Search

Chronic myeloid leukemia (CML) has been regarded as the paradigmatic example of a malignancy defined by a unique molecular event, the BCR-ABL1 oncogene. Decades of research zeroing in on the role of BCR-ABL1 kinase in the pathogenesis of CML have culminated in the development of highly efficacious therapeutics that, like imatinib mesylate, target the oncogenic kinase activity of BCR-ABL1. In recent years, most research efforts in CML have been devoted to developing novel tyrosine kinase inhibitors (TKIs) as well as to elucidating the mechanisms of resistance to imatinib and other TKIs. Nonetheless, primordial aspects of the pathogenesis of CML, such as the mechanisms responsible for the transition from chronic phase to blast crisis, the causes of genomic instability and faulty DNA repair, the phenomenon of stem cell quiescence, the role of tumor suppressors in TKI resistance and CML progression, or the cross-talk between BCR-ABL1 and other oncogenic signaling pathways, still remain poorly understood. Herein, we synthesize the most relevant and current knowledge on such areas of the pathogenesis of CML.

Chronic myelogenous leukemia (CML) evolves from a chronic phase characterized by the Philadelphia chromosome as the sole genetic abnormality into blast crisis, which is often associated with additional chromosomal and molecular secondary changes. Although the pathogenic effects of most CML blast crisis secondary changes are still poorly understood, ample evidence suggests that the phenotype of CML blast crisis cells (enhanced proliferation and survival, differentiation arrest) depends on cooperation of BCR/ABL with genes dysregulated during disease progression. Most genetic abnormalities of CML blast crisis have a direct or indirect effect on p53 or Rb (or both) gene activity, which are primarily required for cell proliferation and survival, but not differentiation. Thus, the differentiation arrest of CML blast crisis cells is a secondary consequence of these abnormalities or is caused by dysregulation of differentiation-regulatory genes (ie, C/EBPalpha). Validation of the critical role of certain secondary changes (ie, loss of p53 or C/EBPalpha function) in murine models of CML blast crisis and in in vitro assays of BCR/ABL transformation of human hematopoietic progenitors might lead to the development of novel therapies based on targeting BCR/ABL and inhibiting or restoring the gene activity gained or lost during disease progression (ie, p53 or C/EBPalpha).

Imatinib is a tyrosine-kinase inhibitor that binds to ABL proteins and induces cytogenetic remissions in patients with chronic myeloid leukemia (CML). In these patients measuring response by molecular techniques is clearly required. We determined the cytogenetic and molecular response (CgR, MR) to imatinib in 191 patients with late chronic-phase Philadelphia-positive (Ph+) CML, previously treated with interferon alpha. MR was assessed with real-time quantitative (TaqMan) reverse transcription-polymerase chain reaction and was expressed as the ratio between BCR/ABL and beta 2-microglobulin x 100, the lowest level of detectability of the method being 0.00001. A complete CgR (CCgR) was achieved in 85 (44%) of 191 patients and was maintained for 2 years in 67 (79%) of 85 patients. A reduction of the transcript level of more than 2 logs was achieved in all but 9 patients with CCgR versus none of 23 with partial CgR. In the CCgRs the median value of the MR was 0.0008 after 12 months and 0.0001 after 24 months, with the transcript level undetectable in 22 cases. We conclude that in CCgRs the degree of MR may vary from 2 to more than 4 logs, and that there is a progressive decrease of transcript level by time. Only 1 of 22 negative cases has had a relapse as yet.

