Suboptimal Response to or Failure of Imatinib Treatment for Chronic Myeloid Leukemia: What Is the Optimal Strategy?

Abstract

Treatment responses to imatinib vary among patients with chronic myeloid leukemia (CML), and definitions of treatment failure and suboptimal response have been published. This article discusses monitoring and treatment of patients with CML after failure of or suboptimal response to imatinib therapy. We reviewed articles listed on PubMed from January 1, 2002, to July 31, 2008, and abstracts from the 2007 Annual Meeting of the American Society of Hematology. Search terms used were chronic myeloid/myelogenous leukemia, imatinib, and BCR-ABL. To enable early recognition of suboptimal responses, patients should be frequently monitored according to published guidelines, including cytogenetic analysis every 6 months until a complete response is achieved and molecular monitoring every 3 months from the start of therapy or monthly if an increasing BCR-ABL1 transcript level is detected. Mutational analysis of BCR-ABL1 may assist with treatment selection. A recent survey suggests that a notable proportion of physicians do not follow treatment guidelines and that broader communication is required. Recent recommendations state that, in patients whose response to imatinib at 400 mg/d is suboptimal, the dose should be increased, whereas alternative therapies, such as dasatinib, nilotinib, and allogeneic stem cell transplant (in eligible patients), and imatinib dose escalation should be considered after imatinib failure. However, clinical data are lacking to confirm this sequence of treatments, and introducing alternative therapies at an earlier stage of treatment, for example, after a suboptimal response, may produce better long-term outcomes in a higher proportion of patients. Patient and disease characteristics should be carefully considered to optimize treatment strategy for CML.

CCyR = complete cytogenetic response; CHR = complete hematologic response; CML = chronic myeloid leukemia; CP = chronic phase; ELN = European LeukemiaNet; IRIS = International Randomized Study of Interferon vs STI571; MCyR = major cytogenetic response; MMR = major molecular response; PCR = polymerase chain reaction; SCT = stem cell transplant; START = Src/Abl1 Tyrosine Kinase Inhibition Activity: Research Trials of Dasatinib; TKI = tyrosine kinase inhibitor

Chronic myeloid leukemia (CML) was the first neoplastic disease associated with a chromosomal aberration.¹ The Philadelphia chromosome is a reciprocal t(9;22)(q34;q11) translocation that results in the production of the Bcr-Abl1 fusion protein (previously termed Bcr-Abl), a constitutively active tyrosine kinase. Elucidation of the molecular pathogenesis of CML led to the development of the tyrosine kinase inhibitor (TKI) STI-571, now known as imatinib mesylate.

The Food and Drug Administration first approved imatinib in May 2001 for the treatment of patients with chronic-phase (CP) CML after failure of interferon alfa therapy and for the treatment of advanced CML. In December 2002, after positive results from the International Randomized Study of Interferon vs STI571 (IRIS), imatinib was approved for the treatment of patients with newly diagnosed CP-CML.^2,3 Use of imatinib as first-line therapy has resulted in complete cytogenetic responses (CCyRs) in 65% to 85% of patients with CML.^4-6

Despite the successes of imatinib in treating CML, some patients' responses to treatment are inadequate. In the IRIS trial, approximately 30% of patients with newly diagnosed CP-CML who were randomized to receive imatinib did not achieve a CCyR within 1 year of treatment.⁶ In addition, approximately 10% of patients experienced relapse during 5 years of follow-up, including approximately 10% who had achieved a CCyR.⁴ Higher rates of treatment failure occur in patients with accelerated- or blast-phase disease.^7,8

Treatment guidelines for CML focus on how to treat patients in whom imatinib therapy has failed. Numerous studies of CML suggest that earlier responses to imatinib therapy correlate with better outcomes.^4,5,9-14 Therefore, definitions of both failure of and suboptimal response to imatinib have been proposed.^15,16 The European LeukemiaNet (ELN) guidelines describe a suboptimal response as patients who “may still have a substantial benefit from continuing imatinib treatment but that the long-term outcome is not likely to be optimal, so the patient becomes eligible for other treatments.”¹⁵ Recent studies have confirmed that patients whose responses are suboptimal based on ELN definitions have an inferior prognosis compared with patients whose response is optimal.^17,18 Currently, few clinical data are available to support treatment selection after suboptimal response to imatinib.

