Flumatinib versus Imatinib for Newly Diagnosed Chronic Phase Chronic Myeloid Leukemia: A Phase III, Randomized, Open-label, Multi-center FESTnd Study

Over the past decades, BCR-ABL1 tyrosine kinase inhibitors (TKI) as the initial treatment for chronic phase Philadelphia chromosome–positive (Ph+) chronic myeloid leukemia (CML-CP) have dramatically turned CML-CP into a manageable and potentially curable chronic disease. The life expectancy of patients with chronic myeloid leukemia (CML) with optimal response is now approaching that of general population (1).

Although the long-term outcomes of more than 10 years of follow-up of imatinib in recent updates of both the IRIS and CML-IV study were very promising (2, 3), newer TKIs including second-generation TKIs (2G-TKIs), such as nilotinib, dasatinib, bosutinib, and radotinib, have been more attractive due to their higher rates of responses, and faster and deeper responses compared with imatinib (4–8). There was a trend for small improvement in relative survival with deeper cumulative response achieved within 1 year such that 10-year relative survival was 88.2% for patients irrespective of response status, 92.1% for those patients with complete cytogenetic response (CCyR, defined as 0% Ph+ metaphases by conventional cytogenetics), 94.2% for those achieving major molecular response (MMR, BCR-ABL1/ABL1 ≤ 0.1% on the international scale), 94% for patients with MR4.5 (BCR-ABL1/ABL1 ≤ 0.0032% on the international scale), and 97.3% for patients with undetectable transcripts (9). Regardless of treatment approach, confirmed MR4.5 at 4 years predicted significantly higher survival probabilities than BCR-ABL transcription level on the international scale 0.1%–1% (10). As described previously (11), the strongest and least controversial benefit of deep molecular response (DMR) for patients is the potential for treatment discontinuation. A sustained DMR (typically defined as MR4.5 sustained for at least 2 years and ideally longer) is an important prerequisite for considering an attempt at treatment discontinuation (12). The 5-year cumulative rate of MR4.5 is significantly higher with 2G-TKIs than with imatinib (3, 6). A multivariate analysis showed that among factors associated with achievement of sustained MR4.5, treatment with a 2G-TKI was an independent favorable predictive factor for the probability of achievement of sustained MR4.5 (13). Such responses were also achieved significantly earlier with 2G-TKIs (14). This suggests that considerably more patients receiving a 2G-TKI would be eligible for an attempt at treatment discontinuation and that criteria for treatment discontinuation would be reached earlier.

Despite these favorable improvements of efficacy, adverse events (AE) associated with both imatinib and 2G-TKIs are increasingly important as TKI therapy is lifelong for most patients. Long-term use of 2G-TKIs has been associated with AEs such as pleural effusion and cardiovascular events (15, 16), while imatinib has been shown to cause both acute and chronic decline in glomerular filtration rate (17), which may increase disease morbidity or mortality, and even less severe AEs associated with long-term use of imatinib, such as fatigue or musculoskeletal pain, may affect patients' quality of life (QoL; ref. 18). Thus, room for improvement for a better TKI should not only refer to efficacy, but also safety.

Flumatinib mesylate is a newer and orally available TKI with higher selectivity and potency against BCR-ABL1 kinase versus imatinib. Aimed at optimizing imatinib's properties, the phenyl ring of imatinib was replaced with a pyridine group and a trifluoromethyl group was introduced on the basis of the molecular framework of imatinib (19). Flumatinib demonstrated that it blocked BCR-ABL1 kinase autophosphorylation with much more potent activity than did imatinib, while it inhibited the activity of c-Kit and PDGFR kinase with less potency than imatinib and had no effect on the phosphorylation of EGFR, VEGFR, c-Src, or HER2 kinase (20). Flumatinib also demonstrated that its trifluoromethyl was accommodated within the hydrophobic pocket of ABL kinase and strongly interacted with residues I293, L298, L354, and V379 in ABL kinase via hydrophobic interactions, while there was no hydrophobic interaction between imatinib and the hydrophobic pocket in ABL kinase (21). Flumatinib even showed a higher potency against wild-type BCR-ABL1 kinase than nilotinib because the position of the fluorine of flumatinib was more suitable for the hydrophobic pocket in ABL kinase than that of nilotinib (20–22). The better performance of flumatinib than both imatinib and nilotinib also benefited from the pyridine in flumatinib (21). Furthermore, flumatinib showed higher potency against mutant BCR-ABL1 kinase in vitro, especially for mutations in the ATP binding region of ABL kinase (such as V299L, F317L, and F317I; ref. 21).

