Blast crisis is the terminal phase of CML and refractory to therapy. The causative event in the initiation of CML is the Philadelphia translocation t(9;22) resulting in the fusion of genomic sequences to form the BCR-ABL oncogene which codes for a constitutively active tyrosine kinase mediating transformation. Gleevec is an orally administered inhibitor of the BCR-ABL tyrosine kinase. In a phase II trial, 260 CML patients in myeloid blast crisis were enrolled between 7/1999 and 6/2000 and treated with Gleevec in 27 centers in 6 countries (USA, Germany, Italy, France, UK, Switzerland) starting at daily doses of 400 mg (n=37) or 600 mg (n=223). Central review of baseline data confirmed a diagnosis of blast crisis in 229 patients, in 148 (65%) of whom blast crisis was newly diagnosed while 81 (35%) patients had received prior therapy for advanced CML (other than IFN, hydroxyurea or low dose ara-C). Median duration of therapy was 4 months(>6 months in 34% of patients). Gleevec induced hematologic response in 119 patients (52%). Response was sustained (> 4 weeks) in 66 patients (29%) including complete hematologic response in 16 patients (7%) and 7 patients (3%) who had <5% bone marrow blasts without recovery of peripheral blood counts. Cytogenetic response was detected in 36 patients (16%) including 15 (7%) complete (Ph+ 0%) and 21 (9%) partial remissions (Ph+ 1-35%). At the time of analysis (data cut-off 31 January 2001), median duration of hematologic response was 8.3 months, and median survival 6.9 months. At 6 months 65% of responders were relapse-free. Prognostic factors for hematologic response were age >60 years, hemoglobin >10g/dl, peripheral blood blasts <50%, and platelets >100x109/L. Hematologic response at 3 months was a significant prognostic factor for survival. Gleevec was well tolerated in these critically ill patients. Non-hematologic adverse reactions were frequent and included nausea, edema, vomiting, muscle cramps, diarrhea, and rash but were generally mild or moderate in severity and rarely required discontinuation of treatment. Newly occurring grade 3/4 neutropenia and thrombocytopenia were seen in 63% and 60% of patients, respectively. Only 7 patients discontinued therapy because of cytopenia, 4 of which were suspected to be drug-related. With the exception of one fatal hemorrhage, deaths due to infection or bleeding were seen only in patients with uncontrolled leukemia. Gleevec is a significant advance for the treatment of CML patients in myeloid blast crisis though emerging data on resistance mechanisms strongly suggest the need to combine Gleevec with chemotherapy and/or stem cell transplantation. The information in this abstract will be updated to reflect data available as of 31 July 2001.

Toxicity profile was similar to that seen with hyper-CVAD alone. Tumor lysis syndrome with renal failure required hemodialysis in one patient. Molecular studies with quantitative RT-PCR for bcr-abl are ongoing. In summary, hyper-CVAD with STI571 is active, feasible and well tolerated. Longer follow up is needed to determine durability of responses.

Gleevec™ is an orally administered selective inhibitor of the Bcr-Abl tyrosine kinase approved for use in patients with CML after failure of interferon therapy. This update provides additional follow-up on 532 patients enrolled in late chronic phase CML (cut-off 31 January 2001). Treatment was with Gleevec™ 400 mg/day. Of the 532 enrolled patients, 152 were hematologically resistant or refractory to interferon, 188 were cytogenetic failures, and 192 intolerant of interferon. The median duration of treatment for this update was 347 days (range 16 to 418). Overall 92% (n=490) of patients continue on study. Among the 454 (85%) patients for whom the diagnosis of chronic phase CML was confirmed upon a central review, Gleevec™ induced major cytogenetic responses (complete plus partial) in 55% of patients (n=248), of which 36% were complete. Only 8% (n=20) lost their major cytogenetic response thereafter. Complete hematologic response was reported in 91% of patients (n=415). The estimated probability of being free from progression to accelerated or blastic phase at 12 months was 91%, and estimated survival at 12 months was 98%. In a multivariate analysis of prognostic factors, platelet < 450 x 10⁹/L), hemoglobin 120 g/L, bone marrow blasts < 5%, time from diagnosis of CML< 1 year, and disease group were associated with a better rate of cytogenetic response. In a landmark analysis, the achievement of a major cytogenetic response at 3 months was associated with an improved progression-free survival. Safety was analysed on the whole population of 532 patients. The most frequently reported events were mild nausea, edema, and muscle cramps. Grade 3&4 neutropenia and thrombopenia were seen in 34% and 18% of patients. Drug-related adverse reactions caused withdrawal of 9 patients, and no treatment-related deaths occurred. Prolonged therapy with Gleevec™ is feasible, effective, and well tolerated. This information will be updated to reflect an additional six months of data in these patients.

The hallmark of chronic myeloid leukemia (CML) is the BCR-ABL fusion gene, which is formed as a result of the Philadelphia translocation, t(9;22) in the majority of patients. Glivec, a BCR-ABL tyrosine kinase inhibitor, has shown significant anti-CML efficacy in all phases. We treated a 33-year-old female in accelerated phase with Glivec 600 mg orally daily as a part of Multicenter Expanded Access Program. She was first diagnosed with chronic phase Ph-positive CML in December 1992, treated with hydroxiurea and alpha interferon, with no cytogenetics response. In January 2001, she started with fever and massive splenomegaly. Her hemoglobin was 10,0 g/dl, platelets 297x10⁹/L, WBC 21,8x10⁹/L, with 30% of basophils. The bone marrow aspirate showed 5,6% of blasts. The conventional cytogenetics reveled an additional chromosomic abnormality further than Ph-chromosome [add(17)(p11), +21]. The patient was started on 600 mg daily of Glivec and achieved a clinical and morphological CR after a briefly myelosupression on third month. She achieve complete hematological response six months later. The cytogenetics showed 20/20 metaphases positive for Ph chromosome, with no additional chromosomic abnormality. CML is a disease that progresses inexorably to accelerated and blast crisis, often within 5 years of diagnosis. The mechanism underlying this evolution remain unknown. Deletion or inactivation of p53 (17p) have been reported. Some studies support the concept of genomic instability in BCR-ABL transformed cells. It is possible that the anti-apoptotic effect of BCR-ABL favors an inexact DNA repair. The Glivec has demonstrated effects on BCR-ABL positive leukemic cells, including inhibition bcr/abl tyrosine phosphorylation, block of cellular proliferation and induction of apoptosis. Targeting a critical molecular abnormality, even in advanced stage of disease, is a useful strategy. Combination trials with other active agents will need to achieve maximal therapeutic benefits and reduction of resistance.

