Further evidence for the existence of a clonal Ph-negative stage in some cases of Ph-positive chronic myelocytic leukemia.

The Ph chromosome abnormality is involved in the pathogenesis of almost all patients with chronic myelocytic leukemia (CML). Previous studies on the B-lymphoid cell lineage in two patients with Ph-positive CML suggest that there may also be a clonal Ph-negative stage in CML and that the Ph-positive stage arises by subclonal expansion. To determine whether this is a frequent or a rare occurrence, 14 additional glucose-6-phosphate dehydrogenase (G6PD)-heterozygous patients with CML were studied. In five of these patients there was a statistically significant excess of Ph-negative B-lymphoid cell lines expressing the same G6PD type expressed in the corresponding CML clone. In no case was an excess of B-lymphoid lines expressing the opposite G6PD type recovered. These data provide further evidence that in some patients the Ph chromosome arises in a pluripotent stem cell from a pre-existing Ph-negative clone that enjoys a growth advantage.

Recombinant human erythropoietin therapy for anemic cancer patients on combination chemotherapy.

BACKGROUND: Patients with advanced cancer frequently experience clinically significant anemia, which is often exacerbated by myelosuppressive chemotherapy. Consistent with the anemia of chronic disease, studies have documented serum erythropoietin levels that are inappropriately low for the degree of anemia in cancer patients. Myelosuppressive chemotherapy impairs erythropoiesis, which may not fully recover between treatment cycles. Recombinant human erythropoietin (rHuEPO) has been used safely and effectively to treat anemia in AIDS patients receiving zidovudine (AZT) and in patients with chronic renal failure. PURPOSE: This study was designed to evaluate the clinical role of rHuEPO in reducing symptomatic anemia in patients with advanced cancer who were receiving myelosuppressive chemotherapy (excluding cisplatin). METHODS: We studied 153 anemic cancer patients receiving cyclic combination chemotherapy in a prospective multicenter, double-blind, placebo-controlled trial. The patients were randomly assigned to receive either rHuEPO (150 U/kg) or placebo subcutaneously three times a week for a maximum of 12 weeks or until the hematocrit level increased to 38%-40%. If the hematocrit reached this target level before 12 weeks, the rHuEPO dose could be reduced to maintain the hematocrit at that level for the duration of the study. Response to rHuEPO therapy was assessed by measuring changes in hematocrit level, transfusion requirements, and quality of life. Quality-of-life assessment was based on patients' responses to questionnaires before and after the courses of therapy. RESULTS: The increase in hematocrit in the rHuEPO-treated group compared with hematocrit in the placebo-treated group was statistically significant (P = .0001) as measured by percentage point of change from baseline to final evaluation, by an increase in hematocrit level of six percentage points or more unrelated to transfusion, and by a rise in hematocrit level to 38% or more unrelated to transfusion. There was a trend toward the reduction in mean units of blood transfused per patient during months 2 and 3 of therapy combined in rHuEPO-treated patients compared with placebo-treated patients (0.91 U versus 1.65 U; P = .056). In addition, rHuEPO-treated patients experienced a statistically significant improvement in energy level and ability to perform daily activities (P < or = .05). The two treatment groups showed no statistically significant differences in toxic effects except for increased incidence of diaphoresis (P < .05) and diarrhea (P = .05) in the rHuEPO-treated group. CONCLUSIONS: We conclude that rHuEPO is safe and effective for reversing anemia related to advanced cancer or to chemotherapy for cancer.

Temporal association of marrow eosinophilia with inversion of chromosome 16 in recurrent blast crises of chronic myelogenous leukemia.

We report a patient with Ph+ chronic myelogenous leukemia (CML) whose recurrent blast crises were associated with marrow eosinophilia and inv(16). After intensive chemotherapy, for each blast crisis, the patient reentered chronic phase with disappearance of both the inv(16) and the eosinophilia.

Evidence for clonal development of Wilms' tumor.

To assess the clonality of Wilms' tumor, glucose-6-phosphate dehydrogenase (G6PD) enzymes were studied in normal and tumor tissue from 11 black girls who were heterozygous for G6PD. Normal tissues expressed both A and B type G6PD, whereas only a single G6PD enzyme was found in all tumor specimens. These data support the clonal nature of Wilms' tumor. In the one patient with bilateral disease, type B G6PD was found in both a recurrence and a subsequent tumor in the contralateral kidney. This finding is consistent with either the chance occurrence of the same G6PD in independent tumors or persistence of the original malignant clone. Another patient, who presented with the nephroblastomatosis complex (a precursor of Wilms' tumor), also had only type B enzyme detected. Further studies in patients with bilateral disease or the nephroblastomatosis complex, including the use of molecular biologic probes, are needed to test the hypothesis that Wilms' tumor in these cases arises from a somatic mutation as a second event in persons with an underlying genetic alteration.

