Nilotinib after imatinib first-line: a real-life longitudinal cohort of patients with chronic myeloid leukaemia in chronic phase

Nilotinib after imatinib first-line: a real-life longitudinal cohort

of patients with chronic myeloid leukaemia in chronic phase

Pascale Cony-Makhoul,1,2 Martine Gardembas,2,3

Val
erie Coiteux,2,4_{Nathalie Carpentier,}5 C
ecile Pommier,6_{Isabelle Violet,}5 Philippe Quittet2,7

and Marc G. Berger,2,8,9on behalf of TARGET-RMC Investigators

1_{Centre Hospitalier Annecy Genevois, Pringy,}

France,2Fi LMC group, Institut Bergoni
e, Bor-deaux, France,3Centre Hospitalier Universitaire d’Angers, Angers, France,4Centre Hospitalier R
egional Universitaire de Lille 2, Lille,5Novartis Pharma, Rueil-Malmaison,6IT&M Stats, Neuilly-Sur-Seine,7Centre Hospitalier Universi-taire de Montpellier, Montpellier, France,

8_{Service d’H
ematologie Biologique, Centre}

Hospi-talier Universitaire de Clermont-Ferrand, H ^opi-tal Estaing, and9_{EA7453 CHELTER, Universit
e}

Clermont Auvergne, Clermont-Ferrand, France Received 12 July 2017; accepted for publication 26 September 2017

Correspondence: Dr Pascale Cony-Makhoul, Centre Hospitalier Annecy Genevois, 1 avenue de l’H^opital, BP 90074 Epagny Metz-Tessy, 74374 Pringy, Cedex, France.

E-mail: pconymakhoul@ch-annecygenevois.fr

Summary

This prospective, observational study enrolled 150 adult patients with chronic myeloid leukaemia (CML) in chronic phase (CP) treated with nilo-tinib as second-line after imanilo-tinib, in a real life setting in France. Two-thirds of patients switched to nilotinib treatment due to lack of imatinib efficacy. Of 146 evaluable patients, 16 (11
0%) (95% confidence interval: 6
4–17
2%) achieved uMR4

, defined as undetectable molecular disease in cDNA with MR4 sensitivity (≥10 000 ABL1 transcripts) at 18 months and confirmed at 24 months (primary endpoint). Among patients without major molecular response (MMR) or deep molecular response (DMR) at study entry, 66
3% achieved MMR and 44
2% DMR within a median of 5
7 and 6
24 months, respectively. Fifty-three patients (36
3%) have prema-turely terminated the study before 24 months of follow-up, primarily due to nilotinib treatment discontinuation (n= 43; 29
5%), mainly motivated by treatment intolerance (n = 27; 18
5%) and inefficacy (n = 10; 6
8%). The most frequent extra-haematological adverse events (AEs) reported as related to treatment with nilotinib were pruritus (16
4%), asthenia (13
7%) and dry skin (13
0%). Ischaemic cardiovascular AEs were reported in 18 patients (12
3%). This French nationwide large cohort adds valuable infor-mation to the body of evidence on the efficiency and safety of nilotinib in the treatment of patients with CP-CML.

Keywords: chronic myeloid leukaemia, real-life practice, nilotinib, second line treatment, BCR-ABL1.

In the last decade, the landscape of chronic myeloid leukae-mia (CML) management changed dramatically due to the advent of tyrosine kinase inhibitors (TKIs) targeting the BCR-ABL1 oncoprotein. Imatinib mesylate was the first TKI approved for newly diagnosed patients with CML. However, the results of the IRIS (International Randomized Study of Interferon and STI571) study (Druker et al, 2006) showed a 5-year resistance rate to imatinib of 14% and 4% of patients who discontinued therapy owing to an adverse event (AE). Imatinib resistance is related to mutations in the kinase domain of BCR-ABL1 gene in about one-third of the resis-tant patients (O’Brien et al, 2003; Jabbour et al, 2006; Sover-ini et al, 2006; Hehlmann et al, 2007). Second-generation TKIs offer therapeutic alternatives for those patients (Bhamidipati et al, 2013).

Nilotinib, an orally bioavailable selective BCR-ABL1 sec-ond-generation TKI, is 30-fold more potent than imatinib, as

observed in preclinical models, and overcomes some imatinib resistant BCR-ABL1 mutations (Weisberg et al, 2006). A phase 2, open-label study evaluated the efficacy and safety of 400 mg nilotinib administered orally twice daily to 280 patients with Philadelphia chromosome-positive (Ph+) CML in chronic phase (CP) after imatinib failure or intolerance (Kantarjian et al, 2007). At 6 months, the rate of major cyto-genetic response (MCyR) (Ph ≤ 35%) was 48%, with 31% complete cytogenetic response (CCyR) (Ph = 0%). The esti-mated survival at 12 months was 95%. Adverse events reported in this phase 2 study were mostly mild to moderate, and there was minimal cross-intolerance with imatinib (Kan-tarjian et al, 2007); the promising data from this phase 2 study supported the granting of market authorization for nilotinib.

The objective of the TARGET-RMC (Tasigna As second-line theRapy in Glivec-resistant or intolerant CML PC

patiEnts in currenT practice-complete molecular response) study was to describe, in a real-life setting, the management of patients with CML-CP treated with nilotinib as second-line, and evaluate the cytogenetic and molecular response rates including undetectable disease with≥10 000 ABL1 tran-scripts (uMR4), and time to reach the response. The study also evaluated the safety and tolerance of nilotinib during the 24 months of follow-up.

Methods

The TARGET-RMC study was a prospective, observational, descriptive study. This large cohort of 150 patients with CML-CP was performed at 50 active sites with hospital-based haematologists in France. Patients were included from Jan-uary 2011 to JanJan-uary 2014.

The study was conducted in accordance with the Declara-tion of Helsinki, Good Epidemiological Practice (GEP) Guidelines of April 2007, and European and French regula-tions and was approved by respective ethics committees and the French Data Protection Authority (Commission Natio-nale de l’Informatique et des Libert
es, CNIL) prior to the start of the study. Written informed consent was obtained from the participating subjects prior to study entry.