A subgroup of CMMoL patients can present with t(5;12)(q33;p13) translocation. Cloning of the breakpoint reveals the presence of a fusion between TEL (an Ets family transcription factor) and platelet derived growth factor receptor (PDGFbR) genes, coding for a protein with tyrosine kinase (TK) activity. The TEL/PDGFbR consists of 154 amino acids of TEL fused to transmembrane and cytoplasmic domains of PDGFbR. TEL-PDGFbR fusion gene produce a rapidly fatal myeloproliferative disorder in mice resembling human CMMoL. Gleevec is a TK inhibitor with proved activity against bcr-abl TK in patients with chronic myeloid leukemia (CML), and has also been shown to inhibit the TEL-PDGFbR TK in vitro. Based on these observations, and the clinical similarities between some CMMoL and CML patients, a pilot study was conducted to test the efficacy of Gleevec in patients with CMMoL. Since the t(5;12) translocation is a rare finding in CMMoL, its presence was not included in the eligibility criteria. Median age of 15 CMMoL patients registered on study was 68 years, with 3 females and 12 males. All had markedly hypercellular marrows, median cellularity being 90%. Eleven patients had normal cytogenetics, 1 had -7, 1 had del 20q, 1 had t(6;13) and 1 had del 11q. Median white blood count (WBC) was 14,000/ul (range: 3-226,000/ul) with 10/15 patients having proliferative CMMoL or WBC of 13,000/ul or greater. All treatment for CMMoL save supportive care was stopped for at least 4 weeks prior to starting Gleevec at 400mg q day for 16 weeks. Nine patients stopped Gleevec between 2-7 weeks, 1 due to fluid retention, and 8 due to disease progression or continuous increase in WBC. Of 6 remaining patients, 2 are too early for evaluation, and the other 4 have shown no response despite receiving Gleevec for 11, 13, 16 and 16 weeks, but their disease has remained stable. All four with stable disease had normal cytogenetics. Four patients have died as a consequence of their disease during the course of this study, although none of the deaths were considered unexpected, or related to the drug in any way. Interestingly, one CMMoL patient with a complex karyotype including del(12)(p11.2;p13) treated off study with similar doses of Gleevec showed a dramatic improvement. Her WBC fell from 79,000/ul to 8,000/ul along with decrease in hepatomegaly. We conclude that Gleevec has little or no efficacy in at least those CMMoL patients who present without specific cytogenetic abnormalities. The full range of cytogenetic abnormalities which may render the cells sensitive to Gleevec remain to be defined and may include others such as del(12) in addition to the t(5;12) associated with TEL-PDGFbR fusion protein. We suggest that future trials of this agent in CMMoL be restricted to cases presenting with cytogenetic abnormalities associated with chromosome 12 since the precise relationship between chromosome 12 abnormalities and the mechanism of response remains unknown.

Imatinib mesylate (Gleevec, formerly known as STI571) is a tyrosine kinase inhibitor approved for the treatment of certain patients (pts) with chronic myelogenous leukemia (CML). It has been demonstrated that Gleevec can induce both hematologic and cytogenetic responses in pts with stable phase CML who are refractory to or intolerant of interferon. Substantial activity has also been noted in more advanced CML as well as in Philadelphia chromosome positive (Ph+) acute lymphoblastic leukemia (ALL). Gleevec's activity and safety has been less well examined in pts with bcr/abl positive malignancies that have recurred following allogeneic transplantation. We have treated 16 pts with Gleevec for relapse of either CML (N=12) or Ph+ ALL (N=4) after allogeneic transplantation. Median pt age was 44 yrs (range, 28-59). Eleven pts were male and 5 were female. All had received marrow transplants, 10 from related and 6 from unrelated donors. The median interval between transplant and relapse was 12 months (range, 2-24 mo) and between relapse and initiation of Gleevec 2 months (range, 1-96 mo). Five pts had previously received donor lymphocyte infusions for their relapse; two such pts had responded to and then relapsed after DLI while three had never responded. At the time of relapse after transplant, fifteen of the 16 pts had karyotypic abnormalities in addition to the Philadelphia chromosome. Five of 12 CML pts initiated Gleevec for blast crisis (BC), while 7/12 clincally appeared to be in stable phase (SP) but were classified as accelerated phase (AP) based primarily upon cytogenetic clonal evolution. Five of these seven had evidence of hematologic relapse while 2 had cytogenetic relapse only. The initial dose of Gleevec was 600 mg/day for 14 patients, 400 mg/day for 2 pts. In general, treatment was well tolerated. On therapy, four pts (3 CML-BC, 1 ALL) developed liver function test abnormalities prompting discontinuation of the drug. In 3 cases, LFT abnormalities were thought most likely to be secondary to GVHD, while in one pt secondary to leukemic infiltration. These four patients had all undergone abrupt withdrawal of immune suppression a short time before initiating Gleevec, perhaps precipitating GVHD. Two pts (CML-BC) developed severe marrow hypoplasia on Gleevec, one of whom became negative for the Ph chromosome by FISH analysis. Hematopoiesis was restored in this patient with a non-myeloablative transplant from the initial donor. The three other CML-BC pts did not achieve a hematologic or cytogenetic response and all died of progressive disease. Six of seven earlier phase CML (SP/AP) pts have achieved a complete hematologic response after 1 month of therapy. After 3 months of Gleevec, 4/5 evaluable pts achieved a cytogenetic response (1 major and 3 complete). All three complete responders are negative for bcr/abl by fluorescent in situ hybridization(FISH) at 3 months. Three of 4 Ph+-ALL patients achieved a hematologic response, and 1 of these achieved a complete cytogenetic response at 3 months. Two of these three responding pts have relapsed, however, while receiving Gleevec. We concluded that Gleevec is generally well tolerated in pts who relapse after allogeneic transplantation. Complete cytogenetic and molecular responses can be obtained, especially in those pts with less advanced disease. Gleevec should be evaluated as an alternative or adjunct to DLI in a prospective trial for pts with Ph+ leukemias who relapse after allogeneic BMT.