The aims of this article are to briefly discuss effective monitoring of CML for early identification of patients whose disease is resistant or whose response is suboptimal, to summarize relevant data on treatment options, and to discuss clinical scenarios when alternative strategies might be used. Relevant articles published from January 1, 2002, to July 31, 2008, were identified through a PubMed search using the following terms: chronic myeloid leukemia, chronic myelogenous leukemia, imatinib, and BCR-ABL. Abstracts presented at the 2007 Annual Meeting of the American Society of Hematology were also reviewed.

WORKING DEFINITIONS FOR SUBOPTIMAL RESPONSE TO AND FAILURE OF IMATINIB THERAPY

Criteria have been published to define inadequate responses to imatinib treatment in patients with CP-CML (Table 1).^15,16 The ELN guidelines, published in 2006, define a suboptimal response as less than a complete hematologic response (CHR) at 3 months, less than a major cytogenetic response (MCyR) at 6 months, less than a CCyR at 12 months, or less than a major molecular response (MMR) at 18 months.¹⁵ Imatinib therapy failure is defined as the absence of any hematologic response at 3 months, less than a CHR or absence of any cytogenetic response at 6 months, less than an MCyR at 12 months, less than a CCyR at 18 months, or loss of CHR or CCyR at any time.¹⁵ These milestones are also cited in recently updated European Society for Medical Oncology recommendations for CML.¹⁹ Similar guidelines have been proposed by the US National Comprehensive Cancer Network (updated in 2008).¹⁶ However, because recent data have shown no significant survival advantage associated with achieving an MMR at 18 months for those already in CCyR, this may not represent a clinically useful milestone for suboptimal response.^4,9,18

TABLE 1.

Recommended Frequencies of Response Assessment in Patients With Chronic Myeloid Leukemia and Definitions of Resistance (Suboptimal Response or Failure) During Initial Imatinib Therapy^a

For patients with advanced-phase CML, response milestones have not been suggested, and treatment strategy involves initial imatinib treatment with the aim of proceeding to allogeneic stem cell transplant (SCT).

EFFECTIVE ASSESSMENT OF RESPONSES TO THERAPY

The purpose of regular ongoing assessments for CML is to identify early those patients who are not responding optimally so that an alternative treatment strategy can be considered. Several tests are available that provide information on responses to drug therapy for CML, including full blood cell count (hematologic response), cytogenetic analysis (conventional or fluorescent in situ hybridization), and real-time quantitative polymerase chain reaction (PCR) for BCR-ABL1 messenger RNA (molecular response).²⁰ An important consideration for effective monitoring is the optimal frequency of assessment. Guidelines recommend the following: hematologic responses should be assessed every 2 weeks until a CHR is achieved, then every 3 months; cytogenetic responses should be assessed every 6 months until a CCyR is achieved, then every 12 months; and molecular responses should be assessed every 3 months or monthly if an increasing BCR-ABL1 transcript level is detected (Table 1).^15,16,21 Although the definition of an increasing level is not fully established, different studies have found that a BCR-ABL1 transcript increase of 2- or 3-fold, including loss of MMR among those who had achieved MMR, was associated with the development of BCR-ABL1 mutations and relapse^22,23; however, in another study, a serial increase was more reliable for predicting mutations.²⁴

The use of molecular testing is less widespread than other methods. Because of its high sensitivity, molecular monitoring enables further follow-up once CCyR has been achieved, and most often, this method is used only in patients with a CCyR. However, patients who do not achieve early reductions in BCR-ABL1 transcript levels during the first 6 months of therapy have a significantly lower probability of achieving a subsequent MMR and a higher probability of experiencing disease progression.^12-14,25 Therefore, early molecular monitoring after initiation of treatment helps to identify patients at higher risk of suboptimal response to or failure of imatinib therapy. Several practical issues exist regarding molecular monitoring. For BCR-ABL1 transcript levels to be interpreted in individual patients according to established definitions of suboptimal response or failure, laboratories are required to establish a baseline level that is standardized using the international scale.^20,21 In addition, variability among centers can be introduced because of methodological differences. However, when molecular assessments are performed at a center that uses optimized methods,²¹ BCR-ABL1 levels provide clinically important information on patient status.

