Potential mechanisms of action of interferon-alpha in CML.

Treatment with interferon-alpha (IFN-alpha) adequately controls the leukemic cell mass in the majority of newly diagnosed patients with chronic myeloid leukemia (CML). However, the degree of response ranges from no 'hematologic' response to complete suppression of the leukemic clone. The mechanism(s) by which IFN-alpha elicits these responses is unknown, but in vitro studies have indicated that IFN-alpha might function by (1) selective toxicity against the leukemic clone, (2) enhancement of 'immune' regulation, and (3) modulation of bone marrow microenvironmental regulation of hematopoiesis. Using in vitro clonogenic assays we were unable to demonstrate that IFN-alpha selectively inhibited the proliferation of CML progenitor cells. We also found no difference in the expression of LFA-3 on normal or CML CD34+ cells. However, by panning and co-culturing hematopoietic cells on monolayers of bone marrow stromal cells, grown with and without IFN-alpha, we found that IFN-alpha enhanced the adhesion of CML progenitors to stromal cells, whereas adhesion by normal progenitor cells was essentially unaffected. This enhanced adhesion by CML progenitor cells was associated with a reduction in neuraminic acid levels in the extracellular matrix overlying stromal cells. Therefore, it is possible that one of the mechanisms by which IFN-alpha exerts its regulatory effect on the leukemic clone is through enhancement of hematopoietic cell-microenvironmental cell interactions.

Interferon-alpha overrides the deficient adhesion of chronic myeloid leukemia primitive progenitor cells to bone marrow stromal cells.

Primitive blast colony-forming cells (BI-CFC) from chronic myeloid leukemia (CML) patients are defective in their attachment to bone marrow-derived stromal cells compared with normal BI-CFC. We investigated the effect of recombinant interferon-alpha 2a (IFN-alpha) on this interaction between hematopoietic progenitor cells and bone marrow-derived stromal cells by culturing normal stromal cells with IFN-alpha (50 to 5,000 U/mL). At 50 U/mL we found that: (1) the capacity of stromal cells to bind two types of CML primitive progenitor cells (BI-CFC and long-term culture-initiating cells) was increased; and (2) the amount of sulfated glycosaminoglycans (GAGs) in the stromal layer was increased. However, sulfated GAGs were not directly involved in binding CML BI-CFC, unlike binding by normal BI-CFC, which is sulfated GAG-dependent. Neuraminidase-treated control stromal cells bound an increased number of CML BI-CFC, reproducing the effect of IFN-alpha, whereas the binding to IFN-alpha-treated stromal cells was unaffected by neuraminidase treatment. Thus, the enhanced attachment by primitive CML progenitor cells to INF-alpha-treated stromal cells might be due to changes in the neuraminic acid composition in the stromal cell layer. Our in vitro evidence may provide insights into the mechanism of action of IFN-alpha in vivo. Prolonged administration may alter the marrow microenvironment in some patients such that it can restrain the aberrant proliferation of Philadelphia chromosome (Ph)-positive stem cells while permitting Ph-negative stem cells to function normally.

Interferon-alpha alters the distribution of CFU-GM between the adherent and nonadherent compartments in long-term cultures of chronic myeloid leukemia marrow.

We studied the effect of recombinant interferon-alpha 2a (IFN alpha) on the interaction between stromal cells and granulocyte-macrophage colony-forming units (CFU-GM) from the marrow of normal individuals and chronic myeloid leukemia (CML) patients in chronic phase in long-term bone marrow cultures using preformed stromal layers. These stromal layers were established with marrow cells from normal allogeneic donors and grown to confluence in the presence or absence of IFN alpha at low concentration (100 U/ml). The number of CML CFU-GM localized within IFN alpha-treated stromal layers was significantly greater than the corresponding number localized within control stromal layers. Conversely, the distribution of normal CFU-GM between the adherent and nonadherent compartments was unaffected by IFN alpha treatment of the stromal layer. Preincubation of CML marrow cells with IFN alpha did not alter this distribution, so the observed effect of IFN alpha must be due to a primary action on stromal layers. Thus, in addition to its well known antiproliferative effect, IFN alpha enhances the capacity of marrow stromal cells to bind and/or retain CML progenitor cells, and the resulting restoration of normal regulatory control may be the basis for the selectivity of IFN alpha in CML.

The effects of interferon-alpha on the proliferation of CML progenitor cells in vitro are not related to the precise position of the M-BCR breakpoint.