The median CQPCR ratio was 35% if no CG response, 18% with minor CG, 6% with partial CG response, and .89% with complete CG response (p < 0.001). In patients in complete CG response the CQPCR ratio ranged from 0.03 % to 100%. 92 samples (41%) had CQPCR ratio < 1%. 125 pts has simultaneous CQPCR in peripheral blood and marrow samples which showed an excellent correlation.(r=.965; p < 0.001; concordance rate 88%). CQPCR median values in patients in CG CR on therapy over time were: 3 months (N=17)- 5.0%; 6 months (N=25)- 1.0%; 6 months (N=44)- 0.89%; 12 months(N=31)- 1.3%; 15 months (N=33)- 0.89%.

Blast crisis (BC) is the most advanced stage of chronic myelogenous leukemia (CML) and highly refractory to therapy. CML is caused by expression of the chimeric BCR-ABL tyrosine kinase, the product of the t(9;22) Philadelphia (Ph) translocation. BC is marked by karyotypic evolution, or the accumulation of multiple genetic abnormalities. STI571 (imatinib mesylate, Glivec) is an orally administered selective inhibitor of the BCR-ABL tyrosine kinase with efficacy in BC. We sought to determine the impact of biological factors, i.e. the individual immunophenotype, additional chromosomal aberrations, and the BCR-ABL fusion type on the response to STI571 and survival. 59 patients (27 m, 32 f, median age 56, range 32-78 years) were treated for myeloid BC with an initial dose of 400-600 mg STI571/day in two consecutive phase II protocols (0102 and 0115). Myeloid phenotype was confirmed by flow cytometry and/or immunostaining (APAAP method) of peripheral blood (PB) and/or bone marrow (BM) blasts in all cases, requiring myeloperoxidase positivity and the presence of at least two standard myeloid markers. In 49 patients, an extended analysis of markers for specific lineages was performed. Thirty patients (51%) presented myeloid markers only, six (10%) expressed additional B-lymphoblastic, seven (12%) megakaryoblastic, and eight (14%) monoblastic markers. Sustained hematologic response was observed in 15/39 evaluable patients, in 9/25 patients with myeloid markers only, 2/3 patients with coexpression of B-lymphoid markers, 1/6 megakaryoblastic, and 3/7 monoblastic crises. Median survival was 270 days in all patients, 346 days in patients with myeloid markers only, 270 days in patients with lymphatic coexpression (ns), 120 days in megakaryoblastic crises (p=0.026), and not yet reached in monoblastic crises (ns). Cytogenetic analysis of BM and/or PB metaphases was performed in 53 cases. Twenty-two Ph+ patients lacking karyotypic evolution had a median survival of 270 days. In 31 cases (58%) additional aberrations have been observed (median survival 176 days, ns), being aneuploidy, n=19 (132 days, ns); trisomy 8, n=8 (276 days, ns); 2nd Ph chromosome, n=4 (99 days, p=0.021); aberrations of chromosome 17, n=11 (98 days, p=0.067). The type of the BCR-ABL fusion transcript was determined by multiplex RT-PCR in 59 cases. Forty-six patients (78%) expressed b3a2, 11 b2a2 (19%), one b3a2 & b2a2, and one e1a2 BCR-ABL transcripts. Patients expressing the b3a2 transcript type survived significantly longer than b2a2 patients (276 vs 108 days, p=0.01). We conclude that in myeloid BC, immunophenotype, specific additional cytogenetic aberrations, and the BCR-ABL fusion type may be associated with response to STI571. Expression of megakaryoblastic markers, a 2nd Ph chromosome, and the b2a2 BCR-ABL fusion transcript are negative prognostic factors with regard to survival. This data should encourage the collection of potential prognostic factors associated with the biology of BC in larger trials. Combination of STI571 with chemotherapy and/or stem cell transplantation might be considered in unfavourable situations.

