Isolated thrombocytopenia after bone marrow transplantation was investigated in 65 fully engrafted patients surviving at least 60 days posttransplant. Twenty-four patients (37%) developed this complication, which occurred most frequently in patients receiving pretransplant preparation with total body irradiation or busulfan. Two distinct thrombocytopenic syndromes were identified: (1) transient thrombocytopenia (nine patients), in which a normal platelet count (greater than 100,000/microL) was initially established by day +40 but then diminished to less than 10,000 to 45,000/microL on day +40 to +70, with subsequent resolution of the thrombocytopenia by day +90; (2) chronic thrombocytopenia (15 patients), in which a platelet count greater than 100,000/microL was not achieved at any time during the first four months posttransplant, despite the simultaneous presence of normal granulocyte and reticulocyte counts. Although the transient syndrome did not adversely affect prognosis, the chronic syndrome carried a high mortality (21% actuarial survival at 1,000 days posttransplant compared with 67% survival for all patients, P less than .01) and had a high association with both severe (grades 3 to 4) acute graft-versus-host disease (GVHD) and chronic GVHD. In three of nine patients with transient thrombocytopenia, a temporal association with trimethoprim-sulfamethoxazole administration was observed, whereas in all other patients, no drug association could be found. Bone marrow biopsies in those patients with drug-associated thrombocytopenia showed decreased numbers of megakaryocytes, whereas biopsies in the remainder of the transiently thrombocytopenic patients demonstrated adequate numbers of platelet precursors, suggesting peripheral platelet destruction or ineffective thrombopoiesis. Biopsies in the chronic thrombocytopenic patients included those with and without adequate numbers of platelet precursors, although the association with chronic GVHD was strongest in patients demonstrating normal numbers of megakaryocytes. We conclude that isolated thrombocytopenia represents a significant complication of bone marrow transplantation, particularly in patients receiving hematopoietic ablative preparatory regimens, and that it is the chronic, not the transient, thrombocytopenic syndrome that is associated with an adverse patient prognosis.

Paroxysmal nocturnal hemoglobinuria (PNH) involves the proliferation of an abnormal and possibly premalignant hematopoietic stem cell. Successful treatment of PNH by marrow grafting requires that the PNH clone be eradicated by the pretransplant conditioning regimen. Four patients with PNH-associated marrow aplasia were transplanted with marrow from their HLA-matched, MLR-nonreactive siblings. Three patients were conditioned with cyclophosphamide, procarbazine, and antithymocyte serum (CTX/PCZ/ATS), and one was conditioned with busulfan/CTX/PCZ/ATS. Persistent complete engraftment of myeloid, lymphoid, and erythroid cell lines was demonstrated in all four patients by DNA sequence polymorphism analysis or cytogenetics, and RBC typing. There was no recurrence of the abnormal clone of cells for up to five years after transplantation despite the use of a conditioning regimen in three of them, which is not usually associated with permanent marrow aplasia. Bone marrow transplantation is a curative therapy in patients whose illness is severe enough to warrant the risk.

We have studied the effect of removing donor T cells by treatment with the monoclonal antibody Leu-1 and complement before marrow transplantation on the regeneration of functionally competent T lymphocytes in the blood at selected times after transplant. Using sensitive limiting-dilution methods that allow us to enumerate helper, cytotoxic, and proliferating T lymphocyte precursors, we report that regeneration of a functional T cell compartment is more severely impaired for the first 180 days after transplantation in those patients given T cell-depleted bone marrow than in recipients of untreated marrow. After this first 6 months, however, patients given T cell- depleted bone marrow had blood T cell frequencies comparable to those observed in patients given untreated marrow. Diminished frequencies of reactive T cells in recipients of depleted marrow could leave them more susceptible to infection or to the recurrence of neoplastic cells.

We prospectively documented the development of a fatal, secondarily acquired severe immunodeficiency in a 19-year-old man who underwent uncomplicated bone marrow transplantation. He had no graft v host disease (GVHD) and had normal recovery of his immune system as determined by lymphocyte phenotyping, mitogenic responses of his peripheral blood lymphocytes, and his ability to secrete immunoglobulin. This alteration in immunity was associated with the acquisition of antibody to HTLV-III. His only risk factor for the development of HTLV-III infection was the transfusions he had received during the transplant and recovery period. Two of his 54 transfusions were from an asymptomatic individual at high risk for acquired immunodeficiency syndrome (AIDS), who was subsequently found to be seropositive for anti-HTLV-III and from whom HTLV-III was isolated. The loss of immunocompetence in patients without chronic GVHD disease is unusual, and our data support the view that this patient's immunodeficiency was due to HTLV-III. When bone marrow transplant recipients without chronic GVHD develop late opportunistic infections, consideration should be given to transfusion-associated AIDS.

The role of natural killer (NK) lymphocytes in the regulation of human hematopoiesis is controversial. NK-mediated inhibition of colony formation of hematopoietic progenitor cells has been irregularly reported for various cell lineages. In an effort to clarify such disparate findings, we studied the interaction of clearly defined NK and partially purified progenitor cell populations. Cell sorter purified CD16 positive blood NK cells and enriched autologous marrow progenitors were co-incubated at various lymphocyte to marrow cell ratios and then cultured in methylcellulose. There was no inhibition of myeloid, erythroid, or mixed colony formation. Similarly, activation of CD16 positive lymphocytes by interleukin-2 (IL-2) before co-incubation and co-culture did not result in inhibition of colony formation. Furthermore, in a newly designed assay system, we demonstrated that NK cells, which did not modulate colony-formation, remained capable of recognizing and killing rare K562 target cells seeded within the marrow cell population. Our results indicate that unstimulated and IL-2 activated isolated blood NK cells coexist with functioning autologous marrow progenitors in vitro.