Letters to the Editor
(Table 1). The levels of total PKR were similar in all of the MDS samples analyzed when compared with BMMC from healthy donors (Figures 1a and 2a). In contrast, while BMMC from healthy donors contained very low levels of p-PKR, an increase in the amount of p-PKR was detected in low-risk MDS patients (Figure 1b and Table 1). Interestingly, high-risk MDS cells showed not only an increased amount, but also a different subcellular localization, of p-PKR (Figure 2b and Table 1). In fact, in these patients, about 50% of BMMC displayed a nuclear localization of p-PKR, whereas in low risk MDS patients and in healthy donors the staining was exclusively detected in the cytoplasm (Figure 1 and Table 1). Our findings suggest that PKR could be somehow involved in the increased apoptosis seen in low risk MDS patients. Indeed, BMMC from low-risk MDS patients show increased levels of p-eIF2a when compared with healthy donors (data not shown). As to nuclear localization of p-PKR in BMMC from high-risk MDS patients, who are characterized by a lower apoptotic rate, it has been hypothesized that the nuclear translocation may be a mechanism to sequester active kinases, thus preventing upregulation of cytosolic signaling pathways. This is the case for phosphoinositide-dependent protein kinase-1, whose nuclear localization has been suggested to inhibit Akt-mediating signalling.8 Therefore, it might be that active (phosphorylated) PKR is sequestered in the nucleus of BMMC from high risk MDS patients, who are characterized by a lower apoptotic rate. However, we could not rule out that nuclear localization of p-PKR in high-risk MDS patients is somehow linked to disease progression to AML. Further investigations are therefore needed to fully understand the functional role of PKR in both low-risk and high-risk MDS patients, nevertheless our findings suggest that also this protein kinase could be involved in the pathogenesis of MDS.
Acknowledgements This work was supported by grants from: CARISBO Foundation, Progetti Strategici Universita` di Bologna EF2006.
MY Follo1, C Finelli2, S Mongiorgi1, C Clissa2, C Bosi2,3, G Martinelli2, WL Blalock1, L Cocco1 and AM Martelli1,4
1
Dipartimento di Scienze Anatomiche Umane e Fisiopatologia dell’Apparato Locomotore, Sezione di Anatomia, Cell Signaling Laboratory, Universita` di Bologna, Bologna, Italy; 2 Dipartimento di Ematologia e Scienze Oncologiche ‘L e A Sera`gnoli’, Universita` di Bologna, Bologna, Italy; 3 Hematology Unit, Ospedale Civile di Piacenza, Piacenza, Italy and 4 Istituto di Genetica Molecolare del CNR, c/o IOR, Bologna, Italy E-mail:
[email protected]
2269
References 1 Corey SJ, Minden MD, Barber DL, Kantarjian H, Wang JC, Schimmer AD. Myelodysplastic syndromes: the complexity of stem-cell diseases. Nature Rev Cancer 2007; 7: 118–129. 2 Martelli AM, Nyakern M, Tabellini G, Bortul R, Tazzari PL, Evangelisti C et al. Phosphoinositide 3-kinase/Akt signaling pathway and its therapeutical implications for human acute myeloid leukemia. Leukemia 2006; 20: 911–928. 3 Doepfner KT, Spertini O, Arcaro A. Autocrine insulin-like growth factor-I signaling promotes growth and survival of human acute myeloid leukemia cells via the phosphoinositide 3-kinase/Akt pathway. Leukemia 2007; 21: 1921–1930. 4 Nyakern M, Tazzari PL, Finelli C, Bosi C, Follo MY, Grafone T et al. Frequent elevation of Akt kinase phosphorylation in blood marrow and peripheral blood mononuclear cells from high-risk myelodysplastic syndrome patients. Leukemia 2006; 20: 230–238. 5 Follo MY, Mongiorgi S, Bosi C, Cappellini A, Finelli C, Chiarini F et al. The Akt/mammalian target of rapamycin signal transduction pathway is activated in high-risk myelodysplastic syndromes and influences cell survival and proliferation. Cancer Res 2007; 67: 4287–4294. 6 Follo MY, Finelli C, Bosi C, Martinelli G, Mongiorgi S, Baccarani M et al. PI-PLCbeta-1 and activated Akt levels are linked to azacitidine responsiveness in high-risk myelodysplastic syndromes. Leukemia 2008; 22: 198–200. 7 Garcia MA, Gil J, Ventoso I, Guerra S, Domingo E, Rivas C et al. Impact of protein kinase PKR in cell biology: from antiviral to antiproliferative action. Microbiol Mol Biol Rev 2006; 70: 1032–1060. 8 Lim MA, Kikani CK, Wick MJ, Dong LQ. Nuclear translocation of 30 phosphoinositide-dependent protein kinase 1 (PDK-1): a potential regulatory mechanism for PDK-1 function. Proc Natl Acad Sci USA 2003; 100: 14006–14011.