Chronic myelogenous leukemia (CML) and acute leukemias with the Philadelphia chromosome - t(9;22) (Ph1) have constitutively activated bcr-abl tyrosine kinase activity which is an important molecular basis of these diseases. Childhood Ph1 leukemias are very difficult to cure when resistance to conventional therapy or relapse post allogeneic bone marrow transplantation occurs. Gleevec™ (imatinib mesylate, STI-571), an inhibitor of abl tyrosine kinase has shown clinical activity in adult Ph1 leukemias with minimal adverse side effects. Nine children, median age 7.4 (4-17) years with CML in chronic phase (3), CML relapsed after allogeneic bone marrow transplant (2) or Ph1 acute lymphoblastic leukemia (ALL) relapsed after conventional chemotherapy (4) have been treated at our institution. Four received an initial dosage of 160 mg/m², in two of whom the dose was subsequently doubled, three patients 260 mg/m² and two patients 340 mg/m², given by single daily oral dose and continued until evidence of disease progression or limiting toxicity. No other chemotherapeutic drugs were given concurrently. All received at least 6 weeks of drug, sufficient to evaluate anti-leukemia effect. All had hematological responses based on greater than 50% decrease in peripheral blast count (ALL), white cell count (CML) and/or percentage of blasts in bone marrow. No dose limiting toxicities have been observed. There have been three instances of transient grade III elevation of liver function tests. Three patients had mild nausea (grade I) not requiring dose modification. One patient had myelosuppression requiring transient discontinuation of the drug. Drug was discontinued in two ALL patients with disease progression after six and eight weeks and in one ALL patient after 14 months with continued leukemia, myelosuppression and fungal disease. One ALL patient died of progressive fungal pneumonia after six weeks while still on therapy. One patient (CML post transplant) discontinued drug after 12 weeks while in morphologic remission prior to treatment with donor leukocyte infusion therapy. This patient subsequently had an isolated CNS relapse. Four patients, 3 chronic phase CML and 1 CML post transplant, continue on medication for up to twenty-four months (median 17 months), 2 have no Ph1 noted on the most recent bone marrow cytogenetics evaluation. These preliminary results show that Gleevec™ can be safely given to children at doses that have clinical effect on Ph1 leukemias. Further prospective cooperative group studies are planned to better define the efficacy and optimal dosage.

Translocation t(9;22)(q34;q11) occurs in approximately one third of adult ALL patients and confers an unfavorable prognosis. Even the introduction of imatinib mesylate (Gleevec®) as a single agent has not altered the outcome of patients with relapsed or refractory ALL with t(9;22) (Druker et al., N Engl J Med; 2001, 344:1038). We sought to determine the nature and prognostic significance of secondary cytogenetic changes in newly diagnosed adults with ALL and t(9;22). We studied 90 patients with t(9;22) who enrolled onto CALGB protocol 8461, a prospective study of cytogenetics in ALL. All patients were treated on front-line CALGB clinical protocols. Secondary cytogenetic changes were seen in 64 (71%) of the patients. They included in order of descending frequency: +der(22)t(9;22), +21, abn(9p), hyperdiploidy (50 chromosomes), +8, +X, -7, loss (8p), +8q, loss(7p) and gain(1q). More than one clone was present in 26 (29%) of the patients, and 67 (74%) had at least one normal metaphase cell detected in their karyotype. There were eight balanced and 16 unbalanced translocations previously unreported in ALL with t(9;22). One balanced translocation, t(2;7)(p11;p13), was present in two cases in this series and has never been reported in any other malignancy. Further, one unbalanced translocation, der(18)t(8;18)(q11;p11), was found in two cases in this series and has been reported in the literature in a single case with another malignancy. Patients <30 years old had a significantly better outcome, but the only long-term survivors were patients who underwent allogeneic transplantation. The presence of secondary cytogenetic changes was not associated with clinical pretreatment parameters or outcome in patients with ALL with t(9;22) treated on CALGB protocols. It will be of interest to see if newly diagnosed adult ALL patients with t(9;22) only will respond differently to Gleevec® as compared to t(9;22) patients with additional cytogenetic aberrations.

The deregulated tyrosine kinase activity of the BCR-ABL fusion gene is generally accepted as the primary cause of chronic myeloid leukemia (CML). We are currently using the new Abl tyrosine kinase inhibitor STI571 (imatinib mesylate, Glivec®) to treat CML patients in various phases of CML. Here we report the quantitative real time PCR (QRT-PCR) data for 31 patients (14 male, 17 female, median age 52.3 years [range 37.6 - 67.5]) with CML in chronic phase who achieved complete cytogenetic remission (CCR) after treatment with STI571. The median time to achieve CCR was 23 weeks (range - 12 to 64). CCR was defined as the absence of the Ph chromosome in at least 20 marrow metaphases. QRT-PCR assay could detect both b3a2 and b2a2 BCR-ABL transcripts in a single reaction. Triplicate QRT-PCR analyses were performed on blood specimens using ABL transcripts as the endogenous control. The ABL assay was performed in duplicate for each sample and the result reported as BCR-ABL/ABL percentage ratio. BCR-ABL transcripts were detected by QRT-PCR in all 31 patients with a median BCR-ABL/ABL ratio of 0.052% (range - negative to 0.7). In some cases the QRT-PCR data identified those patients who had achieved CCR before the cytogenetic data were available, thus providing further evidence of correlation between CCR status and BCR-ABL/ABL ratio values. A gradual reduction in BCR-ABL/ABL ratio was observed among patients for whom serial samples were available. In one patient BCR-ABL transcripts were determined to be <1.0 with an ABL value of 4.1x10⁴, which we report as undetectable, 23 weeks post STI571 treatment. In an additional 4 patients, the BCR-ABL/ABL ratio was <0.001% but these patients were still positive by nested RT-PCR. To date none of the patients revealed an increase in BCR-ABL/ABL ratio subsequent to a reduction. The fact that the BCR-ABL/ABL ratio was <2.0% in all CCR patients was consistent with our earlier observations in patients treated by IFN or allografting which showed that the Ph chromosome could not be detected in marrow metaphases when the BCR-ABL/ABL ratio was below 2.0%. Overall, our data confirm the value of QRT-PCR to monitor patient response to therapy. We should be able to define a threshold QRT-PCR level above which the finding of cytogenetic relapse is likely. In the absence of clear molecular prognostic indicators the MRD data may help to identify patients who are likely to do well on ST571 or develop resistance to it.

