvested. A mean of 11 +/- 2 X 10(4) CFU-GM/kg body weight were cryopreserved after each leukapheresis, representing 5 times the number of CFU-GM considered necessary for successful autologous haemopoietic reconstitution. Haemopoietic stem cell viability was assessed after varying periods of cryopreservation. There was no significant stem cell loss after up to 24 months storage. Thus, it is possible to collect and cryopreserve large numbers of CFU-GM and by inference pluripotent haemopoietic stem cells from the peripheral blood of patients with ANLL during very early remission. The possible biological and therapeutic implications are discussed.

SUMMARY The effect of intravenous iron dextran on serum ferritin levels was observed in two patients with iron deficiency anaemia. Serum ferritin levels rose sharply and reached peak levels seven to nine days after infusion when at least 90 %of the infused iron had been removed from the plasma.A linear relation was shown for each patient between the logarithm of the serum ferritin levels and the logarithm of the calculated cellular non-haem iron levels. Material and methods CASE 2A 76-year-old man with anaemia (Hb 8·9 gjd!) had severe atrophic gastritis demonstrated by gastroscopy. MCV, serum iron, transferrin, and ferritin levels were consistent with iron deficiency. There was no history of chronic blood loss, and faecal blood loss studies were normal. He had been transfused to a Hb level of 11·0 gldl three days before iron therapy was given. Since oral iron would be poorly absorbed because of gastritis, 20 mI of iron dextran (Imferon ®, Fisons) was infused intravenously with 500 ml of normal saline, a total dose of 1 g of elemental iron. Serum ferritin was estimated by a two-site immunoradiometric assay, 3 and serum iron as recommended by the International Committee for Standardisation in Haernatolcgy," incubating for 2 hours with chromagen solution to ensure complete measurement of Imferon ® iron. Haematological measurements were made on a Model S Coulter Counter. Red cell mass, total blood volume, and plasma volume were determined using patient's red cells labelled with oICr.°T otal plasma iron was calculated from the serum iron level and the plasma volume. Iron utilised for Hb synthesis was calculated from the formula: changeinHb Iron (mg) = 100 x Blood volume (mI) x 3·4 Cellular non-haem iron (CNHI) was calculated to be the amount of iron removed from the plasma less that required for the observed Hb synthesis. Iron stores before infusion were assumed to be zero for the purpose of these calculations. The levels of CNHI in case I were calculated for only the first II 215 Patients CASE 1

No abstract