Supplementary Information accompanies the paper on the Leukemia website (http://www.nature.com/leu)
An activating KRAS mutation in imatinib-resistant chronic myeloid leukemia
Leukemia (2008) 22, 2269–2272; doi:10.1038/leu.2008.124; published online 29 May 2008
Mutations in the kinase domain of BCR-ABL that impair drug binding are the best characterized mechanism of imatinib resistance in patients with chronic myeloid leukemia (CML) or Philadelphia chromosome (Ph)-positive acute lymphoblastic leukemia.1 Some patients without kinase domain mutationbased resistance exhibit increased BCR-ABL expression.2 However, in a sizable proportion of patients, neither of these mechanisms is demonstrable.3 Causes of imatinib resistance implicated in such patients include insufficient intracellular drug levels due to increased efflux or reduced influx or sequestration
of imatinib through increased plasma protein binding.2 Further, some kinase domain mutation-negative patients appear to have truly BCR-ABL-independent imatinib resistance, for example, through the activation of SRC family kinases or adaptive granulocyte-macrophage colony-stimulating factor secretion.4,5 A myeloproliferative disease with features of CML can be induced by a variety of genetic lesions, including activating mutations of RAS, PTPN11, JAK2 and receptor tyrosine kinases such as FLT3. We therefore hypothesized that patients with acquired imatinib resistance but no evidence of BCR-ABL mutations or gene amplification may have activating mutations in such imatinib-insensitive pathways or that they may have acquired resistance mutations in the imatinib targets KIT and platelet-derived growth factor receptor a/b. Leukemia
Letters to the Editor
2270 We identified a total of 17 imatinib-resistant CML patients (nine males and eight females) with a median age of 57 years (range: 34–69 years) (Table 1) who had no evidence of a BCRABL mutation on standard diagnostic sequencing. At the time of the study, cytogenetic analysis revealed a median of 40% (range: 10–100%) Ph-positive metaphases, fluorescence in situ hybridization showed a median of 43% (range: 1.5–98.5%) BCR-ABL-positive interphases, and quantitative PCR showed a BCR-ABL/G6PD median ratio of 2.5% (range: 0.52–210%). Out of the 17 patients, 10 never achieved a complete cytogenetic response before this study and seven had lost this level of response at the time of sampling. Sequence analysis of BCR-ABL covering the Cap, SH3, SH2 and kinase domains of ABL (exons 1–9) confirmed the absence of mutations. Mutation screening of KRAS (exons 3–5), NRAS (exons 2–5), PTPN11 (exons 3, 4, 8 and 13), KIT (entire coding region), FLT3 (exons 14 and 20),
Table 1
JAK2 (exon 14), PDGFRa (exons 12, 14 and 18) and PDGFRb (exons 11 and 17) revealed several previously described polymorphisms, but no novel mutations were detected. One of 17 patients (5.8%) harbored a missense mutation (C-T) in exon 3 of KRAS leading to replacement of threonine 58 with isoleucine (T58I). Approximately 30% of amplicons were mutant, as confirmed by bidirectional sequencing of three independently amplified PCR products. The patient found to harbor the T58I KRAS mutation was a 62-year-old man with an 8-year history of chronic phase CML. He received initial therapy with interferon-a and cytarabine. In 2001, he was started on imatinib 400 mg daily and achieved a minor cytogenetic response that he subsequently lost, probably due to the appearance of a clone with a second copy of the Ph chromosome (Table 2). As a result, he was enrolled in a study combining imatinib 800 mg daily with arsenic trioxide.