In the unpublished phase Ia (in patients with CML in the accelerated phase or blast phase) and phase Ib (in TKI treatment–naïve patients with CML-CP) studies, flumatinib showed its substantial antileukemic activity and acceptable tolerability in patients with CML in the accelerated phase, blast phase, and chronic phase, and the dose-limiting toxicities occurred at 400 mg twice daily [Common Terminology Criteria for Adverse Events (CTCAE) grade 3 diarrhea] and 1,200 mg once daily (CTCAE grade 3 diarrhea). After single dosing, the average time to reach maximum plasma concentration (T_max) was 2 hours and the average elimination half-life (T_½) was 18 hours.

In an investigator-initiated, phase II, multi-center (four sites), randomized, and open-label trial, 24 newly diagnosed patients with CML-CP were treated with flumatinib 400 and 600 mg, both once daily, or imatinib 400 mg once daily for 24 weeks. The MMR rate of flumatinib 600 mg once daily group was higher than that of imatinib group after 24 weeks (P = 0.017). The rate of BCR-ABL1/ABL1 ^IS ≤10% in flumatinib 600 mg once daily group was significantly higher than that in imatinib group (P = 0.002; ref. 23). Similarly, in an extended Chinese SFDA-registered, unpublished phase II (19 sites), randomized, open-label, multi-center study, flumatinib at a dose of both 400 and 600 mg provided a better efficacy and tolerability compared with imatinib at a dose of 400 mg for the treatment of patients with newly diagnosed CML-CP. Also, as first-line treatment for newly diagnosed CML-CP, flumatinib at a dose of 600 mg once daily was more effective than that at a dose of 400 mg once daily, with a similar safety profile. This phase II study had to be extended to a registered phase III study due to sample size insufficiency.

In this phase III, randomized, open-label, multi-center trial, called evaluating Flumatinib Efficacy and Safety in clinical Trials–Newly Diagnosed CML-CP Patients (FESTnd) study, we compared the efficacy and safety of flumatinib (at a dose of 600 mg once daily on empty stomach) with that of imatinib (at a dose of 400 mg once daily with meal) in patients with newly diagnosed CML-CP, with the rates of MMR at 6 months as the primary endpoint.

Adult patients ages between 18 and 75 years within 6 months after the diagnosis of CML-CP at 25 sites were eligible. Diagnosis was confirmed by the presence of at least one Ph+ metaphase cell via conventional cytogenetic assessment, and/or the presence of P210 BCR-ABL1 transcripts via molecular assessment. Chronic-phase CML was defined according to the European LeukemiaNet recommendations (24, 25). Patients needed to have an adequate organ function with an Eastern Cooperative Oncology Group (ECOG) performance status of at least 2 (26).

Like the ENESTnd trial (27), patients were excluded if they had received treatment with a TKI before study entry (except imatinib for ≤2 weeks) or any medical treatment for CML for more than 2 weeks (except hydroxyurea or anagrelide). Patients with impaired cardiac function were excluded, including uncontrolled angina, clinically documented myocardial infarction during last 12 months, low left ventricular ejection fraction <45%, complete left bundle branch block, use of a ventricular-paced pacemaker, congenital long QT syndrome or a known family history of long QT syndrome, QTcF > 450 milliseconds (male) or QTcF > 470 milliseconds (female), presence of clinically significant ventricular or atrial tachyarrhythmias, clinically significant resting bradycardia (<50 beats/minute), and congestive heart failure. The prescription of therapeutic coumarin derivatives, drugs that block or stimulate the activity of the liver enzyme cytochrome P450-3A4, or any medication with the potential to prolong the QT interval were prohibited. Patients who had to be treated with any of these medications were excluded.

