Lack of clinical efficacy of imatinib in metastatic melanoma

This two-centre phase-II trial aimed at investigating the efficacy of imatinib in metastasised melanoma patients in correlation to the tumour expression profile of the imatinib targets c-kit and platelet-derived growth factor receptor (PDGF-R). The primary study end point was objective response according to RECIST, secondary end points were safety, overall and progression-free survival. In all, 18 patients with treatment-refractory advanced melanoma received imatinib 800 mg day(-1). In 16 evaluable patients no objective responses could be observed. The median overall survival was 3.9 months, the median time to progression was 1.9 months. Tumour biopsy specimens were obtained from 12 patients prior to imatinib therapy and analysed for c-kit, PDGF-Ralpha and -Rbeta expression by immunohistochemistry. In four cases, cell lines established from these tumour specimens were tested for the antiproliferative effects of imatinib and for functional mutations of genes encoding the imatinib target molecules. The tumour specimens stained positive for CD117/c-kit in nine out of 12 cases (75%), for PDGF-Ralpha in seven out of 12 cases (58%) and for PDGF-Rbeta in eight out of 12 cases (67%). The melanoma cell lines showed a heterogenous expression of the imatinib target molecules without functional mutations in the corresponding amino-acid sequences. In vitro imatinib treatment of the cell lines showed no antiproliferative effect. In conclusion, this study did not reveal an efficacy of imatinib in advanced metastatic melanoma, regardless of the expression pattern of the imatinib target molecules c-kit and PDGF-R.

Compensatory PI3-kinase/Akt/mTor activation regulates imatinib resistance development

BCR/ABL-kinase mutations frequently mediate clinical resistance to the selective tyrosine kinase inhibitor Imatinib mesylate (IM, Gleevec). However, mechanisms that promote survival of BCR/ABL-positive cells before clinically overt IM resistance occurs have poorly been defined so far. Here, we demonstrate that IM-treatment activated the phosphatidylinositol 3-kinase (PI3K)/Akt/mammalian target of rapamycin (mTor)-pathway in BCR/ABL-positive LAMA-cells and primary leukemia cells in vitro, as well as in a chronic phase CML patient in vivo. In fact, PI3K/Akt-activation critically mediated survival during the early phase of IM resistance development before manifestation of BCR/ABL-dependent strong IM resistance such as through a kinase mutation. Accordingly, inhibition of IM-induced Akt activation using mTor inhibitors and Akt-specific siRNA effectively antagonized development of incipient IM-resistance in vitro. In contrast, IM-resistant chronic myeloid leukemia (CML) patients with BCR/ABL kinase mutations (n=15), and IM-refractory BCR/ABL-positive acute lymphatic leukemia patients (n=2) displayed inconsistent and kinase mutation-independent autonomous patterns of Akt-pathway activation, and mTor-inhibition overcame IM resistance only if Akt was strongly activated. Together, an IM-induced compensatory Akt/mTor activation may represent a novel mechanism for the persistence of BCR/ABL-positive cells in IM-treated patients. Treatment with mTor inhibitors may thus be particularly effective in IM-sensitive patients, whereas Akt-pathway activation variably contributes to clinically overt IM resistance.

Early reduction of BCR-ABL mRNA transcript levels predicts cytogenetic response in chronic phase CML patients treated with imatinib after failure of interferon alpha

The degree of tumor load reduction as measured by cytogenetic response is an important prognostic factor for chronic myelogenous leukemia (CML) patients on therapy. We sought to determine whether BCR-ABL transcript levels can predict chromosomal response. Residual disease was evaluated in 120 CML patients in chronic phase (CP) treated with the selective tyrosine kinase inhibitor imatinib after resistance or intolerance to interferon alpha (IFN). Median time of therapy was 401 days (range 111-704). BCR-ABL and total ABL transcripts were measured in 486 peripheral blood (PB) specimens with a real time RT-PCR approach using fluorescent-labeled hybridization probes (LightCycler technology) and results were expressed as the ratio BCR-ABL/ABL. Cytogenetic response was determined in 3-monthly intervals: From 101 evaluable patients, 42 achieved a complete (CR, 0% Philadelphia chromosome (Ph)- positive metaphases), 18 a partial (PR, 1-34% Ph+), 13 a minor (MR, 35-94% Ph+), and 26 no response (NR, >94% Ph+). All PB samples were RT-PCR positive. The proportion of Ph+ metaphases and simultaneous BCR-ABL/ABL ratios correlated with r = 0.74, P < 0.0001. In order to investigate whether early molecular analysis may predict cytogenetic response, quantitative RT-PCR data obtained after 1 and 2 months of therapy were compared with cytogenetic response at 6 months. BCR-ABL/ABL ratios after 1 month were not predictive, but results after 2 months correlated with the consecutive cytogenetic response (P = 0.0008). The probability for a major cytogenetic response was significantly higher in patients with a BCR-ABL/ABL ratio <20% after 2 months of imatinib therapy. We conclude that: (1) quantitative determination of residual disease with real time RT-PCR is a reliable and sensitive method to monitor CML patients on imatinib therapy; (2) BCR-ABL/ABL ratios correlate well with cytogenetic response; (3) in IFN-pretreated patients all complete responders to imatinib have evidence of residual disease with the limited follow-up available; and (4) cytogenetic response at 6 months of therapy in CP patients is predictable with real time RT-PCR at 2 months.

