Imatinib Resistance

Resistance to IM can be defined at different levels and different time points. These categories are outlined in Table 1. Intrinsic or primary resistance to IM can be defined when initial IM therapy does not result in a hematologic or cytogenetic response. Secondary or acquired resistance or relapse is defined as a loss of IM efficacy after an initial response to therapy. Each of these types of resistance can be defined at the hematologic,

Table 1 Definitions of Resistance3

Primary (intrinsic) Secondary (acquired) Subcategories of resistance Hematologic Chronic phase Advanced phases Cytogenetic Molecular a See text for detailed explanations.

cytogenetic, or molecular level. Hematologic resistance must also be defined by the phase of the disease the patient is in. Hematologic resistance in the chronic phase would refer to a worsening of peripheral blood counts or the white blood cell differential or inability to reduce splenomegaly. In the accelerated or blast phase, hematologic resistance would refer to a lack of return to the chronic phase or a hematologic relapse following an initial response to therapy.

At the cytogenetic level, resistance can be defined as a loss or lack of MCR or CCR. At the molecular level, a complete molecular response is defined as undetectable BCR-ABL1 transcripts by PCR or a > 3-log reduction, a BCR-ABL1 transcripts which represents a major molecular remission. Loss of either of these responses would be defined as resistance at the molecular level. However, standardization of measurement of BCR-ABL1 transcripts by PCR has not yet been accomplished, and results in one lab may not necessarily be equivalent to those of another lab. Also, a standardized definition of resistance at the molecular level is not well defined, although many experts consider a half log or one-log increase which is confirmed in a second sample as a reasonable benchmark for loss of molecular response (27).

An international effort is underway to harmonize methodologies for detecting BCR-ABL1 transcripts and kinase domain mutations and adjusting the results so they are standardized from one lab to another (28,29). The availability of a reproducible real-time quantitative PCR (RQ-PCR) has shown in one study that patients with a major molecular response do not have evidence of cytogenetic abnormalities in their bone marrow, and therefore a policy of performing bone marrow biopsies only in patients who have not achieved or have lost a major molecular response would allow many patients to forego the discomfort and expense of multiple bone marrow biopsies (30).

A panel of experts has recently published a consensus opinion based on a literature review as to when to consider patients to have failed IM and to then be considered for alternate therapies. In this study, failure was defined as lack of a hematologic response after 3 months of IM, less than a complete hematologic response or no cytogenetic response (>95% positive PH chromosome) after 6 months of IM, less than an MCR after 12 months, and less than a CCR after 18 months. The development of mutations, loss of a CCR, or loss of a complete hematologic response at any time was also considered a failure. Definitions for a suboptimal response and "warning" situations such as additional chromosome abnormalities were also listed (Table 2) (31).

Table 2

Operational Definition of Failure and Suboptimal Response for Previously Untreated Patients in ECP CML Who Are Treated with 400 Mg IM Daily



Suboptimal response0





High riskd, del9qe, ACAs

in Ph+ cells

3 mo after diagnosis

No HR (stable

Less than CHR


disease or disease


6 mo after diagnosis

Less than CHR, no

Less than PCgR


CgR (Ph+ > 95%)

(Ph+ > 35%)

12 mo after diagnosis

Less than PCgR

Less than CCgR

Less than MMolR

(Ph+ > 35%)

18 mo after diagnosis

Less than CCgR

Less than MMolR



Loss of CHRd, loss

ACA in Ph+ cells\

Any rise in transcript

of CCgR/,

loss of MMolR\

level; other




abnormalities in Ph-


a Failure implies that the patient should be moved to other treatments whenever available. b Suboptimal response implies that the patient may still have a substantial benefit from continuing IM treatment but that the long-term outcome is not likely to be optimal, so the patient becomes eligible for other treatments.

c Warnings imply that the patient should be monitored very carefully and may become eligible for other treatments. The same definitions can be used to define the response after IM dose escalation.

dBy Sokal or Hasford score. HR = hematologic response; ECP = early chronic phase; CHR = complete HR; CgR = cytogenetic response; PCgR = partial CgR (Ph + 1-35%); CCgR = complete CgR (Ph + 0%); MMolR = major molecular response (< 0.10 BCR-ABL1 gene ratio); ACA = additional chromosome abnormalities; NA, not applicable.

e To be confirmed on two occasions unless associated with progression to AP/BC. f To be confirmed on two occasions, unless associated with CHR loss or progression to AP/BC. g High level of insensitivity to IM.

h To be confirmed on two occasions, unless associated with CHR or CCgR loss. ' Low level of insensitivity to IM.