Patient population

Eligible patients were included in the study if they were aged 18 years or older, were in CML-CP, and for whom a treat-ment with nilotinib was planned or prescribed for less than 3 months as second-line after imatinib, owing to the fact that nilotinib is indicated for adult patients resistant or intolerant to prior therapy that included imatinib (http://www.ema.eu ropa.eu/docs/en_GB/document_library/EPAR_-_Product_Inf ormation/human/000798/WC500034394.pdf). Exclusion crite-ria were: previous treatment with another second-generation TKI, treatment with nilotinib for more than 3 months, and disease in accelerated phase or blast crisis. Investigators were asked to include consecutive patients responding to inclu-sion/exclusion criteria. There was no log for screened non-enrolled patients.

Data collection

Patients were followed up over 24 months. Information col-lected by the physician at study entry and around 3, 6, 12, 18 and 24 months (M3, M6, M12, M18 and M24, respec-tively) using patient medical records. Given that the present study was observational, the prescription (dosage) or the change of study treatment, nilotinib, as well as the treatment duration and concomitant medications were left to the dis-cretion of the haematologist according to patients’ routine management. As per approved European Summary of Pro-duct Characteristics, nilotinib should be administered twice daily at a total daily dose of 800 mg, taken every 12 h

without food (http://www.ema.europa.eu/docs/en_GB/docu ment_library/EPAR_-_Product_Information/human/000798/ WC500034394.pdf).

Assessments

At the time of design of this study, the primary endpoint was previously known as the “complete molecular response rate” (undetectable transcript with a sensitivity level ≥104, later defined as uMR4_{), around 18 months of treatment,} con-firmed by a subsequent polymerase chain reaction (PCR) test around 24 months. Following a conservative approach, only patients with uMR4 at 18 months, with a confirmed result later, around 24 months, were considered as reaching the primary endpoint. Patients who prematurely discontinued the study were all counted as treatment failure, regardless of the reason. In 2013, the European Leukaemia Net (ELN) rec-ommendations (Baccarani et al, 2013), according to Cross et al (2012) highlighted that the term “complete molecular response” should be avoided and substituted with the term “molecularly undetectable leukaemia”, with specification of the number of the control gene transcript copies; thus, in this publication we replaced the term “complete molecular response” used in the TARGET-RMC protocol to uMR4_{, i.e.,} undetectable molecular disease in cDNA with MR4_sensitivity (≥10 000 ABL1 transcripts). The other evaluated responses included the rate of patients with no CCyR at study entry who subsequently obtained a CCyR during the study and the time needed to reach such response. Actually, as molecular biology testing was more frequently performed than cytoge-netics, and since MR2is acknowledged to approximate CCyR (Baccarani et al, 2006; National Comprehensive Cancer Net-work (NCCN), 2017), it was also decided, together with the study Scientific Committee that MR2would be used instead of CCyR to describe the results of the TARGET-RMC sub-groups and responses.

Other endpoints consisted of the rate of patients with a major molecular response (MMR, defined by an BCR-ABL1 transcript ≤0
1%) around 18 months, time to obtain this MMR, relationship between early MMR (within the first 6 months of treatment) and achievement of uMR4_. More-over, the rate of patients progressing to an accelerated phase or a blast crisis and the discontinuation rate of nilo-tinib were evaluated.

Molecular response as second-line treatment was defined as follows: MR2, based on PCR testing BCR-ABL1/ABL1 ≤1%; MR3_{or MMR, based on PCR testing BCR-ABL1/ABL1} ≤0
1%; deep molecular response (DMR), based on either detectable disease with PCR testing BCR-ABL1/ABL1≤0
01% or undetectable disease in cDNA with ≥10 000 ABL1 tran-scripts; and uMR4, undetectable disease in cDNA with ≥10 000 ABL1 transcripts (Baccarani et al, 2009). At the start of the study, BCR-ABL1 transcript levels were commonly expressed on the International Scale in France, as molecular standardization was extended to most laboratories.

Safety assessments

Evaluation of nilotinib tolerance was based on the frequency of all AEs as well as serious adverse events (SAEs) reported by the patients and/or investigators during the study. SAEs were defined according to the International Conference on Harmonisation Guidelines. (ICH Steering Committee, 2003) The investigator determined whether the event was related to treatment.

Statistical considerations

In order to estimate an anticipated proportion of 10% patients with uMR4at 18 months confirmed after 2 years of follow-up, with a confidence interval (CI) of 5%, 138 evaluable patients were needed. To account for an estimated 8% of study discontinuation or missing data, 150 patients had to be included in the study.

The “Reference Population” used for the data analysis encompassed all patients included in the study and treated with nilotinib, with available information after baseline.

Data analysis was descriptive. Quantitative variables were described in terms of number and percentage of missing val-ues, the number of observed valval-ues, minimum, maximum, median, interquartile range (IQR), mean and standard devia-tion (SD). Qualitative variables were described in terms of number and percentage of missing values, the number of observed values, absolute proportion and percentage by modality.

The uMR4 rate was analysed including the whole treated population (patients who left the study prematurely were deemed as treatment failure); only patients who were lost to follow-up immediately after inclusion were excluded from the study analysis. Analyses were also performed on prede-fined subgroups, by age (< or ≥65 years), treatment duration with imatinib (> or ≤18 months), reason for discontinuation of imatinib (lack of efficacy, intolerance or both), transcript levels at inclusion (patients with or without MR2, MMR and DMR (MR4 or deeper, detectable or undetectable), and the initial daily dose of nilotinib (800, 600 or<600 mg/day). The Statistical Analysis System (SAS), version 9.2 (SAS Institute Inc, Cary, NC, USA), was used for all analyses.

Results

Overall, 150 patients were included in this observational cohort study, among whom four patients were excluded; two were untreated and two for lack of information after base-line. Thus, a total of 146 were analysed (Reference Popula-tion). Patients evaluable for the primary endpoint without major protocol deviation (n= 141) constituted the Per-Pro-tocol (PP) Population (Fig 1). In the reference population, 53 patients (36
3%) prematurely terminated the study, essen-tially due to nilotinib treatment discontinuation (43 patients; 29
5%), which was motivated by treatment inefficacy in 10

patients (6
8%), treatment intolerance in 27 patients (18
5%) and other reasons in 6 patients (4
1%) (Fig 1).

The median (IQR) age of the patients was 61 (47– 68) years, with a sex ratio (male/female) of 1
35. The median (IQR) time from diagnosis of CML to inclusion in the study was 26 (13–68) months (Table I). Two-thirds of patients were treated with imatinib for more than 18 months and almost two-thirds of patients have stopped imatinib treat-ment due to lack of efficacy (Table I).