The study was designed by representatives of the sponsor, Hansoh Pharmaceuticals, with inputs from the investigators on the study management committee. The study protocol and its amendments were approved by the institutional review board at each site, and the trial was conducted in accordance with the protocol and the Declaration of Helsinki and the International Conference on Harmonisation Guidelines for Good Clinical Practice. All patients provided written informed consent. The data were collected with the use of outsourcing data management systems and were analyzed and interpreted by an outsourcing biostatistician (Yuanyuan Kong, Clinical Epidemiology and EBM Unit, Beijing Friendship Hospital, Capital Medical University, Beijing, China) in close collaboration with the other investigators. A data and safety monitoring board reviewed the trial data and made recommendations regarding the continuation of the study.

Patients with newly diagnosed CML-CP were randomly assigned in a 1:1 ratio to receive flumatinib (at a dose of 600 mg once daily on empty stomach) or imatinib (at a dose of 400 mg once daily with meal; Supplementary Fig. S1). Similar to the ENESTnd trial (27), randomization was stratified according to the Sokal risk score at the time of diagnosis. The Sokal score is based on age, spleen size, and peripheral blood platelet count and blast count (28). Patients are classified as being low risk (Sokal score < 0.8), intermediate risk (0.8–1.2), or high risk (>1.2).

On the basis of investigators' decision, dose reduction and withholding of both flumatinib and imatinib due to intolerable hematologic or nonhematologic toxicities were allowed until the toxicities had resolved, and then the treatment medications were resumed. Therapy could be terminated because of treatment failure (including progression), intolerable AEs, or other reasons. Unlike the IRIS study (29), cross-over was not permitted.

Molecular response was assessed for BCR-ABL1 by means of RT-PCR at baseline and every 3 months after treatment. Conventional bone marrow cytogenetic analyses were performed at baseline and at months 3, 6, and 12. BCR-ABL mutation was assessed by Sanger sequencing in a patient with disease progression and treatment failure. Complete blood counts were measured at baseline and monthly after treatment, until study completion (28 December 2017 for the last recruited patient).

The primary efficacy endpoint was the rate of MMR at 6 months, defined as a BCR-ABL1 transcript level ≤0.1% in peripheral blood on RT-PCR assay, as expressed on the international scale (30–34). RT-PCR assays were performed in a central laboratory (KingMed, with kit provided by MolecularMD), with the ABL gene as reference gene. The assay was standardized through an exchange of samples from patients with the molecular laboratory in Adelaide, Australia. The international scale conversion factor was 0.81. Consideration of earlier timepoint (6 months) to the attainment of MMR as the primary efficacy endpoint was because the rate of DMR achievement was significantly higher and the time to DMR was significantly faster for patients with MMR by 6 months compared with those with MMR at >6 months (35, 36). These findings suggested that DMR was strongly related to MMR by 6 months, and clinical data obtained to date indicate that for many patients who achieved DMR on TKI therapy, treatment-free remission is a safe treatment goal (37).

The secondary endpoints included rates of molecular responses over early term, including the rates of early molecular response (EMR, BCR-ABL1/ABL1 ≤ 10% on the international scale), MR2 (BCR-ABL1/ABL1 ≤ 1% on the international scale), and MMR at 3 months; rates of MMR at 9 and 12 months; rates of MR4 (BCR-ABL1/ABL1 ≤ 0.01% on the international scale) and MR4.5 at 6, 9, and 12 months; rates of CCyR at 3, 6, and 12 months; time to MMR and CCyR; and progression (24) to accelerated phase/blast crisis (AP/BC) or death due to CML while on treatment.

Cytogenetic assays with chromosome banding analysis (CBA) of Giemsa-stained metaphases from bone marrow cells were performed in local laboratories at each site.

AEs were assessed continuously for all treated patients and were graded according to the CTCAE, version 4.03, of the NCI (evs.nci.nih.gov/ftp1/CTCAE/CTCAE_4.03/CTCAE_4.03_2010-06-14_QuickReference_5 × 7.pdf).