Reprinted with permission from: Baccarani M, Saglio G, Goldman J et al. Evolving concepts in the management of chronic myeloid leukemia: recommendations from an expert panel on behalf of the European Leukemia Net (31).

Multiple mechanisms of resistance to IM have been defined in recent years. Four broad mechanisms have been characterized (Table 3). In vitro models that studied resistance predicted several of these mechanisms. CML cell lines that are BCR-ABL1 positive and murine hematopoietic cells that have been transformed with a BCR-ABL1 gene and then exposed to IM have subsequently developed resistance and have been used to predict several mechanisms (32).

In a mouse model of IM resistance, it was demonstrated that in vivo tumors were resistant to IM but retained in vitro sensitivity (33). Several clinical studies have suggested

Table 3

Mechanisms of Imatinib Resistance

Decreased intracellular drug levels o plasma binding by alpha 1-acid glycoprotein o differential expression of transporter proteins (MDR-1, hOCTl) o pharmacologic interactions Increased expression of BCR-ABL kinase from genomic amplification BCR-ABL independent mechanisms Clonal evolution (non- BCR-ABL-dependent mechanism) o Clonal evolution o Aneuploidy o Over-expression of SRC family kinases Mutations in ABL kinase of BCR-ABL affecting drug interaction or kinase activity.

Reprinted with permission from U.S Healthcare Communications, LLC. Litzow MR, Tefferi A. Chronic Myeloid Leukemia: Problems Propel Progress. The American Journal of Hematol-ogy/Oncology, 2007; 6(5) supplement 7:19-22.

that plasma binding of a-1 acid glycoprotein correlates with clinical responses to IM, although none of these studies have clearly distinguished a cause-and-effect relationship

Transporter proteins have been shown in in vitro models to contribute to IM resistance

(32.36). This is known to be mediated in some instances by the multi-drug resistant gene (MDR1) producing the p-glycoprotein, where p-glycoprotein expression was associated with a decrease in intracellular IM levels and development of resistance (37).

Recently, other transporter proteins mediating influx, the organic cation transporters (OCT), have been identified. It has been shown that the OCT-1 influx protein mediates transport of IM into cells and reduced OCT-1 activity appears to be a cause for low in vitro sensitivity of CML cells to IM (38,39).

Down-regulation of T-cell protein tyrosine phosphatase in IM-resistant cells may represent a novel mechanism for IM resistance (40). A recently published analysis of data from the IRIS trial demonstrated that trough blood levels of IM and its active major metabolite, CGP74588, correlated with achievement of CCR and MMR (41).

Gene amplification of the BCR-ABL1 kinase may be associated with development of IM resistance. The presence of multiple copies of the BCR-ABL1 gene in interphase nuclei can be demonstrated by fluorescence in situ hybridization (FISH) (40,42). In one series, more than half of the patients with IM resistance had evidence of clonal evolution with the development of additional chromosome abnormalities. Paired cytogenetic analyses performed at the beginning of IM therapy and at the time of resistance demonstrated this evolution.

Chromosomal abnormalities included the presence of aneuploidy, a second Ph chromosome, and trisomy 8. The loss of one p53 allele by an alteration of the short arm of chromosome 17 was seen in seven patients, and new reciprocal translocations were seen in two patients. In eight cases, multiple cytogenetic abnormalities were also present (40).

Loss of the p53 tumor suppressor gene has been shown to impede the anti-leukemic response to BCR-ABL1 inhibition (43). Another mechanism for resistance that is independent of BCR-ABL1 and that has been demonstrated in vitro is over-expression of SRC-related kinases such as LYN (44). It appears that kinases from the SRC family mediate signaling of BCR-ABL1. This has been demonstrated in vitro with the use of SRC kinase inhibitors and SRC mutants that are kinase defective (45,46,47).

If one takes CML cells and cultures them in the presence of IM, or obtains cells from patients who have progressed while on IM therapy, a decrease in BCR-ABL1 mRNA or protein levels is seen with a concomitant increase in SRC family kinases (44). If one inhibits LYN expression by RNA interference in IM-resistant CML cells, survival and proliferation of these cells is inhibited (48). Additional support for the role of SRC kinases in some patients with IM resistance is that IM is unable to directly inhibit SRC kinases but does so only through its effect on BCR-ABL1 (49).

