The platelet derived growth factor receptor beta (PDGFRB) is a receptor tyrosine kinase involved in mitogenesis, cytoskeletal rearrangements and chemotaxis, particularly of mesenchymal cells. Ligand binding to PDGFRB results in dimerization, kinase activation and autophosphorylation, which trigger a cascade of intracellular signaling events via STATS 1 and 5. Rearrangement of the gene encoding PDGFRB, located at 5q31, has been described in chronic myeloproliferative diseases (MPD), myelodysplasia (MDS) and AML. The best known of these is the t(5;12) MPD in which the partner gene on chromosome 12 is ETV6 (TEL) resulting in the TEL-PDGFRB fusion protein. The clinical phenotype of MPD associated with t(5;12) chromosomal abnormalities is heterogeneous but there are some common features, such as splenomegaly, peripheral blood and bone marrow eosinophilia and monocytosis. Transformation to blastic crisis is well recognised. Interferon-a has been beneficial in some patients, and has induced complete cytogenetic remissions. STI571 (imatinib mesylate, Gleevec) is a tyrosine kinase inhibitor with specificity for the ATP binding site of ABL, c-KIT and PDGFRB. We treated two patients with MPD characterized by the presence of t(5;12) and rearrangement of PDGFRB with STI571. Patient 1 is a 50 year old male in whom leukocytosis was an incidental finding during a period of depression. He was asymptomatic, but had mild hepatosplenomegaly, a WCC of 50 x 10⁹/l and peripheral blood eosinophilia and monocytosis. Further characterisation of the t(5;12) confirmed the presence of the classical TEL-PDGFRB fusion gene. Patient 2 presented at the age of 6 years with an infiltrative skin rash and peripheral blood eosinophilia. He was later found to have MPD/MDS with a t(5;12) chromosomal rearrangement. His disease failed to respond to hydroxyurea and interferon-a. His skin condition progressed until by the age of 19 years almost the entire skin surface was involved with erythematous raised infiltrative lesions and extensive ulceration. Further study of the t(5;12) revealed PDGFRB rearrangement with an as yet unidentified partner gene. Peripheral blood and bone marrow derived mononuclear cells displayed in vitro sensitivity to STI571 in liquid culture. Both patients were entered into the Novartis study STIB2225 for patients with life-threatening diseases involving STI571 sensitive tyrosine kinases. He started treatment with STI571 400 mg daily. In both patients the leukocyte counts normalized with loss of eosinophilia within 7 days of starting treatment. At 4 weeks bone marrow metaphases from patients 1 and 2 were 50% and 100% 46,XY respectively. Patient 1 became 100% 46,XY by 12 weeks of treatment. The skin lesions of Patient 2 responded rapidly with decreased erythroderma, flattening and healing of ulceration. To speed recovery he underwent cadaveric skin grafting. Both patients remain well and free of side-effects 6 months after starting therapy. STI571 appears to be an effective therapy for the management of hematological malignancies involving the PDGFRB.

