Failure of purged autologous bone marrow transplantation in high risk acute lymphoblastic leukaemia in first complete remission.

Patients in first remission of acute lymphoblastic leukaemia (ALL) considered to be at high risk of relapse were offered autologous bone marrow transplantation (ABMT) using purged marrow as a therapeutic alternative to cranial irradiation and maintenance chemotherapy. Twenty-seven bone marrows taken in remission, were purged using monoclonal antibodies (anti CD7 for T lineage and anti CD10 and/or anti CD19 for B lineage leukaemias) plus rabbit complement. Retrospective analysis of 19 purged marrows by immunophenotyping or immunoglobulin gene rearrangement studies demonstrated no evidence of disease. Engraftment was seen in 26 of the patients. No correlation was found between the numbers of infused nucleated cells or colony forming units-granulocyte-macrophage (CFU-GM) and subsequent engraftment kinetics. The actuarial disease-free survival (DFS) is 32% at 7 years (median follow-up 3.4 years). There were two transplant related deaths (actuarial risk 8%); the main cause of treatment failure has been disease recurrence with an overall actuarial risk of 67%; 76% for T-ALL (five of nine), 62% for common ALL (five of 10), two of five pre B and none of three patients with B-ALL. In these 27 high risk patients in vitro purging of remission marrow as part of ABMT appears not to improve patient outcome, although confirmation of this would require a randomized trial.

Standardization of T-cell depletion in HLA matched bone marrow transplantation.

The IBM 2991 Blood Cell Processor has been used to isolate a mononuclear cell (MNC) fraction from the marrow of 31 allogeneic donors. The MNC fraction was then incubated with a combination of two murine monoclonal antibodies MBG6 (CD6) and RFT8 (CD8) followed by two rounds of treatment with rabbit complement resulting in a marrow inoculum significantly reduced in the number of T-lymphocytes. We report here new specifications for the use of Ficoll-Metrizoate, the method used to calculate T-lymphocyte depletion and the details of our attempts to improve T-depletion. Following marrow transplantation with this T-depleted fraction, 29 patients are evaluable for engraftment, one patient failed to engraft and one died too early for evaluation. Twenty-two had no acute graft-versus-host disease (aGvHD), at a minimum of 60 d, six had grade I acute GvHD and one grade III. No correlation was found between the absolute number of MNC infused and time to engraftment, nor any relationship between the number of residual viable T-lymphocytes in the infused marrow and the incidence of GvHD, but the patient with the most severe aGvHD also had the highest number of T-lymphocytes infused.

Standardization of the processing of human bone marrow for allogeneic transplantation.

Allogeneic bone marrow transplantation has been undertaken within this centre on 62 patients with acute or chronic leukaemia. Employing the standard separation protocol described, all bone marrows were processed on the 2991 Blood Cell Processor to isolate the 'buffy coat' cells and subsequently the mononuclear cell component using a density separation medium of Ficoll-metrizoate. Following the mononuclear cell separation, the cells were identified for their T lymphocyte component using a combination of two murine monoclonal antibodies, MBG6 reacting with a pan T cell antigen (CD6) and RFT8 detecting the 'cytotoxic/supressor' cell antigen (CD8). The numerical results for nucleated cells, red blood cells, T lymphocytes and colony forming units-granulocyte macrophage are presented.

Depletion of T lymphocytes in donor marrow prevents significant graft-versus-host disease in matched allogeneic leukaemic marrow transplant recipients.

For more than 15 years preclinical studies have suggested that acute graft-versus-host disease (aGvHD) might be prevented by the removal of immunocompetent T lymphocytes from the donor marrow inoculum. To test this observation in man 14 patients were given marrows virtually (greater than 99%) depleted of identifiable donor marrow T lymphocytes by the use of a "cocktail" of specific anti-T-cell monoclonal antibodies (MBG6 and RFT8) and rabbit complement. Patients were not given immunosuppressive prophylaxis after bone-marrow transplantation. Moderate to severe (grades II-IV) GvHD was totally prevented. 2 of 13 evaluable patients showed mild (grade I) skin GvHD only. Although peripheral blood recovery was slower than that obtained with other forms of GvHD prophylaxis, no fatal infections occurred. All patients survived the early post-transplant period.

The role of monoclonal antibodies in the prevention of graft versus host disease.

We have previously reported that the elimination of T-lymphocytes (greater than 99%) from the donor bone marrow prevents significant graft versus host disease (GvHD). This has been confirmed by other centres. Many of these groups have applied monoclonal antibodies with cytolytic rabbit complement. In some of these studies mostly patients with fully matched sibling donors have been studied and no further immunosuppression has been given during the regeneration period. The experience here shows that the haemopoietic regeneration (to recover a total leucocyte count of 1.0 X 10(9)/l) has only been minimally delayed (mean 25 days) when compared to patients receiving standard methotrexate (Mtx) GvHD prophylaxis (22 days). The incidence of cytomegalovirus (CMV) infections (14%; none fatal) was lower than that in our previous 54 patients (43% CMV infections, 13% fatal). Furthermore, no fatal pneumonitis was seen. The regeneration of T-cells in our patients has shown normal values of T8-positive cells in the circulation from 45 to 50 days onwards, as opposed to the previous groups of patients whose T8-positive cells regenerated to 3-4 times higher than normal values. These observations indicate that T-cell depletion with monoclonal antibodies is an effective method for preventing GvHD, but further studies are necessary to investigate the cause(s) of a new occasional complication, the rejection of the newly established bone marrow.