The efficacy analysis populations included the intention-to-treat (ITT) population for molecular responses, defined as all patients who received ≥1 dose of study medication (flumatinib, n = 196 and imatinib, n = 197), and the evaluable population for cytogenetic responses, defined as patients who received ≥1 dose of study medication and had available cytogenetic data at each landmark timepoints. One patient in the imatinib arm with Ph+ chromosome but negative P210 BCR-ABL transcript was excluded from the molecular assay. The primary endpoint was tested at a significance level of 0.05 with the use of the Cochran–Mantel–Haenszel test, with adjustment for stratification according to the Sokal score. The times to a CCyR and to a MMR were calculated with the use of the Kaplan–Meier product-limit method in the ITT population and were compared between arms with stratified Cox proportional hazards model according to the Sokal risk. Differences between the treatment groups in times to an event were evaluated with the use of a stratified log-rank test. Response rates were binomial. The safety analysis population was defined as all patients who received ≥1 dose of the study medication (flumatinib, n = 196 and imatinib, n = 198). All reported P values and 95% confidence intervals (CI) were two-sided.

Disease progression was defined as progression to accelerated or blast phases or CML-related death. Treatment failure was defined by the European LeukemiaNet 2009 recommendation (25), including less than complete hematologic response (CHR) at 3 months, no CCyR at 6 months, less than partial cytogenetic response (PCyR, defined as >0%–35% Ph+ metaphases by conventional cytogenetics) at 12 months, loss of CHR, loss of PCyR, or loss of CCyR at any time, appearance of BCR-ABL mutation, and acquisition of additional clonal cytogenetic aberrations in Ph+ clones.

Patients whose 6- and 9-month PCR data were missing, but both the previous (3 months for 6-month PCR data missing and 6 months for 9-month PCR data missing) and next (9 months for 6-month PCR data missing and 12 months for 9-month PCR data missing) PCR evaluations indicated MMR, MR4, or MR4.5, the 6- or 9-month assessments were considered corresponding “responses,” respectively. Patients whose 12-month PCR evaluation was missing, but the previous (9 months) PCR evaluation indicated MMR, MR4, or MR4.5, the 12-month assessments were considered corresponding “responses.” Any patients with 3-month PCR data missing were considered as nonresponders. Missing data for cytogenetic responses at each landmark timepoints were not handled, considering that missing cytogenetic data are frequent in China due to insufficient metaphases or bad patients' adherence to bone marrow aspirations.

SAS version 9.4 was used for all statistical analyses.

From 26 August 2014 to 30 December 2016, 394 eligible patients with newly diagnosed CML-CP were randomly assigned to receive flumatinib 600 mg once daily (n = 196) or imatinib 400 mg once daily (n = 198). The cut-off date for this study was 28 December 2017, based on the 12-month visit for the last recruited patient who underwent randomization.

Baseline characteristics and distributions of the Sokal risk score were well-balanced in the two study groups (Table 1). The median dose intensity of flumatinib was 593 mg per day (interquartile range, 509–600). The median dose intensity of imatinib was 400 mg per day (interquartile range, 353–400). At the time of data cutoff, the proportions of patients receiving a study drug were 81% in the group receiving flumatinib and 81% in the group receiving imatinib (Table 2). By the time of data cutoff, 37 (19%) patients in the group receiving flumatinib and 38 (19%) patients in the group receiving imatinib had discontinued treatment.

At 6 months, the rate of MMR (the primary endpoint) in the ITT population was significantly higher in patients receiving flumatinib than in those receiving imatinib (33.7%; 95% CI, 27.1%–40.8% vs. 18.3%; 95% CI, 13.1%–24.4%; P = 0.0006 for both comparisons; Fig. 1). The rates of MMR in the ITT population at 9 and 12 months were also significantly higher with flumatinib versus imatinib (45.9%; 95% CI, 38.8%–53.2% vs. 30.0%; 95% CI, 23.7%–36.9% at 9 months; P = 0.0011 and 52.6%; 95% CI, 45.3%–59.7% vs. 39.6%; 95% CI, 32.7%–46.8% at 12 months; P = 0.0102, respectively; Fig. 1). These results revealed that MMR was achieved more quickly among patients receiving flumatinib than among those receiving imatinib.