Our laboratory previously described the first documented cases of a specific BCR/ABL kinase domain mutation resulting in resistance to the tyrosine kinase inhibitor Gleevec (STI-571, imatinib mesylate). The T315I mutation was shown to preserve kinase activity yet, based upon the crystal structure of the kinase domain when bound to a derivative of Gleevec, is predicted to result in ineffective binding of the drug to BCR/ABL. In an effort to define the full spectrum of kinase domain mutations in a larger sample size, we sequenced the BCR/ABL kinase domain of 19 additional patients with CML in myeloid blast crisis and one patient with recurrent Ph+ ALL. In 15 of these patients, samples were obtained at the time of relapse after initial normalization of bone marrow blast percentage in response to Gleevec (acquired resistance). The remaining five patients did not achieve normalization of bone marrow blast percentage in response to Gleevec (de novo resistance). In these cases, analysis was performed on samples obtained prior to treatment. To ensure detection of kinase domain mutations that might account for a minority of BCR/ABL expressing cells in the blood, we sequenced ten independent clones from each patient sample. A mutation was considered present only if it was detected by sequence in both cDNA strands. The previously identified T315I mutation was found in two additional patients with acquired resistance, and in one patient with de novo resistance. In conjunction with our previous report of 11 patients (Gorre et al, Science 293:876, 2001), the T315I mutation has now been detected in 9 of 31 patients tested(5/27 myeloid blast crisis, 4/4 with lymphoid blast crisis or Ph+ ALL). Four other mutations, Q252H, E255K, M351T, and G250E were found in six, five, four, and two patients respectively. G250E comprised the dominant clone in two patients with acquired resistance. Examples of several additional mutations, each of which comprised a minority of clones, were observed in both acquired and de novo resistance. One such mutation, Y253F, has been previously shown to activate the kinase activity and transformation potential of c-ABL in vitro (Allen and Weidemann, J. Biol. Chem. 271:19585, 1996). Our findings indicate that BCR/ABL kinase domain mutations are common in both acquired- and de novo- resistant CML blast crisis cases. These mutations may be a reflection of genetic instability associated with disease progression or, possibly, prior treatment exposure. With the exception of the T315I mutation, which we have previously shown to cause in vitro resistance to Gleevec, the significance of these additional mutations in Gleevec drug resistance remains to be defined. Structure/ function analysis of these mutations is ongoing. Our data further support the hypothesis that BCR/ABL activity remains essential even in the blast crisis phase of the disease, and suggest that numerous mutations may be capable of preserving this activity in the presence of Gleevec.

Eleven patients (pts) on a multi-institutional clinical trial utilizing STI571 for the treatment of either interferon refractory/intolerant or accelerated phase CML were identified with major or complete cytogenetic responses. These pts were referred for growth factor mobilization of CD34⁺ autologous PBSC products while maintaining concomitant therapy with imatinib (STI571)for potential future autologous transplantation. 5/11 pts previously had been treated with IFN-alpha for 2 years. 6/11 pts received G-CSF only at 10 mcg/kg qd while 5/11 required altered cytokine schedules with either dose escalation of G-CSF or combination with GM-CSF. Two pts required remobilization efforts. All pts were valuable for mobilization kinetics. The minimal target dose of CD34⁺ progenitors/kg = 2x10⁶/kg. Peripheral blood CD34 counts were monitored to assess onset of stem cell procurement. Results: Median days of growth factor administration = 7 (range 6-16). Median days of apheresis = 3 to achieve target dose of CD34⁺ progenitors (range 3-5). Median CD34⁺ progenitors/kg collected = 2.68 x 10⁶ (range 0.78-7.95). PBSC products were negative for the Ph chromosome in 5/5 evaluated pts. FISH analysis on the PBSC product for the bcr/abl fusion product was completely absent in 3/10 pts, < 5% in 3/10 pts and < 10% in 4/10 pts assessed to date, with a range of 200-1000 interphase cells analyzed. Qualitative PCR on the PBSC product for the bcr/abl gene product was (-) in 1/9 pts, indeterminate in 1/9 pts and (+) in 7/9 pts evaluated. IFN-alpha treatment appeared to impact on mobilization. For 6 patients treated < 2 yrs, mean CD34⁺ progenitors/kg = 4.6 x 10⁶ while for the 5 pts treated 2 yrs, mean CD34⁺ progenitors /kg collected = 2.022 x 10⁶ . Conclusions: CD34⁺ PBSC can successfully be mobilized with growth factors on patients treated concurrently with STI571. Limited prior exposure to IFN-alpha may enhance the mobilization kinetics of these PBSC populations. Successful cryopreservation of Ph ^- (> 90%) enriched PBSC products is feasible and may be more easily obtained than with other chemotherapy mobilization strategies.