We investigated the effects of brief (2 h) and continuous exposure to recombinant interferon-alpha (2a) (rIFN-alpha) on the proliferation of primitive (blast colony-forming cells, Bl-CFC) and committed myeloid progenitor cells (BFU-E and GM-CFC) derived from blood and bone marrow of patients with chronic myeloid leukaemia (CML) and normal subjects. In all three clonogenic assays, rIFN-alpha suppressed colony formation in a dose-dependent manner. No differences were detected in the proliferation of CML or normal Bl-CFC and GM-CFC exposed to rIFN-alpha. Erythroid colony formation by normal, but not by CML BFU-E, was inhibited by relatively low concentrations (100 U/ml) of rIFN-alpha. However, in patients whose blood or marrow contained a mixture of Philadelphia chromosome (Ph)-positive and Ph-negative BFU-E, cytogenetic analysis of individual erythroid colonies showed no differential inhibition by rIFN-alpha. We found no difference in the sensitivity to rIFN-alpha of GM-CFC from patients whose leukaemic cells expressed BCR/ABL mRNA with the b2a2 junction and that of GM-CFC from patients with the b3a2 mRNA. We conclude that (1) rIFN-alpha does not have a significant leukaemia-specific effect on the progenitor cells detected in these assays, and (2) the sensitivity of CML GM-CFC to rIFN-alpha is independent of the type of BCR/ABL message present in the cells. The clinical efficacy of rIFN-alpha could be due to selective toxicity to cells not assayed in this study, to effects on accessory cells or to alterations induced in progenitor cell/stromal cell interactions.

Slow evolution of chronic myeloid leukaemia relapsing after BMT with T-cell depleted donor marrow.

Thirty-three patients with Philadelphia positive (Ph+) chronic myeloid leukaemia (CML) treated in chronic phase by bone marrow transplantation (BMT) with T-cell depleted HLA-identical sibling marrow were evaluable for relapse at a median follow up of 41 months (range 16-59 months). Twenty-six (78%) had Ph+ marrow metaphases demonstrated at some time post BMT. The subsequent pattern of disease was variable. In 15 of these cases haematological relapse occurred within 24 months of BMT. Four patients proceeded to haematological relapse more slowly. Seven patients had only cytogenetic evidence of relapse. Of the 19 patients with haematological relapse, five received second transplants and two survive; 13 of the other 14 survive in chronic phase at median times from allografting and from recognition of haematological relapse of 41 months (range 25-59 months) and 18 months (range 5-36 months) respectively. For these 13 patients disease progression after relapse seems to be relatively indolent. In the four patients we could study, blood lymphocytes were almost all of donor origin. We suggest that even in patients with cytogenetic or haematological evidence of relapse after T-cell depleted BMT, leukaemic cell proliferation may still be restrained to some extent by a graft-versus-leukaemia effect mediated by donor-derived lymphoid cells.

Autografting for patients with chronic myeloid leukaemia in chronic phase: peripheral blood stem cells may have a finite capacity for maintaining haemopoiesis.

We treated 14 patients with Ph-chromosome-positive chronic myeloid leukaemia still in chronic phase by autografting with blood-derived haemopoietic stem cells. Eleven patients were autografted electively after cytoreductive treatment with busulphan (16 mg/kg) and melphalan (60 mg/m2) and three were autografted after marrow cells from HLA-identical sibling donors had failed to engraft. In 13 patients haemopoiesis recovered; one failed to engraft and died 114 d after autografting. Two other patients became pancytopenic and received further stem cell transfusions at 3 and 40 months respectively after first autografting. One patient entered lymphoid transformation and died 14 months after autografting. Twelve patients survive at a median of 41 months (range 24-53) after autografting. Nine of the survivors have required further chemotherapy after autografting and four of the nine were electively autografted on a second occasion. Three patients surviving after autografting for 28, 43 and 53 months respectively have not required further chemotherapy. In two of these patients haemopoiesis is now predominantly Ph-negative. We conclude that autografting in chronic phase might prolong survival in some cases by reducing the size of the leukaemic stem cell population. The fact that initially successful grafts failed in two patients suggests that blood-derived stem cells may have a finite potential for self-replication.

Polymerase chain reaction for detection of residual leukaemia.

The occasional finding of cells positive for the Philadelphia (Ph) chromosome months or years after bone-marrow transplantation for chronic myeloid leukaemia raises the possibility that the Ph-positive clone may never be eradicated. The polymerase chain reaction with probes able to detect the transcript of the bcr/abl hybrid gene at very low levels was used to study marrow cells from seven patients in continuing haematological and cytogenetic remission 5-7 years after allogeneic bone-marrow transplantation for chronic myeloid leukaemia. No evidence of the leukaemic mRNA was found. Thus, it seems that all leukaemic cells were eradicated in these patients and that they are truly cured.