Characteristics of the patient cohort
Patient
Age (years)/ sex
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
62/M 69/M 55/M 63/F 35/F 69/M 45/M 53/F 58/F 54/F 68/F 51/F 34/F 57/M 68/M 76/M 61/M
Time on imatinib (months) 46 65 60 76 ND 48 40 52 69 12 44 70 11 19 11 16 11
Imatinib response (at study onset) Relapse Primary resistance Primary resistance Primary resistance Primary resistance Relapse Relapse Relapse Primary resistance Suboptimal response Primary resistance Relapse Primary resistance Relapse Relapse Primary resistance Suboptimal response
Cytogenetics (% Ph+)
Bone marrow FISH (% BCR-ABL+)
RT-PCR (%)
35 85 55 ND 100 90 40 10 35 30 85 35 100 100 ND 100 20
14.5 64 49 ND 87 76.5 37 1.5 32 24.5 88 35.5 ND 98.5 85 74.5 15
0.98 16 0.55 9.2 1.4 5.3 0.52 0.1 3.2 1.7 7.1 1.6 ND 210 88 6.4 1.8
Abbreviations: ND, not determined; RT-PCR; reverse transcriptase PCR. Primary resistance was defined as the failure to achieve a major cytogenetic response (MCyR, o35% Ph+ metaphases) after at least 12 months of imatinib therapy. Suboptimal response was defined as 1–35% Ph-positive metaphases after 12 months of imatinib treatment. Relapse was defined by a loss of MCyR or CCyR at the time of the study. Fluorescence in situ hybridization (FISH) was performed on unseparated bone marrow interphase cells using the Vysis BCR/ABL dual fusion probe. BCR-ABL was measured in peripheral blood, normalized for expression of G6PD, and expressed in percent (BCR-ABL/G6PD), as described.6
Table 2
Follow-up summary of patient with KRAS T58I mutation
Sample date
Treatment before cytogenetic evaluation (mg/day)
February 2002 December 2002a December 2003a October 2004a June 2005a October 2005a December 2005a March 2006a April 2006a September 2006a December 2006a February 2007a May 2007a
Imatinib 400 Imatinib 400 Imatinib 600–800+ Arsenic trioxide Imatinib 800 + Arsenic trioxide Imatinib 800 Dasatinib 140 Dasatinib 100 Dasatinib 80 Dasatinib 80 Dasatinib 80 Dasatinib 80 Dasatinib 80 Dasatinib 80
Cytogenetics (% Ph+)
Bone marrow FISH (% BCR-ABL+)
KRAS T58I allele (%)
65 100 20 35 85 ND 30 ND 8 15 33 20 95
41.5 78 31.5 14.5 62.5 23 24.5 13.5 10 24 29 51 60.5
0 0 0 31.8 ND 28.8 ND 37.4 ND 34.9 33.1 0 0
BCR-ABL kinase domain Native Native Native Native Native Native Native Native Native Native Native Native Native
ND: Not determined owing to poor sample quality. a The patient samples where the Ph+ cells were double Ph+ due to an extra copy of a derivative of chromosome 22 of BCR-ABL. Data shown here by cytogenetics (metaphases) and fluorescence in situ hybridization (FISH) (interphases) represent total BCR-ABL-positive clones. Leukemia
Letters to the Editor
been observed in these studies.8 To assess the functional relevance of T58I in BCR-ABL-positive cells, we introduced KRAS T58I, KRAS G12D (a strongly activating mutation) and wild-type KRAS into 32D cells expressing BCR-ABL (32Dp210BCRABL) or parental 32D cells. Consistent with previous studies,9 neither wild-type KRAS nor any of the mutants rendered parental 32D cells interleukin-3 independent (Figure 1a). To test whether the KRAS T58I mutant may affect imatinib or dasatinib sensitivity, KRAS T58I-expressing 32Dp210BCRABL cells were grown in the absence of interleukin-3 with graded concentrations of imatinib or dasatinib. Proliferation assays revealed a reduction in sensitivity to both imatinib and dasatinib in 32Dp210BCRABL cells co-expressing wild-type or mutant KRAS. The effect was strongest with G12D, intermediate with T58I and weakest with wild-type KRAS. Notably, 15% of 32Dp210BCRABL cells co-expressing KRAS
He achieved a transient major cytogenetic response. Upon relapse, he was switched to dasatinib. The T58I mutation was absent while the patient was responding to 800 mg of imatinib, but it became detectable before relapse and remained detectable (at approximately 30%) in 5 subsequent samples collected over a period of 18 months (Table 2). During this time, the patient achieved a second major cytogenetic response on dasatinib. Immediately before relapse on dasatinib, the T58I allele became undetectable simultaneously with a rapid increase of the clone with the additional copy of Ph. None of the samples tested were positive for BCR-ABL kinase mutations at any time. T58I is an activating mutation of KRAS that has been previously reported in patients with Noonan syndrome who developed juvenile myelomonocytic leukemia.7 KRAS mutations are associated with various other human cancers including colorectal cancers, though the specific T58I exchange has not
2271
120 32D 32D WT KRAS 32D G12D KRAS 32D T581 KRAS
% Growth Rate
100 80 60 40 20 0 1
0.1
0.01
0.001 0.0001 0.00001
0
IL-3 (ng/ml) 120 32Dp210 32Dp210 WT KRAS 32Dp210 G12D KRAS 32Dp210 T581 KRAS
% Growth Rate
100 80 60 40
32Dp210 32Dp210 WT KRAS 32Dp210 G12D KRAS 32Dp210 T581 KRAS
100 % Growth Rate
120
80 60 40 20
20 0
0 0
0.312 0.625 1.25
2.5
5
10
0
0.5