Epidemiological, genetic, and clinical characterization by age of newly diagnosed acute myeloid leukemia based on an academic population-based registry study (AMLSG BiO)

We describe genetic and clinical characteristics of acute myeloid leukemia (AML) patients according to age from an academic population-based registry. Adult patients with newly diagnosed AML at 63 centers in Germany and Austria were followed within the AMLSG BiO registry (NCT01252485). Between January 1, 2012, and December 31, 2014, data of 3525 patients with AML (45% women) were collected. The median age was 65 years (range 18-94). The comparison of age-specific AML incidence rates with epidemiological cancer registries revealed excellent coverage in patients < 70 years old and good coverage up to the age of 80. The distribution according to the European LeukemiaNet (ELN) risk categorization from 2010 was 20% favorable, 31% intermediate-1, 28% intermediate-2, and 21% adverse. With increasing age, the relative but not the absolute prevalence of patients with ELN favorable and intermediate-1 risk (p < 0.001), with activating FLT3 mutations (p < 0.001), with ECOG performance status < 2 (p < 0.001), and with HCT-CI comorbidity index < 3 (p < 0.001) decreased. Regarding treatment, obesity and favorable risk were associated with an intensive treatment, whereas adverse risk, higher age, and comorbidity index > 0 were associated with non-intensive treatment or best supportive care. The AMLSG BiO registry provides reliable population-based distributions of genetic, clinical, and treatment characteristics according to age.

Dynamics of BCR-ABL mRNA expression in first-line therapy of chronic myelogenous leukemia patients with imatinib or interferon α/ara-C

We sought to determine dynamics of BCR-ABL mRNA expression levels in 139 patients with chronic myelogenous leukemia (CML) in early chronic phase, randomized to receive imatinib (n=69) or interferon (IFN)/Ara-C (n=70). The response was sequentially monitored by cytogenetics from bone marrow metaphases (n=803) and qualitative and quantitative RT-PCR from peripheral blood samples (n=1117). Complete cytogenetic response (CCR) was achieved in 60 (imatinib, 87%) vs 10 patients (IFN/Ara-C, 14%) after a median observation time of 24 months. Within the first year after CCR, best median ratio BCR-ABL/ABL was 0.087%, (imatinib, n=48) vs 0.27% (IFN/Ara-C, n=9, P=0.025). BCR-ABL was undetectable in 25 cases by real-time PCR, but in only four patients by nested PCR. Median best response in patients with relapse after CCR was 0.24% (n=3) as compared to 0.029% in patients with continuous remission (n=52, P=0.029). We conclude that (i) treatment with imatinib in newly diagnosed CML patients is associated with a rapid decrease of BCR-ABL transcript levels; (ii) nested PCR may reveal residual BCR-ABL transcripts in samples that are negative by real-time PCR; (iii) BCR-ABL transcript levels parallel cytogenetic response, and (iv) imatinib is superior to IFN/Ara-C in terms of the speed and degree of molecular responses, but residual disease is rarely eliminated.

Molecular monitoring of response to imatinib (Glivec) in CML patients pretreated with interferon alpha. Low levels of residual disease are associated with continuous remission