We have previously found that P210BCR-ABL increases the adhesion of hematopoietic cell lines to fibronectin by a mechanism that is independent of tyrosine kinase activity. To investigate the pathway(s) by which P210BCR-ABL influences cell adhesion, we used a quantitative cell adhesion device that can discern small changes in cell adhesion to assay P210BCR-ABL with mutations in several critical domains. We expressed P210BCR-ABL mutants in 32D myeloblast cells and found that binding to fibronectin is mediated primarily by the alpha5beta1 integrin. We performed a structure/function analysis to map domains important for cell adhesion. Increased adhesion was mediated by 3 domains: (1) the N-terminal coiled-coil domain that facilitates oligomerization and F-actin localization; (2) bcr sequences between aa 163 to 210; and (3) F-actin localization through the C-terminal actin-binding domain of c-abl. We compared our adhesion results with the ability of these mutants to cause a chronic myelogenous leukemia (CML)-like disease in a murine bone marrow transplantation assay and found that adhesion to fibronectin did not correlate with the ability of these mutants to cause CML. Together, our results suggest that F-actin localization may play a pivotal role in modulating adhesion but that it is dispensable for the development of CML.

Cytogenetic clonal evolution (CE) is a known poor prognostic factor in Philadelphia chromosome-positive chronic myelogenous leukemia (Ph-positive CML). However, its prognostic relevance in the era of imatinib therapy is unknown. We investigated the independent prognostic relevance of CE in 498 patients with Ph-positive CML treated with imatinib for chronic or accelerated phases. One hundred twenty-one patients had CE alone (n = 70) or with other accelerated phase criteria (n = 51). Patients were compared in 4 categories: chronic phase (n = 295), CE only (n = 70), accelerated phase without CE (n = 82), and accelerated phase with CE (n = 51). Statistical methods used established methodologies for univariate and multivariate analyses. In chronic and accelerated phases of CML, CE was not associated with significant differences in major or complete cytogenetic response rates, but it was an independent poor prognostic factor for survival by multivariate analyses in both chronic (P =.005) and accelerated phase (P =.03). Multivariate analyses conducted at the 3-month landmark (including the 3-month cytogenetic response) identified the lack of cytogenetic response at 3 months to be a stronger independent poor prognostic factor for survival than CE for both chronic (major cytogenetic response versus other) and accelerated phase (any cytogenetic response versus other). We conclude that cytogenetic CE is not an important factor for achieving major or complete cytogenetic response with imatinib mesylate therapy, but it is an independent poor prognostic factor for survival in both chronic and accelerated phases of CML. The 3-month cytogenetic response to imatinib mesylate refined the prognostic relevance of such studies in patients on imatinib mesylate therapy.

Natural killer (NK) cells are deficient in patients with chronic myelogenous leukemia (CML), but the mechanisms responsible for the dysfunction are not completely understood. This study reports that CML cells effectively inhibit the baseline and interleukin-2 (IL-2)-induced NK cell cytotoxicity against a CML cell-derived line (K562). A sizable fraction of NK cells subsequently acquired features characteristic of programmed cell death/apoptosis. The CML cell-mediated inhibition of NK cells required triggering of reduced nicotinamide adenine dinucleotide phosphate (NADPH) oxidase-mediated formation of reactive oxygen species (ROS) and was prevented by catalase, a scavenger of ROS, and by histamine, acting via H(2)-receptor-mediated inhibition of ROS production in CML cells. In contrast, nonmalignant neutrophilic granulocytes inhibited NK cells via ROS production without the requirement of exogenous NADPH oxidase-triggering stimuli. We propose that paracrine production of ROS may contribute to the dysfunction of NK cells in CML and that histamine may serve as an autocrine inhibitor of ROS formation in leukemic granulocytes. (Blood. 2000;96:1961-1968)

Patients with Ph+ chronic myelogenous leukemia who relapse after a first allogeneic stem cell transplant still have a possibility of long-term survival. To assess the value of the individual therapeutic options, the factors predicting outcome should be identified. We investigated data from 500 patients who relapsed before July 1996; follow-up was updated during 1998. The actuarial survival from relapse was 34.2% (95% confidence interval [CI]: 29. 9%-38.5%) at 5 years and 23.4% (95% CI: 18.9%-27.9%) at 10 years. Survival after relapse was significantly related to 5 factors: time from diagnosis to transplant (< 2 years vs >/= 2 years), disease phase at transplant (first chronic phase vs other), disease stage at relapse (cytogenetic or chronic phase vs advanced phase), time from transplant to relapse (< 1 year vs >/= 1 year), and donor type (HLA-identical sibling vs volunteer unrelated donor). The effects of individual adverse risk factors were cumulative: The probability of survival at 10 years decreased stepwise from 42% (0 factors), 32% (1 factor), 14% (2 factors), 3% (3 factors), to 0% (4 or 5 factors). Novel strategies for high-risk patients are warranted. We conclude that these 5 factors should be taken into account when comparing results of salvage therapies in patients with Ph+ chronic myeloid leukemia relapsing after allogeneic stem cell transplant.