At study entry, 84
9% of patients had a complete haema-tological response, 59
6% a molecular response MR2 (consid-ered as CCyR), 28
8% a major molecular response (MMR) and 11
6% a DMR; none of the patients (n = 17) with DMR at baseline had started nilotinib before study inclusion (Table II). Moreover, 61
0% of patients started treatment with nilotinib at an initial dose of 800 mg/day, 29
5% at 600 mg/day and 9
6% at a dose of <600 mg/day. The median dose of nilotinib was 800 mg/day in the first 6 months then 600 mg/day until study end. As planned in the protocol, the majority of patients (135/146) started nilotinib treatment close to the enrolment date in the study; only 9 patients had started nilotinib between 1–30 days before inclusion, and 2 patients received nilotinib more than 1 month before starting the study (Table I). The median [IQR] time from study inclusion to nilotinib treatment start was 1
0 day (0
0–3
0).

The median (IQR) nilotinib treatment duration was 23
7 (15
9–24
8) months for all the patients and 10
1 (2
8– 18
7) months for the 46 patients who discontinued the

Fig 1. Patient disposition. *All 5 patients were included in the refer-ence population for safety and efficacy analysis. **Discontinued study before the third month (M3) for protocol deviation. ***1 patient dis-continued nilotinib due to intolerance, the other completed the 24-month study. M, months; N, number of patients in bar graphs.

treatment prematurely (n= 43) or died (n = 3). This dura-tion was roughly similar across all predefined subgroups. In total, 25 patients (17
1%) had at least one temporary discon-tinuation of nilotinib treatment for a median duration of 13
5 days.

Regarding the cytogenetic profile, the proportion of patients having a cytogenetic profile were 14
0% (20/143) at M3, 11
6% (15/129) at M6, 15
3% (19/124) at M12, 8
9% (10/112) at M18 and 5
0% (5/101) at M24. The molecular profile was analysed in all the patients at study entry and the

majority of patients underwent molecular analysis between M6 (114/129 patients; 88
4%) and M24 (94/101; 93
1%); a lower rate was observed at M3 (111/143; 77
6%).

Efficacy

Primary endpoint. Eventually, 16 patients (11
0%) (95% CI: 6
4–17
2%) achieved an uMR4 _{around 18 months of} nilo-tinib treatment, confirmed by subsequent PCR amplification around 24 months. This rate was higher in patients who had switched from imatinib to nilotinib due to intolerance only [12/49 patients; 24
5% (95% CI: 13
3–38
9)], as compared to those who have switched due to inefficacy [4/95; 4
2% (95% CI: 1
2–10
4)]. Higher rates of confirmed uMR4 _{were also} observed in patients with MMR at baseline [10/42 patients; 23
8% (95% CI: 12
1–39
5)] and those with DMR [6/17; 35
3% (95% CI: 14
2–61
7)], compared with patients without MMR [6/104; 5
8% (95% CI: 2
1–12
1)] and without DMR [10/129; 7
8% (95% CI: 3
8–13
8)], respectively, at baseline.

Predefined sensitivity analyses were conducted in 3 popu-lations of patients: the PP Population (n= 141), patients who were evaluated at 18 months (n= 104), and patients evaluated at 18 and 24 months (n= 94). These sensitivity analyses showed comparable results to the primary analysis conducted on the Reference Population, with respective rates of uMR4_{of 11
3% (95% CI: 6
6–17
8), 15
2% (95% CI: 9
0–} 23
6) and 17
8% (95% CI: 10
5–27
3).

Other responses. Molecular responses at study entry and during the study as well as time to response are summarized in Table III. Among the 59 patients without an MR2at study entry, 76
3% (95% CI: 63
4–86
4) achieved MR2 _when trea-ted with nilotinib within a median (IQR) of 3
1 (2
8– 5
6) months. Among the 104 patients without MMR at study entry, 66
3% (95% CI: 56
4–75
3) achieved MMR after treat-ment with nilotinib within a median (IQR) of 5
7 (3
0– 11
3) months.

DMR was achieved in 57/129 patients (44
2%) within a median (IQR) of 6
2 (3
6–12
4) months. Among the 137 patients without uMR4 _{at study entry, 21
2% of patients} achieved uMR4 _{with nilotinib treatment, within a median} (IQR) time of 11
6 (5
6–17
1) months (Table III).

Table I. Demographics and baseline characteristics. N= 146

Male/female, n (%) 84 (57
5)/62 (42
5) Age (years), median (min_{–max), (Q1–Q3)} 61
_{0 (26–93), (47–68)} Time between diagnosis and inclusion

(month), median (Q1–Q3)

26
0 (13–68) Treatment duration with imatinib

(month), median (Q1–Q3)

25
0 (11–63) Reason for switch to nilotinib, n (%)

Resistance to imatinib 82 (56
2) Intolerance to imatinib 49 (33
6)

Both 13 (8
9)

Unmet criterion* 2 (1
4) Other CML previous treatment, n (%)

Interferon-a 11 (7
5) Cytarabine 7 (4
8) Dasatinib* 3 (2
1) Cytogenetic analysis performed

within 3 months prior to inclusion, n (%)

59 (40
4) CCyR reported within 3 months

prior to inclusion, n/n analysed (%)

24/59 (40
₇₎ Nilotinib started in patients, n (%)

Just after study entry (planned treatment with nilotinib)

135 (92
5) Within 30 days prior to study entry 9 (6
2) More than 30 days prior to study entry 2 (1
4) Patients with a nilotinib starting

dose of 800 mg/600 mg/<600 mg (%)

61/29
_5/9
5

% calculated as n/N, unless otherwise specified; CCyR, cytogenetic complete response; CML, chronic myeloid leukaemia; Q, quartile. *Excluded from per protocol population.

Table II. Molecular responses (MR) level at each study visit– reference population.

Inclusion M3 M6 M12 M18 M24

Patients with molecular assessment, N 146 111 114 118 104 94 BCR-ABL1/ABL1≤1% 87 (59
6) 92 (82
9) 106 (93
0) 110 (93
2) 100 (96
2) 90 (95
7) BCR-ABL1/ABL1≤0
1% 42 (28
8) 52 (46
8) 74 (64
9) 88 (74
6) 81 (77
9) 80 (85
1) BCR-ABL1/ABL1≤0
01% * 17 (11
6) 22 (19
8) 38 (33
3) 50 (42
4) 47 (45
2) 46 (48
9) uMR4 _{9 (6
2) 8 (7
2) 14 (12
3) 20 (16
9) 23 (22
1) 23 (24
5)}

Data are n (%) unless otherwise indicated. M, month; MR, molecular response.