A significant proportion of chronic myeloid leukemia (CML) patients achieve a major cytogenetic remission (MCR) to imatinib therapy after failing interferon (IFN) alpha-based protocols. We sought to determine levels of residual disease in patients with MCR using various molecular methods and to establish a relation between residual BCR-ABL transcript levels and rate of relapse in complete cytogenetic remission (CCR). Response was measured by conventional cytogenetic analysis, hypermetaphase and interphase fluorescence in situ hybridization (HM-FISH, IP-FISH) of bone marrow (BM) cells, qualitative nested and quantitative reverse transcriptase polymerase chain reaction (RT-PCR) for BCR-ABL transcripts. We investigated 323 peripheral blood (PB) and BM samples from 48 CML patients who achieved a complete (Ph+ 0%; n=41) or partial (Ph+ 1-34%; n=7) cytogenetic remission after 3-20 months of imatinib therapy. Prior to imatinib, 35 patients were in chronic phase (CP), eight in accelerated phase (AP), four in myeloid and one in lymphoid blast crisis. HM-FISH results correlated with ratios BCR-ABL/ABL in PB and BM. In patients with CCR, residual disease was detectable by HM-FISH (31%), IP-FISH (18%), and RT-PCR (100%). During follow-up, BCR-ABL became undetectable in two patients (one CP, one AP) by both nested and quantitative RT-PCR. CCR is ongoing in 30 evaluable patients, 11 patients have relapsed. At the time of best response, median ratios BCR-ABL/ABL were 2.1% (range 0.82-7.8) in patients with subsequent relapse and 0.075% (range 0-3.9) in patients with ongoing remission (P=0.0011). All 16 CP patients, who achieved ratios BCR-ABL/ABL <0.1% as best molecular response are in continuous remission, while 6/13 patients (46%) with ratios >/=0.1% have relapsed (P=0.0036). We conclude that: (i) in patients with CCR to imatinib, HM-FISH and RT-PCR usually reveal residual BCR-ABL+ cells; (ii) RT-PCR results derived from PB and BM are comparable in CP CML; and (iii) low levels of residual disease with ratios BCR-ABL/ABL &<0.1% are associated with continuous remission.

Acute myeloid leukemia derived from lympho-myeloid clonal hematopoiesis

We studied acute myeloid leukemia (AML) patients with lympho-myeloid clonal hematopoiesis (LM-CH), defined by the presence of DNA methyltransferase 3A (DNMT3A) mutations in both the myeloid and lymphoid T-cell compartment. Diagnostic, complete remission (CR) and relapse samples were sequenced for 34 leukemia-related genes in 171 DNMT3A mutated adult AML patients. AML with LM-CH was found in 40 patients (23%) and was associated with clonal hematopoiesis of indeterminate potential years before AML, older age, secondary AML and more frequent MDS-type co-mutations (TET2, RUNX1 and EZH2). In 82% of AML patients with LM-CH, the preleukemic clone was refractory to chemotherapy and was the founding clone for relapse. Both LM-CH and non-LM-CH MRD-positive AML patients who achieved CR had a high risk of relapse after 10 years (75% and 75%, respectively) compared with patients without clonal hematopoiesis in CR with negative MRD (27% relapse rate). Long-term survival of patients with LM-CH was only seen after allogeneic hematopoietic stem cell transplantation (HSCT). We define AML patients with LM-CH as a distinct high-risk group of AML patients that can be identified at diagnosis through mutation analysis in T cells and should be considered for HSCT.

DNMT3A mutant transcript levels persist in remission and do not predict outcome in patients with acute myeloid leukemia

We investigated the prognostic impact of minimal residual disease (MRD) monitoring in acute myeloid leukemia patients harboring DNA methyltransferase 3A-R882H/-R882C mutations (DNMT3A^mut). MRD was determined by real-time quantitative PCR (RQ-PCR) in 1494 samples of 181 DNMT3A^mut patients. At the time of diagnosis, DNMT3A^mut transcript levels did not correlate with presenting clinical characteristics and concurrent gene mutations as well as the survival end points. In Cox regression analyses, bone marrow (BM) DNMT3A^mut transcript levels (log10-transformed continuous variable) were not associated with the rate of relapse or death. DNMT3A^mut transcript levels were significantly higher in BM than in blood after induction I (P=0.01), induction II (P=0.05), consolidation I (P=0.004) and consolidation II (P=0.008). With regard to the clinically relevant MRD time points, after two cycles of induction and at the end of therapy, DNMT3A^mut transcript levels had no impact on the end point remission duration and overall survival. Of note, only a minority of the patients achieved RQ-PCR negativity, whereas most had constantly high DNMT3A^mut transcript levels, a finding which is consistent with the persistence of clonal hematopoiesis in hematological remission.