Introduction : Gleevec® (STI571, imatinib) is a synthetic pyrimidine derivative which inhibits selectively the tyrosine-kinase activity of c-abl and BCR-ABL. STI571 produces hematological and cytogenetic remission in patients with chronic myelogenous leukemia (CML) in chronic phase but resistance and relapses are observed in the accelerated (AP) or blastic phases (BP). In a mouse model of CML, The plasma protein alpha 1 acid glycoprotein (AGP) has been reported to bind STI571 and to inhibit BCR-ABL kinase activity in a dose dependant manner (Gambacorti-Passerini, JNCI 2000). We investigated in this study the role of AGP levels in patients with CML in BP. Methods : K562 cells and cells from 9 patients with CML in BP enrolled in a multicenter clinical trial sponsored by Novartis Pharmaceuticals (CSTI571 0115) were incubated in presence of STI571 (provided by Novartis Pharma S.A.) alone or associated with various concentration of AGP. Autologous plasma and plasma from donors with high or low levels of AGP were also used (normal value 0.5 to 1.3 mg/ml). Tyrosine phosphorylation ratio was studied by FACS analysis in CML cells from patients in presence or in absence of STI571. AGP dosage was performed in patients at diagnosis and during the first month of treatment and then correlated to early clinical response. Results : AGP at 1 mg/ml and 3 mg/ml completely blocks the induction of apoptosis of K562 cells incubated with STI571 at 1 mM and 10 mM respectively. Similar results were obtained using sera from donors with low (0.98 +/- 0.3 mg/ml) and high (2.6 +/- 0.7 mg/ml) AGP levels. When the competitor erythromycin 100 mM was added in the culture, STI571 sensitivity was restored in the AGP 1 mg/ml group. The median value for AGP plasmatic concentration in the 9 patients with BP-CML was 1.48 mg/ml (range 0.77-2.31) compared to 1.10 mg/ml (range 0.78-1.95) and 0.94 mg/ml (range 0.56 to 1.57) in AP-CML and CP-CML patients respectively . When cultured in RPMI 10% FCS, cells from the 9 patients with BP-CML were sensitive to STI571 1 mM (56 +/- 20% of apoptotic cell at day 3, p<0.01 compared with untreated cells). The tyrosine phosphorylation ratio was identical for all the sample studied (49 +/- 9 %). When cultured in autologous serum and STI571 10 mM, the percentage of apoptotic cells at day 3 was significantly different (p<0,01) compared with untreated cells in the group of patients with low AGP plasmatic concentration. However, in the high AGP level group, no difference could be evidenced (p=0.2). All (3/3) patients with a high AGP level during the first three months failed to respond to STI571 in contrast to the 6 responding patients with a normal AGP level (PR 5/6 and CR 1/6). 4 responding patients subsequently relapsed after the third month. Conclusion : Resistance to STI571 has already been studied in patients who underwent disease progression after an initial response and has been found to be the consequence of mutation or amplification of BCR-ABL. We provide here a possible pharmacological mechanism of primary resistance to STI571 mediated by the binding of STI571 to AGP in patients with CML in BP.

Chronic Myeloid Leukemia (CML) is a chronic myloproliferative disease resulting from the accumulation of myeloid cells due in part to dysregulation of anti-apoptotic signaling. Intracellular pathways that control the survival of CML cells are activated by Bcr-Abl tyrosine kinase. Inhibition of Bcr-Abl tyrosine kinase activity with two different inhibitors imatinib meyslate (Gleevec®) and PD180970 resulted in apoptosis of an HL-60 cell line stably expressing Bcr-Abl (p185,HL-60-Bcr-Abl), K562 cells, and primary leukemic (CD34+) CML blasts. There was a corrresponding reduction in Mcl-1 (a Bcl-2 family protein) mRNA and protein expression. Stable overexpression of Bcr-Abl (p185)in HL-60 cells resulted in increased expression of Mcl-1 and Mcl-1 is constitutively expressed at a high level in K562 (Bcr-Abl positive)cells, both consistent with Bcr-Abl regulation. An important signaling event that contributes to CML survival is activation of STAT5. Bcl-XL was demonstrated to be transcriptionally regulated by STAT5 in a Bcr-Abl-dependent fashion. Transient transfection experiments were performed with a Mcl-1-luciferase (p-203/+10-luc)construct in HL60-Bcr-Abl and K562 cells. These experiments demonstrated that Bcr-Abl overexpression induced STAT5-dependent transcriptional activity from the Mcl-1 promoter. The Bcr-Abl inducible transcription was repressed by co-transfection with dominant-negative STAT5, pre-treatment with Gleevec ®, or pre-treatment with PD180970. These findings suggest that Mcl-1 is a novel anti-apoptotic molecule increased in CML and is transcriptionally regulated by a Bcr-Abl-dependent STAT5 signaling pathway. Because of the important role that Bcr-Abl and STAT5 activation plays in CML survival, anti-apoptotic proteins regulated by these pathways have a high probability of contributing to disease pathogenesis.