A key component of response monitoring in CML is the assessment of BCR-ABL1 mutations. Mutations within the ABL1 kinase domain cause structural changes in the Bcr-Abl1 protein that interfere with binding and inhibition by imatinib.²⁶ The clinical implications of detecting a mutation depend on the testing method. Mutations in minor leukemic cell populations can be detected using highly sensitive methods (eg, PCR or denaturing high-performance liquid chromatography). Because low-level mutations do not inevitably lead to clinical resistance,^27,28 their detection should not prompt any change in therapy unless clinical signs are evident. However, when a mutation can be detected in 20% or more of leukemic clones, which represents the threshold for detection using direct sequencing, this provides a strong indication that clinical resistance is likely to occur.^21,29,30 Furthermore, if direct sequencing detects a highly resistant BCR-ABL1 mutation, which includes, among others, T315I or several mutations within the P-loop region, the patient is unlikely to respond to further imatinib therapy. Guidelines recommend that mutational analysis be performed in any case of suboptimal response or treatment failure, including a confirmed increase in BCR-ABL1 transcript level, to inform subsequent therapeutic choices.^15,16 In the ELN guidelines, development of a highly imatinib-resistant mutation indicates imatinib failure, whereas development of other mutations constitutes a suboptimal response.¹⁵

A recent survey of practice patterns for diagnosis and management of CML among US and European physicians found that, in contrast to the published guidelines, one-third of US physicians and one-quarter of European physicians surveyed believed that the most critical time point for evaluating treatment response to determine a subsequent treatment strategy was at 3 months, rather than at multiple time points.³¹ In particular, physicians were confused about the optimal timings for response assessment, with 15% to 19% reporting that they performed cytogenetic assessment every 12 months and 31% to 38% reporting 6-month PCR testing. Monitoring at these time points rather than at those provided within the guidelines may delay identification of patients whose response is poor. In addition, 73% of US physicians and 50% of European physicians did not use or were unfamiliar with BCR-ABL1 mutational analysis. These findings illustrate that broader communication of guidelines is required.

THERAPEUTIC OPTIONS AFTER SUBOPTIMAL RESPONSE TO AND FAILURE OF IMATINIB THERAPY

Guidelines suggest that alternative treatment strategies may be considered not only after treatment failure but also after suboptimal response.^15,16 After initial treatment with imatinib at 400 mg/d, subsequent therapeutic options include imatinib dose escalation (to 600 or 800 mg/d), dasatinib, nilotinib, allogenic SCT, or clinical trial with an investigational agent.

Imatinib Dose Escalation

Higher doses of imatinib, when tolerated, may be considered for patients in whom imatinib therapy fails or produces suboptimal responses. In the United States, imatinib dose escalation from 400 to 600 mg/d is approved for patients with CP-CML (escalation of 600 to 800 mg/d is approved for advanced disease).³ In the European Union, escalation to 800 mg/d is approved for patients with CP-CML.³

The efficacy of dose-escalated imatinib for patients whose disease is resistant to standard dosing was initially reported in a study of 54 patients with CP-CML.³² Patients with hematologic resistance (no CHR within 3 months) or cytogenetic resistance (no MCyR within 12 months) to standard-dose imatinib (300-400 mg/d) were treated with 600 or 800 mg/d. Of 20 patients treated for hematologic resistance or relapse, 13 (65%) achieved a complete or partial hematologic response. Among the 34 patients with cytogenetic resistance or relapse, 19 (56%) achieved an MCyR. The effects of imatinib dose escalation in the IRIS trial have also been assessed. In a post hoc analysis of patients whose imatinib therapy was escalated from 400 mg/d to 600 or 800 mg/d, 6 (86%) of 7 patients with no CHR at 3 months subsequently achieved a CHR, including 2 of 7 who achieved a CCyR. Also, 4 (50%) of 8 patients improved from a minor cytogenetic response (35%-90% Philadelphia chromosome positive) at 12 months to a CCyR, and 9 (50%) of 18 reachieved an MCyR (3/18 also achieved a CCyR).³³ Preliminary studies of imatinib dose escalation have been performed in patients with a suboptimal response to initial therapy.^34-36 These studies indicate that a proportion of patients can subsequently achieve a CCyR and MMR, although variable response rates were reported in different studies. Longer follow-up is required to confirm these findings, particularly in comparison with patients without modification of therapy.