The most frequent and increasingly best-studied mechanism of resistance to IM in patients with CML are gene mutations in the ABL1 (tyrosine kinase) domain of the BCR-ABL1 gene. In 2001, a single amino acid substitution at a threonine residue of ABL1 kinase domain was described and resulted in substitution of isoleucine for threonine at position 315 (T315I) ofc-ABL1 (33). This amino acid substitution interfered with a hydrogen bond that formed between the ABL1 kinase and IM. The T315I mutation has turned out to be one of the most frequent mutations seen in patients with CML with resistance to imatinib. This altered binding of IM to the BCR-ABL1 kinase appears to confer significant resistance.

A basic understanding of the structure of the BCR-ABL1 chimeric protein is important to understanding how the multiple mutations that have been described in the ABL1 kinase domain cause resistance. The c-ABL1 protein is expressed in two splice forms that are known as 1a and 1b (Fig. 1C). The 1a form is 19 residues shorter than 1b. The 1b form contains a myristoylation site on its second residue. The second residue is a glycine that appears to help regulate enzymatic activity, and its mutations to alanine prevents myristoylation and results in an activated kinase (50). The 1b form also contains a "cap" region that is believed to stabilize the inactive confirmation of the kinase (51). The numbering system used to identify the amino acid residues where mutations occur is based on the shorter 1a form.

Within the c-ABL1 protein, there are three SRC homology domains as described briefly earlier. These include SH1, the kinase domain, which encodes for catalytic function; the SH2 domain, which binds phosphotyrosine-containing peptides; and SH3, which is a negative regulator of kinase activity (Fig. 1C). Toward the N-terminal end of the ABL1 kinase is the adenosine triphosphate-binding portion that is highly conserved with glycine-rich sequences and known as the P-loop. It interacts with IM through hydrogen and van deer Waals bonds. This area spans amino acids 248-256. At the carboxy- or C-terminal end of ABL1 kinase is a flexible activation loop. It is critical for the control of catalytic activity and, as its name implies, changes confirmation depending on whether the molecule is in the inactive or active state. It begins at amino acid 381. See Fig. 2 for details.

Between the P-loop and the activation loop is the catalytic site of ABL1 kinase. It is located in a cleft where IM and other small molecule tyrosine kinase inhibitors bind. The shift of the activation loop between an inactive or closed confirmation and a catalytically active or open confirmation seems to be regulated by the kinase itself in a process known as "auto-inhibition." In the active state, the activation loop flips away from the catalytic region, whereas in the inactive state it is inward toward the catalytic region and appears to serve as a support for substrate binding (52) (Fig. 2). Further details on these structural details have been recently reviewed (51,53,54).

The number of mutations within the ABL1 kinase domain of BCR-ABL1 continues to grow rapidly, and it has recently been estimated that 73 distinct point mutations causing substitutions in 50 amino acids have been found in cells from imatinib-resistant CML patients, with some being more frequently found than others (Fig. 4).

Melo and Chuan have categorized these mutations into four groups, including (1) those which directly impair imatinib binding, (2) those occurring in the P-loop, (3) those within the activation loop, which prevent IM binding (IM is able to bind BCR-ABL1 kinase only when it is in the closed or inactive confirmation, and (4) those within the catalytic domain (Fig. 2) (52).

Substitutions in these single nucleotides that lead to mutations change the amino acids that will, as a result, have varying effects on the confirmation of the ABL1 portion of the BCR-ABL1 protein and its binding to drugs or other substrates. In a study where the crystal structure of IM complexed to the catalytic domain of BCR-ABL1 predicted that resistance would occur with the T315I mutation because the threonine residue at position 315 forms a crucial hydrogen bond between IM and ABL1 kinase (55). In vitro random mutation of the BCR-ABL1 molecule followed by screening for IM resistance demonstrates many of the major mutations that have been identified in patients but also revealed other mutations that illustrate potentially novel mechanisms of acquired IM resistance (12).

Two mutations that confer IM resistance including the T315I and the E255 K appear to enhance the activity of the BCR-ABL1 kinase through an enhanced ability to induce auto-phosphorylation. This observation would suggest that these mutations may

Fig. 4. Map of BCR-ABL kinase domain mutations associated with clinical resistance to imatinib. Abbreviations: P, P-loop; B, imatinib binding site; C, catalytic domain; A, activation loop. Amino-acid substitutions in green indicate mutations detected in 2-10% and in red in >10% of patients with mutations. (Reprinted with permission from Ref. (52)).