Deregulated apoptosis signaling in core-binding factor leukemia differentiates clinically relevant, molecular marker-independent subgroups

Core-binding factor (CBF) leukemias, characterized by translocations t(8;21) or inv(16)/t(16;16) targeting the CBF, constitute acute myeloid leukemia (AML) subgroups with favorable prognosis. However, about 40% of patients relapse and the current classification system does not fully reflect this clinical heterogeneity. Previously, gene expression profiling (GEP) revealed two distinct CBF leukemia subgroups displaying significant outcome differences and identified apoptotic signaling, MAPKinase signaling and chemotherapy-resistance mechanisms among the most significant differentially regulated pathways. We now tested different inhibitors of the respective pathways in a cell line model (six cell lines reflecting the CBF subgroup-specific gene expression alterations), and found apoptotic signaling to be differentiating between the CBF subgroup models. In accordance, primary samples from newly diagnosed CBF AML patients (n=23) also showed differential sensitivity to in vitro treatment with a Smac mimetic such as BV6, an antagonist of inhibitor of apoptosis (IAP) proteins, and ABT-737, a BCL2 inhibitor. Furthermore, GEP revealed the BV6-resistant cases to resemble the previously identified unfavorable CBF subgroup. Thus, our current findings show deregulated IAP expression and apoptotic signaling to differentiate clinically relevant CBF subgroups, which were independent of known molecular markers, thereby providing a starting point for novel therapeutic approaches.

Frequent polymorphism in BCR exon b2 identified in BCR-ABL positive and negative individuals using fluorescent hybridization probes

Recently, a polymorphic base in exon 13 of the BCR gene (exon b2 of the major breakpoint cluster region) has been identified in the eighth position before the junctional region of BCR-ABL cDNA. Cytosine replaces thymidine; the corresponding triplets are AAT (T allele) and AAC (C allele), respectively, both coding for asparagine. Therefore, this polymorphism has no implication in the primary structure of BCR and BCR-ABL proteins. However, since the alteration is located close to the fusion region it may have a significant influence on the annealing of PCR primers, probes for real time PCR, and antisense oligonucleotides. We have developed a RT-PCR-based screening method to easily identify polymorphic BCR and BCR-ABL alleles in CML patients and normal individuals in order to estimate their frequency. After amplification from cDNA, a melting curve of a specific fluorogenic probe mapping to the 3' end of BCR exon b2 and spanning the polymorphism readily discriminates between normal and polymorphic BCR and BCR-ABL alleles. This reporter probe is 3' labeled with fluorescein and placed next to 5' LC Red640-labeled anchor probes mapping to the 5' ends of BCR exon b3 or ABL exon a2 so that resonance energy transfer occurs when the probes are hybridized (LightCycler technology). T and C alleles were discriminated by a melting temperature difference of the reporter probe of 3.2 K. We have investigated cDNAs derived from leukocytes from seven cell lines and a total of 229 individuals: normal donors, n = 15; BCR-ABL negative chronic myeloproliferative disorders, n=30; BCR-ABL negative acute leukemias, n= 11; b2a2BCR-ABL positive CML, n = 93; and b3a2BCR-ABL positive CML, n= 80. The frequency of the C allele was 33.0% in BCR-ABL negative individuals, 30.6% in b2a2BCR-ABL, and 23.8% in b3a2BCR-ABL positive CML. In CML patients, 27.7% of BCR-ABL and 27.2% of BCR alleles had the C allele (NS). In total, 132 of 458 (28.8%) exons b2 of BCR or BCR-ABL alleles demonstrated this polymorphism. We conclude that a thymidine/cytosine replacement occurs frequently in BCR exon b2. Probes for real time quantitative RT-PCR should be designed not to map to the critical region in order to avoid underestimation of the number of BCR-ABL transcripts.

Imatinib restores expression of CD62L in BCR-ABL-positive cells

Chronic myelogenous leukemia (CML) is characterized by aberrant trafficking of malignant hematopoietic progenitor cells in the peripheral blood. Expression of the cell adhesion molecule CD62L was reported to be significantly lower in CML patients than in normal controls. We studied whether the transcription of CD62L in CML cells is dependent on the activity of the BCR-ABL tyrosine kinase. Following addition of the Abelson (ABL) tyrosine kinase inhibitor imatinib (formerly STI571) to two BCR-ABL-positive cell lines (BV173, SD-1), we observed a dose-dependent increase in CD62L RNA levels of up to 45-fold by a quantitative, real-time polymerase chain reaction and an increase in the amount of cell surface-bound CD62L of up to 18-fold by quantitative flow cytometry, respectively. These data are validated by an increased CD62L expression in the bone marrow of patients (n=6) with advanced CML who received imatinib. Restoration of defective cell adhesion mediated via the CD62L pathway may be one mechanism of action of imatinib in BCR-ABL-positive leukemias.

