Prenatal diagnosis of hematologic diseases can now be performed with fetal blood, fetal amniotic fluid cell DNA, and fetal chorionic villi DNA. Some hemoglobinopathies can be detected by all three methods, and the choice will depend on the available obstetric and laboratory techniques, as well as the time of presentation of the pregnancy. Hopefully, further development of molecular probes and techniques will soon expand these options to all of the globin disorders. Detection of coagulation disorders in utero currently requires samples of pure fetal blood. Gene cloning is accomplished for some (factor IX and antithrombin III) and is underway for others (factor VIII), and further investigation is necessary to determine whether deficiencies in these gene products are due to gene deletion or to mutant genes linked to polymorphic restriction enzyme sites of diagnostic use. Thus, molecular biology may be applied to prenatal diagnosis of the clotting problems, but this has not yet been accomplished. Disorders affecting the number and/or function of erythrocytes, leukocytes, and platelets can be diagnosed by analysis of fetal blood. Blood samples will continue to be required until more is known about the molecular biology of hematopoiesis. Syndromes that can be diagnosed by chromosome studies should be revealed in cultures of amniotic fluid cells, fetal blood lymphocytes, and chorionic villi cells. Cultured cells can be examined for karyotypes, Y-chromatin, spontaneous or induced chromosome breakage, DNA repair, SCEs, and translocations. The techniques for culturing amniotic cells and fetal blood white cells are established, and those for growing cells from chorionic villi are improving rapidly. Direct preparations of cells from villi only may suffice for some of the above analyses. The study of hematologic disease in utero has thus come full circle, from the use of amniotic cells to determine the sex in X-linked disorders, to fetal blood sampling for the analysis of gene products, then back to amniocentesis for DNA, and now earlier in gestation to chorionic villi. All of this has occurred in less than ten years, and it is anticipated that developments in the next ten years will be equally dramatic. The future should bring all prenatal testing into the first trimester, use molecular probes, and provide for both early diagnosis and early treatment of genetic hematologic disease.

Bleeding and thrombosis are major causes of morbidity and mortality in patients with myeloproliferative disorders. The significance of uncontrolled polycythemia as a risk factor for thrombosis in these patients has been established. However, the role of thrombocytosis in the pathogenesis of hemostatic complications remains controversial. Abnormalities of platelet function and prolongation of the bleeding time occur in a highly variable number of cases. Specific platelet defects that have been identified in the myeloproliferative defects include abnormal platelet morphology, acquired storage pool disease, platelet membrane abnormalities, and abnormal arachidonic acid metabolism. Causal relationships between any of these specific abnormalities and either bleeding or thrombosis have not been clearly established. The therapeutic efficacy of myelosuppression to reduce the platelet count in patients with thrombocytosis and the role of antiplatelet drugs in the myeloproliferative disorders are controversial issues.

Platelet activity may be involved in tumor metastasis. The tumor cells, after detachment from the primary site, adhere to vascular endothelium at distant sites and proliferate. Platelets form aggregates with tumor cells in circulation, facilitating their adhesion to the vascular endothelium. Formation of platelet-tumor cell aggregates and their sequestration in various end-organs may result in thrombocytopenia. Certain tumor cell lines directly stimulate platelet activity, some by releasing platelet-aggregating material, a urea-extractable membrane component, others by release of cathepsin, and still others by undefined mechanisms. The direct effect of platelets on tumor cells may be of pathogenic significance. For example, platelet-derived factors stimulate growth of some tumors, whereas others increase vascular permeability and thus facilitate migration of tumor cells across the vessel wall. Lack of these platelet factors, as in thrombocytopenic animals, may indeed inhibit tumor metastasis. Arachidonic acid metabolism in platelets and the vessel walls may contribute to metastatic process. In particular, thromboxane A2 and prostacyclin generation capabilities appear to be important in modulating platelet-tumor cell deposition and growth. To alter the metastatic process, several preliminary trials of platelet-inhibitory agents have been performed. However, the results of these trials have been equivocal, perhaps related to nonspecific effects of these agents on arachidonic acid metabolism. Studies directed at specific pathways of platelet-vessel wall interaction on some tumors appear promising. These newer agents may be of therapeutic value in man.

A prominent phenotypic abnormality of human acute myelogenous leukemia cells is the inability of the cells to differentiate to functional mature cells; instead, the cells are blocked at an early stage of development and remain in the proliferative pool and rapidly accumulate. Investigation of the induction of myeloid leukemic cell differentiation has made recent advances with the development of several human myelogenous leukemia cell lines. The lines provide models to study the biology of myeloid differentiation and to identify inducers of differentiation of myeloid leukemic blood cells. This review critically examines the inducers of leukemic cell differentiation and their potential therapeutic importance.