Imatinib mesylate (Gleevec, formerly STI571), has been shown to be a safe and effective treatment for chronic myelogenous leukemia (CML). In chronic phase patients who have failed therapy with interferon, over 50% of patients achieve major cytogenetic responses, while the remainder of patients are predominantly Ph chromosome positive. One postulated mechanism for the lack of cytogenetic responses is that quiescent hematopoietic stem cells may be resistant to the proapoptotic effects of STI571. Even in patients with major cytogenetic responses who remain positive for Bcr-Abl by molecular testing, it is possible that these quiescent cells could contribute to the persistence of residual leukemic cells. To determine whether changes in the proportion of cycling cells affect the capability of STI571 to induce apoptosis, we sought a method that was capable of inducing a reversible cell cycle arrest without toxicity. Unfortunately, most methods to arrest cells in G1 also result in cellular cytotoxicity. However, cells can be reversibly blocked at the G1/S boundary using a double thymidine (Tdr) block without toxicity. As STI571 requires at least 48 hrs of continuous exposure to irreversibly induce apoptosis, preliminary experiments were performed to determine the survival of K562 cells, a CML blast crisis cell line, for 48 hrs with ongoing exposure to Tdr after an initial blockade at the G1/S boundary. Thus, K562 cells were exposed to 1 mM Tdr for 2x16 hrs with an intermediate release of 10 hrs to recruit cells from late S and G2/M. This was followed by an additional 48 hr incubation in 1 mM Tdr. Cell count experiments demonstrated a complete block to exponential cell growth and by trypan blue staining, no loss of cell viability was observed. Staining with propidium iodide and subsequent FACS-analysis revealed that with 24hr to 48hr continuous exposure to 1mM Tdr after the initial blockade, > 80% of cells were arrested in S-phase. Detection of apoptosis by FACS-analysis of activated caspase-3 showed low levels of activation after 48hrs compared to control cells (6.6% ± 4.6% vs 1.4% ± 0.11%). This was confirmed using a TUNEL assay. Treatment of K562 cells with 1mM Tdr and 1mM STI571, started after the completion of the sequential 2x16hrs Tdr-block, resulted in activation of caspase-3 in 13.6% ± 3.1% of the treated fraction, whereas treatment with STI571 alone activated caspase-3 in 8.7% ± 0.4% of the cells. TUNEL-positivity did not differ in either treatment group (8.7% [STI571-Tdr-combination] vs 7.3% [STI571 only]). Treatment with STI571 for up to 3 days after the release from the 48hr cell cycle arrest also did not show different levels of apoptotic activation as compared to cells that were not subject to cell cycle arrest. Our data demonstrate that cell cycle blockade does not protect cells from the proapoptotic effects of STI571. This observation suggests that cellular proliferation is not a requirement for STI571 to induce apoptosis and that attempts to recruit cells into the cell cycle would not be an effective means of increasing cytogenetic or molecular response rates to STI571.

The goal of CML therapy is to eradicate the leukemic clone containing the Philadelphia (Ph) chromosome. While a PCR negative status can be achieved following allogeneic transplantation, such molecular remissions are rarely observed with interferon-based treatment. Gleevec (STI-571, imatinib mesylate) induces major cytogenetic remissions in 50% of chronic phase CML patients who fail interferon and thus represents the most effective drug for the induction of a minimally residual disease state. During the past decade, two-step nested RT-PCR of the BCR/ABL fusion mRNA has become established as the most sensitive qualitative assay for the detection of minimal residual disease, but due to the increased sample handling required, the assay is particularly susceptible to contamination, thereby yielding falsely positive results. MGB real-time RT-PCR uses a BCR/ABL probe which binds the minor groove of DNA and increases the specificity, sensitivity and overall accuracy of PCR. By combining reverse transcription, real time PCR and signal analysis in a single closed tube, MGB RT-PCR provides a potential platform for a robust, highly sensitive, quantitative assay that minimizes the risk of contamination. We compared the accuracy of quantitative MGB RT-PCR to an established quantitative RT-PCR assay (Radich et al, Blood 98, 2001, in press) using a known standard dilution series and a limited number of defined clinical samples. We found the two assays to be equivalent. We assessed minimal residual disease in 23 patients treated at UCLA as part of a phase II clinical trial who achieved complete Ph-negative, FISH-negative cytogenetic remissions. We analyzed the ratio of BCR/ABL to ABL mRNA copies (as well as to GAPDH) present in RNA extracted from blood or bone marrow. On at least one occasion, nine of these 23 patients had no detectable BCR/ABL mRNA by the quantitative MGB assay, confirming that profound remissions can be achieved in a subset of patients treated with Gleevec, although many of these remained detectable qualitatively by two-step nested PCR. In patients with detectable BCR/ABL mRNA, the levels were generally 3-4 orders of magnitude below those observed in newly diagnosed patients. Serial quantitative measurements of BCR/ABL mRNA in this patient cohort and comparisons to Ph-negative, FISH-negative patients treated with interferon or allogeneic transplantation are ongoing. For the growing population of CML patients whose disease is in complete cytogenetic remission as a result of Gleevec, we found that the quantitative MGB RT-PCR assay offers the sensitivity, reproducibility and precision requisite for appropriate clinical management decisions.