Potential issues with imatinib dose escalation have been reported. In 2 studies, investigators reported that responses after a dose increase to 600 or 800 mg/d after failure or relapse may be transient.^37,38 However, in a recent study with longer follow-up, durable responses were reported, and median time to relapse among patients whose disease responded to dose escalation was 68 months.³⁴ Poor responses to dose escalation have been observed among patients who had no cytogenetic response (100% Philadelphia chromosome positive) to initial treatment, with 0% to 6% of such patients achieving any improvement in cytogenetic response.^37,39 Higher doses of imatinib may also be associated with decreased tolerability.⁴⁰ Imatinib dose escalation is likely to be most useful in patients who have low-level resistance to imatinib (eg, those with insufficient plasma levels, BCR-ABL1 gene amplification, or weakly resistant mutations such as M351T).^15,41,42 However, many common BCR-ABL1 mutations (eg, P-loop mutations G250E, Y253F/H, and E255K/V or others, including T315I, F359V, or H396R) are more resistant to imatinib and are unlikely to be overcome by a dose increase of 1.5- to 2-fold.¹⁵ Because of these issues, alternative treatments should be considered for patients in whom imatinib therapy fails or for patients whose responses are suboptimal.

Dasatinib

Dasatinib is a potent, orally available inhibitor of the Bcr-Abl1 and Src family kinases that is approved for use in patients with all phases of CML who have resistance or intolerance to imatinib.^43,44 Dasatinib has 325-fold greater potency than imatinib against unmutated Bcr-Abl1 in vitro.⁴⁵ The efficacy of dasatinib in patients with imatinib resistance or intolerance was shown in the Src/Abl1 Tyrosine Kinase Inhibition Activity: Research Trials of Dasatinib (START) program, which now has more than 2 years of follow-up (Table 2).^46-56 In particular, in the randomized START-R trial of patients with CP-CML with resistance (failure) to 400 or 600 mg/d of imatinib, dasatinib treatment demonstrated superior responses to imatinib dose escalation to 800 mg/d.^49,53 After 12 weeks of treatment (primary end point), dasatinib treatment resulted in higher rates of MCyR (36% vs 29%; P=.40) and CCyR (22% vs 8%; P=.041), and differences in response rates were further increased with extended follow-up. Notably, responses to dasatinib were more durable (Figure⁴⁹), including a significantly extended time to treatment failure (24-month rates, 59% vs 18%; P<.0001) and progression-free survival (24-month rates, 86% vs 65%; P=.0012). Among patients who did not achieve a prior cytogenetic response during previous imatinib treatment, 51% achieved an MCyR while taking dasatinib compared with 7% taking high-dose imatinib.⁵³ In addition, MCyR rates were higher for dasatinib in patients with (46% vs 27%) or without (58% vs 34%) any baseline BCR-ABL1 mutation.

TABLE 2.

Patients' Hematologic and Cytogenetic Responses to Dasatinib and Nilotinib After Imatinib Therapy Failure (Resistance or Intolerance)^a

FIGURE.