*i.e., Deep molecular response, based on either detectable disease with polymerase chain reaction testing BCR-ABL1/ABL1_{≤0
01% or undetectable} disease in cDNA with≥10 000 ABL1 transcripts; uMR4, undetectable disease with≥10 000 ABL1 transcripts.

Predefined subgroup analyses showed that the rate of MMR was greater in patients with at least MR2 at study entry [36/45 (80
0%); 95% CI: 65
4–90
4] versus those with-out MR2 [33/59 (55
9%); 95% CI: 42
4–68
6], and in this subgroup of patients with MR2 at study entry 51
4% achieved DMR within a median time of 5
67 months. More-over, the rate of uMR4 achieved once during the study per-iod was higher in patients having MMR at baseline [14/33 (42
4%); 95% CI: 25
5–60
8] as compared to those without MMR [15/104 (14
4%); 95% CI: 8
3–22
7]. In addition, the rate of uMR4 was higher among patients with early MMR, defined as MMR reached in the first 6 months of treatment with nilotinib (11/37 patients; 29
7%), as compared to those not having an early MMR (4/67; 6
0%), (P = 0
0023). There were no other noticeable differences between the predefined subgroups.

Table II describes the molecular response rates observed at inclusion and at each visit. The distribution of the total number of patients with documented molecular profiles according to the transcript BCR-ABL1/ABL1 ratio at each study visit (inclusion, M3, M6, M12, M18, M24) is illus-trated in Fig 2. During treatment with nilotinib, only one patient progressed to accelerated phase, 1
9 months after ini-tiation of nilotinib, and no progression to blast crisis was observed.

Safety

A total of 119/146 patients (81
5%) reported at least one AE. This corresponded to 574 reported AEs, of which 111 were SAEs. Grade 3 AEs were reported in 45 patients (30
8%) and 3 patients (2
1%) had Grade 4 AEs. Thirty patients (20
5%) reported at least a nilotinib-related SAE. There was no nota-ble difference observed in the whole incidence of AEs with respect to age, starting dose of nilotinib, and imatinib dis-continuation for intolerance or inefficacy. An increased ten-dency of SAEs in patients treated with an initial dose of

nilotinib 800 mg/day was observed compared to patients on initial dose of 600 mg/day.

The most frequent (≥10% of patients) extra-haematological Treatment Emergent Adverse Events (TEAEs) that were con-sidered to be related to treatment with nilotinib were pruritus (16
4%), asthaenia (13
7%) and dry skin (13
0%). TEAEs related or not to treatment with nilotinib observed in≥5% of patients are summarized in Table SI. Nilotinib treatment-related diabetes mellitus was reported in four patients (2
7%) and was assessed as a SAE in three of those patients (2
1%).

Ischaemic cardiovascular TEAEs were reported in 18 patients (12
3%) and are summarized in Table IV. The dis-tribution of these ischaemic cardiovascular TEAEs was simi-lar between the predefined subgroups, except for patients aged ≥65 years who had a higher incidence (8/59 patients; 13
6%) compared with those <65 years of age (5/87; 5
7%).

Haematological toxicity related to treatment with nilotinib was observed in four patients and was poorly documented: cytopenia (one patient), neutropenia (one patient), thrombo-cytopenia (one patient), anaemia (two patients) and unspeci-fied haemotoxicity (one patient).

Grade 3/4 AEs, related or not to nilotinib treatment, were observed in≥1% of patients and are summarized in Table SII.

Three patients died during the study: one from sudden death with cardiac arrest, 8 days after starting treatment with nilotinib (not proven but may be related to cardiac rhythm abnormalities), the second following a cerebrovascular stroke which occurred 15 months after the start of treatment, and the third after acute lower limb ischaemia which occurred 19 months after the start of treatment. A fourth patient died 4 months after stopping nilotinib treatment, following aggra-vation of a grade 4 amyotrophy; this was considered to be the consequence of a lower limb ischaemia that was aggra-vated during the nilotinib treatment period.

Inefficacy of the drug was reported as an AE in 10 patients (6
8%). A total of 27 patients (18
5%) discontinued the study due to nilotinib intolerance.

Table III. Molecular responses (MR) achievement in patients without documented MR level at baseline– reference population.

Analysed

population N

Patients without MR level at baseline n (%)

Response during study* Time to response (months) n (%) 95% CI (%; %) n Median IQR MR2† 146 59 (40
4) 45 (76
3) (63
4; 86
4) 45 3
12 2
8–5
6 MMR† 146 104 (71
2) 69 (66
3) (56
4; 75
3) 69 5
68 3
0–11
3 DMR† 146 129 (88
3) 57 (44
2) (35
5; 53
2) 57 6
24 3
6–12
4 uMR4_{† 146 137 (93
8) 29 (21
2) (14
7; 29
0) 29 11
60 5
6–17
1}

CI, confidence interval; DMR, deep molecular response; IQR, interquartile range; MMR, major molecular response; MR, molecular response; uMR4_{, undetectable disease with≥10 000 ABL1 transcripts.}

*Includes patients having achieved response during the study; % calculated as number of response during study/n patients without MR level at baseline.

†MR2 _{(equivalent to complete cytogenetic response) based on polymerase chain reaction (PCR) testing BCR-ABL1/ABL1≤1%; MR}3_{or MMR,}

based on PCR testing BCR-ABL1/ABL1≤0
1%; MR4or DMR based on either detectable disease with PCR testing BCR-ABL1/ABL1≤0
01% or undetectable disease in cDNA with_{≥10 000 ABL1 transcripts; uMR4, undetectable disease with ≥10 000 ABL1 transcripts.}

Discussion

The TARGET-RMC study allowed the observation, in a real-life setting, of a large cohort of patients with CML-CP for whom treatment with nilotinib was prescribed as second-line therapy after imatinib, with a follow-up of 24 months. In the IRIS study, after a median follow-up of 11 years, nearly half of patients stopped imatinib (49
2%), all causes taken together (Hochhaus et al, 2017). Our study confirms the known rea-sons for switching from first-line imatinib to second-line/sec-ond-generation TKI (2G-TKI) treatment, with approximately two-third of the patients switching for inefficacy and one-third for intolerance. It also confirms the efficacy of nilotinib in this setting in routine clinical practice, as demonstrated in previous clinical trials (Kantarjian et al, 2011; Nicolini et al, 2012). In addition, it provides information on molecular response rates and response kinetics that were not documented in those pre-vious trials. Finally, the safety observed during the study was in line with the known safety profile of nilotinib (http://www.e ma.europa.eu/docs/en_GB/document_library/EPAR_-_Produc t_Information/human/000798/WC500034394.pdf), and no new signal was detected.