Chemokines are a family of related proteins that regulate leukocyte infiltration into inflamed tissue and play important roles in disease processes. Among the biologic activities of chemokines is inhibition of proliferation of normal hematopoietic progenitors. However, chemokines that inhibit normal progenitors rarely inhibit proliferation of hematopoietic progenitors from patients with chronic myelogenous leukemia (CML). We and others recently cloned a subfamily of CC chemokines that share similar amino-terminal peptide sequences and a remarkable ability to chemoattract T cells. These chemokines, Exodus-1/LARC/MIP-3alpha, Exodus-2/SLC/6Ckine/TCA4, and Exodus-3/CKbeta11/MIP-3beta, were found to inhibit proliferation of normal human marrow progenitors. The study described here found that these chemokines also inhibited the proliferation of progenitors in every sample of marrow from patients with CML that was tested. This demonstration of consistent inhibition of CML progenitor proliferation makes the 3 Exodus chemokines unique among chemokines. (Blood. 2000;95:1506-1508)

In a single institution, we have used recombinant interferon- (IFN-) to treat 116 newly diagnosed Philadelphia-positive (Ph+) chronic myeloid leukemia (CML) patients and analyzed the predictive factors for response and survival. The patients whose median age was 50.3 years (range, 9 to 70) were administered IFN- (5 million units/m2/d) subcutaneously. The IFN- dose was subsequently adjusted to maintain the white blood cell and platelet counts between 1.5 and 5 x 10(9)/L, 50 and 100 x 10(9)/L, respectively. At diagnosis, the Sokal score was used to classify the patients into three groups: low (n = 57), intermediate (n = 42), and high risk (n = 16). A complete hematological response (CHR) was achieved in 93 cases (80.2%). Of the 116 patients, 113 were available for cytogenetic evaluation. Fifty patients (43%) achieved a major cytogenetic response (MCR) (=65% marrow Ph- cells), 37 of them having a complete cytogenetic response (CCR). The estimated 5-year survival of the 116 patients was 68% +/- 11% (95% confidence interval [CI]) with a median follow-up of 42 months (range, 3 to 114) and 85% +/- 11% (95% CI) with a median follow-up of 30.9 (range, 3 to 111) when patients were censored at the time of transplantation. Event-free survival at 5 years (adding death and transplant as event) was 46% +/- 11% (95% CI). Using proportional hazards regression to study time-dependent variables, we confirmed that the most significant factor associated with survival was the cytogenetic response (MCR or CCR) (P <.0001). This factor was independent compared with the Sokal score and baseline variables used to calculate the Sokal score. Moreover, using either univariate or multivariate analysis, the achievement of CHR within 3 months was strongly correlated with MCR (P <.0001). Minimum cytogenetic response (mCR, ie, at least 5% of Ph- metaphases) at 3 months was also a significant predictive factor for MCR (P <.0001). These results show that IFN- can induce a high rate of hematological and cytogenetic response when administered in doses leading to myelosuppression. Factors such as the achievement of CHR and mCR within 3 months could be useful to identify early those patients who will not respond to IFN- and who need alternative treatments such as stem cell transplantation.