Allogeneic bone marrow transplantation: the monitoring of granulocyte macrophage colonies following the collection of bone marrow mononuclear cells and after the subsequent in-vitro cytolysis of OKT3 positive lymphocytes.

Marrow nucleated cells from eight normal allogeneic donors was layered on Ficoll-Metrizoate to isolate the mononuclear cell fraction. The cells were then washed to remove Ficoll-Metrizoate and coagulation factors prior to resuspension in a balanced salt solution and the addition of the murine anti-human T-lymphocyte monoclonal antibody OKT3 and rabbit complement. The procedures were assessed for their effect on mononuclear cell viability (mean recovery 84.4%); the ability of the cells to proliferate granulocyte-macrophage colonies (mean recovery 57.4%); the in-vitro T-lymphocytolysis (mean 75.7%) and the removal of rabbit complement (greater than 99%). Following marrow transplantation with this treated mononuclear fraction the mean day to recovery of greater than or equal to 1.0 X 10(9)/l leucocytes was 20 d, with three patients developing greater than or equal to Grade II acute graft versus host disease (GvHD). Thus, treatment of donor marrow with OKT3 and complement in a large volume system was not detrimental to subsequent engraftment, nor effective in complete T-lymphocytolysis, nor in prevention of severe GvHD.

GCT-conditioned medium: an unsuitable stimulus for monitoring granulocyte--macrophage colony-forming cells in cryopreserved bone marrow.

Assays of granulocyte-macrophage colony-forming cells provide a means of testing the viability of cryopreserved bone marrow cells intended for autologous transplantation. We have compared two different sources of granulocyte-macrophage colony-stimulating activity, giant cell tumor-conditioned medium (GCT-CM) and peripheral blood leukocyte feeder layers, to determine whether the former is a suitable substitute for leukocyte feeder layers in the assay. The results show that GCT-CM, while providing a comparable stimulus to that provided by leukocyte feeder layers for colony formation by fresh bone marrow samples, is an inadequate stimulus when cryopreserved bone marrow samples are cultured. GCT-CM is not therefore suitable for use in monitoring cryopreserved bone marrow, since there is gross underestimation of the number of colony-forming cells present when this stimulus is used. The results suggest that great care should be taken when selecting alternative sources of granulocyte-macrophage colony-stimulating activity for culture of cryopreserved material.

A technique for the concentration of nucleated bone marrow cells for in vitro manipulation or cryopreservation using the IBM 2991 blood cell processor.

27 bone marrow aspirates were processed on the IBM 2991 Blood Cell Processor to achieve a reduction in volume (greater than 80%) and in red blood cell contamination (greater than 75%), without loss (less than 20%) of nucleated cells. The procedure was used to concentrate nucleated bone marrow cells either prior to cryopreservation for subsequent autologous transplantation, or prior to incubation with a murine monoclonal antibody (OKT3) for allogeneic transplantation. We conclude that the procedures used for the concentration, cryopreservation, thawing and infusion do not adversely affect the viability of cells as assessed by in vitro culture (CFU-GM). Of the marrows processed, 12 have been reinfused and resulted in successful engraftment.

ABO-incompatible bone-marrow transplantation: removal of red blood cells from donor marrow avoiding recipient antibody depletion.

Eight patients with acute leukaemia undergoing allogeneic bone-marrow transplantation from ABO-incompatible donors received red-cell-depleted donor marrow without any procedure to diminish their anti-ABO antibody titres. Successful marrow red-cell removal (mean 98.8%) was achieved by means of a large-volume separation technique on Ficoll-Metrizoate in the IBM 2991 blood-cell processor. Clinically significant ABO-haemolytic reaction was prevented in all patients, and there was neither failure of engraftment nor rejection. This approach used alone is satisfactory for most ABO-incompatible marrow-transplant recipients, although combining this with some method of recipient antibody depletion, such as plasma exchange, is recommended in the occasional patient with high anti-ABO titres.

A technique for rapid isolation of bone marrow mononuclear cells using Ficoll-Metrizoate and the IBM 2991 blood cell processor.

Marrow from seven normal donors and patients has been layered onto Ficoll-Metrizoate (FM) under pressure in the IBM 2991 blood cell processor to isolate the mononuclear cell (MNC) population prior to allogeneic transplantation or cryopreservation. This separation method, which takes less than 90 min, is a further development since our previous report detailing the use of the IBM 2991 to produce a concentrated marrow 'buffy coat' for infusion (Gilmore & Prentice, 1981). By adding FM to the system, marrow stem cells are further concentrated in a small volume with removal of unwanted granulocytes and red blood cells. This facilitates in in vitro treatment of marrow with monoclonal antibodies (Granger et al. 1982) or drugs, for either the selective elimination of malignant cells prior to autologous bone marrow transplantation (BMT), or T lymphocytes in an attempt to prevent graft versus host disease (GvHD) in allogeneic BMT. Five of the seven marrows processed by this procedure have thus far been infused into lethally irradiated recipients with engraftment (allogeneic); the other two marrows have been cryopreserved.