Duration of major cytogenetic response in the randomized phase 2 trial of dasatinib vs high-dose imatinib in patients with chronic-phase chronic myeloid leukemia after failure of standard-dose imatinib therapy. From Blood.⁴⁹

In vitro studies have shown that dasatinib can bind and inhibit all Bcr-Abl1 mutants except for T315I.^45,57 The ability of dasatinib to overcome imatinib-resistant mutations may be related to its higher potency compared with imatinib and its predicted ability, unlike imatinib, to bind active and inactive conformations of Bcr-Abl1.⁵⁸ Clinical resistance to dasatinib has been associated with mutations in amino acids V299, T315, and F317.^59-62

On the basis of findings of a phase 3 dose-optimization trial, which showed equivalent efficacy and improved tolerability, the approved dosage of dasatinib in patients with CP-CML and imatinib failure is now 100 mg/d.^63,64 Eligibility criteria for the dose-optimization trial included ELN definitions of suboptimal response (ie, lack of CHR, MCyR, and CCyR at 3, 6, and 12 months, respectively), indicating that dasatinib is also effective in this patient group. A trial comparing the efficacy of dasatinib or imatinib dose escalation in patients with suboptimal response to 400 mg/d of imatinib is currently recruiting patients (NCT00320190).

Nilotinib

Nilotinib is an alternative second-generation drug that was developed by modifying the chemical structure of imatinib, thereby increasing the specificity for Bcr-Abl1. Nilotinib is 20- to 50-fold more potent than imatinib against Bcr-Abl1 in vitro and, like imatinib, binds only to the inactive conformation of the enzyme, preventing it from adopting the catalytically active conformation.^45,65 Across several phase 2 studies,^54-56,66,67 nilotinib has shown efficacy in patients with CML after imatinib resistance or intolerance (Table 2)^50-56 and has subsequently been approved for treating patients with CP-CML or accelerated-phase CML who have imatinib resistance or intolerance.⁶⁸ Nilotinib has been associated with treatment responses in patients in whom treatment with both imatinib and dasatinib has failed.⁶⁹ Among 29 patients with CP-CML, 28% achieved an MCyR with nilotinib treatment, including 11% with a CCyR. Among patients with accelerated-phase (n=16) or blast-phase CML (n=25), an MCyR was achieved by 13% and 12%, respectively.

In vitro studies show that nilotinib inhibits all Bcr-Abl1 mutants, except for T315I, although certain mutations occurring in the P-loop region (Y253F/H, E255K/V) and in amino acid F359 are 10- to 35-fold less sensitive to nilotinib than unmutated Bcr-Abl1.⁴⁵ Clinical studies have mirrored in vitro findings, with mutations in amino acids Y253, E255, T315, or F359 found in patients in whom nilotinib failed,⁶² and no CCyRs were recorded among patients with any of these mutations during a mutational analysis of the phase 2 trial in CP-CML.⁷⁰

Nilotinib may be a treatment option for patients with suboptimal responses to imatinib. To specifically investigate the efficacy of nilotinib therapy in this patient group compared with an escalated dose of imatinib (800 mg/d), a phase 3, randomized, open-label trial of patients with CP-CML who have had a suboptimal cytogenetic response to imatinib, 400 mg/d, is under way (NCT00519090).

Choosing Between Dasatinib and Nilotinib

Currently, no guidelines are available to assist with the selection of second-line TKI treatment. Several factors, including response rates, type of imatinib-resistant BCR-ABL1 mutation, and tolerability, could influence treatment choice after suboptimal response or imatinib failure. However, any comparisons between dasatinib and nilotinib data should be made with extreme caution given that differences exist among patient populations and designs of clinical trials.

On the basis of the data from phase 2 clinical trials, response rates are similar. For patients with CP-CML, cumulative CCyR rates for dasatinib vs nilotinib were 33% vs 31% at 6 months and 49% vs 41% at 12 months, respectively.^46,54,66,71 For patients with accelerated-phase CML, the 6-month CHR rates were 33% vs 26%, and the CCyR rates were 22% vs 16% for dasatinib vs nilotinib.^47,55 Currently, dasatinib but not nilotinib is approved for use in patients with blast-phase CML.