The median age of patients in TARGET-RMC was similar to those in a recent CML-CP French cross-sectional epidemi-ological survey (62 years) (Etienne et al, 2016) and were older than those in other second line nilotinib clinical studies (Kantarjian et al, 2011; Nicolini et al, 2012; Hughes et al, 2014; Gora-Tybor et al, 2015; Kuo et al, 2015; Rea et al, 2015; Miyamura et al, 2016).

At the time of TARGET-RMC protocol design, the term “complete molecular response” was used, and the definition included undetectable BCR-ABL1 transcript with at least 10000 ABL1 copies, according to ELN 2009 definitions (Bac-carani et al, 2009). Due to evidence that leukaemic stem cells persisted even in patients without any BCR-ABL1 signal in PCR (Ross et al, 2010; Chomel et al, 2011), the 2013 revised

ELN recommendations (Baccarani et al, 2013) highlighted that the term “complete molecular response” should be avoided and substituted with the term “molecularly unde-tectable leukaemia”, with specification of the number of the control gene transcript copies, according to Cross et al (2012). In TARGET-RMC publication, we replaced the term “complete molecular response” used in the initial protocol by “undetectable molecular disease in cDNA with MR4 sensi-tivity (≥10 000 ABL1 transcripts) (uMR4_).

The proportion of patients who achieved the primary end-point, confirmed uMR4at 18 and 24 months, was consistent with the study hypothesis. This proportion was, as expected, higher in patients who had stopped imatinib treatment due to intolerance than those who stopped imatinib due to inefficacy. Higher rates of confirmed uMR4_{were also observed in patients} with MMR at baseline compared with patients without MMR.

Of patients without DMR at study entry, 44
2% achieved DMR within a median time of 6
24 months in TARGET-RMC. Of patients with MR2 (equivalent to CCyR) at study entry, 51
4% achieved DMR within a median time of 5
67 months.

According to ELN recommendations (Baccarani et al, 2013), optimal response to second-line therapy at 12 months in case of imatinib failure is MR2 (BCR-ABL1/ABL1 <1%) and/or CCyR and, at any time, MMR (BCR-ABL1/ABL1 ≤0
1%). In the present study, the MMR rate was higher (66
3%) and more quickly achieved (median time 5
7 months) compared with the results in heavily pre-treated patients of the ENACT (Expanding Nilotinib Access in Clinical Trials) study, in which 37% of the French subset of patients (n= 168) achieved a MMR by 12 months (Nicolini et al, 2012).

Nilotinib was approved for second-line treatment after ima-tinib-resistance or -intolerance on the basis of the initial results of a phase 2 open-label study (Kantarjian et al, 2007). In the 24-month follow-up analysis (Kantarjian et al, 2011), CCyR was achieved in 51% of imatinib-intolerant and 41% of

(A) (B)

Fig 2. (A) Patient distribution (%) according to the BCR-ABL1/ABL1 transcript rate at each study visit. (B) Primary objective, uMR4

imatinib-resistant patients (enrolment of imatinib-intolerant patients in MCyR or better at study entry was not permitted, resulting in the imatinib-intolerant population having disease characteristics more closely resembling imatinib resistance). The recent French epidemiological survey in unselected patients with CML-CP (Etienne et al, 2016) showed that the cytogenetic response was less investigated than the molecular response. On the other hand, ELN guidelines (Baccarani et al, 2013) indicate that the possibility of using a molecular test alone is sufficient to evaluate response to treatment after CCyR and MMR achievement. Assuming that MR2 is currently acknowledged to approximate CCyR, 76
3% of patients with-out MR2 at entry achieved MR2 within a median of 3
1 months in the TARGET-RMC study. These results are also in line with those seen in LASOR (Randomized Phase Lll Study of Imatinib Dose Optimization versus Nilotinib in CML Patients With Suboptimal Response to Imatinib) (Cortes et al, 2016a). Due to the strong association between molecular-response (MR) milestones and long-term treatment outcomes, molecular results now guide clinical decision-making and play an essential role in the clinical management of CML patients (Baccarani et al, 2006; National National Comprehensive Can-cer Network (NCCN, 2017).

The safety profile of nilotinib evolved after the first market authorization with recommendations to assess and monitor lipid profiles, blood glucose levels and cardiovascular risk factors before and during treatment with nilotinib which appeared in August 2013, February 2014 and May 2014, respectively, in the European Tasigna SmPC (http://www.e ma.europa.eu/docs/en_GB/document_library/EPAR_-_Produc t_Information/human/000798/WC500034394.pdf).

In the present observational study, ischaemic cardiovascu-lar TEAEs were reported in 18 patients and, as expected, occurred more frequently in older patients. In the LASOR study (Cortes et al, 2016a), five patients in the nilotinib group versus none in the imatinib group had cardiovascular events during the study. No incidence of peripheral arterial occlusive disorder was reported in the NOVEL (Nilotinib in patients with CP or AP Philadelphia chrOmosome positiVe (Ph+) chronic myElogenous Leukaemia) study (Kuo et al, 2015) which enrolled patients younger (median age of 47 years) than those in TARGET-RMC.

The safety profile of nilotinib presented in the TARGET-RMC study population complies with its known safety pro-file observed in other cohorts of patients switching from imatinib (Nicolini et al, 2012; Hughes et al, 2014; Cortes et al, 2016a,b; Miyamura et al, 2016).

The proportion of patients who discontinued nilotinib treatment due to AEs or drug-related toxicity has been reported to range from 4
7% to 19% (Kantarjian et al, 2011; Nicolini et al, 2012; Kuo et al, 2015; Cortes et al, 2016a,b). Study discontinuation motivated by treatment inefficacy was low (10 patients; 6
8%), compared to treatment intolerance (27 patients; 18
5%) in the TARGET-RMC study. No pattern was noted regarding nilotinib treatment-discontinuation based on inefficacy or intolerance across nilotinib studies.