We conclude, that (i) quantitative determination of residual disease with real time RT-PCR is a reliable and sensitive method to monitor CML pts. on STI571 therapy; (ii) both ABL and G6PD are appropriate genes for standardization of the assay; (iii) ratios BCR-ABL/ABL correlate well with cytogenetic response; (iv) all complete responders to STI571 have evidence of residual disease with the limited follow-up available; and (v) cytogenetic response at six months of therapy in CP pts. is predictable with real time RT-PCR at two months.

Selective inhibition of the BCR-ABL tyrosine kinase by STI571 (imatinib, Glivec) is a promising new therapeutic strategy in patients (pts.) with chronic myelogenous leukemia (CML). Despite significant hematologic and cytogenetic responses, resistance has been observed in a proportion of pts. after variable periods of STI571 monotherapy. We investigated mechanisms of resistance in 42 pts. (21 m, 21 f; median age 60, range 26-70 years) refractory or resistant to STI571 (400-800 mg p.o./day) out of a total of 291 pts. treated with STI571 for CML in chronic phase (CP, n=137), accelerated phase (AP, n=80), myeloid blast crisis (BC, n=73), and lymphoid BC (n=1). Pts. were recruited into six multicenter phase II studies. STI571 resistance developed in 28 myeloid BC, one lymphoid BC, nine AP, and four CP pts.. The median duration of therapy was 123 days (range 13-741). Prior to STI571 treatment and at the time of resistance we determined the expression level of BCR-ABL transcripts in peripheral blood leukocytes by quantitative RT-PCR, the number of genomic BCR-ABL copies by interphase fluorescence in situ hybridization (IP-FISH), and evaluated for karyotypic evolution by metaphase cytogenetics. Lastly, the BCR-ABL tyrosine kinase domain was sequenced from cDNAs derived from resistant blasts. Results: (i) Although the median level of BCR-ABL transcripts expressed as the ratio BCR-ABL/G6PD was not significantly changed at the time of resistance (5.3%, range 0.07-100) as compared to pretherapy (4.6%, range 0.1-43), 3/21 pts. showed a >10fold increase in BCR-ABL levels; (ii) genomic amplification of BCR-ABL was found in 2/21 evaluated pts. investigated by IP-FISH; (iii) additional chromosomal aberrations resulting in clonal evolution were observed in 9/24 pts., of whom five developed aneuploidy. (iv) Point mutations of the ABL tyrosine kinase domain were detected in 6/40 pts.. Mutations led to amino acid substitutions (Y253F, n=1; E255K, n=2; E255V, n=1; T315I, n=2) which change the conformation of the ATP binding site. Mutations have been confirmed by DNA restriction digest analysis and are not found in pretherapy samples. Reactivation of BCR-ABL kinase activity was confirmed by CRKL immunoblotting in five pts. with point mutations demonstrating insensitivity to STI571 with a median proportion of phosphorylated CRKL of 63% (range 38-77%). ABL kinase assays demonstrated an increase of the IC50 for STI571 from 0.025 mM for wildtype ABL to >0.5 mM for E255K and E255V, and 0.30 mM for T315V. In total, molecular or cytogenetic causes of resistance have been identified in 17/42 pts.. We conclude that various mechanisms of resistance to STI571 exist which lead to clonal selection of resistant cells and ultimately result in hematologic resistance to STI571 monotherapy. These observations offer additional possibilities for clinical resistance beyond previous reports and warrant trials combining STI571 with other agents.