CML recurring with advanced disease stage after allogeneic stem cell transplantation (alloSCT) is associated with poor treatment outcome. Glivec (imatinib mesylate, formerly STI571) is a selective Abl tyrosine kinase inhibitor displaying promising single agent activity in advanced Ph+ acute leukemia patients in phase I and II studies. We therefore examined the clinical effects of Glivec in 13 patients with Ph+CML who relapsed after a previous alloSCT and were enrolled in successive multicenter, phase II trials of Glivec supported by Novartis. At the time of relapse and immediately prior to study entry, 8 patients were in blastic transformation, 1 patient presented with a chloroma, 3 patients were in accelerated and one patient was in chronic phase. The duration between alloSCT and relapse was a median of 6.5 months (2-50 months.). The patient with chloromas had already received Glived prior to SCT, with limited response. Assessment of efficacy was based on bone marrow morphology, cytogenetic and FISH analysis and measurement of donor chimerism. A complete hematologic response (CR) was achieved in 7 of 13 patients (4/9 in BC, 2/3 in AP and in 1 CP) within a median time of 1 month (0.5-3 months) of STI571. One AP patient failed to achieve a BM response despite clearance of peripheral blood blasts. A meningeal relapse occured in 1 patient with bilineage BC in CR after 7 weeks, which responded to i.th. chemotherapy and continued STI571. CR is ongoing in 6 of 7 pts. with a median duration of 9 months (5-17 months) of Glivec. All 6 patients with an ongoing hematologic response also showed a complete cytogenetic/FISH response, which persists after a median of 8 months. 3 patients achieved a complete molecular response on the basis of quantitative RT-PCR analysis (Taqman). The levels of donor chimerism increased from median 36% at relapse to above 98% within 12 weeks of starting Glivec in the responding patients. The most frequent treatment-related side effects were generally mild to moderate gastrointestinal discomfort and edema. Neutropenic colitis occured in one patient with only 7% donor chimerism prior to Glivec, which resolved after granulocyte transfusions, prolonged G-CSF and interruption of STI571. 2 patients who were refractory to STI57 developed a subdural hygroma/hematoma, which required surgical decompression in one patient; a possible contribution of STI571 could not be excluded. In conclusion, a significant proportion of patients with Ph+CML relapsing with advanced stage disease subsequent to an alloSCT achieve a complete hematologic and cytogenetic remission in response to single agent therapy with Glivec. Responses are associated with reestablishment of complete donor chimerism and may be prolonged. Glivec in conjunction with alloSCT appears to be an effective treatment for advanced stage CML. Further investigations in prospective clinical trials are warrented, some of which are ongoing.

STI571 (imatinib Mesylate) represents one of the major innovation in cancer therapy over the last few years. Targeting the ATP binding site of the protein domain, STI571 has been shown to inhibit effectively the BCR-ABL, c-kit and PDGF-R tyrosine kinases. In patients with CML, while STI571 induces durable remissions in chronic phase of the disease, responses have been shown to be transient in blast crisis, with relapses occurring within 12 months in >95% of patients despite continued drug treatment. However, recent clinical evidences suggest that, over a longer period of uninterrupted treatment, hematologic resistance to STI571 can appear also during the chronic phase of the disease, therefore representing an emerging major clinical problem. For this reason, understanding the mechanisms of resistance to STI571, at this point, is a crucial research challenge. To date, BCR-ABL gene amplification, point mutation of ABL, and Pgp overexpression, are cellular events that have been demonstrated in STI571-resistant cell lines and STI571-resistant patient-derived blast cells, while acquisition of compensatory mutations in genes other than BCR-ABL, as well as altered drug metabolism and/or transport, represent other mechanisms that have been hypothesized. From three BCR-ABL cell lines, by culturing them to gradually increasing concentrations of STI571, we developed sublines showing different levels of resistance to this drug:1 m M for p210^bcr/abl KBM5R, 0.7 m M p210^bcr/abl KBM7R, and 0.3 m M for p190^bcr/abl Z119R. After extraction of total RNA from parental and resistant cells, cDNA was synthesized by reverse transcription, labeled with Cy3 and Cy5 fluorescent dyes, and hybridized to CG8 pathway array, which have been designed at the Cancer Genomics Core Laboratory of MD Anderson. This array consists of a total of 1344 double spotted genes of known function, all involved in major cellular pathway. In particular, we aimed to compare the gene expression pattern of each STI571-resistant cell line to its parental strain, to unveil possible molecular changes involved in the mechanisms of resistance. 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. Comparing parental and resistant cells, several differentially expressed genes were identified. In particular, we documented up-regulation of signal transduction and transcription factors (e.g. HMG proteins, zinc finger proteins, nuclear antigen Sp100, V-crk oncogene), as well as VEGF-B, in both KBM5R and KBM7R. Furthermore, KBM5R showed up-regulation of the interferon consensus binding protein 1 and DNA repair enzymes, whereas there was a significant down-regulation of the Hck kinase. In the Z119R we identified up-regulation of several genes involved in signal transduction (e.g. TYRO3 protein tyrosine kinase, RAB5B, and MAPKK2) as well as in transcription modulation (e.g. GATA-binding proteins) and cell cycle regulation (e.g. cyclin-dependent kinases). Although confirmation by more specific molecular approaches is warranted, these findings suggest that the acquisition of resistance to STI571 entails multiple molecular events.

The Prognosis of patients with chronic myelogenous leukemia in blastic phase (CML-B) and relapsed/refractory philadelphia positive acute lymphoblastic leukemia (Ph+ALL) is extremely poor. Gleevec (imatinib mesylate, STI571, Novartis Pharmaceuticals) is a potent and selective inhibitor of the BCR/ABL tyrosine kinase that has substantial single agent activity in CML-B and Ph+ALL. Although the majority of patients respond to Gleevec and often rapidly, we sought to systematically determine the kinetics of hematological responses in this patient population. This analysis might assist in determining the length of a therapeutic trial with Gleevec as well as the optimal timing for the addition of other agents once maximal cytoreduction has been achieved. Data from 40 patients with CML-B and refractory/relapsed Ph+ALL treated at OHSU on phase 1 or 2 multi-institutional clinical trials were analyzed. Patients included in the analysis had WBC> 10x10³ cells/mm³ and peripheral blasts >5%. The WBC at presentation ranged from 14.6-85.8(Ph+ALL), 51.3-210.8 (Lymphoid blast) and 15.9-189.4 x10³ cells/mm³ (myeloid blast). The peripheral blast% were 14-88%(Ph+ALL), 7-72%(lymphoid blast) and 14-90%(myeloid blast). A WBC response was defined as a reduction in the WBC to <10,000 cells/mm³ and a peripheral blast response was defined as 0% blast in the peripheral blood. Bone marrow response was defined as <15% marrow blasts. 32 of 40 treated patients were evaluable for kinetics of response. 5 patients were excluded due to low WBC and peripheral blast% and 3 patients demonstrated progressive disease. The median time to WBC response in the Ph+ALL and lymphoid blast crisis patients were 3 (range 2-5) and 4.5 (range 2-7) days. The median time to peripheral blast response in the Ph+ALL and lymphoid blast crisis patients were 6.5 (range 1-14) and 8 (range 4-10)days. In contrast, the median to WBC response in the myeloid blast crisis patients was 8.5 (range 1-38) days and the median to peripheral blast response was 16 (range 8-28) days after initiation of therapy. 3/5 (60%) of lymphoid blast crisis, 6/13 (46%) Ph+ALL and 9/14 (64%) myeloid blast crisis patients demonstrated bone marrow responses. Despite the rapidity of responses, tumor lysis syndrome was not seen in these treated patients. The BCR/ABL tyrosine kinase inhibitor, Gleevec, has substantial single agent activity in CML-B and Ph+ALL. The time to peripheral blood and bone marrow response is short with a notable difference in kinetics of response between the lymphoid and myeloid variants of the disease. Our data suggests that 2 weeks is a sufficient length for a therapeutic trial of Gleevec in Ph+ALL and lymphoid blast crisis patients, however, up to 4 weeks is required for response to occur in myeloid blast crisis patients. Peripheral blood responses do not always predict for marrow responses. Nevertheless, the high rate of response and ability to safely cytoreduce patients suggests that Gleevec could be used as initial therapy in patients with these malignancies, and may be exploited to develop novel therapeutic approaches for these otherwise poor prognosis patient population.