The approved daily dose of nilotinib in second-line treat-ment is 800 mg. In the present study, 61% of patients initi-ated treatment at this recommended dose, and at the end of study, the median dose was 600 mg/day. Of note, the indica-tion of nilotinib in first line treatment was approved in 2011 based on a total daily dose of 600 mg. In addition, and to our knowledge, there are only two recent studies document-ing results with a 600 mg/day dose of nilotinib post-imatinib treatment: the ENRICH study (Cortes et al, 2016b) and the ENESTPath study (Rea et al, 2015), neither of which included imatinib-resistant patients.

This study has some limitations inherent to its observa-tional non-intervenobserva-tional nature, particularly the absence of a comparator arm, and no centralized molecular response anal-yses. On the other hand, there was no log for screened non-enrolled patients and the enrolment period for patients was relatively long, which may affect the homogeneity of the data in a growing therapeutic area; however, this may be inevita-ble in a rare disease such as CML with a high number of competitive clinical trials. These limitations are balanced with a relatively low amount of missing data, probably related to a clinically adequate study design and monitoring, which could minimize bias in the analysis. Finally, the change in clinical practice and monitoring recommendations (Baccarani et al, 2013), driven by the possibility of using a molecular test alone as sufficient to evaluate response to treatment, yielded less frequent performance of cytogenetic tests, which limited the analysis of the cytogenetic responses in the pre-sent study and consequently the indirect comparison to the main endpoints of the pivotal study. This practice was

Table IV. Ischaemic cardiovascular adverse events regardless of nilotinib treatment relationship– reference population.

Preferred term

N= 146 N= 146 Any grade Grade 3–4 Patients with at least 1 TEAE 18 (12
3) 13 (8
9) Peripheral arterial occlusive disease 6 (4
1) 4 (2
0) Angina pectoris 2 (1
4) 2 (1
4) Cerebrovascular accident 2 (1
4) 1 (0
7) Ischaemic stroke 2 (1
4) 1 (0
7) Peripheral ischaemia 2 (1
4) 1 (0
7) Arterial therapeutic procedure 1 (0
7) 0 Arterial thrombosis 1 (0
₇₎ 1 (0
₇₎ Cerebral ischaemia 1 (0
7) 1 (0
7) Coronary artery bypass 1 (0
7) 1 (0
7) Infarction 1 (0
7) 1 (0
7) Peripheral artery thrombosis 1 (0
7) 1 (0
7) Data are number of patients and percentage: n (%); % calculated as n/N. Adverse events (AEs) were considered emergent if they occurred after the administration of the first dose of nilotinib. An AE coded with the same Preferred Term (MedDRA dictionary: https://tools. meddra.org/wbb/login.aspx) was counted only once per patient. TEAE, treatment emergent adverse event.

emphasized by the molecular standardization of local labora-tories and reimbursement policy in France.

In summary, the French nationwide large cohort TAR-GET-RMC study adds valuable information to the body of evidence on the efficiency of nilotinib in the treatment of patients with CML in chronic phase, for whom treatment with imatinib was stopped due to lack of efficacy or intoler-ance. Real-life data in patients with CML-CP who receive nilotinib as first line treatment are currently under investiga-tion in another ongoing observainvestiga-tional study.

Authors contribution

Marc Berger, Valerie Coiteux, Pascale Cony-Makhoul, Martine Gardembas and Philippe Quittet were investigators and mem-bers of the scientific committee, and were involved in study design and data analysis. Nathalie Carpentier (study project manager), Cecile Pommier (responsible statistician) and Isa-belle Violet (drug safety manager) were involved in data analy-sis. All authors were involved in the writing, reviewing of the manuscript and approved the final version to be published.

Conflicts of interest disclosure

Pascale Cony-Makhoul received fees from Novartis as mem-ber of the scientific committee of this study and consulting fees from Bristol-Myers Squibb (BMS), Pfizer and Novartis outside the submitted work and fees as speaker from BMS. Martine Gardembas received fees from Novartis as member of the scientific committee of this study and consulting fees from BMS and Incyte. Val
erie Coiteux received fees from Novartis as member of the scientific committee of this study,

consulting fees from Novartis and BMS and as speaker from Novartis, BMS, Ariad and Pfizer. Isabelle Violet, Nathalie Carpentier are employees of Novartis, France. C
ecile Pom-mier has no conflict of interest to disclose. Philippe Quittet received fees from Novartis as member of the scientific com-mittee of this study, consulting fees from Novartis, BMS and Pfizer and honoraria from Novartis outside the submitted work. Marc Berger received fees from Novartis as member of the scientific committee of this study, grant from Genzyme Sanofi, and consulting fees from Shire and Incyte.

Acknowledgements

The authors would like to thank AMITIS SA, France for logistic support and data management; Delphine Aubert, statistician of EffiStat, for statistical support; and Thomas Rohban, M.D. of partner 4 health, for providing medical writing support in accordance with Good Publication Prac-tice (GPP3) guidelines. These activities as well as the study were funded by Novartis Pharma France. A full list of Inves-tigators is given in Appendix S1.

Supporting Information

Additional Supporting Information may be found in the online version of this article:

Appendix S1. Investigators list (France)– TARGET RMC Study.

Table SI. TEAEs related or not to treatment with nilotinib observed in≥5% of patients.

Table SII. Grade 3/4 AEs, related or not to treatment with nilotinib, observed in≥1% of patients.

References

Baccarani, M., Saglio, G., Goldman, J., Hochhaus, A., Simonsson, B., Appelbaum, F., Apperley, J., Cervantes, F., Cortes, J., Deininger, M., Grat-wohl, A., Guilhot, F., Horowitz, M., Hughes, T., Kantarjian, H., Larson, R., Niederwieser, D., Sil-ver, R., Hehlmann, R.; European LeukemiaNet. (2006) Evolving concepts in the management of chronic myeloid leukemia: recommendations from an expert panel on behalf of the European LeukemiaNet. Blood, 108, 1809–1820. Baccarani, M., Cortes, J., Pane, F., Niederwieser,

D., Saglio, G., Apperley, J., Cervantes, F., Dei-ninger, M., Gratwohl, A., Guilhot, F., Hochhaus, A., Horowitz, M., Hughes, T., Kantarjian, H., Larson, R., Radich, J., Simonsson, B., Silver, R.T., Goldman, J., Hehlmann, R.; European LeukemiaNet. (2009) Chronic myeloid leukemia: an update of concepts and management recom-mendations of European LeukemiaNet. Journal of Clinical Oncology, 27, 6041–6051.