Unique human neutrophil populations are defined by monoclonal antibody ED12F8C10.

A mouse IgG1 monoclonal antibody ED12F8C10 (C10) binds a constant percentage of peripheral blood neutrophils in the same individual when studied over time, defining a distinct subset of neutrophils in all normal individuals studied to date. Bone marrow studies confirm that the heterogeneity is present to the same degree at all stages of neutrophil development from the myelocyte to the mature neutrophil. Neither in vivo nor in vitro activation of neutrophils explains or significantly alters the relative percentages of C10-positive and -negative neutrophils in the same individual. With both activation and exudation, however, expression of the C10-defined epitope increases in intensity in the C10 binding subpopulation. Studies of NBT reduction, phagocytosis, adherence, light scattering characteristics, and monoclonal antibody surface binding have failed to demonstrate physical or functional differences between the C10-defined populations. We examined C10 binding in patients with different defects of phagocyte function. In two patients with neutrophil-specific granule deficiency, less than 1% of the neutrophils were found to be C10 positive, while neutrophils from a patient with idiopathic leukemoid reaction and recurrent infections demonstrated greater than 99% C10 binding. Although the present study does not delineate the physiologic significance of C10 binding heterogeneity, it firmly supports the concept of neutrophil heterogeneity at the level of surface antigen expression.

Human B cell lines can be triggered to secrete an interleukin 2-like molecule.

To determine whether human B cells can be triggered to secrete interleukin 2 (IL-2), 19 tumor cell lines derived from patients with undifferentiated lymphomas of Burkitt's and non-Burkitt's types and 6 normal lymphoblastoid cell lines were tested. Cells were grown in the presence or absence of the new tumor promoter teleocidin, and culture supernatants were assayed for IL-2 activity using the standard CTLL-2 assay. Teleocidin (10 ng/ml) triggered IL-2 secretion in 7/8 (87%) EBV-negative lymphoma cell lines of American origin and in 6/6 (100%) normal lymphoblastoid cell lines, but in only 1/6 (16%) EBV-positive tumor cell lines of American origin. Teleocidin had no effect on 5/5 (0%) African Burkitt's cell lines. IL-2 secretion was not detected in control supernatants. IL-2 secretion correlated with the induction of IgM secretion and was linked to both EBV status and karyotype. The following similarities in the functional biological characteristics of T cell and B cell IL-2 suggest that B cell IL-2 is not a factor which mimics IL-2 activity in the CTLL-2 assay: (i) neutralization of IL-2 by anti-IL-2 monoclonal antibody (DMS-1); (ii) elution of IL-2 following its adsorption to CTLL-2 cells; (iii) determination of the MW of IL-2 by SDS-PAGE and Western blot analysis; and (iv) ability of B cell IL-2 to support T cell proliferation and blocking of this activity by anti-tac monoclonal antibody. cDNA probes for T cell IL-2, however, did not detect IL-2 mRNA in B cells. The cell lines were also found to constitutively express IL-2 receptors detected by anti-tac monoclonal antibody, and to secrete soluble IL-2 receptors measured by ELISA. Our results imply that under certain circumstances, B cells can be triggered to secrete IL-2 or an IL-2-like molecule and thus influence T cell activation and proliferation.

Clonal development, stem-cell differentiation, and clinical remissions in acute nonlymphocytic leukemia.

To determine whether acute nonlymphocytic leukemia develops clonally, to study the pattern of differentiation of the involved stem cells, and to determine whether clinical remissions are true remissions, we studied 27 patients with acute nonlymphocytic leukemia who were heterozygous for the X-chromosome-linked glucose-6-phosphate dehydrogenase. In each case, leukemic blast cells manifested only one type of glucose-6-phosphate dehydrogenase, indicating that the malignant process had developed from a single cell. In six elderly patients, circulating erythrocytes, platelets, or both expressed only the glucose-6-phosphate dehydrogenase found in blast cells, indicating that these leukemias had arisen from stem cells with multipotent differentiative expression. In 16 younger adults and children, erythroid cells and platelets were predominantly derived from normal stem cells. In three other cases, the stem cell that gave rise to leukemic blasts apparently also gave rise to erythroid progenitors but not to mature erythrocytes. Heterogeneity was also found during remissions. In 8 of 13 patients, restoration of nonclonal hemopoiesis and repopulation of the marrow by normal stem cells was observed during remission. In the other five patients, marrow stem cells remained partially or completely clonal, even during remission. These data indicate that acute nonlymphocytic leukemia is a heterogeneous disease with respect to differentiation of the stem cells involved by leukemia and the nature of remissions.