Although all blood cells are derived from hematopoietic stem cells, the regulation of this production system is only partially understood. Negative feedback control mediated by erythropoietin and thrombopoietin regulates erythrocyte and platelet production, respectively, but the regulation of leukocyte levels is less well understood. The local regulatory mechanisms within the hematopoietic stem cells are also not well characterized at this point. Because of their dynamic character, cyclical neutropenia and other periodic hematological disorders offer a rare opportunity to more fully understand the nature of these regulatory processes. We review the salient clinical and laboratory features of cyclical neutropenia (and the less common disorders periodic chronic myelogenous leukemia, periodic auto-immune hemolytic anemia, polycythemia vera, aplastic anemia, and cyclical thrombocytopenia) and the insight into these diseases afforded by mathematical modeling. We argue that the available evidence indicates that the locus of the defect in most of these dynamic diseases is at the stem cell level (auto-immune hemolytic anemia and cyclical thrombocytopenia seem to be the exceptions). Abnormal responses to growth factors or accelerated cell loss through apoptosis may play an important role in the genesis of these disorders., (Copyright 1998 by The American Society of Hematology.)

Treatment options for patients diagnosed with chronic myelogenous leukemia (CML) in chronic phase (CP) who lack a suitable related donor for marrow transplantation include hydroxyurea, interferon-alpha (IFN-alpha), or transplantation from an unrelated donor (URD). Most studies support the view that treatment with IFN-alpha results in prolonged survival compared with hydroxyurea therapy. Some patients are offered URD transplantation as a second-line treatment; however, the impact of pretransplant IFN-alpha on the outcome of URD transplantation is uncertain. To address this question, we evaluated the effect of pretransplant IFN-alpha therapy in 184 patients undergoing URD transplantation for CML in CP at a single center. Of the 184 patients, 114 did not receive IFN-alpha, whereas 22, 23, and 25 patients received IFN-alpha for, respectively, 1 to 5, 6 to 12, and more than 12 months before transplant. Pretransplant IFN-alpha therapy administered for > or = 6 months was associated with an increased risk of severe (grades III-IV) acute graft-versus-host disease (GVHD; relative risk [RR], 3.0; 95% confidence interval [CI], 1.4 to 6.2; P = .004) and mortality (RR, 2. 1; 95% CI, 1.3 to 3.5; P = .003) relative to less than 6 months or no IFN-alpha therapy. Increased mortality occurred between 100 and 365 days after transplant (P = .005), was limited to patients with severe acute GVHD, and was due to chronic GVHD refractory to immunosuppressive therapy. Other variables associated with mortality included HLA-DRB1 or DQB1 (but not HLA-A or B) mismatched donors, age greater than 50 years, weight > or = 110% of ideal body weight, and the absence of cytomegalovirus (CMV) or fungal prophylaxis. For patients treated with IFN-alpha for less than 6 months before transplant, who were < or = 50 years of age, received a HLA-A, B, DRB1, and DQB1 matched URD transplant, and received CMV and fungal prophylaxis after transplant (n = 48), survival was 87% +/- 5% at 5 years. These data provide a rationale for immediate transplantation in preference to extended treatment with IFN-alpha when the patient is < or = 50 years of age and has an HLA-compatible unrelated volunteer donor.

Interferon-alpha (IFN-alpha) is considered the standard therapy for chronic myeloid leukemia (CML) patients not suitable for allogeneic stem cell transplantation. From 1987 through 1992, 195 patients in the Benelux with recent untreated CML were randomized between low-dose IFN-alpha2b (3 MIU, 5 days/wk) or hydroxyurea alone (control group). The white blood cell count had to be kept less than 10 x 10(9)/L in both arms; to this end, the IFN group received additional hydroxyurea, if necessary. The complete hematologic responses at 6 months in the IFN group were 62%, versus 38% in the control group. In the IFN group, a complete hematologic response at 6 months predicted a better survival (P = .001), but such a tendency was also seen in the control group (P = .07). Cytogenetic responses in the IFN group yielded 9% complete responders, 7% partial responders (<35% Ph+), and 24% minor responders (36% to 95% Ph+). The quality of cytogenetic response within the first 24 months was highly predictive for survival (P = .002). Twenty-four patients discontinued IFN-alpha because of side effects, but they did this at a long median interval of 17.6 months; the remaining patients did not require dose adaptations. Although the hematologic and cytogenetic responses in the IFN group were higher than in the control group, the duration of chronic phase from randomization was not statistically different with 53 and 44 months in the IFN and control group, respectively. Also, no advantage for survival calculated from diagnosis was seen for the IFN group (median, 64 months) compared with the control group (median, 68 months).