Imatinib mesylate has induced cytogenetic and molecular remission in patients (pts) with CML in a Phase II trial. Moreover, some pts had disease progression due to high tumor load. Autologous BMT has induced some cytogenetic responses by reducing the amount of Ph+ cells posttransplant. We report here 2 CML male pts who underwent a PBSCT due to blast phase and received as part of an expanded access protocol from Novartis Biociências AS, STI-571 after transplant. Both pts received ICE regimen for stem cell mobilization and had stem cell harvested during Chronic Phase. Busulfan (16mg/kg) plus Melfalan (140mg/m²) was given as a conditioning regimen and growth factor support with G-CSF was provided for both.Pt#1 developed disseminated Fusarium sp infection during recovery of neutropenia after ICE treatment. The Autologous PBSCT was delayed for 2 years due to a concern of Fusarium sp. reactivation posttransplant. Pt#2 developed disease progression and PBSCT was performed in blast phase. CD34+ cells infused were 1.67x10⁶ /kg and 2.5x10⁶, respectively. Engraftment data were as follow: ANC>500/mm3 +15 &+17 platelets >20K +42 and +50. Pt#1 had a decrease to 17% of Ph+cells early after transplant. Pt#2 maintained 100% Ph+ cells. Both received STI-571 at a beggining dose of 600mg daily. Pt#2 developed myelotoxicity grade III and interstitial pneumonitis and the drug was interrupted for 15 days. Pt#1 revealed disappearence of Ph+on day 90 post STI571. Pt#2 remains Ph100%+ to date. Although these are individual observations and the exact value of STI post transplant needs to be carefully and properly assessed in adequately designed clinical trials, autologous PBSCT followed by STI 571 may be a strategy to reduce the high Ph+ cells in CML patients.

The onset of accelerated phase or blast crisis of chronic myeloid leukemia (CML) is usually associated with the acquisition of new chromosome abnormalities in addition to the t(9;22)(q34;q11) which is characteristic of chronic phase (CP) CML. We describe the case of a 67 year old man who was diagnosed with CML in CP in February 1996 and transformed 5 years later with multiple secondary cytogenetic abnormalities including BCR/ABL amplification after 6 months treatment with the ABL-specific tyrosine kinase inhibitor STI571 (imatinib, Glivec). He was treated initially with hydroxyurea and interferon and subsequently with interferon and cytosine arabinoside over the 4 years following diagnosis, achieving an excellent hematological but no cytogenetic response. In January 2001, 5 years after diagnosis, he was commenced on STI571 400mg/day as part of the interferon-refractory chronic phase CML study. Bone marrow examination prior to study entry showed <1% blast cells and 40 cells with 46,XY,t(9;22)(q34;q11). At three months, the bone marrow was unremarkable with mild hypocellularity and no cytogenetic response. At six months, the patient presented in acute renal failure and was found to be in myeloid blast crisis with 85% bone marrow blasts. Cytogenetic studies showed the presence of 5 cell lines derived from the t(9;22) stem line with no obvious Philadelphia chromosome present but with a large marker chromosome containing a homogeneously staining region which FISH studies showed to contain multiple copies of the BCR/ABL fusion gene. Most interphase cells contained 7 to 8 copies of the fusion gene and all of these appeared in metaphases to be contained within the marker chromosome. The patient died shortly after transformation. The common secondary abnormalities appearing in CML in blast crisis are a derivative 22 or Philadelphia chromosome, an isochromosome of 17q and trisomies of 8 and 19. Amplification of BCR/ABL is a rare finding with only two cases reported in patients not on STI571 who presented with extramedullary blast crisis (Metze-Heidemann et al, Genes Chromosomes and Cancer, May 2001) and in three patients with advanced stage CML who underwent disease progression whilst on STI571 (Gorre et al, Science, 3 August 2001). Our patient represents another example of BCR/ABL amplification manifesting whilst on STI571 and illustrates that amplification provides a potent mechanism for the development of resistance to STI571 therapy in CML.