A myelodysplastic syndrome with marrow eosinophilia terminating in acute nonlymphocytic leukemia, associated with an abnormal chromosome 16.

A patient presented with a myelodysplastic syndrome and bone marrow eosinophilia that evolved six months later into an acute nonlymphocytic leukemia (ANLL). Cytogenetic analyses of the bone marrow revealed 86% of the metaphases with 45,X-Y,inv(16)(p13;q22),t(11;17) (q11;q25),del(21)(q13) and 14% of the metaphases with the same abnormalities but with a Y chromosome. The association of ANLL, bone marrow eosinophilia, and abnormal chromosome 16 has previously been reported and has been suggested to have a favorable prognosis. Our patient is unique in that ANLL was preceded by a preleukemic phase associated with bone marrow eosinophilia. When complete remission was achieved, the bone marrow cytogenetics returned to normal, and the eosinophilia disappeared.

Aspergillus myositis in a patient with a myelodysplastic syndrome.

We present the case of an elderly man who, while being treated with corticosteroids for a myelodysplastic syndrome, developed myositis of the calf due to Aspergillus fumigatus. Despite therapy with amphotericin B the myositis failed to resolve and he died. At autopsy, a localized necrotizing myositis of the right calf was found with no evidence of disseminated Aspergillus infection. Myositis in the setting of disseminated candidiasis or cryptococcosis has been previously reported. This case is unique in that it is the first reported case of localized fungal myositis and of myositis caused by Aspergillus.

Histiocytosis X: report of an oral soft tissue lesion without bony involvement.

The case of a 65-year-old white man with painful oral soft tissue granulomatous lesions of histiocytosis X is reported. The clinical course and diagnostic and therapeutic measures are described. The manifestation of symptomatic oral soft lesions with no definable lesions of bone and the age of the patient are not consistent with the usual presentation of this disease, and thus emphasize its clinical variability. The rationale for the therapeutic regimen and the prognosis are reviewed.

Heterogeneity of B-cell growth factor receptor reactivity in healthy donors and in patients with chronic lymphatic leukemia: relationship to B-cell-derived lymphokines.

Twenty-five long-term B-cell lines were studied for B-BCGF activity. The cell lines were cultured in the presence or absence of the new tumor promoter teleocidin, and control and teleocidin-treated derived supernatants were cocultured with purified B cells obtained from healthy donors and patients with B chronic lymphatic leukemia (B-CLL), in the presence of anti-mu. In attempt to delineate the role of other B-cell lymphokines in promoting proliferation of activated B cells, the supernatants were also studied for interleukin 1 (IL-1), interleukin 2 (IL-2), and interferon gamma (IFN-gamma). The effect of B-cell-derived lymphokines on the proliferation of activated B cells obtained from the 14 donors was heterogeneous, and three types of response were observed: In four healthy donors there was induction of B-cell proliferation by B-cell lymphokines derived from both control cells and teleocidin-treated cells. In cells obtained from the other five healthy donors there was induction of B-cell proliferation by B-cell lymphokines derived from teleocidin-treated cells. In the five B-CLL patients, B-cell proliferative response to B-cell lymphokines derived from both control cells and teleocidin-activated cells was absent. Comparison of B-BCGF reactivity to T-BCGF reactivity demonstrated that B-CLL B lymphocytes did not respond to either B-BCGF or T-BCGF, whereas normal B cells responded to T-BCGF and may proliferate upon stimulation with B-cell-derived IL-2 and/or B-BCGF. These results suggest heterogeneity of B-BCGF receptor reactivity in B lymphocytes derived from healthy donors, and lack of both B-BCGF and T-BCGF receptor reactivities in B lymphocytes derived from B-CLL patients; B-cell-derived lymphokines influence normal B-cell response but not leukemic B cells; B-BCGF optimal effect is in large part due to other B-cell lymphokines, especially B-cell-derived IL-2; the possible existence of various B-BCGFs.

A neutrophil membrane marker reveals two groups of chronic myelogenous leukemia and its absence may be a marker of disease progression.