Randomized phase-II trial evaluating induction therapy with idarubicin and etoposide plus sequential or concurrent azacitidine and maintenance therapy with azacitidine

The aim of this randomized phase-II study was to evaluate the effect of substituting cytarabine by azacitidine in intensive induction therapy of patients with acute myeloid leukemia (AML). Patients were randomized to four induction schedules for two cycles: STANDARD (idarubicin, cytarabine, etoposide); and azacitidine given prior (PRIOR), concurrently (CONCURRENT), or after (AFTER) therapy with idarubicin and etoposide. Consolidation therapy consisted of allogeneic hematopoietic-cell transplantation or three courses of high-dose cytarabine followed by 2-year maintenance therapy with azacitidine in the azacitidine-arms. AML with CBFB-MYH11, RUNX1-RUNX1T1, mutated NPM1, and FLT3-ITD were excluded and accrued to genotype-specific trials. The primary end point was response to induction therapy. The statistical design was based on an optimal two-stage design applied for each arm separately. During the first stage, 104 patients (median age 62.6, range 18-82 years) were randomized; the study arms PRIOR and CONCURRENT were terminated early due to inefficacy. After randomization of 268 patients, all azacitidine-containing arms showed inferior response rates compared to STANDARD. Event-free and overall survival were significantly inferior in the azacitidine-containing arms compared to the standard arm (p < 0.001 and p = 0.03, respectively). The data from this trial do not support the substitution of cytarabine by azacitidine in intensive induction therapy.

MicroRNA 181a (miR-181a) expression as a prognosticator in cytogenetically normal acute myeloid leukemia (CN AML)

7001

Background: We showed recently that CEBPA mutations (mut) in CN AML are associated with better outcome and a unique microRNA expression profile, including miR-181a upregulation. Here we tested if miR-181a expression can predict outcome independently. Methods: We analyzed 187 de novo CN AML adult patients (pts) aged <60 years (y; median 45) similarly treated on CALGB 9621 and 19808. Of these, 122 had molecular high risk [FLT3-ITD or NPM1 wild type (wt)] and 65 low risk (no FLT3-ITD, NPM1 mut) CN AML. FLT3, NPM1, CEBPA, MLL, and WT1 mutations, and ERG and BAALC expression were analyzed centrally. miR-181a expression was measured in pretherapy marrow using OSUCCC v3.0 arrays. The mean of 2 miR-181a probe log intensities was used as a continuous variable for analyses. Results: Higher miR-181a levels (miR-181a↑) were associated with CEBPA mut, NPM1 wt, no FLT3-TKD, lower ERG expression, higher %FAB M1/M2, lower WBC and age, higher blood blasts, and lower % gum hypertrophy. miR-181a↑ tended to associate with more complete remissions (CRs; p = .07) and significantly associated with longer disease-free (DFS; p = .05) and overall (OS; p = .01) survival (median follow-up 6.5 y for pts alive). A stronger prognostic impact of miR-181a was observed in the molecular high risk group, where miR-181a↑ predicted more CRs (p = .03), and longer DFS (p = .0002) and OS (p = .0001). In multivariable analyses of the molecular high risk group, miR-181a↑independently predicted CR, and longer DFS and OS (Table). For descriptive purposes, we dichotomized pts at the median miR-181a expression value. For high v low miR-181a expressers, CR rates were 84% vs 72% and 5 y DFS and OS rates 43% vs 18% and 48% vs 19%, respectively. Conclusions: miR-181a expression is a prognostic marker in CN AML, mainly in the molecular high risk group, where it predicts outcome independently of other variables including CEBPA mutations. As miR-181a↑ confer better treatment response, novel approaches increasing miR-181a levels might benefit not only CN but also other AML pts.

[Table: see text]

No significant financial relationships to disclose.

Outcome prediction in adult core binding factor (CBF) acute myeloid leukemia (AML) with gene expression profiling: A Cancer and Leukemia Group B (CALGB) study