To elucidate the genetic events that may play important roles in the progression of chronic myelocytic leukemia (CML), we performed allelotype analysis in 30 patients with CML as the disease transformed to accelerated phase or blast crisis (21 myeloid and 9 lymphoid cases). DNAs were extracted from slides of bone marrow smears or from freshly isolated bone marrow mononuclear cells. The DNAs from the same individuals in both chronic phase and either blast crisis or accelerated phase were analyzed at 82 microsatellite markers, which mapped to each of the autosomal arms except the short arms of the acrocentric chromosomes. Loss of heterozygosity (LOH) on at least one locus was observed in 21 of the 30 cases (70%) as the disease progressed. Frequent allelic loss of > or = 20% of the informative cases was observed on chromosome arms 1p (35%), 7p (21%), 19p (20%), and 20q (29%). Allelic losses were also analyzed according to phenotypes. LOH of > or = 20% was detected on 1p (29%), 18p (20%), and 20q (27%) in myeloid blast crisis, and on 1p (50%), 4p (25%), 7p (43%), 9p (29%), 18q (25%), 19p (43%), and 20q (33%) in lymphoid blast crisis. Serial cytogenetic information was available for most of our cases with LOH on these arms, and only one case had loss of both chromosomes 9 and 20. Fractional allelic loss, calculated for each sample as the total number of chromosomal arms lost/total number of arms with information, showed a median value of 0.06 and a mean of 0.098 (range 0 to 0.60). These results suggest that tumor suppressor genes especially on 1p, 7p, 19p, and 20q probably have an important role in the progression to blast crisis of CML.

Although the Philadelphia chromosome (Ph1) has been identified as an adverse prognostic factor in acute lymphoblastic leukemia (ALL), little is known about the incidence and clinical course of relapsed Ph1-positive ALL in children. The incidence was determined by screening of 170 consecutive children with first bone marrow relapse of ALL using the reverse transcriptase-polymerase chain reaction (RT-PCR) and comparison, with cytogenetic analysis. Among these 170 children, 20 (12%) were found to be BCR-ABL-positive, representing a rate that is about three times higher than that reported for newly diagnosed ALL. Ten of the cases were identified by RT-PCR only. In none of the 21 patients with T-cell immunophenotypes could an expression of the BCR-ABL mRNA be detected. BCR-ABL positivity was associated with a significantly shorter duration of first remission (P = .0086) and higher white blood cell (P = .0157) and blast cell counts (P = .0304) at relapse diagnosis. All patients were treated according to the ALL-REZ BFM 87 and 90 relapse trials of the BFM Relapse Study Group. The intensive multiagent chemotherapy induced a second complete remission in only 60% of children with BCR-ABL-positive ALL compared with in 91% of those without BCR-ABL expression (P = .0023). The prognosis of BCR-ABL-positive ALL in children is poor, with a probability of event-free survival at 2 years of 8% versus 50% in those without BCR-ABL mRNA or cytogenetic analysis should become part of the routine diagnostic panel for children with newly diagnosed ALL and is fundamental for children presenting with an early bone marrow relapse.