The prognosis of patients with Ph+/bcr-abl+ ALL who relapse after alloSCT is poor. Glivec (imatinib mesylate) is an inhibitor of the ABL tyrosine kinase with potent antileukemic activity in advanced Ph+ALL, although the duration of response is usually short. The clinical effects of Glivec on Ph+ALL recurring after alloSCT have not been established. We analysed 20 consecutive Ph+ALL patients who relapsed subsequent to alloSCT and were enrolled in multicenter clinical trials of Glivec (supported by Novartis). 2 pts. had received Glivec previously to enable transplantation. Glivec as a single agent induced a CR with PB recovery in 11 pts. (55%) and a complete leukemic response with persistent cytopenias in 4 pts. (20%). 5 pts. were refractory, including 1 early death on day 11 due to generalized leukemic organ infiltration. In CR patients, Ph+ cells became undetectable by cytogenetic and FISH analysis. Donor chimerism levels in responding patients increased from a pre-study median of 83% in PB and 64% in BM to 98% in both PB and BM within four weeks of starting Glivec. Concomitant treatment with immunosuppressive agents, antiviral and antifungal agents was feasible without apparent severe drug interactions. 10 of 15 responding patients relapsed after a median treatment duration of 5 months (range 8-33 mos.), one pat. died in CR at 3 mos.of transplant-related causes. A complete remission is ongoing in 4 pts. after 6, 10, 46 and 78 weeks on Glivec, respectively. One patient remains in complete molecular remission, based on quantitative RT-PCR (Taqman), after 1.5 years of treatment. In conclusion, Glivec is highly effective as initial treatment of relapsed Ph+ALL subsequent to alloSCT, with a favorable safety profile. A prolonged CR is achieved in a small subset of patients, and molecular remissions are rare. Additional therapeutic modalities are required to prevent relapses in the majority of patients with advanced Ph+ALL; these will be explored in ongoing and future prospective clinical trials.

STI571 (imatinib mesylate) is an Abl tyrosine kinase inhibitor currently in clinical trials for patients with CML in various disease phases. Between January 2000 and June 2001 77 patients in chronic phase treated previously with interferon-alfa (IFN) received STI571 at the Hammersmith Hospital. The median age was 52 years (range 21-77); 49% males and 51% females. The median interval between diagnosis of CML and inclusion in the trial was 4 years (range 1 to 9 years). The median follow-up after starting STI571 was 236 days (range 42-474). 16% of the patients were hematologically resistant to IFN, 44% were cytogenetically resistant and 40% were intolerant. 15% of patients had previously undergone autologous stem cell transplantation. After 450 days on STI571 the actuarial risk of progression to accelerated (AP) or blastic phase (BP) was 32%. AP was defined as 10-19% blasts in peripheral blood or bone marrow or a peripheral basophil count >20%. BP was defined as >20% blasts in peripheral blood or bone marrow or presence of extramedullary blast cell tumors. At the time of most recent follow-up 19% of patients had complete cytogenetic responses (CCR, 0% Ph-positive metaphases), 21% had major cytogenetic responses (Ph-positive metaphases 1<35), 3% had partial cytogenetic responses (Ph-positive metaphases 35<65), 15% had minimal cytogenetic response (Ph-positive metaphases 65<95) and 42% had no cytogenetic response (Ph-positive metaphases 95%). Based on the presence or absence of cytogenetic response after 6 months of STI571 (defined as the achievement of at least minimal cytogenetic response) and the development of grade III-IV cytopenias (defined as neutropenia or thrombocytopenia on one or more occasion during the first 6 months) we identified 4 patient groups whose prognoses were significantly different at one year (p=0.02, Cox regression analysis) as follows: (A) Cytogenetic responders with no cytopenias (44%), actuarial risk of progression of 3%. (B) Cytogenetic responders with cytopenias (15%), actuarial risk of progression of 15%. (C) Cytogenetic non-responders with no cytopenias (15%), actuarial risk of progression of 26%, and (D) Patients with no cytogenetic response and cytopenias (26%), actuarial risk of progression 67%. We conclude that consideration of the degree of cytogenetic response and the presence or absence of significant cytopenias permits patients treated with STI571 to be classified into one of four distinct prognostic groups.