The most common adverse events of imatinib 400 mg reported from the IRIS trial [4] and the Japanese CML202 trial [17] were edema (peripheral and periorbital edema) (60% and 49%), muscle cramps (49% and 17%), diarrhea (45% and 16%), nausea (50% and 22%), musculoskeletal pain (47% and 21%), rash (40% and 40%), fatigue (39% and 24%), and headache (37% and 8%), respectively. Incidence of grade 3 or 4 adverse events in both trials was neutropenia (17%, 18%), thrombocytopenia (9%, 12%), anemia (4%, 6%), and elevated liver enzymes (5%, 4%), respectively. Several types of vascular adverse events have been reported in patients receiving second- or third-generation BCR-ABL1 TKIs [21]. However, imatinib exhibits a favorable long-term safety profile. The vascular safety of imatinib therapy is well documented and occurrence of congestive heart failure has been rare (1.7%) [22]. Newly occurring or worsening grade 3 or 4 hematologic or biochemical adverse events were infrequent both after 2 and 4 years of treatment. Imatinib is generally well tolerated; most adverse events were grade 1 or 2 in severity and tended to be most frequent in the first year of treatment. Discontinuation of imatinib for drug-related adverse events was required in <4% of patients over a median follow-up duration of 60 months.

The recommended starting dose of imatinib for patients with CML-CP is 400 mg/day, while that for patients with CML-AP or BP is 600 mg/day. Efficacy evaluation of imatinib 800 mg/day was performed in a phase 3, randomized study (TOPS trial). Newly diagnosed patients with CML-CP were treated with either 800 mg/day or 400 mg/day of imatinib [23, 24]. The primary end point was MMR at 12 months. MMR rates were similar in both treatment arms at 12 months (51.6% vs. 50.2%; P = 0.77) and 42 months (75.8% vs. 79.0%; P = 0.48). Further, no significant between-group difference was observed with respect to EFS, PFS, or OS. However, patients who tolerated more than 600 mg/day of imatinib showed a higher response rate.

In the ELN study, 216 patients with newly diagnosed CML at high Sokal risk were randomly assigned to receive imatinib 800 or 400 mg/day for at least 1 year [25]. The primary end point of CCyR at 12 months was comparable between the two treatment arms, as were the OS, PFS, and EFS. In the high-dose arm, the median average daily dose was 720 mg (range: 350–800 mg).

In a phase 2 study (TIDEL trial) of higher-dose imatinib, patients with newly diagnosed CML-CP initiated on imatinib 600 mg/day were shifted to 800 mg/day, if the prespecified response criteria were not met [26]. The primary end points were MMR, MCyR, and CCyR at 2 years. Both TIDEL trial and IRIS trial showed comparable MMR rates at 12 months (47% and 40%, respectively). However, the MMR at 24 months was higher in TIDEL trial (73% vs. 55% in the IRIS trial). Superior response in patients who were able to tolerate imatinib at 600 mg suggests that early dose intensity may be critical to achieve optimal response in CML-CP.

Overall, the current body of evidence alone does not justify extensive use of higher-dose imatinib (800 mg daily) as a frontline therapy in patients with CML-CP. Indeed, treatment adherence appears to be more important than initiation of treatment at a higher dose.

Subgroup analyses of a Japanese phase 2 study (CML202 trial) were performed according to the mean daily dose during the first 24 months of treatment: ≥360 mg (400 mg group; n = 294), 270–359 mg (300 mg group; n = 90) and <270 mg (200 mg group; n = 67). This was because imatinib dosage was reduced in many patients due mainly to adverse events [17]. No significant between-group differences were observed in OS and EFS between the 300- and 400-mg groups. These results indicate that long-term outcomes in small, elderly, and/or female patients who received 300 mg imatinib daily were similar to those who received 400 mg/day. However, cumulative rates of cytogenetic (92% vs. 98%; P = 0.018) or molecular response (78% vs. 87%, P = 0.017) at 7 years in the 300-mg group were inferior to those in the 400-mg group. Further, survival and efficacy was markedly inferior in the 200-mg group. Therefore, excessive dose reduction (<300 mg/day) should be avoided even in patients who are intolerant to imatinib 400 mg/day or in those who have a small body size.

IFNα can induce sustained cytogenetic response in some patients with CML-CP [27]. Furthermore, Essers et al. reported that acute stimulation with IFNα activates dormant hematopoietic stem cells, while chronic stimulation leads to their exhaustion [28]. In a German CML Study IV, a randomized 5-arm trial designed to optimize imatinib therapy, 465 patients were randomized to imatinib alone or imatinib plus IFNα 2a or 2b [29]. Imatinib + IFNα failed to show superiority over imatinib monotherapy with respect to MMR and 3-year OS.

In a French phase 3 study (SPIRIT trial), 636 newly diagnosed patients with CML-CP were randomly assigned to receive imatinib (400 mg daily) alone, imatinib (400 mg) plus cytarabine or pegylated interferon (peginterferon) α-2a (90 μg weekly), or imatinib 600 mg daily alone [30]. The primary end points were EFS, PFS, and OS. At 12 months, the rate of molecular response was significantly higher in the imatinib + peginterferon α-2a group than that in imatinib 400 mg alone group (30% vs. 14%; P = 0.001). However, the combination was associated with substantial toxic effect, because of which peginterferon α was discontinued in 46% of the patients.