7011

Background: In CBF AML with t(8;21)(q22;q22) or inv(16)(p13q22)/t(16;16)(p13;q22) [abbreviated inv(16)], KIT mutations (mutKIT) and, in inv(16), trisomy 22 predict outcome and may guide the development of novel risk-adapted therapies. However, prognosis of patients (pts) lacking the aforementioned markers is less clear. Therefore, we profiled gene expression in t(8;21) (n=22) or sole inv(16) (n=25) pts who lacked mutKIT to identify signatures predictive of outcome. All pts were treated on CALGB trials incorporating consolidation therapy with multiple courses of higher dose cytarabine. Methods: Gene expression profiling was performed using Affymetrix U133 plus 2.0 arrays on diagnostic samples. As differences in gene expression distinguished all t(8;21) pts from all inv(16) pts, indicating two different biological entities, we pursued outcome prediction separately for each cytogenetic group. Gene expression-based outcome predictors for event-free survival (EFS) were constructed using a cross-validated prediction algorithm. Results: Among t(8;21) pts, EFS for predicted good (n=13) and poor (n=9) outcome groups differed strikingly (P=0.005; estimated 3-year EFS rates: 69% v 11%). Prediction was correct for 77% of pts. Among sole inv(16) pts, EFS for predicted good (n=18) and poor (n=7) outcome groups also differed (P=0.08; 3-year EFS rates: 78% v 29%). Prediction was correct for 76% of pts. FLT3 mutations appeared not to account for differences in EFS between the predicted groups; only the predicted outcome groups were associated with EFS (all baseline clinical characteristics at P>0.10). Pts with predicted poor outcome had higher expression of genes with leukemogenic potential such as WT1 [t(8;21) and inv(16)], CCNA1 [t(8;21)] and the oncogene MYCN [inv(16)]. Conclusions: Gene expression profiling improves outcome prediction in CBF AML pts lacking the known prognostic markers. Future studies should explore the clinical usefulness of targeting products of over- expressed genes, such as WT1 encoding a potential target for immunotherapy in AML.

No significant financial relationships to disclose.

High BAALC expression associates with other molecular prognostic markers, poor outcome and a distinct gene expression signature in cytogenetically normal acute myeloid leukemia (CN AML): A Cancer and Leukemia Group B (CALGB) study

7013

Background: High BAALC expression predicts poor outcome in CN AML patients (pts). Yet, little is known about BAALC's function or its relation to other prognostic markers. We evaluated BAALC expression in the context of molecular markers and clinical outcome in 172 de novo CN AML adults, aged <60 years (y) treated on similar CALGB protocols (9621 and 19808). Methods: BAALC expression was measured by quantitative real-time RT-PCR in pretreatment blood samples. Pts were grouped as high (n=86) or low BAALC (n=86) expressers using the median expression value, by protocol, as a cutoff. Gene expression profiling (Affymetrix U133 plus 2.0 GeneChip) was performed on high (n=26) and low (n=24) pts. Results: BAALC expression was associated with several molecular markers ( Table ). Complete remission (CR) rate was lower in high BAALC pts (79% v 90%), for whom achieving CR was almost 4 times less likely than for low BAALC pts (P=.02; odds ratio=0.27), after adjusting for age (P=.004), ERG expression (P=.05) and white blood cell count (WBC; P=.04). With a median follow-up of 4.3 y, high BAALC pts had shorter overall survival (OS) (P=.002; 3-y OS: 42% v 59%). In a multivariable model, high BAALC provided further adverse prognostic information (P=.06; hazard ratio=1.66) independent of FLT3 ITD (P<.001), WBC (P=.007) and NPM1 (P=.02). A gene expression signature was identified consisting of 312 probe sets differentially expressed (P<.001) between BAALC groups. High BAALC was associated with overexpression of, among other genes, PROM1, CD34 and KIT indicating a less differentiated phenotype in these pts. Conclusions: Although associated with other prognostic markers, high BAALC expression independently predicts outcome and is associated with a distinct gene expression profile, potentially identifying new therapeutic targets in this pt subset.

No significant financial relationships to disclose.

[Table: see text]

FLT3 D835/I836 mutations predict worse disease-free survival (DFS) in younger adults with cytogenetically normal acute myeloid leukemia (CN AML) without FLT3 internal tandem duplications (ITD): A Cancer and Leukemia Group B (CALGB) study