Baccarani, M., Deininger, M.W., Rosti, G., Hoch-haus, A., Soverini, S., Apperley, J.F., Cervantes, F., Clark, R.E., Cortes, J.E., Guilhot, F.,

Hjorth-Hansen, H., Hughes, T.P., Kantarjian, H.M., Kim, D.W., Larson, R.A., Lipton, J.H., Mahon, F.X., Martinelli, G., Mayer, J., M€uller, M.C., Niederwieser, D., Pane, F., Radich, J.P., Rous-selot, P., Saglio, G., Saußele, S., Schiffer, C., Sil-ver, R., Simonsson, B., Steegmann, J.L., Goldman, J.M. & Hehlmann, R. (2013) Euro-pean LeukemiaNet recommendations for the management of chronic myeloid leukemia: 2013. Blood, 122, 872–884.

Bhamidipati, P.K., Kantarjian, H., Cortes, J., Cor-nelison, A.M. & Jabbour, E. (2013) Management of imatinib-resistant patients with chronic mye-loid leukemia. Therapeutic Advances in Hematol-ogy, 4, 103–117.

Chomel, J.C., Bonnet, M.L., Sorel, N., Bertrand, A., Meunier, M.C., Fichelson, S., Melkus, M., Bennaceur-Griscelli, A., Guilhot, F. & Turhan, A.G. (2011) Leukemic stem cell persistence in chronic myeloid leukemia patients with sus-tained undetectable molecular residual disease. Blood, 118, 3657–3660.

Cortes, J.E., De Souza, C.A., Ayala, M., Lopez, J.L., Bullorsky, E., Shah, S., Huang, X., Babu, K.G., Abdulkadyrov, K., de Oliveira, J.S.R., Shen, Z.X.,

Sacha, T., Bendit, I., Liang, Z., Owugah, T., Szczudlo, T., Khanna, S., Fellague-Chebra, R. & le Coutre, P.D. (2016a) Switching to nilotinib versus imatinib dose escalation in patients with chronic myeloid leukaemia in chronic phase with suboptimal response to imatinib (LASOR): a randomised, open-label trial. The Lancet Hae-matology, 3, e581–e591.

Cortes, J.E., Lipton, J.H., Miller, C.B., Busque, L., Akard, L.P., Pinilla-Ibarz, J., Keir, C., Warsi, G., Lin, F.P. & Mauro, M.J. (2016b) Evaluating the impact of a switch to nilotinib on imatinib-related chronic low-grade adverse events in patients with CML-CP: the ENRICH study. Clinical Lymphoma, Myeloma & Leukemia, 16, 286–296.

Cross, N.C., White, H.E., M€uller, M.C., Saglio, G. & Hochhaus, A. (2012) Standardized definitions of molecular response in chronic myeloid leuke-mia. Leukemia, 26, 2172–2175.

Druker, B.J., Guilhot, F., O’Brien, S.G., Gath-mann, I., Kantarjian, H., GatterGath-mann, N., Dei-ninger, M.W., Silver, R.T., Goldman, J.M., Stone, R.M., Cervantes, F., Hochhaus, A., Pow-ell, B.L., Gabrilove, J.L., Rousselot, P., Reiffers,

J., Cornelissen, J.J., Hughes, T., Agis, H., Fis-cher, T., Verhoef, G., Shepherd, J., Saglio, G., Gratwohl, A., Nielsen, J.L., Radich, J.P., Simon-sson, B., Taylor, K., Baccarani, M., So, C., Let-vak, L., Larson, R.A. & Investigators, I.R.I.S. (2006) Five-year follow-up of patients receiving imatinib for chronic myeloid leukemia. The New England Journal of Medicine, 355, 2408– 2417.

Etienne, G., Huguet, F., Guerci-Bresler, A., Nicol-ini, F.E., Maloisel, F., Coiteux, V., Dauriac, C., Carpentier, N., Bourdeix, I., Tulliez, M. & Cony-Makhoul, P. (2016) Impact of ELN rec-ommendations in the management of first-line treated chronic myeloid leukaemia patients: a French cross-sectional study. British Journal of Haematology, 174, 71–80.

Gora-Tybor, J., Medras, E., Calbecka, M., Kolk-owska-Lesniak, A., Ponikowska-Szyba, E., Robak, T. & Jamroziak, K. (2015) Real-life com-parison of severe vascular events and other non-hematological complications in patients with chronic myeloid leukemia undergoing second-line nilotinib or dasatinib treatment. Leukemia & Lymphoma, 56, 2309–2314.

Hehlmann, R., Hochhaus, A. & Baccarani, M.; for European LeukemiaNet. (2007) Chronic myeloid leukaemia. Lancet, 370, 342–350.

Hochhaus, A., Larson, R.A., Guilhot, F., Radich, J.P., Branford, S., Hughes, T.P., Cortes, J.E. & Guilhot, F. (2017) Longterm outcomes of ima-tinib treatment for chronic myeloid leukemia. The New England Journal of Medicine, 376, 917– 927.

Hughes, T.P., Lipton, J.H., Spector, N., Cervantes, F., Pasquini, R., Clementino, N.C., Dorlhiac Lla-cer, P.E., Schwarer, A.P., Mahon, F.X., Rea, D., Branford, S., Purkayastha, D., Collins, L., Szczu-dlo, T. & Leber, B. (2014) Deep molecular responses achieved in patients with CML-CP who are switched to nilotinib after long-term imatinib. Blood, 124, 729–736.

ICH Steering Committee. (2003) ICH Harmonised Tripartite Guideline. Post-approval safety data management: Definitions and standards for expedited reporting, E2D. (2003). Available at: https://www.ich.org/fileadmin/Public_Web_Site/ ICH_Products/Guidelines/Efficacy/E2D/Step4/ E2D_Guideline.pdf. [Accessed 16 October, 2017].

Jabbour, E., Kantarjian, H., Jones, D., Talpaz, M., Bekele, N., O’Brien, S., Zhou, X., Luthra, R., Garcia-Manero, G., Giles, F., Rios, M.B., Ver-stovsek, S. & Cortes, J. (2006) Frequency and clinical significance of BCR-ABL mutations in patients with chronic myeloid leukemia treated

with imatinib mesylate. Leukemia, 20, 1767– 1773.