The Nordic randomized phase 2 study compared imatinib (400 mg daily) monotherapy with imatinib plus peginterferon α-2b (50 μg weekly) after imatinib induction in patients with CML-CP with low and intermediate Sokal risk [31]. The primary end point of MMR rate at 12 months was significantly higher in the former group, when compared with imatinib-alone group (82% vs. 54%, P = 0.002). Peginterferon α-2b was discontinued in 61% of patients in the combination arm. Therefore, the combination of peginterferon α with imatinib, despite the increase in associated toxicity, seems to enhance deep molecular response rates and should offer more possibilities for TKI therapy. These results suggested that addition of peginterferon-α added value to imatinib treatment but were not associated with superior PFS or OS.

Oral bioavailability of imatinib is 98%, and its half-life is approximately 20 h. Therefore, imatinib is administered orally in a once daily regime in patients with CML. Imatinib is metabolized mainly by CYP3A4 isoenzyme. The pharmacokinetics of imatinib tends to show considerable interpatient variability [32]. The mean plasma trough level of imatinib at day 28 was slightly higher in females than that in males (1078 ± 515 ng/mL vs. 921 ± 531 ng/mL). A weak correlation of imatinib trough levels with body weight, body surface area, and the age of patients was observable. However, Larson et al. suggested that effects of these factors on imatinib trough levels are not likely to be clinically significant due to the large interpatient variability in plasma trough concentrations.

The imatinib trough levels were significantly higher in patients who achieved CCyR than those who did not (mean, 1009 ng/mL vs. 812 ng/mL, P = 0.01). Thus, an imatinib trough level of ˃1000 ng/mL appears to be important for achievement of CcyR. Picard et al., too, reported a threshold of 1002 ng/mL to achieve MMR [33].

Predictive variables for risk of progression and probability of drug discontinuation are required for patients with CML. Three prognostic classifications are currently available, and two of these (the Sokal score and the Euro [Hasford] score) have been used in studies of TKIs [34, 35]. However, the Sokal score was developed for patients treated with conventional chemotherapy, while the Euro score was based on IFNα therapy. In the European Treatment and Outcome Study for CML (EUTOS), a new prognostic risk score based on a study of 2060 imatinib-treated patients was developed to predict the probability of achieving CCyR within 18 months (Table 3.3) [36]. The EUTOS score relies on the percentage of basophils and spleen size to differentiate high-risk from low-risk patients. Other variables (age, platelet count, blast cells, and eosinophils) used in the previous classifications were later shown not to affect the response to imatinib. The EUTOS score predicted that 34% of high-risk patients would fail to achieve a CCyR in 18 months and also predicted a significant difference in PFS (82% for high-risk vs. 90% for low-risk patients, P = 0.006).

In the German CML Study IV, the relationship between comorbidities at diagnosis and overall prognosis were assessed using the Charlson Comorbidity Index (CCI) [37]. Age was associated with a higher CCI score in 863 patients, while comorbidities had an impact on OS. However, comorbidities had no negative effect on response rates and progression to advanced phases. Therefore, OS may not be appropriate for assessment of outcome measures for TKI therapy in CML-CP.

As biological variables, translocation type (standard and variant), additional chromosomal alteration (del der 9, trisomy 8, isochromosome 17, additional loss of material from 22q, and double Ph), transcript type, and baseline BCR-ABL1 kinase domain mutations have been reported. Regarding the leukemic stem cell burden, Mustjoki et al. reported a correlation between the proportion of Ph-positive cells in CD34⁺ CD38⁻ fraction at diagnosis and the cytogenetic and molecular response [38].

Novel biological prognostic biomarkers for clonal evolution of CML have been identified by use of gene expression profiles, whole exome sequencing, and protein-level profiling.

Radich et al. identified 3000 genes that showed a significant association with the disease phase on DNA microarray analysis [39]. Their findings suggest the progression of CP CML to advanced phase CML as a two-step process. Patients with gene expression patterns of advanced disease may benefit from more aggressive therapies. Recently, targeted deep sequencing has been performed in Ph-positive clones, and mutations of ASXL1, DNMT3A, RUNX1, and TET2 have been identified [40]. These BCR-ABL1-independent gene mutations in patients with CML may be important cofactors in the clonal evolution of CML.

Monitoring the response to imatinib requires absolute and differential blood counts, cytogenetic tests, and molecular testing for BCR-ABL1 transcript level and for ABL1 tyrosine kinase domain mutations [7]. According to ELN recommendations for imatinib therapy, blood counts and differentials are required frequently during the first 3 months until CHR has been confirmed. Cytogenetic testing is required at 3 and 6 months, every 6 months thereafter until confirmation of CCyR, and every 12 months thereafter. Real-time quantitative polymerase chain reaction (RQ-PCR) for assessment of BCR-ABL1 transcript levels is recommended every 3 months until MMR has been achieved and confirmed and, at least, every 6 months thereafter (Table 3.4).