We are hereby reporting a case of CML complicated with myeloid blast transformation, managed successfully with Gleevec (STI-571, imatinib) therapeutic approach. Remarkably, the described patient has achieved so far a 2 year survival since the onset of his blast crisis. This is a patient who was diagnosed with chronic phase CML in 6/97 at the age of 47, when he presented with fatigue, moderate splenomegaly and a WBC of 105,000/mm3. Cytogenetics showed complex translocation involving chromosomes 6, 9 and 22, positive for the Philadelphia chromosome. He had no allogeneic donors. He was treated with interferon-a and ara-c, requiring periodically anagrelide for control of thrombocytosis. He had no cytogenetic response. In September of 1999 (51th month since diagnosis), he evolved into blast crisis, with anemia, thrombocytopenia, with a bone marrow biopsy showing 35% blasts, and additional complex cytogenetic abnormalities, involving chromosomes 3, 4 and 17. He enrolled in a Gleevec (600 mg) trial for blast crisis (Novartis). He developed cytopenia lasting several weeks followed by complete hematologic recovery. On the third month of the treatment his BM contained only 6% blasts, whereas on the 6th month he had normal appearing BM biopsy and normal karyotype. He became negative by rt-PCR for the detection of the bcrabl translocation (stringency=1:100,000). In an effort to take advantage of his excellent response, he had BM harvested for the possibility of future use as source of hematopoietic cells. The patient was clinically doing well, however a repeat BM done one year after the initiation of Gleevec treatment (10/00) showed slight myeloid preponderance; karyotype and FISH were consistent with reemergence of the blast CML clone in 34% of the cells. He subsequently underwent high dose chemotherapy (Bu/Cy) followed by autologous BM transplant (12/00) using the previously stored bone marrow, harvested in CR. The patient tolerated this treatment well. He has been receiving maintenance with interferon-alpha (5 million units, five days per week) as well as Gleevec and he is currently in hematologic response, although he remains PCR (+). Gleevec produced a dramatic response in this patient, enabling bone marrow harvest and transplant, which we believe has further benefited this patient. This case report illustrates the usefulness of Gleevec in treating such patients that may enable favorable long term outcomes.

The effectiveness of the tyrosine kinase inhibitor imatinib (STI571, Glivec®) in CML has been established since the drug first entered clinical trials in 1998. We have examined morphologic changes in the bone marrow and blood in 27 patients with CML 3 months after commencement of STI571. Eighteen patients had CML in chronic phase. Nine patients had accelerated (AP) or blastic phase (BP) disease. The median patient age was 56 years (range 20-75) and median interval from diagnosis to commencement of STI571 was 50 months (1-134). At three months, 25 of 27 patients achieved complete hematologic response (CHR) as defined by WBC count <10x10⁹/l and platelet count <450x10⁹/l. In 25 the marrow cellularity was reduced by a median of 56% (5-95%). In the other two patients there was an increase in cellularity of 25% and 15%; one of these two missed 3 weeks of treatment due to a skin rash and the other had interferon induced marrow hypoplasia prior to commencement of STI571. There was a reduction in granulocytic cells and an increase of erythroid cells with a consequent median 2.2 fold decrease in the myeloid to erythroid ratio. The number of megakaryocytes was reduced in 18/24 patients. Importantly the reduction in megakaryocyte number was accompanied by an increase in the proportion of morphologically normal appearing megakaryocytes in 21 of 24 patients. In three patients the number and the morphology of the megakaryocytes could not be assessed. The proportion of Ph+ metaphases was reduced by a median of 77% in CP and 48% in AP and BP patients. Ten of 14 CP and 3 of 4 AP and BP patients with assessable metaphases had a reduction in Ph+ cells. There was no statistical correlation between the degree of reduction of cellularity and cytogenetic response (r²=0.01). In summary, in CML patients treated with 3 months of STI571, there was a significant clearance of leukaemic cells as evidenced by CHR in the blood together with reduced marrow cellularity, restoration of morphologically normal haemopoiesis and cytogenetic response.