Patients in the flumatinib arm had significantly higher rates of DMRs and achieved DMRs faster compared with those in the imatinib arm. Compared with patients in the imatinib arm, more patients in the flumatinib arm achieved a DMR of MR4 at 6, 9, and 12 months (8.7%, 95% CI, 5.1%–13.5% vs. 3.6%, 95% CI, 1.4%–7.2% at 6 months, P = 0.0358; 16.8%, 95% CI, 11.9%–22.8% vs. 5.1%, 95% CI, 2.5%–9.1% at 9 months, P = 0.0002; and 23.0%, 95% CI, 17.3%–29.5% vs. 11.7%, 95% CI, 7.6%–17.0% at 12 months, P = 0.0034, respectively), as well as a DMR of MR4.5 numerically at 6 months (4.1%, 95% CI, 1.8%–7.9% vs. 1.5%, 95% CI, 0.3%–4.4%; and at 9 months, respectively), and with statistically significant at 9 and 12 months (6.1%, 95% CI, 3.2%–10.5% vs. 2.0%, 95% CI, 0.6%–5.1% at 9 months, P = 0.0429 and 10.7%, 95% CI, 6.8%–15.9% vs. 3.6%, 95% CI, 1.4%–7.2% at 12 months, P = 0.0064; Fig. 1).

Molecular responses over early term were superior among patients receiving flumatinib than those receiving imatinib. Rates of EMR, MR2, and MMR were higher among patients receiving flumatinib than those receiving imatinib at 3 months (for EMR: 82.1%, 95% CI, 76.1%–87.2% vs. 53.3%, 95% CI, 46.1%–60.4%, P < 0.0001; for MR2: 41.8%, 95% CI, 34.9%–49.1% vs. 12.7%, 95% CI, 8.4%–18.2%, P < 0.0001; and for MMR: 8.2%, 95% CI, 4.7%–12.9% vs. 2.0%, 95% CI, 0.6%–5.1%, P = 0.059, respectively; Fig. 2).

The rates of CCyR among cytogenetic data available patients at 3, 6, and 12 months were significantly higher for flumatinib than for imatinib (63.1%, 95% CI, 54.2%–71.4% vs. 42.5%, 95% CI, 34.7%–50.6% at 3 months, P = 0.0002; 85.0%, 95% CI, 78.0%–90.4% vs. 61.3%, 95% CI, 53.2%–68.8% at 6 months, P < 0.0001; and 91.4%, 95% CI, 85.5%–95.5% vs. 79.3%, 95% CI, 72.0%–85.5% at 12 months, P = 0.05, respectively; Fig. 3). These results also revealed that patients receiving flumatinib achieved a CCyR faster than those receiving imatinib. a total of 239 frequencies with CBA data missing were recorded in both arms. The recorded reasons for missing CBA data included missing bone marrow aspiration or inadequate bone marrow sample, unsuccessful bone marrow cell culture, and metaphases less than 20. On the basis of these recorded reasons, there were no obvious systematic differences between the missing CBA data and the observed CBA data thus, the missing CBA data in the FESTnd primarily met missing completely at random (38). Similar to radotinib in the RERISE study (8), frequencies of missing CBA data in the flumatinib arm (169 frequencies, 62.34%) were numerically more than that in the imatinib arm (90 frequencies, 37.66%), but without statistical significance (P = 0.4432).

The rates of MMR and CCyR at 12 months were higher among patients receiving flumatinib than among patients receiving imatinib across low and intermediate Sokal risk group, which were statistically significant, except for MMR in low Sokal risk group (Supplementary Fig. S2). The sample sizes in high Sokal risk in both arms were too small to evaluate the difference of molecular and cytogenetic responses between the two arms. The small sample size in high Sokal risk group was similar to that in ENESTChina (39), perhaps because more and more patients in high Sokal risk group are receiving 2G-TKIs which have been launched in China.

Time to MMR and CCyR was faster with flumatinib compared with imatinib (MMR: HR, 1.64, 95% CI, 1.22–2.20, Stratified Cox Model P = 0.0012; CCyR: HR, 1.51, 95% CI, 1.20–1.91, Stratified Cox Model P = 0.0005; Supplementary Fig. S3A and S3B).

By 12 months, the rates of progression to the AP/BC or death due to CML while on treatment were slightly higher in the imatinib arm (n = 4) than in the flumatinib arm (n = 0; 0% vs. 2%, P = 0.047; Supplementary Fig. S3C).