Fig. 4. Map of BCR-ABL kinase domain mutations associated with clinical resistance to imatinib. Abbreviations: P, P-loop; B, imatinib binding site; C, catalytic domain; A, activation loop. Amino-acid substitutions in green indicate mutations detected in 2-10% and in red in >10% of patients with mutations. (Reprinted with permission from Ref. (52)).

confer a growth advantage for CML cells, even in the absence of selective pressure from treatment with IM (56).

It is currently believed that mutations are not induced by IM, but rather exist before therapy is initiated, albeit at a low level, and then emerge as drug-sensitive cells are eliminated. This observation is reminiscent of the emergence of antibiotic-resistant bacteria. Indeed, multiple studies have now shown that mutations in the BCR-ABL1 gene that confer IM resistance are detectable prior to treatment with IM (57,58,59, 60). Prior to therapy, these mutant clones are sometimes difficult to detect, but in other instances are more readily apparent. This again raises the question of whether these higher level mutations may confer a growth or survival advantage for the leukemic clone, even without therapy (61). However, some studies have shown that not all mutations detected before initiation of therapy were subsequently expanded by therapy with IM (60).

This suggests that other factors may be necessary for development of IM resistance or that mutations may be present in cells that are unable to differentiate. The presence of a mutation in a patient may also not be the sole cause of resistance because multiple mechanisms may be contributing to resistance. Therefore, the presence of a mutation in a patient should not automatically be assumed to be the cause of resistance and needs to be interpreted within the overall clinical context (62). Additionally, not all mutations confer resistance to IM. One study found that 5 of 17 BCR-ABL1 kinase domain mutants remained sensitive to IM (63).

Several studies have suggested that patients with IM resistance secondary to mutations in the P-loop have a poorer prognosis than patients with mutations in other portions of the ABL1 kinase domain. A study from Australia detected mutations in 27 of 144 patients at 17 different residues. Twenty-four of these 27 patients (89%) developed acquired resistance. Thirteen of these 24 had mutations in the P-loop, and 12 of these patients died with a median survival of only 4.5 months after mutation detection.

In the 14 patients with mutations outside the P-loop, there were only three deaths, and the median follow-up was 11 months (64). Studies from France and Italy have confirmed these findings of a significantly poorer outcome in patients with P-loop mutations (65,66). These same studies showed a worse outcome for patients with the T315I mutation as well. One study, however, was unable to demonstrate a worse prognosis for patients with P-loop mutations and suggested that the prognosis of patients who fail IM is multifactorial in nature (67).

Not surprisingly, mutations are more frequently found in patients in the late chronic, accelerated, or blast phase of CML and also in patients with additional chromosome abnormalities (60).

Oligonucleotide microarray analysis of bone marrow samples from patients with IM-sensitive and IM-resistant CML have shown the ability to distinguish IM sensitivity and resistance. Differential gene expression patterns have the potential to identify new gene and protein targets for treatment and may have the potential to be used as a screening tool to identify patients with resistance prior to therapy (68,69, 70,71,72).

Mutation detection in the BCR-ABL1 gene has varying sensitivity based on the technique utilized. Methodologies vary from direct sequencing of the kinase domain (64), use of denaturing high-performance liquid chromatography (73), a flow cytometric measurement of downstream targets of BCR-ABZ (74), restriction fragment length polymorphism (RFLP)-based assays (75), and various polymerase chain reaction-based assays including those involving allele-specific oligonucleotides (57), peptide nucleic acid (PNA)-based clamping techniques (76), identification of single nucleotide polymorphisms utilizing microarrays (77), and real-time quantitative PCR (78).

In this last study, a two-fold rise in BCR-ABL1 expression by RQ-PCR predicted a mutation in a kinase domain of BCR-ABL1 in 61% of patients either at the time of the rise in BCR-ABL1 expression or within 3 months of the rise (78). However, another study did not predict the presence of mutations in patients who had a single two-fold or greater rise in BCR-ABL1 transcripts. Rather, the development of rising BCR-ABL1 transcript levels was necessary to reliably pick up a mutation. The authors concluded that a serial rise was more reliable than a single rise (79). Eleven mutations were detected in 10 out of 82 patients in this study.

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook

Post a comment