The application of banding techniques to the study of tumor chromosomes during the past 10 yr has yielded extensive data defining the types of chromosome changes that occur in human hematopoietic and other tumors. Chromosome changes characterize most tumors and exhibit a high degree of nonrandomness; in some tumors, this nonrandomness is defined by the tumor-inducing agents, while in others, it is defined by target cells of tumorigenesis. These observations led to the suggestion that chromosome abnormalities impart a selective advantage to the cells in which they occur and hence are of importance in the development of tumors. The mechanisms by which cells carrying chromosome abnormalities gain selective advantage are beginning to become apparent as the recently acquired data on the molecular genetics of neoplastic transformation are considered in conjunction with cytogenetic data. Activation of cellular oncogenes and overproduction of their products has been shown to be a key step in some types of neoplastic transformation. Chromosome abnormality is suggested to accomplish this step by either causing alterations in oncogene dosage or by activating normally quiescent oncogenes by bringing them into the transcriptional control of active genes. The paradigm for the latter model is the development of human and murine B-cell neoplasms in which specific translocations transfer c-myc from its constitutive site to a site next to the immunoglobulin genes. The chromosomal positions of several oncogenes have now been determined, and the elucidation of their fate in association with chromosome abnormalities in tumor cells can be expected to clarify mechanisms of oncogenesis.

Fifteen compounds reported to be inhibitors of gelation or sickling were studied by standard methods. These tests included (1) the determination of the solubility of deoxyhemoglobin S or Csat, (2) evaluation of sickling in whole SS blood at various pO2s, (3) measurement of the oxygen affinity of hemoglobin and blood, and (4) examination of red cell indices and morphology. Among the 4 noncovalent agents tested, butylurea was the most potent inhibitor of gelation and sickling in vitro; however, relatively high concentrations were required compared to the covalent agents. In the latter group, bis-(3,5 dibromosalicyl)-fumarate, nitrogen mustard, and dimethyladipimidate were especially effective inhibitors of gelation and/or sickling. All of these compounds require further development before they can be considered for clinical use.

The selectivity of monoclonal antibody J-5 (anti-gp 100, common ALL antigen) for normal and leukemic hemopoietic cells has been investigated. J-5 gave concordant reactions with rabbit anti-cALL, coredistributed on the cell surface, and precipitated a similar if not identical glycoprotein from leukemic and normal tissue. Normal, immature lymphoid cells reactive with J-5 were detected in bone marrow and in fetal thymus. In marrow they were largely coincident with the TdT+ population. J-5 defines a major subgroup of ALL (common ALL) with a favorable prognosis. Of 853 non-ALL acute leukemias investigated, 80 were J-5 positive. These included 14 cases diagnosed as AML, 51 TdT+ blast crises of CGL, and 15 cases diagnosed as "AUL." Of the 14 J-5+ AML, 13 were subsequently rediagnosed either as cALL (10 cases) or mixed lymphoid-myeloid leukemias (3 cases). One-hundred forty-three cases of mature lymphoid cell leukemia (91 B, 52 T) were investigated with J-5; 3 cases only, of disseminated B lymphoma, were positive, albeit weakly. A higher proportion of follicular lymphomas are, however, J-5 positive when studied in sections of biopsy material. A similar pattern of selective reactivity was observed in a series of leukemia/lymphoma cell lines. These studies emphasize the diagnostic value of monoclonal anti-cALL reagents.

A small quantity of extracellular calcium is required for the stimulation of lymphocytes by mitogens such as plant lectins. Lectin binding to the lymphocyte surface and early postbinding events that eventually lead to DNA synthesis are calcium dependent. Mitogenic lectins such as PHA and Con-A rapidly increase the size of an exchangeable pool of cell calcium and cause a smaller rise in intracellular ionized calcium. The increase in ionized calcium is so small (100-200 nM), however, that no increase in total cell calcium is measurable. When lymphocytes are stimulated by a lectin, the rate of calcium entry into the cell increases, but the plasma membrane calcium extrusion pump can prevent the total cell calcium from increasing measurably. The calcium ionophore A23187 is a lymphocyte mitogen and causes an increase in the exchangeable, ionized, and total cell calcium. The former two effects may be causal in mitogenesis; the latter effect is cytotoxic. With A23187 treatment, the rate of calcium influx exceeds the maximum rate of the plasma membrane extrusion pump and cell calcium increases in proportion to the concentration of A23187. The mitochondria, by virtue of their high membrane potential, provide a sink for the buffering of cytoplasmic calcium after A23187 treatment. Thus, the plasma membrane or mitochondria regulate the distribution of lymphocyte calcium when the cell is stimulated by mitogenic lectins or ionophores. The evidence strongly suggests that an alteration in calcium pools or an increase in cytoplasmic ionized calcium plays a role in the initiation of the biochemical reactions that lead to mitogen-induced lymphocyte proliferation in vitro and, perhaps, to the immune response.

Major advances have been made in the past 10 yr in both the understanding of the biologic characteristics of acute nonlymphocytic leukemia and in the treatment of patients with this disease. Advances in the biologic characteristics include: a better understanding of the nature of leukemic cell proliferation and differentiation; a clearer description of the morphological, histochemical, and ultrastructural characteristics of leukemic cells; a recognition that a high percentage of patients may have specific cytogenetic abnormalities; and a recognition that biochemical differences exist between acute nonlymphocytic leukemia (ANLL) and acute lymphoblastic leukemia (ALL). Today, over 70% of children with ANLL can be induced into a complete remission and over 25% are remaining in a continuous remission for over 2 yr. In spite of these improved results, the best method of extending remissions is unknown. It is unlikely that better results of therapy will be achieved in the future by tailoring the treatment according to the biologic characteristics of the patient, since it appears that ANLL is a heterogeneous group of diseases.