We present a case of a 44 year old Korean American lady who initially presented with a morbiliform rash over her arms and chest for 2 months. She had not been ill previously and had no other positive findings on physical examination. The rash was initially thought to be an allergic reaction and treated as such without response. Subsequently she developed night sweats, and a complete blood count revealed WBC >120,000 with >80% circulating blasts. Bone marrow biopsy was consistent with AML FAB class M0 based on negative myeloperoxidase and NASDCE stains. Flow cytometry was positive for CD13 and CD 33, and further supported this classification. Cytogenetics revealed a translocation between the short arm of 12 and a chromosomal segment of unknown origin. A skin biopsy confirmed leukemia cutis which, to our knowledge, has not been reported in a patient with M0 AML.She was treated with standard induction chemotherapy of 7+3 with a cytarabine and daunorubicin, to which she initially responded. By day 7 the absolute number of circulating blasts began increasing, and she was started on high dose cytarabine with weekly vincristine and prednisone. This did not clear the blasts from her peripheral blood, and on day 18 she was treated with gemtuzumab ozogomycin. This did reduce her total WBC count, but there was persistence of circulating blasts. PCR analysis for BCR/ABL was requested on peripheral blood, and this was interpreted as positive. On day 30 she was started onimatinib mesylate (STI571). By day 40, she entered a remission. Three weeks after starting STI571 she was moderately thrombocytopenic with a normal WBC and no circulating blasts. She is preparing to undergo a sibling-matched allogeneic transplant. This case illustrates an example of a patient with AML M0, who was Philadelphia chromosome negative but had a cryptic Bcr/abl translocation and obtained a remission with STI571. A retrospective analysis of patients with AML reported by ECOG (Paietta et all-Leukemia-1998) suggests that the incidence of Ph chromosome in AML is very low(0.9%), but this group of patients represents a clinically distinct entity with poor outcomes. All six patients who had the Ph chromosome in the ECOG data failed the standard induction regimen with anthracycline and cytarabine. Molecular analysis for BCR/ABL was not performed, and may have been underreported. It would therefore be important to analyse all AML patients who fail to respond to standard induction therapy for Bcr/abl.Based on the above example we make a case for molecular testing for BCR/ABL in chemoresistant and relapsed AML patients given the advent of targeted therapy for this situation. We suggest future studies in patients with AML should assess the molecular evidence of BCR/ABL to determine the true incidence.

Large Phase II studies have demonstrated the effectiveness of STI571 (imatinib mesylate) in patients (pts) with CML in chronic phase, although data regarding the value of dose escalation in pts with suboptimal response to the initial dose are lacking. Follow-up data are available on 16 pts (13 chronic phase, 3 accelerated) in whom the dose of STI571 was increased, treated at a single institution as part of a multicenter trial of pts previously treated with interferon. All but 2 of the pts in chronic phase originally received 400 mg/day with an increase to 800mg/day; the other 2 pts were increased to 600mg/day. The patients in accelerated phase started at 600 mg/day with an increase to 800 mg/day. 3 pts had their dose increased after 3 months due to lack of hematologic response (HR). Primary indications for dose increase in the other 13 pts (who received therapy for 6-15 months before dose escalation) were lack of cytogenetic response (CytoR) in 3 pts, loss of previous CytoR in 1 pt, and clonal evolution (CE) with cytogenetic abnormalites in addition to Philadelphia chromosome (Ph) in 9 pts. Of the 3 pts with lack of CytoR, 2 have had a major CytoR (now only 25-35% Ph+ cells 6 months after dose increase), and the other, who initially had an excellent clinical response in aggressive accelerated phase, did not respond cytogenetically or clinically to the dose increase from 600 to 800 mg. One pt who lost his initial partial CytoR had a major CytoR to dose increase (5% Ph+ cells) with associated thrombocytopenia. Subsequently, his Ph+ clone has increased, perhaps due to stopping the STI571 for almost a month to permit surgery. Of the 9 pts with CE, 5 had CytoR after dose increase. In 2 cases, this is a complete CytoR; in the other 3 pts, there was either reduction or elimination of the sub-clone, with persistence of the Ph chromsome in all remaining cells (i.e., a return to chronic phase). 2 of 3 pts with no initial HR achieved HR, with either minor or transient CytoR. A number of patients experienced significantly increased fluid retention at the higher dose. 5 of the pts who improved after dose increase had a BSA > 2.2, raising the question of whether patient size needs to be considered in selecting the initial dose of STI571. Summary: Some pts with chronic phase CML who have inadequate HR or CytoR or evidence of clonal evolution may benefit from dose increase to 800 mg/day. Examination of a larger cohort of pts with consideration of recipient size would be of interest.