The genomic breakpoint in a patient with Philadelphia-positive acute leukemia is 5' of the breakpoint cluster region.

We report a case of acute leukemia in which studies at presentation showed both myeloid and lymphoid cell surface markers. At relapse membrane markers studies were consistent with a leukemia of B-lymphoid lineage. However, immunoglobulin (Ig) and T cell receptor (TCR) beta chain genes were both found in a rearranged configuration. The majority of metaphases from the leukemic cells at presentation showed the Philadelphia chromosome, t(9;22)(q34;q11), whereas a minority were normal. At relapse both Ph-positive and -negative metaphases were still present in the bone marrow but some of the Ph-negative metaphases had acquired an additional chromosome #19 [47,XY, + 19]. Southern analysis of DNA from leukemic bone marrow cells at diagnosis showed no rearrangement of breakpoint cluster region (bcr). There was no bcr-abl chimeric mRNA typical of Ph-positive chronic myeloid leukemia (CML). However, the cells expressed an abl-related protein of Mr 190 kd with enhanced tyrosine kinase activity. Leukemic cell metaphases were studied by the technique of in situ hybridization with probes for C-lambda, sis, abl, and 5' bcr. The c-abl probe mapped to chromosome 22q11 in Ph-positive metaphases. The 5' bcr probe mapped to 9q+ in the Ph-positive metaphases and the C-lambda gene mapped to the Ph chromosome. Thus, the genomic breakpoint in this patient must lie upstream of the BCR defined by study of Ph-positive CML and downstream of the C-lambda gene locus. We speculate that the Ph-negative cells in this patient may represent a leukemic proliferation susceptible to acquisition of specific chromosomal changes.

Cytogenetic events after bone marrow transplantation for chronic myeloid leukemia in chronic phase.

Forty-eight patients treated by allogeneic bone marrow transplantation (BMT) for Philadelphia (Ph) chromosome-positive chronic myeloid leukemia in chronic phase had serial cytogenetic studies of marrow performed at intervals after transplant. Twenty patients received marrow cells from donors of opposite sex. Ph+ marrow metaphases were identified in 24 of 48 (50%) of patients after BMT; they were first seen early (within 1 year) in 16 cases and late (greater than 1 year after BMT) in eight cases. Ph-positivity after BMT occurred more commonly in recipients of T-depleted than nondepleted marrow (19 of 28 v 5 of 20). In 4 cases the Ph+ metaphases were found only transiently after BMT; in 11 cases the Ph+ metaphases have persisted but hematologic relapse has not ensued; in 9 cases the finding of Ph+ metaphases coincided with or preceded hematologic relapse. Chromosomes in cells of donor origin had morphological abnormalities in two cases. No relapses were identified in cells of donor origin. Our data suggest that the relationship between cells of recipient and donor origin is complex: cure of leukemia may depend on factors that operate for some months or years after BMT.

Complete remission after autografting for chronic myeloid leukaemia.

A previously untreated 31-yr old female with Ph-positive chronic myeloid leukaemia (CML) received busulphan and melphalan at high dosage followed by an autograft of peripheral blood stem cells collected 4 weeks earlier. Though recovering haemopoiesis was at first mainly Ph-positive, Ph-negative haemopoiesis predominated at 12 months and has persisted until the most recent study (24 months post-autograft). Her haemopoietic cells now show no rearrangement of the bcr gene, unlike the cells collected at diagnosis. Autografting carried out soon after diagnosis could be valuable for obtaining cytogenetic remissions in other patients with CML.

Rearrangement of the bcr gene in Philadelphia chromosome-negative chronic myeloid leukemia.

We studied the clinical, hematologic, cytogenetic, and molecular biologic features of seven patients with Philadelphia (Ph1) chromosome-negative chronic myeloid leukemia (CML). In five cases the hematologic findings were indistinguishable from those of patients with classical Ph1-positive disease. Myeloid cells were studied by chromosome-banding techniques. One patient had a masked Ph1 chromosome (with translocation t(4;9;22)), one had a deletion involving chromosome 16, and one had a small minority population of 22q- cells without 9q+ but otherwise normal metaphases; metaphases from the other four patients were entirely normal. DNA prepared from the myeloid cells was digested with the restriction enzymes EcoRI, HindIII, BamHI and BglII. Southern analysis using a 0.6-kb fragment of the breakpoint cluster region (bcr) gene showed the presence in each patient's DNA of a germline fragment together with a rearranged fragment or fragments with at least one of the restriction enzymes. We conclude that genomic changes in the bcr gene characteristic of CML can be present in the absence of a Ph1 chromosome.