BCR-ABL1 mutational analysis may be a more useful guide. Because mutations in specific amino acids have been associated with clinical resistance to either dasatinib (V299, F317)^59-62 or nilotinib (Y253, E255, F359),^62-70 treatment with the agent in question may not be optimal if these mutations are detected during imatinib treatment. Patients with T315I mutations would be unlikely to benefit from dasatinib or nilotinib.

Dasatinib and nilotinib are both generally well tolerated, and in most cases, adverse events resolve with drug interruption and/or dose reduction. With both agents, cytopenias were the most common adverse event, and in patients with CP-CML, similar 6-month rates of grade 3/4 thrombocytopenia and neutropenia were reported with dasatinib, 100 mg/d (22% and 33%, respectively), and nilotinib (29% for each). Pleural effusions are rare with nilotinib therapy but did occur in 7% of patients with CML-CP treated with dasatinib at 100 mg/d (grade 3/4 in 1%),⁶³ with higher rates observed in patients with advanced disease (14%-23%) who received 70 mg twice daily.^47,48 Possible risk factors for dasatinib-associated pleural effusion include a history of cardiac disease, hypertension, or autoimmune disease.^72,73 Nilotinib is associated with biochemical abnormalities, with grade 3/4 lipase and bilirubin level elevations reported in 9% and 14% of patients with CP-CML, respectively.⁶⁶ Other nonhematologic events are broadly similar with both agents, with diarrhea perhaps more common with dasatinib, rash and pruritus more common with nilotinib, and grade 3/4 events rare with both drugs. A regulatory warning exists for nilotinib that relates to QT prolongation and sudden deaths during clinical trials.⁶⁸ During 6 months of follow-up of patients with CP-CML, discontinuation rates after adverse events were 4% with dasatinib at 100 mg/d and 15% with nilotinib.^63,66 Because of increased bioavailability, nilotinib-treated patients should avoid food 2 hours before and 1 hour after taking their tablets,⁶⁸ which may be inconvenient. No clinically relevant food effects have been reported for dasatinib.

Allogeneic Hematopoietic SCT

Allogeneic SCT is the only potentially curative treatment for CML. However, since the development of modern TKIs, allogeneic SCT is more commonly performed after patients have received imatinib and second-generation agents and is an important treatment option after treatment failure.⁷⁴ In a recent single-institution study, 2-year survival rates for patients with CP-CML in whom imatinib treatment had failed were better when subsequent therapy was dasatinib or nilotinib vs allogeneic SCT (100% vs 72%).⁷⁵ Prior imatinib treatment does not adversely affect overall survival, progression-free survival, or nonrelapse mortality after allogeneic SCT,⁷⁶ and second-generation TKIs do not appear to increase transplant-related toxic effects.⁷⁷ A recent study found that prior imatinib therapy was associated with better survival among patients with newly diagnosed CP-CML.⁷⁸ Allogeneic SCT is probably the best available therapy for eligible patients with a T315I mutation. Although allogeneic SCT has been suggested as a potential therapeutic option after suboptimal response in some guidelines,^15,19 limited clinical data are available specifically for this patient group.

Investigational Agents

Several novel agents are currently being investigated in CML, with some in clinical trials. These agents include the TKIs bosutinib (SKI-606) and INNO-406 (NS-187), neither of which is able to inhibit T315I-mutated Bcr-Abl1.^79-82 A number of agents that target the difficult to treat T315I mutant are being developed, including MK-0457 and XL228.^83,84 Nonkinase inhibitors are also being used in clinical trials. These include homoharringtonine (a protein synthesis inhibitor), tipifarnib and lonafarnib (farnesyl transferase inhibitors), and decitabine (a DNA methylation inhibitor).^85-88 Current guidelines suggest that patients may be considered for clinical trials with novel agents after failure of imatinib or other approved therapies.

Reconsidering Treatment Strategies

Response goals in CML and the time points when alternative treatment should be considered are an area of debate. Recent IRIS data have shown that patients who achieved a CCyR at any time had good long-term outcomes regardless of when the CCyR was achieved. However, patients who did not achieve a CCyR had higher rates of disease progression, a substantially shorter duration of MCyR, and lower survival rates.⁸⁹ In addition, it is well established that patients who do not respond within a given time frame have an increased risk of progression.^4,5,9-14 These data provide a clear rationale for modifying treatment in patients who do not experience an optimal response.