The prognosis in adult acute lymphoblastic is poor in the Mexican population. This may be due to unfavourable biologic factors intrinsic to this population. The presence of the Ph translocation is an accepted unfavorable prognostic factor. We tested the presence of Ph chromosome by conventional cytogenetics and the BCR-ABL fusion with interphase fluorescent in situ hybridization (FISH) in 17 new patients seen at our Department from September 2000 to June 2001. Median age: 18 years (range 16 to 59), male 6, female 11. They were classified by morphology as L-2 (FAB), all were myeloperoxidase negative by cytochemistry and flow cytometry. By immunphenotype 12 cases marked as of B lineage, 4 as T and 1 was not done. Nine (52.9%) patients were BCR-ABL +, 5 (29.5%) were negative and in 3 (17.6%) hybridization was not successful. The median of the percentage of positive BCR/ABL nuc ISH was 8% (range 6 to 97%). We used as positive controls 7 newly diagnosed patients with Chronic Myelogenous Leukemia, in all of them BCR-BCL by FISH was positive; median 47%(range;12% to 68%). In 5 healthy subjects FISH was negative median 1% (range 0.5% to 3.2%). We studied with conventional cytogenetics the 9 patients BCR-BCL +: The karyotype was normal in 4, not evaluable in 3 and Ph+ in 2. In the CML patients, the karyiotype was Ph+ in 6 and in 1 was not evaluable. All the normal controls had a normal karyotype.This small study suggests that 1).- The BCR-ABL fusion gene is present in approximately 50% of the patients with myeloperoxidase negative acute leukemia seen in our Institution. 2).- Fish techniques are more sensitive than conventional cytogenetics and should be done in every patient with ALL. This is important because they should be treated with hematopietic stem cell transplantation or entered into clinical trials with new drugs like imatinib. We are accruing more patients to strengthen our hypothesis and we are planning to test our material with PCR.

In order to determine the pronostic factors for the response to imatinib (STI571), we have analysed the data from 75 CML patients (M/F : 44/31) median age 55.7 (20.9-77.7) treated in a single institution. The patients were treated orally (400 mg daily) and included in the 0110 and 0113 protocols (Novartis Pharma). All the patients were in chronic phase CML according to Novartis's protocols criteria, but 26/75 patients were considered in accelerated phase according to the IBMTR's criteria. The median time from diagnosis was 37.8 months (7.4-276.6). The Sokal score at the time of diagnosis was used to classify the population into three groups : low (n=28/74), intermediate (n=25/74) and high risk (n=21/74). All the patients were treated with Interferon a (INF) before STI571; therefore, Hasford score was also used to classify pts in three groups : low risk (n=25/73), intermediate risk (n=39/73) and high risk (n=9/73). Twenty eight patients received autologous peripheral blood stem cell transplantation (PBSCT) before, the median time between transplantation and STI571 was 33.3 months (11.9-91). The therapy had to be stopped for hematological toxicity (grade III) in 26/75 pts (34.66%) which occurred more frequently in patients previously treated by PBSCT (16/28, 57.14%) than for other patients (10/47, 21.74%)(p=0.006). The median follow up of this study was 7.34 months (0-16.9). A complete hematological response (CHR) was achieved in 69/72 (95.83%) evaluable patients after a median time of 0.69 months (0-5.51 ). After a median time of 5.51 months (0-11.02), 37/55 evaluable patients achieved a major cytogenetic response (MCR) including 21 patients with a complete cytogenetic response (CCR) (median time to obtain CCR : 11.05 months, range : 2.75-11.25).The cumulative probability to obtain MCR and CCR at 12 months is 83.9%±14% and 58.3%±20% (95% CI) respectively. By univariate log rank analysis, a number of patient-related and disease-related variables were tested for their association with CCR induced by STI571 therapy : age, sex, white blood count (WBC), platelet count, hematocrite, spleen size, peripheral