10532

Background: The adverse prognostic impact of FLT3 ITD is well-recognized in younger adults with CN AML, whereas the clinical relevance of D835/I836 mutations in the second tyrosine kinase domain of FLT3 (FLT3 TKD), that occur in 5%-14% of patients (pts), remains controversial. We evaluated the prognostic impact of FLT3 TKD, alone and with other prognostic markers, in 139 de novo CN AML adults, aged <60 years, without FLT3 ITD, and treated on similar CALGB protocols (9621 and 19808). Methods: FLT3 TKD was detected by RT-PCR/EcoRV digestion. Affymetrix U133 plus 2.0 gene expression profiles derived from FLT3 TKD+ and FLT3 wild-type (FLT3 WT) pts with available RNA were compared. Results: Relative to FLT3 WT pts, FLT3 TKD+ pts [n=16 (11.5%)] presented with higher white blood cell counts (WBC) (median: 57.2 v 14.6 × 109/L; P<.001), % bone marrow (BM) blasts (median: 75.5 v 63; P=.001), and % blood blasts (median: 62.5 v 46; P=.03). All 16 FLT3 TKD+ pts achieved complete remission versus 85% of FLT3 WT pts (P=.13). With a median follow-up of 4.6 years, FLT3 TKD+ pts had worse DFS (P=.01; 3-year DFS rates: 31% v 60%) and a trend for worse overall survival (OS) (P=.17, 3-year OS rates: 40% v 65%) compared to FLT3 WT pts. Other variables moderately associated with a worse DFS (P<.10) included higher % BM blasts and NPM1 WT. Adjusting for these two variables, FLT3 TKD remained significantly associated with adverse DFS (P=.02; hazard ratio=2.3). A gene expression signature consisting of 333 probe sets (P<.001) differentiated FLT3 TKD+ (n=9) from FLT3 WT (n=50) pts. Genes represented in this signature include those involved in cell motility (e.g., C3AR1, VNN1, VNN2, ADAM8; all over-expressed in FLT3 TKD+), development (e.g., HOXB5, HOXB6; both over-expressed in FLT3 TKD+) and stress response (e.g., MPO; under-expressed in FLT3 TKD+). Conclusions: FLT3 TKD appears to be an independent predictor for DFS among younger CN AML adult pts without FLT3 ITD. Gene expression profiling provides additional insight into the biology of FLT3 TKD+ AML and may lead to identification of additional targets for improved therapy.

No significant financial relationships to disclose.

Mutations of KIT tyrosine kinase (TK) gene predict relapse in adult patients (pts) with core binding factor acute myeloid leukemia (CBF AML): A Cancer and Leukemia Group B (CALGB) study

2

Background: Multi-course high-dose cytarabine (HDAC) has largely improved the outcome of CBF AML pts with t(8;21)(q22;q22) and inv(16)(p13q22). Yet, ∼50% of pts relapse within 5 years (yrs), indicating the need for markers identifying high-risk pts who require more aggressive and/or novel therapies. KIT mutations (mKIT) of exons 17 (mKIT17) and 8 (mKIT8) are good candidates to be such markers. Previous studies showed that mKIT17 impacted adversely on outcome of t(8;21), but not inv(16), pts. Methods: Sixty-one pts with inv(16) and 49 with t(8;21), assigned to HDAC consolidation on CALGB protocols,were analyzed at diagnosis for mKIT17 and mKIT8 by denaturing high-performance liquid chromatography and direct sequencing. The median follow-up was 5.3 yrs. Results: Among pts with inv(16), 10 had mKIT17 (3 with concurrent mKIT8) and 8 sole mKIT8; among pts with t(8;21), 9 had mKIT17 (2 with mKIT8) and 2 sole mKIT8. In each cytogenetic group, complete remission rates were similar for pts with wild-type KIT (wtKIT) and those with mKIT or mKIT17. In both groups, cumulative incidence of relapse (CIR) was higher for pts with any mKIT [inv(16), P=.050, 5-yr CIR 56% vs 29%; t(8;21), P=.017, 5-yr CIR 70% vs 36%], and also for the subsets of pts with mKIT17 [inv(16), P=.002, 5-yr CIR 80% vs 29%; t(8;21), P=.014, 5-yr CIR 75% vs 36%] compared to wtKIT. Neither mKIT nor mKIT17 significantly affected overall survival (OS). In multivariable analyses (see table ), mKIT [in t(8;21)] and mKIT17 [in inv(16) and t(8;21)] independently predicted worse CIR; worse OS correlated significantly with mKIT only in inv(16) pts. Conclusions: We show for the first time that mKIT17 confer significantly higher relapse rates in inv(16) pts and confirm their adverse impact on t(8;21) pts. Thus, this subset of pts may benefit from more aggressive treatment and/or, since mKIT17 lead to constitutive KIT activation, from therapy that includes TK inhibitors, such as imatinib or PKC412.

[Table: see text]

No significant financial relationships to disclose.