Recent guidelines suggest that patients who have a suboptimal response to imatinib at 400 mg/d should be considered for imatinib dose escalation as tolerated; alternative therapies should be used only after imatinib failure, with imatinib dose escalation also appropriate at this point if not already considered.^15,16,19 However, data are limited to support this sequence of treatments, and further investigation is necessary to determine which potential response milestones would be optimal when switching from imatinib to a second-line agent. Clinical trial data have confirmed the efficacy of dasatinib and nilotinib in patients with imatinib resistance or intolerance and the superiority of dasatinib vs imatinib dose escalation after imatinib resistance. An earlier treatment switch to second-line agents (ie, after suboptimal response) could possibly result in more favorable long-term outcomes than with dose-escalated imatinib; however, no clinical data are currently available to support this hypothesis. Interestingly, a recent analysis of dasatinib trial data in CP-CML found that earlier initiation of treatment with dasatinib after relapse while taking imatinib (ie, after loss of MCyR rather than after loss of both MCyR and CHR) led to higher response rates and extended progression-free survival.⁹⁰

Patients with specific disease characteristics may derive the greatest benefit from an earlier switch in treatment. Response rates to imatinib dose escalation are low among patients who achieved no cytogenetic response with initial imatinib treatment.^37,39 In the START-R trial, response rates among patients with no prior cytogenetic response were substantially higher with dasatinib than with imatinib dose escalation.⁵³ BCR-ABL1 mutations should also be considered during treatment decisions. In START-R, response rates among patients with mutations were higher with dasatinib than with imatinib dose escalation, indicating that second-line TKIs may provide a better overall chance of response. As previously discussed, the characteristics of specific mutations within BCR-ABL1 may assist with treatment selection. In ELN guidelines, development of a highly imatinib-resistant BCR-ABL1 mutation indicates that imatinib treatment should be discontinued.¹⁵ However, detecting any mutation during imatinib treatment has been shown to predict subsequent relapse,⁹¹ providing an early indication of patients who are likely to require a subsequent switch in treatment and supporting the detection of a mutation as a definition of a suboptimal response. Patients with a T315I mutation are resistant to all currently approved TKIs and should be considered for allogeneic SCT, a clinical trial with a novel agent, or older therapies (eg, interferon alfa).

A recent study has hinted at possible goals for second-line treatment. Of patients with CP-CML who received dasatinib (n=70) or nilotinib (n=43) after failure of imatinib therapy, those who achieved an MCyR within 12 months after imatinib treatment had a significantly higher survival rate than patients with a minor cytogenetic response or CHR (1-year survival rates, 97% vs 84%; P=.02).⁹² Moreover, less than 10% of patients with no cytogenetic response of any level within 3 to 6 months achieved an MCyR at 12 months, suggesting that these patients might require a switch to third-line treatment. However, additional studies are required to confirm these findings.

CONCLUSION

To identify patients whose disease is not responding optimally to imatinib, physicians must monitor patients regularly using methods recommended in published guidelines; however, molecular and mutational assessments should be interpreted with due caution. Definitions of failure and suboptimal response during imatinib therapy for CP-CML have been established, but the treatment strategies that should be used are less clear. Imatinib dose escalation is an option; however, clinical data that show an advantage over alternative treatment options are lacking. Dasatinib has proven efficacy compared with high-dose imatinib in patients whose disease is resistant to standard-dose imatinib and has been used to treat patients with a suboptimal response to imatinib during clinical trials. Nilotinib has proven efficacy in patients whose disease is resistant to standard-dose imatinib; however, its relative benefits compared with dose escalation are not yet proven. Treatment selection should be guided, if possible, by patient or disease characteristics. Clinical trials are required to determine the relative efficacy of dasatinib, nilotinib, and dose-escalated imatinib in patients with a suboptimal response to initial imatinib treatment. Until additional clinical data are available, treatment selection and time points for switching treatment are likely to remain areas of debate.