Drug-related AEs associated with both treatments were primarily grade 1 or grade 2 events (Table 3). In the flumatinib arm, 10.2% of patients discontinued therapy because of AEs compared with 6.1% in the imatinib arm. The most common AEs leading to discontinuation of flumatinib were thrombocytopenia (3.5%) and elevation of aminotransferase (3.1%). The most common AEs leading to discontinuation of imatinib were thrombocytopenia (2.0%) and rash (1.0%). Flumatinib demonstrated significantly lower rates of AEs of edema, pain in extremities, rash, neutropenia, anemia, and hypophosphatemia. Conversely, higher rate of diarrhea occurred in the flumatinib arm, so did the alanine transaminase elevation (Supplementary Fig. S4). Although the rate of diarrhea was higher in the flumatinib group, 49% of episodes in the flumatinib arm had a duration of less than 1 day and 62% less than 2 days. Ninety-three percent of diarrhea events in the flumatinib arm were CTCAE grade 1, and dose interruption due to diarrhea occurred in 4 patients in the flumatinib arm versus 2 patients in the imatinib arm. No treatment discontinuation because of diarrhea occurred in both arms. No electrocardiogram corrected QT interval prolongation was reported in the flumatinib arm, but three cases were reported in the imatinib arm. Pericardial and pleural effusions were rare in both arms. Among patients receiving flumatinib, one died from a cerebral hemorrhage, which was unrelated to both CML and thrombocytopenia (Supplementary Fig. S4).

Reasons for discontinuation among patients receiving flumatinib and imatinib, respectively, included AEs (10% vs. 6%), disease progression (0% vs. 2%), subject withdrew consent (2% vs. 3%), death (<1% vs. 0%), lost to follow-up (<1% vs. 0%), treatment failure (5% vs. 6%), and other reasons (1% vs. 3%; Table 2).

Results of the FESTnd study, with 12 months of follow-up, and median duration of treatment of 335 and 334 days in the flumatinib arm and imatinib arm, respectively, indicate improved outcomes with flumatinib versus imatinib as initial therapy for patients with CML-CP, as demonstrated by higher rates of EMR, MMR, CCyR, and DMRs with flumatinib at each landmark timepoint, including the rate of MMR at 6 months as the primary endpoint. Furthermore, responses were achieved significantly faster and deeper with flumatinib than with imatinib and flumatinib was more effective than imatinib across low and intermediate Sokal risk group. Branford and colleagues reported that patients with newly diagnosed CML-CP who achieved MMR within 6 months were associated with significantly higher rates of and more sustained MR4.5 (35, 40), and clinical data published to date indicate that for many patients who attained DMRs on TKI therapy, treatment-free survival is a safe treatment goal (37).

Results from the FESTnd study are consistent with similar studies that compared nilotinib (ENESTnd and ENESTChina), dasatinib (DASISION), and bosutinib (BELA and BFORE) with imatinib as first-line treatment of CML and demonstrated higher response rates and earlier and deeper responses with 2G-TKI therapies. At 3 months, the rate of EMR with flumatinib (82.1%) was similar to bosutinib (75.2%; ref. 4), dasatinib (84%; ref. 5), and nilotinib (82%; ref. 39); at 6 months, the rate of MMR with flumatinib (35.2%) was comparable with bosutinib (35.0%; ref. 4), dasatinib (27%; ref. 41), and nilotinib (33%, 300 mg twice daily and 30%, 400 mg twice daily; ref. 27); at 9 months, the rate of MMR with flumatinib (49.5%) approximated to bosutinib (42.3%; ref. 4), dasatinib (39%; ref. 5), and nilotinib (43%, 300 mg twice daily and 38%, 400 mg twice daily; ref. 27); and at 12 months, the rate of MMR with flumatinib (52.6%) was close to bosutinib (47.2%; ref. 4), dasatinib (46%; ref. 42), and nilotinib (44%, 300 mg twice daily and 43%, 400 mg twice daily; ref. 27).

Although the follow-up period in the FESTnd study was limited, it has been seen that the rates of progression to the accelerated phase or blast crisis were lower in the flumatinib arm than in the imatinib arm, showing that the higher rate of molecular responses, including MMRs and DMRs, with flumatinib may improve disease control in patients with newly diagnosed CML-CP, which might improve the long-term outcomes in patients receiving flumatinib as first-line therapy for CML-CP.