1
120
120
100
100
80 60 40 32Dp210 32Dp210 WT KRAS 32Dp210 G12D KRAS 32Dp210 T581 KRAS
20
5
10
50
100
Dasatinib (nM)
0
% of Annexin Positive Cells
% of Annexin Positive Cells
Imatinib (µM)
80 60 40
32Dp210 32Dp210 WT KRAS 32Dp210 G12D KRAS 32Dp210 T581 KRAS
20 0
0
0.312 0.625 1.25
2.5
Imatinib (µM)
5
10
0
0.5
1
5
10
50
100
Dasatinib (nM)
Figure 1 The KRAS T58I mutation is activating and confers increased resistance to imatinib and dasatinib in BCR-ABL-positive cells. (a) T58I, G12D and wild-type (WT) KRAS were stably expressed in 32D cells and interleukin-3 independence was tested. (b and c) BCR-ABL-positive 32D cells stably expressing mutant or WT KRAS were tested for sensitivity to imatinib and dasatinib in cell proliferation assays and (d and e) for apoptosis in response to imatinib and dasatinib. Values shown are the mean value from two independent experiments performed in triplicate. Leukemia
Letters to the Editor
2272 T58I retained viability at 10 mM imatinib (Figure 1b) or 100 nM dasatinib (Figure 1c) over 3 days in culture as compared to 0% of 32Dp210BCRABL cells expressing wild-type KRAS. Consistent with this, there was less apoptosis in 32Dp210BCRABL cells coexpressing KRAS T58I (85%) as compared to those co-expressing wild-type KRAS (100%) at intermediate concentrations of imatinib (1.25 mM) (Figure 1d) or dasatinib (10 nM) (Figure 1e) after 2 days in culture. Differences were less pronounced at higher concentrations of either drug. No effect was observed in 32D parental cells expressing wild-type or mutant KRAS (data not shown). Notably, our demonstration of KRAS T58I in a patient with imatinib resistance is analogous to findings by Lievre et al.,10 who identified other types of KRAS mutations in association with resistance to cetuximab, a drug targeting epidermal growth factor receptor in colorectal cancer. Thus, escape from the effects of tyrosine kinase inhibitor therapy by the activation of alternative pathways can occur in the setting of solid tumors as well as leukemia. Evidently, alternative therapies are required to treat this kind of resistance. In the case of RAS mutations, this could include Farnesyl transferase inhibitors or agents targeting the mitogen-activated protein kinase pathway.11 The transient appearance of a cell clone with an activating KRAS mutation could have two explanations. One possibility is that the KRAS mutant clone may have arisen independently of the CML clone and may be comparable to the Ph-negative but cytogenetically abnormal clones seen in some 5–10% of CML patients treated with imatinib, which are frequently transient.12 Alternatively, as suggested by the temporal correlation with cytogenetic relapse, it is possible that the KRAS T58I mutant conferred partial resistance to imatinib, which promoted relapse and prevented the extinction of the clone on subsequent dasatinib therapy. Consistent with this, we found that co-expression of KRAS T58I and BCR-ABL in 32D cells reduced imatinib and dasatinib sensitivity, though not to the same degree as the strongly activating KRAS G12D mutation. Ultimately, a second clone, cytogenetically characterized by a second copy of the Ph chromosome, became dominant and suppressed the KRAS T58I clone below the detection threshold of direct sequencing. Distinguishing between these two possibilities would require single cell or colony assays, but unfortunately no suitable archived material is available. Overall, our data show that activating RAS mutations may contribute to imatinib resistance in some CML patients who relapse with native BCR-ABL, although these mutations are apparently not a common mechanism of drug resistance in CML. In contrast, we found no evidence of mutations in PTPN11, JAK2, FLT3 or resistance mutations in the activation loops of KIT or PDGFRa/b in any of the 17 patients. Given that the cohort under study is small, however, it remains possible that such mutations would be detected in a larger group of patients. Additionally, the proportion of BCR-ABLpositive interphases was o50% in 7/17 samples. As direct sequencing was used for screening of KIT, KRAS, NRAS and JAK2, mutations present in the BCR-ABL-positive cells may have been missed, given that a mutant allele must be present at 420–25% to be detectable by this technique. This is less likely for mutations of PTPN11, FLT3 and PDGFRa/b, which were analyzed by D-HPLC with a sensitivity of approximately 10%. Irrespective of this, our data suggest that BCR-ABL mutation-negative resistance is not due to a frequent or universal mutation in one of the genes under study. Thus, in the majority of patients, other yet unknown
Leukemia
mechanisms must account for resistance, and elucidating these mechanisms will be crucial for developing new therapeutic approaches.11
Acknowledgements This study was supported in part by NHLBI Grant HL082978-01 (MWD) and by the Leukemia and Lymphoma Society (MWD).