Improving the molecular risk classification for younger (<60 years) de novo cytogenetically normal acute myeloid leukemia (CN AML) patients (pts)

7002

Background: CN AML pts are currently stratified into Low-risk [FLT3-ITD negative (neg)/NPM1 mutated (mut)] and High-risk [FLT3-ITD positive (pos) or NPM1 wild type (wt)] groups (FLT3-ITD/NPM1-only classification). We recently showed that low ERG expression and CEBPAmut identify pts with better outcome within, respectively, the Low- and High-risk groups, and that WT1mut confers poor outcome regardless of FLT3-ITD/NPM1 status. Here, we assess if adding CEBPA and WT1 mutation and ERG expression testing improves the currently used CN AML molecular risk classification. Methods: FLT3, NPM1, CEBPA and WT1 mutations and ERG and BAALC expression were tested at diagnosis in 143 CN AML adults enrolled on CALGB treatment protocols 9621 and 19808. Pts were divided into two molecular risk groups: i) CALGB Group I that included Low-risk pts with low ERG & High-risk pts with CEBPAmut and ii) CALGB Group II that included WT1mut pts & Low-risk pts with high ERG & High- risk pts with CEBPAwt. Results: CALGB Group I (n=56) v Group II (n=87) had more complete remissions (CRs) (P=.005; 96% v 79%), and longer disease-free (DFS; P<.0001; 5 year (y) 69% v 21%) and overall (OS; P<.0001; 5 y 70% v 31%) survival [median follow-up for pts alive 6 y]. In multivariable (MV) analyses, Group I predicted higher rate of CR (P=.02), and longer DFS (P<.0001) and OS (P=.0002), after correcting for other variables (Table). In contrast, for the same cohort of pts grouped by the FLT3-ITD/NPM1-only classification, CRs were 94% v 82% and 5 y DFS 59% v 32% and OS 67% v 36% in the Low- v High-risk groups. Based on the Akaike Information Criterion, MV models for DFS and OS using the CALGB risk Groups were better than those using the FLT3-ITD/NPM1-only risk groups. Conclusions: Prognostic classification of younger de novo CN AML pts is improved by adding CEBPA and WT1 mutation and ERG expression testing. While mutational analyses are ready for use in clinical trials, quantification of ERG expression is yet to be standardized.

[Table: see text]

No significant financial relationships to disclose.

A phase I study of GTI-2040 (G), an antisense to ribonucleotide reductase (RNR), in combination with high-dose AraC (HiDAC) in acute myeloid leukemia (AML)

6561

Background: RNR converts ribonucleotides to deoxyribonucleotides for DNA synthesis. AraC is converted into AraC triphosphate (AraCTP) and competes with deoxycytidine for DNA incorporation. We hypothesized that RNR downregulation by G leads to lower deoxycytidine levels, preferential AraCTP incorporation into DNA and increased cytotoxicity. A CTEP-sponsored Phase I dose escalation study of G +HiDAC in relapsed/refractory AML tested this hypothesis. Methods: Cohort I (18–59 yrs) received G (dose level (DL) 1: 3.5 mg/m2/d) by continuous IV infusion (CIVI) on d 1–6 + AraC IV q12 hrs on d 2, 4, 6 (DL1: 2500 mg/m2/dose). Cohort II (≥60 yrs) received G CIVI on d 1–6 +AraC IV on d 2 −6 (DL1: 1500 mg/m2/d). An ELISA-based assay measured plasma and intracellular concentration (IC) of G. Results: To date, cohort I included 9 pts with relapsed and 9 with refractory AML; 9 had intermediate and 9 adverse risk cytogenetics (CyG); 8 received prior HiDAC. Cohort II included 10 pts with relapsed and 6 with refractory AML; 8 pts had intermediate and 8 high risk CyG; 5 pts received prior HiDAC. Toxicities were comparable to HiDAC alone. The younger pts had higher AUC and longer t1/2. Of 16 pts evaluable in cohort I (median time to 1st relapse 6 mos), 6 had complete remission (CR) and 1 incomplete CR (no disease and incomplete blood count recovery). In cohort II, no responses were observed. At 120 hrs of antisense infusion, median G IC in marrow cells was higher (i.e., 175 vs75 nM) in younger than in older pts. A median 50% decrease in RNR protein was noted in 5/9 and 5/10 pts in cohort I and II, respectively. In cohort I, a median 50% decrease and 200% increase in RNR was noted in CR (n=4) and non-responder (NR; n=9) pts, respectively. In cohort II R2 downregulation did not predict response. In cohort I 62% of the ICs was in nucleus and 21.2% in cytoplasm in CR pts (n=3) vs. 20.3% and 53.5% in NR pts(n=5). Conclusions: G/HiDAC is feasible. Robust plasma and IC levels of G and target downregulation are achievable in vivo. Responses (41%) were observed only in the younger cohort, where CR pts had higher G nuclear IC and target downregulation than NR pts. Dose escalation continues in the younger cohort to establish a dose for Phase II trials.

No significant financial relationships to disclose.