Overall, flumatinib demonstrated significantly lower rates of AEs of edema, pain in extremities, rash, neutropenia, anemia, and hypophosphatemia. On the contrary, higher rates of tolerable and manageable diarrhea and alanine transaminase elevation with flumatinib, than imatinib, were observed in the FESTnd study. In the flumatinib arm, most of diarrhea events were very transient and manageable without leading to treatment discontinuation. Liver function abnormalities were also manageable in the flumatinib arm. Although TKI therapy has made CML-CP an almost curable disease, they are frequently associated with long-term AEs. Efficace and colleagues reported that the following AEs were among the top 10 factors affecting Health-related Quality of Life (HRQoL), fatigue, muscle cramps, joint pains, swelling in certain parts of the body, dry or itchy skin, diarrhea, nausea and abdominal pains/cramps, and weight gain/loss (43). With long-term treatment, even low-grade AEs can substantially impact QoL and reduce treatment adherence (44). Off-target effects (OTE) of TKIs, which have been defined as secondary effects that cannot be explained by the inhibition of BCR-ABL1 tyrosine kinase, may contribute to the differences in AE profiles observed with different TKIs (45). The difference of AE profile between flumatinib and imatinib may be associated with the fact that flumatinib is highly selectivity against BCR-ABL1 tyrosine kinase. Flumatinib has less OTEs because it inhibits the phosphorylation of c-Kit and PDGFR with less potency than imatinib and had no effect on the phosphorylation of EGFR, VEGFR, c-Src, or HER2 (20). However, the sustainability of AEs, especially the occurrence of cardiovascular events, is not adequately studied because of the limited follow-up of 1-year data. Thus, a longer period of follow-up is required to evaluate the safety.

Collectively, the FESTnd study suggested that flumatinib at a dose of 600 mg has an efficacy profile that is superior to imatinib and similar to 2G-TKIs among patients with newly diagnosed CML-CP. These results suggest that flumatinib can be an alternative for patients with previously untreated CML-CP.

Further studies will be necessary to provide information on the responses among high Sokal risk group, the durability of responses, the development of treatment resistance, and the long-term follow-up of both survival and safety profile of flumatinib in the first-line setting.

J. Wang reports grants from Celgene and personal fees from AbbVie (advisor) outside the submitted work. No disclosures were reported by the other authors.

L. Zhang: Resources, writing-review and editing. L. Meng: Resources, supervision, writing-review and editing. B. Liu: Resources, supervision, validation, writing-review and editing. Y. Zhang: Resources, supervision, writing-review and editing. H. Zhu: Resources, validation, writing-review and editing. J. Cui: Resources, writing-review and editing. A. Sun: Resources, writing-review and editing. Y. Hu: Resources, writing-review and editing. J. Jin: Resources, writing-review and editing. H. Jiang: Resources, writing-review and editing. X. Zhang: Resources, writing-review and editing. Y. Li: Resources, writing-review and editing. L. Liu: Resources, writing-review and editing. W. Zhang: Resources, writing-review and editing. X. Liu: Resources, writing-review and editing. J. Gu: Resources, writing-review and editing. J. Qiao: Resources, writing-review and editing. G. Ouyang: Resources, writing-review and editing. X. Liu: Resources, writing-review and editing. J. Luo: Resources, writing-review and editing. M. Jiang: Resources, writing-review and editing. X. Xie: Resources, writing-review and editing. J. Li: Resources, writing-review and editing. C. Zhao: Resources, writing-review and editing. M. Zhang: Resources, writing-review and editing. T. Yang: Resources, writing-review and editing. J. Wang: Conceptualization, resources, supervision, validation, writing-review and editing.

This study was sponsored by Hansoh Pharmaceuticals. The first draft of the article was written by a medical writer employed by an independent company with funding provided by Hansoh Pharmaceuticals. All authors and representatives of the sponsor reviewed and amended the article and vouched for the completeness and integrity of the reported data. The authors also certified that the study, as reported, conformed with the protocol (as amended) and statistical analysis plan. We wish to acknowledge all participating patients and their families, as well as all of the investigators, research nurses, study coordinators, and operations staff.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.