CD44 is a widely expressed, multifunctional, cell-surface glycoprotein that has been implicated in the regulation of normal hematopoiesis. In addition, expression of particular isoforms of CD44 has been associated with malignant transformation and/or the acquisition of metastatic potential. In this study, we used two recently developed monoclonal anti-CD44 antibodies, one reactive with an epitope shared by many CD44 isoforms and the other with an epitope unique to CD44 isoforms containing amino acids encoded by the alternatively spliced exon v10, to compare the expression of CD44 on primitive hematopoietic cells from the marrow of normal individuals and their neoplastic counterparts present in the peripheral blood of patients with chronic myeloid leukemia (CML). Multiparameter fluorescence-activated cell sorter (FACS) analysis and cell sorting studies showed that CD44 is normally expressed at high to very high levels on both long-term culture-initiating cells (LTC-IC) and granulopoietic colony-forming cells (granulocyte-macrophage colony-forming units [CFU-GM]). In contrast, primitive erythropoietic progenitors (burst-forming units-erythroid [BFU-E]) in normal marrow were more homogeneous in their expression of CD44, and very few (less than 5%) showed the very high levels of CD44 seen on 20% to 25% of LTC-IC and CFU-GM. Antibody staining showed the expression of exon v10-containing CD44 isoforms to be restricted to a small subpopulation (4% to 8%) of morphologically recognizable mature (CD34-) myeloid cells within the light-density fraction of normal marrow cells. Reverse transcription-polymerase chain reaction (RT-PCR) analysis showed the presence of two exon v10-containing mRNA species. In CML, a significantly greater proportion of the circulating neoplastic CFU-GM expressed very high levels of CD44, and these CFU-GM were accompanied by an increased number of light density v10+ cells, including some that coexpressed CD34. Nonmalignant hematopoietic progenitors mobilized by prior chemotherapy and growth factor treatment of patients with Hodgkin's disease or acute myeloid leukemia in remission showed no changes in CD44 expression relative to normal marrow progenitors. These results provide evidence of early differentiation-associated changes in CD44 expression during normal hematopoiesis in vivo that may be deregulated in the neoplastic clone of patients with CML.

A patient with typical Philadelphia chromosome (Ph1)-positive chronic myelocytic leukemia (CML) was studied during sequential phases of disease: (1) initial chronic phase; (2) myeloid blast crisis; (3) second chronic phase; and (4) accelerated disease. A point mutation in the coding sequence of the p53 gene first appeared concomitantly with the blast crisis and then disappeared with the re-establishment of a second chronic phase. The chromosomal concomitant of the molecular alteration was a deletion of 17p. These observations suggest that abnormalities of the p53 anti-oncogene are temporally related to the clinical progression of some cases of CML and are probably responsible for the development of blast crisis in these cases.

Juvenile chronic myelogenous leukemia (JCML) is a good model for the study of myeloproliferation because JCML hematopoietic progenitor cells grow in vitro at very low cell densities without the addition of exogenous stimulus. Previous studies have demonstrated that this proliferation is dependent on granulocyte-macrophage colony-stimulating factor (GM-CSF), and that removal of monocytes from the cell population before culture eliminates this "spontaneous" myeloproliferation, suggesting a paracrine role of monocyte stimulation. However, subsequent studies have shown that increased GM-CSF production from the JCML monocytes is not a consistent finding and therefore not a plausible sole mechanism. In examining hematopoietic growth factor dose-response curves, both JCML GM and erythroid nonadherent progenitor cell populations displayed a marked and selective hypersensitivity to GM-CSF. Responses to interleukin-3 and G-CSF were identical to control dose-response curves. This is the first demonstration of a myeloid leukemia in which hypersensitivity to a specific growth factor appears to be involved in the pathogenesis of the disease.

We studied the in vitro effects of lymphokine-activated killer (LAK) cells from the peripheral blood of chronic myeloid leukemia (CML) patients after allogeneic and syngeneic bone marrow transplantation (BMT). LAK cells were generated by incubating peripheral blood mononuclear cells from patients post-BMT with recombinant interleukin-2 (IL-2) (500 U/mL) in 10% AB serum for 7 days. They were phenotyped and tested for activity in a standard 4-hour 51Cr release assay (n = 37) and in a CFU-GM assay (n = 24). We found that the LAK cells were mainly activated natural killer cells, but some were CD3+ T cells. In the 51Cr release assay LAK cells from 20 of 33 (61%) allogeneic and 2 of 4 syngeneic recipients killed recipient CML cells and in 22 of 37 (60%) cases also killed the HLA disparate CML cells. In the CFU-GM assay the LAK cells incubated together with the CML cells in liquid culture before plating inhibited (P less than .05) colony growth in 16 of 22 allogeneic and 2 of 2 syngeneic recipients. Cell-cell contact was necessary for optimal effect. There was little or no inhibition of proliferation of donor marrow CFU-GM. This in vitro graft-versus-leukemia (GVL) effect could also be demonstrated after LAK effectors were depleted of CD3+ T cells. It was inducible in recipients of both T cell-depleted and T cell-replete donor marrow and in recipients with or without graft-versus-host disease. These results suggest that a major histocompatibility complex-unrestricted GVL effect is inducible following allogeneic and syngeneic BMT. The use of IL-2/LAK cells after BMT could reduce the risk of relapse.