An IgG1 monoclonal antibody, 31D8, that recognizes normal neutrophil (PMN) membranes, was used to study PMN from patients with chronic myelogenous leukemia (CML). Nineteen patients with Philadelphia chromosome positive CML were followed over a ten-month period and compared with 23 normals, six patients with leukemoid reactions, and eight patients with phagocytic cell defects. The percentage of PMN binding of 31D8 among normal subjects was variable about a normal distribution with an average of 95 +/- 2% of cells binding 31D8. In contrast, there were two groups of CML patients: in 14 patients 88 +/- 3% PMN bound 31D8 while in the remaining five patients only 6 +/- 6% PMN bound 31D8. PMN 31D8 binding was normal in the control patient groups. Control antibodies 7C3 (binds to PMN precursors) and OKM1 (binds to the CR3 (iC3b) receptor) bound normally to CML neutrophils. Functionally, CML cells had normal chemotaxis to several stimuli and normal superoxide generation to phorbol myristate acetate. However, superoxide production in response to fmet-leu-phe was significantly less in 31D8 negative CML PMN than both 31D8 positive CML PMN and normal PMN which contained 85% 31D8 positive and 15% 31D8 negative PMN. Clinically, 2 of 14 CML patients with 31D8 positive PMN were in blast crisis (one extramedullary) at the time of study and the other 12 patients remained clinically stable in the chronic phase during the ten months of study. In contrast, one of five patients with 31D8 negative PMN was in blast crisis at the time of study and all four of the remaining patients progressed to either the accelerated phase or blast crisis. Three of these patients died of their disease eight to ten months after their initial study. Thus, failure of CML cells to bind 31D8 may be useful for predicting which patients are likely to progress to the accelerated phase or blast crisis.

Burkitt's cells can be triggered by teleocidin to secrete interferon-gamma.

The secretion of interferon (IFN)-gamma by T lymphocytes is mediated by the synthesis of interleukin 2 (IL-2) and the availability of IL-2 receptors. Since some Burkitt's lymphoma lines express Tac antigen and can be triggered to secrete IL-2 following activation with the new tumor promoter teleocidin, we addressed the question of whether the induction of IL-2 by B lymphocytes is accompanied by the induction of IFN-gamma. IFN-gamma has not been detected in any of the 25 cell lines studied, and following stimulation with teleocidin, we triggered the synthesis of IFN-gamma in JLP(C), a pre-Burkitt's cell line. The mechanism of IFN-gamma secretion by B lymphocytes is not clear. Our findings demonstrate that the synthesis of IL-2 by B cells is not accompanied by IFN-gamma and suggest that the synthesis of IFN-gamma is not mediated by IL-2 or IL-1 or B-cell growth factor. Neutralization studies have shown that IFN-gamma secretion is not accompanied by the induction of IFN-alpha or IFN-beta. Our data imply that B cells can be triggered to secrete IFN-gamma under certain circumstances. Whether similar function occurs in vivo is not known.

Heterogeneity of B cell involvement in acute nonlymphocytic leukemia.

In order to study the pattern of B cell involvement in acute nonlymphocytic leukemia (ANLL), multiple B lymphoid cell lines were established by Epstein-Barr virus transformation of peripheral blood mononuclear cells from two patients with the disease who were heterozygous for the X chromosome-linked glucose-6-phosphate dehydrogenase (G6PD). In one patient, the progenitor cells involved by the leukemia exhibited multipotent differentiative expression, whereas in the other patient the cells showed differentiative expression restricted to the granulocytic pathway. In the patient whose abnormal clone showed multipotent expression, the ratio of B-A G6PD in B lymphoid cell lines was skewed in the direction of type B (the enzyme characteristic of the leukemia clone) and significantly different from the 1:1 ratio expected. It is, therefore, likely that the neoplastic event occurred in a stem cell common to the lymphoid series as well as to the myeloid series. In contrast, evidence for B cell involvement was not detected in the patient whose ANLL progenitor cells exhibited restricted differentiative expression. These findings underscore the heterogeneity of ANLL. Clinically and morphologically similar malignancies in these two patients originated in progenitors with different patterns of stem cell differentiative expression. This difference may reflect differences in cause and pathogenesis.

Hydroxyurea interferes negatively with triglyceride measurement by a glycerol oxidase method.

Measured triglyceride concentrations were extremely low (less than 100 mg/L) in the serum of some patients who were receiving hydroxyurea for myeloproliferative diseases. The assay being used to quantify triglycerides was a "cascaded" enzymatic method involving (a) lipase, to generate glycerol from triglycerides; (b) glycerol oxidase, to convert glycerol to glyceraldehyde, with generation of hydrogen peroxide; and (c) peroxidase, which acts on the hydrogen peroxide with subsequent coupled generation of a red-violet quinone (reagent system used in the Technicon RA-1000). Hydroxyurea added to serum samples appeared to inhibit the action of glycerol oxidase, with a stoichiometric relation to the concentration of substrate (a decrease of roughly 2.4 mmol/L in measured triglyceride per 1 mmol of hydroxyurea per liter). A different enzymatic assay for triglycerides, which involves glycerol kinase (Beckman Instruments) did not show this effect of hydroxyurea.