eosinophils, basophils, peripheral blast cells (PB), Sokal score, Hasford score, time between diagnosis and STI571, PBSCT, time between PBSCT and STI571, IFN, hematological or cytogenetical resistance and transitory response before STI571, minor cytogenetic response (i;e at least of 5% Ph negative metaphases) on Day 0 before STI571 : Age (p=0.034), WBC (p=0.051), hematocrite (p=0.054), eosinophils (p=0.011) and obtention of minor cytogenetic response (p=0.0005) were found to influence the CCR rates. Using multivariate log rank analysis, three factors were confirmed to be related with the probability of achievement of CCR at 12 months : Age51 years with p=0.01 and RR (95%CI)=3.8 (1.3-11.3); PB cells at diagnosis with p=0.03 and RR (95%CI)=0.28 (range : 0.1-0.9) and minor cytogenetic response Day 0 before STI571 therapy with p=0.0001 and RR (95%CI)=9.77 (2.39-39.9). So from this study, we conclude : 1)STI571 induces more hematological toxicity for the patients previously treated by autologous transplantation than for the other patients, 2)The presence of Ph negative cells at the time of starting STI571 therapy is the strongest predictive factor for the response.

The Bcr-Abl chimeric protein plays a central role in the pathogenesis of Philadelphia-positive chronic myeloid leukemia (CML) by its independent tyrosine kinase activity. Consequently, multiple signaling pathways are activated in the malignant cell proliferation and differentiation, which may affect the expression of several genes. STI571 (imatinib Mesylate) is a specific tyrosine kinase inhibitor that blocks Bcr-Abl protein activation by targeting the ATP binding site of the protein domain, representing a successful therapy in patients with chronic myeloid leukemia (CML). By CG8 pathway microarray, which has been designed at the Cancer Genomics Core Laboratory of MD Anderson, we investigated the alterations in gene expression of p210^bcr/abl KBM5 leukemia cell line after 2 and 12 hours of exposure to 0.5 m M STI571 (IC₅₀). After extraction of total RNA from the cells, cDNA was synthesized by reverse transcription, labeled with Cy3 and Cy5 fluorescent dyes, and hybridized to the microarray, in which are spotted 1344 well-characterized genes, all involved in major cellular pathways. Following image acquisition and quantification of the signal intensity, resulting data were processed for normalization, background-correction, and subsequent statistical analysis. To correctly determine gene expression profiles on each array, we applied a statistical method in which the log ratio of gene expression levels between samples has been rescaled to account for the observed variability, and differences were analyzed by t-statistics. In comparison with untreated cells, after 12 h of drug exposure, 86 genes were shown to have different level of expression. However, 76 of those were already and consistently up- or down-regulated after 2 h of treatment. In particular, we identified up-regulation of genes encoding for transcription factors and DNA binding proteins (e.g. HMGs, E2F, Aml-1), cell cycle regulator proteins (e.g. CDK4, CDK8), signal transduction proteins (e.g. dual specificity phosphatase 4, death-assocaited protein 6), growth factors and growth factor receptors (e.g. TPO, FGF, VIPR1). We observed steady down-regulation of genes encoding for structural proteins (e.g. integrin-alpha 5 subunit, CD54), transcription regulatory genes (e.g. BCL3), and DNA-binding proteins. Up-regulation of Hck, and down-regulation of genes such as nuclear antigen Sp100, MAD2, and nucleophosmin were transiently detected after 2 h of drug exposure. In contrast, up-regulation of genes encoding for PTPRC and vimentin, and down-regulation of NM23A, PCNA and ornithine decarboxylase 1, were detected only after 12 h of treatment. In summary, our results suggest that even only after 2 hours of treatment, the expression of a number of genes involved in different cell pathways is regulated by the blockade of BCR-ABL tyrosine-kinase activity. These genes may represent potential targets for additional therapeutic interventions.