Kantarjian, H.M., Giles, F., Gattermann, N., Bhalla, K., Alimena, G., Palandri, F., Ossenkop-pele, G.J., Nicolini, F.E., O’Brien, S.G., Litzow, M., Bhatia, R., Cervantes, F., Haque, A., Shou, Y., Resta, D.J., Weitzman, A., Hochhaus, A. & le Coutre, P. (2007) Nilotinib (formerly AMN107), a highly selective BCR-ABL tyrosine kinase inhibitor, is effective in patients with Philadelphia chromosome-positive chronic myelogenous leukemia in chronic phase follow-ing imatinib resistance and intolerance. Blood, 110, 3540–3546.

Kantarjian, H.M., Giles, F.J., Bhalla, K.N., Pinilla-Ibarz, J., Larson, R.A., Gattermann, N., Ott-mann, O.G., Hochhaus, A., Radich, J.P., Saglio, G., Hughes, T.P., Martinelli, G., Kim, D.W., Shou, Y., Gallagher, N.J., Blakesley, R., Bac-carani, M., Cortes, J. & le Coutre, P.D. (2011) Nilotinib is effective in patients with chronic myeloid leukemia in chronic phase after ima-tinib resistance or intolerance: 24-month follow-up results. Blood, 117, 1141–1145.

Kuo, C.Y., Wang, P.N., Hwang, W.L., Tzeng, C.H., Bai, L.Y., Tang, J.L., Chang, M.C., Lin, S.F., Chen, T.Y., Chen, Y.C., Tan, T.D., Hseih, C.Y., Lin, C., Darko, M. & Cheng-Shyong, C. (2015) Safety and efficacy of nilotinib (NIL) in patients (Pts) with chronic phase (CP) or accelerated phase (AP) Philadelphia chromosome positive (Ph+) chronic myelogenous leukemia (CML) with resistance or intolerance to imatinib mesy-late (IM): results from the multicenter, observa-tional NOVEL study in Taiwan. Blood (ASH Annual Meeting Abstracts), 126, 1593. Miyamura, K., Miyamoto, T., Tanimoto, M.,

Yamamoto, K., Kimura, S., Kawaguchi, T., Mat-sumura, I., Hata, T., Tsurumi, H., Saito, S., Hino, M., Tadokoro, S., Meguro, K., Hyodo, H., Yamamoto, M., Kubo, K., Tsukada, J., Kondo, M., Aoki, M., Okada, H., Yanada, M., Ohya-shiki, K. & Taniwaki, M. (2016) Switching to nilotinib in patients with chronic myeloid leuke-mia in chronic phase with molecular suboptimal response to frontline imatinib: SENSOR final results and BIM polymorphism substudy. Leuke-mia Research, 51, 11–18.

National Comprehensive Cancer Network (NCCN). (2017) Clinical Practice Guidelines in Oncology: Chronic Myelogenous Leukemia. Ver-sion 1.2018. Available at: https://www.nccn.org/ professionals/physician_gls/pdf/cml.pdf [Accessed 16 October, 2017].

Nicolini, F.E., Turkina, A., Shen, Z.X., Gallagher, N., Jootar, S., Powell, B.L., De Souza, C., Zheng,

M., Szczudlo, T. & le Coutre, P. (2012) Expand-ing Nilotinib Access in Clinical Trials (ENACT): an open-label, multicenter study of oral nilo-tinib in adult patients with imanilo-tinib-resistant or imatinib-intolerant Philadelphia chromosome-positive chronic myeloid leukemia in the chronic phase. Cancer, 118, 118–126.

O’Brien, S.G., Guilhot, F., Larson, R.A., Gath-mann, I., Baccarani, M., Cervantes, F., Cornelis-sen, J.J., Fischer, T., Hochhaus, A., Hughes, T., Lechner, K., Nielsen, J.L., Rousselot, P., Reiffers, J., Saglio, G., Shepherd, J., Simonsson, B., Grat-wohl, A., Goldman, J., Kantarjian, H., Taylor, K., Verhoef, G., Bolton, A.E., Capdeville, R. & Druker, B.J.; IRIS Investigators (2003) Imatinib compared with interferon and low-dose cytara-bine for newly diagnosed chronic-phase chronic myeloid leukemia. The New England Journal of Medicine, 348, 994–1004.

Rea, D., Rosti, G., Cross, C., Cross, N., Hellman, A., Niederwieser, D., Pungolino, E., Falzetti, F., Pregno, P., Orlandi, E.M., Medina Almeida, A., Illes, A., Fernandez, C., Dezzani, L., Pellegrino, A., Costantini, C., Walasek, C., Saglio, G., Steeg-mann, J.L. & Baccarani, M. (2015) Enestpath: a phase III study to assess the effect of nilotinib treatment duration on treatment-free remission (TFR) in chronic phasechronic myeloid leuke-mia (CP-CML) patients (pts) previously treated with imatinib: interim analysis from the first year of induction phase. Blood (ASH Annual Meeting Abstracts), 126, 4040.

Ross, D.M., Branford, S., Seymour, J.F., Schwarer, A.P., Arthur, C., Bartley, P.A., Slader, C., Field, C., Dang, P., Filshie, R.J., Mills, A.K., Grigg, A.P., Melo, J.V. & Hughes, T.P. (2010) Patients with chronic myeloid leukemia who maintain a complete molecular response after stopping ima-tinib treatment have evidence of persistent leu-kemia by DNA PCR. Leuleu-kemia, 24, 1719–1724. Soverini, S., Colarossi, S., Gnani, A., Rosti, G.,

Castagnetti, F., Poerio, A., Iacobucci, I., Ama-bile, M., Abruzzese, E., Orlandi, E., Radaelli, F., Ciccone, F., Tiribelli, M., di Lorenzo, R., Carac-ciolo, C., Izzo, B., Pane, F., Saglio, G., Baccarani, M. & Martinelli, G.; GIMEMA Working Party on Chronic Myeloid Leukemia. (2006) Contri-bution of ABL kinase domain mutations to ima-tinib resistance in different subsets of Philadelphia-positive patients: by the GIMEMA Working Party on Chronic Myeloid Leukemia. Clinical Cancer Research, 12, 7374–7379. Weisberg, E., Manley, P., Mestan, J., Cowan-Jacob,

S., Ray, A. & Griffin, J.D. (2006) AMN107 (nilo-tinib): a novel and selective inhibitor of BCR-ABL. British Journal of Cancer, 94, 1765–1769.