A Agarwal1,2, CA Eide1,2, A Harlow1,3,4, AS Corbin1,2, MJ Mauro1, BJ Druker1,2, CL Corless1,3,4, MC Heinrich1,3 and MW Deininger1 1 Division of Hematology and Medical Oncology, Oregon Health & Science University, Portland, OR, USA; 2 Division of Hematology and Medical Oncology, Howard Hughes Medical Institute, Portland, OR, USA; 3 Portland Veterans Affairs Medical Center, Portland, OR, USA and 4 Department of Pathology, Oregon Health & Science University, Portland, OR, USA E-mail:
[email protected]
References 1 O0 Hare T, Eide CA, Deininger MW. Bcr-Abl kinase domain mutations, drug resistance, and the road to a cure for chronic myeloid leukemia. Blood 2007; 110: 2242–2249. 2 Mahon FX, Deininger MW, Schultheis B, Chabrol J, Reiffers J, Goldman JM et al. Selection and characterization of BCR-ABL positive cell lines with differential sensitivity to the tyrosine kinase inhibitor STI571: diverse mechanisms of resistance. Blood 2000; 96: 1070–1079. 3 Ren R. Mechanisms of BCR-ABL in the pathogenesis of chronic myelogenous leukaemia. Nature reviews 2005; 5: 172–183. 4 Wang Y, Cai D, Brendel C, Barett C, Erben P, Manley PW et al. Adaptive secretion of granulocyte-macrophage colony-stimulating factor (GM-CSF) mediates imatinib and nilotinib resistance in BCR/ ABL+ progenitors via JAK-2/STAT-5 pathway activation. Blood 2007; 109: 2147–2155. 5 Donato NJ, Wu JY, Stapley J, Gallick G, Lin H, Arlinghaus R et al. BCR-ABL independence and LYN kinase overexpression in chronic myelogenous leukemia cells selected for resistance to STI571. Blood 2003; 101: 690–698. 6 Press RD, Love Z, Tronnes AA, Yang R, Tran T, Mongoue-Tchokote S et al. BCR-ABL mRNA levels at and after the time of a complete cytogenetic response (CCR) predict the duration of CCR in imatinib mesylate-treated patients with CML. Blood 2006; 107: 4250–4256. 7 Schubbert S, Zenker M, Rowe SL, Boll S, Klein C, Bollag G et al. Germline KRAS mutations cause Noonan syndrome. Nature genetics 2006; 38: 331–336. 8 Sjoblom T, Jones S, Wood LD, Parsons DW, Lin J, Barber TD et al. The consensus coding sequences of human breast and colorectal cancers. Science (New York, NY) 2006; 314: 268–274. 9 Soddu S, Blandino G, Scardigli R, Martinelli R, Rizzo MG, Crescenzi M et al. Wild-type p53 induces diverse effects in 32D cells expressing different oncogenes. Mol cell biol 1996; 16: 487–495. 10 Lievre A, Bachet JB, Boige V, Cayre A, Le Corre D, Buc E et al. KRAS mutations as an independent prognostic factor in patients with advanced colorectal cancer treated with cetuximab. J Clin Oncol 2008; 26: 374–379. 11 Jabbour E, Cortes JE, Giles FJ, O’Brien S, Kantarjian HM. Current and emerging treatment options in chronic myeloid leukemia. Cancer 2007; 109: 2171–2181. 12 Bumm T, Muller C, Al-Ali HK, Krohn K, Shepherd P, Schmidt E et al. Emergence of clonal cytogenetic abnormalities in Ph- cells in some CML patients in cytogenetic remission to imatinib but restoration of polyclonal hematopoiesis in the majority. Blood 2003; 101: 1941–1949.