A new human myeloid cell line has been established recently from the bone marrow cells of a patient with chronic myelogenous leukemia in blast crisis. The active proliferation and survival of the cells in RPMI 1640 medium containing fetal calf serum are clearly dependent on the presence of either natural or recombinant human granulocyte-macrophage colony-stimulating factor (rhGM-CSF). Despite permanent culturing in rhGM-CSF (100 U/mL), the cells do not differentiate and bear the myelomonocytic surface markers CD34, CD13, CD36, as well as HLA-DR, but not CD3, CD7, CD10, CD11b, CD14, CD20, or CD42b. The predominant karyotype, apart from tetraploidy in several cells, is 45, XX, -9, -17, -19, -22, 7p-, 9q+ (der t[9;22]), der (13q), with three additional marker chromosomes, from which one was observed in the patient's leukemic cells. On BglII-digested DNA, Southern blot analysis with bcr 5' as the probe detected two additional hybridizing restriction fragments of 8.6 and 11.0 kilobase pairs.

Juvenile chronic myelogenous leukemia (JCML) is a rare myeloproliferative disorder of early childhood that is clinically and cytogenically distinct from the well-recognized adult type of chronic myeloid leukemia. Unlike the adult disease, growth of hematopoietic progenitors from peripheral blood (PB) occurs in the absence of exogenous stimulus even at low cell densities. This so-called "spontaneous" growth can be abrogated by adherent cell depletion and appears to depend on production of endogenous growth factors. We studied seven children with JCML to determine the nature of endogenous stimulators. With isolated PB mononuclear cells (PBMNCs) and a 3H-thymidine (3H-TdR) incorporation assay, JCML cells were shown to incorporate high levels of 3H-TdR when cultured in the absence of stimulus even at low cell densities. When neutralizing antisera prepared against each of the four known colony-stimulating factors (CSFs), GM-CSF, G-CSF, M-CSF, and interleukin-3 (IL-3), as well as antisera against interleukin-1 (alpha and beta) and tumor necrosis factor (TNF) were added to these cultures, only the antisera against recombinant human GM-CSF (rhGM-CSF) consistently resulted in significant inhibition of cell proliferation, achieving up to 72% inhibition of 3H-TdR incorporation in one case. Monoclonal antibodies (MoAbs) against rhGM-CSF resulted in a similar and highly significant degree of inhibition. A marked inhibitory effect of rhGM-CSF antiserum on "spontaneous" growth of PB CFU-GM derived colonies in semisolid medium was also demonstrated in four of five patients studied (87% to 90% inhibition). Production of growth factors by highly enriched JCML monocytes was variable. When initially studied in five of the seven patients, the monocytes from three of the patients revealed increased release of IL-1-like activities; two patients had levels similar to those of controls. One patient with normal levels when initially studied was later shown to have markedly increased amounts of IL-1-like activities in a second preparation of monocyte-conditioned medium (MCM). High levels of GM-CSF were detected in the initial MCM from one patient, but this may have indirectly reflected elevated IL-1-like activities present in the MCM. IL-3 and M-CSF levels were either low or undetectable in the patients studied as compared with MCM prepared with normal adult monocytes. These results clearly implicate GM-CSF as the primary endogenous regulator of JCML cell proliferation in culture and suggest that this malignant myeloproliferative disease may in part result from paracrine stimulation of marrow progenitor cells by growth factors/cytokines secreted by the malignant monocytes.