Blood Journal
Leading the way in experimental and clinical research in hematology

Bosutinib safety and management of toxicity in leukemia patients with resistance or intolerance to imatinib and other tyrosine kinase inhibitors

  1. Hagop M. Kantarjian1,
  2. Jorge E. Cortes1,
  3. Dong-Wook Kim2,
  4. H. Jean Khoury3,
  5. Tim H. Brümmendorf4,5,
  6. Kimmo Porkka6,
  7. Giovanni Martinelli7,
  8. Simon Durrant8,
  9. Eric Leip9,
  10. Virginia Kelly10,
  11. Kathleen Turnbull11,
  12. Nadine Besson12, and
  13. Carlo Gambacorti-Passerini13
  1. 1Department of Leukemia, University of Texas MD Anderson Cancer Center, Houston, TX;
  2. 2Department of Hematology, Seoul St. Mary’s Hospital, Seoul, South Korea;
  3. 3Department of Hematology and Medical Oncology, Winship Cancer Institute of Emory University, Atlanta, GA;
  4. 4Hematology, Oncology, and Stem Cell Transplantation, Universitätsklinikum Aachen, Rheinisch-Westfaelische Technische Hochschule Aachen, Aachen, Germany;
  5. 5Hematology, Oncology, and Stem Cell Transplantation, Universitätsklinikum Hamburg-Eppendorf, Hamburg, Germany;
  6. 6Division of Hematology, Helsinki University Central Hospital, Helsinki, Finland;
  7. 7Department of Experimental, Diagnostic and Specialty Medicine, University of Bologna, Bologna, Italy;
  8. 8Bone Marrow Transplant Unit, Royal Brisbane Hospital, Herston, Australia;
  9. 9Oncology Clinical Statistics,
  10. 10Oncology Clinical Research, and
  11. 11Oncology Clinical Development, Pfizer Inc., Cambridge, MA;
  12. 12Pfizer Global Research and Development, Paris, France; and
  13. 13Clinical Research Unit, University of Milano-Bicocca, Monza, Italy
  1. Presented in part in abstract form at the annual meeting of the American Society of Hematology, December 7-10, 2013, New Orleans, LA.

This article has an Erratum 124(6):981

Key Points

  • Bosutinib had manageable toxicity and acceptable tolerability, with adverse events distinct from that of other tyrosine kinase inhibitors.

  • Adverse events were primarily characterized by early-onset gastrointestinal events that were low grade, transient, and manageable.


Bosutinib is an oral, dual SRC/ABL tyrosine kinase inhibitor (TKI) with clinical activity in Philadelphia chromosome–positive (Ph+) leukemia. We assessed the safety and tolerability of bosutinib 500 mg per day in a phase 1/2 study in chronic-phase (CP) chronic myeloid leukemia (CML) or advanced Ph+ leukemia following resistance/intolerance to imatinib and possibly other TKIs. Patient cohorts included second-line CP CML (n = 286), third-/fourth-line CP CML (n = 118), and advanced leukemia (n = 166). Median bosutinib duration was 11.1 (range, 0.03-83.4) months. Treatment-emergent adverse events (TEAEs) in each cohort were primarily gastrointestinal (diarrhea [86%/83%/74%], nausea [46%/48%/48%], and vomiting [37%/38%/43%]). Diarrhea presented early, with few (8%) patients experiencing grade 3/4 events; dose reduction due to diarrhea occurred in 6% of affected patients. Grade 3/4 myelosuppression TEAEs were reported in 41% of patients; among affected patients, 46% were managed with bosutinib interruption and 32% with dose reduction. Alanine aminotransferase elevation TEAEs occurred in 17% of patients (grade 3/4, 7%); among patients managed with dose interruption, bosutinib rechallenge was successful in 74%. Bosutinib demonstrated acceptable safety with manageable toxicities in Ph+ leukemia. This trial (NCT00261846) was registered at (this manuscript is based on a different data snapshot from that in


Tyrosine kinase inhibitors (TKIs) designed to inhibit the BCR-ABL oncoprotein are the backbone of treatment of all phases of Philadelphia chromosome–positive (Ph+) chronic myeloid leukemia (CML)1,2 and acute lymphoblastic leukemia (ALL).3 However, TKIs are associated with toxicity that may prevent patients from maintaining drug intensity, limiting therapeutic benefit. Indefinite TKI treatment duration makes tolerability and manageability of these adverse events (AEs) essential to therapeutic success. Familiarity with these AEs can aid monitoring and early identification of drug toxicity and appropriate intervention, including TKI dose modifications and concomitant medication support.

Imatinib, a TKI with specificity for BCR-ABL, as well as KIT and platelet-derived growth factor receptor (PDGFR), has been the standard of care for CML patients.4,5 However, many patients cannot tolerate imatinib because of toxicities, including gastrointestinal symptoms, arthralgia/myalgia, rash, fatigue, and myelosuppression.6-8 Intolerance also occurs with the second-generation TKIs dasatinib9 and nilotinib10 as first-line CML treatment. Thus, safety and tolerability of each TKI may influence treatment selection.

Bosutinib (SKI-606) is an oral, dual competitive SRC and ABL TKI with minimal activity against PDGFR or KIT.11,12 In a phase 1/2 study, bosutinib demonstrated efficacy in all phases of CML previously treated with imatinib alone or imatinib followed by dasatinib and/or nilotinib.13-15

Bosutinib was associated with acceptable safety and tolerability across cohorts; mild or moderate gastrointestinal events and rash were the most common AEs.14,15 Although myelosuppression is universally observed during TKI therapy for CML and Ph+ ALL, nonhematologic AEs associated with bosutinib appear distinct from those of imatinib, dasatinib, and nilotinib.16-18 The current analysis from the same phase 1/2 study characterizes toxicities associated with bosutinib and describes toxicity management in Ph+ leukemia patients. Toxicity was assessed in patients receiving bosutinib as chronic-phase (CP) second-line (CP2L) or third-/fourth-line (CP3L) therapy and in patients with advanced (ADV) disease, including accelerated-phase (AP) or blast-phase (BP) CML and ALL.


Study design

This was an open-label, 2-part, multicenter, phase 1/2 study. Part 1 was a phase 1 dose-escalation study that determined a recommended dose of bosutinib 500 mg per day in primarily imatinib-resistant CP CML patients.14 No dose-limiting toxicities occurred in the 400- and 500-mg cohorts; in the 600-mg cohort, 1 of 12 patients experienced a dose-limiting toxicity (grade 3 rash, nausea, and vomiting) and additional patients experienced grade 2 alanine aminotransferase (ALT) elevation, grade 2 rash, and grade 3 diarrhea.14 Bosutinib 500 mg per day was selected as the part 2 starting dose, despite not reaching a protocol-defined maximum tolerated dose because of observed AEs with 600 mg per day. Clinical benefit was observed at all doses.

Part 2 is a phase 2 safety and efficacy evaluation of bosutinib 500 mg per day in CP, AP, or BP CML or Ph+ ALL patients with resistance or intolerance to imatinib and possibly dasatinib and/or nilotinib. Dose escalation to 600 mg per day was permitted for lack of efficacy (complete hematologic response not reached by week 8 or complete cytogenetic response not reached by week 12) if no drug-related grade 3/4 AE had occurred. Methodology and overall study results were reported previously for CP patients.14,15

The protocol was approved by the central or institutional review board for each study site, and informed consent was obtained in accordance with the Declaration of Helsinki.


Patients were aged 18 years or older with cytogenetic or polymerase chain reaction confirmation of Ph+ CML or ALL, resistant to full-dose imatinib (CP CML, ≥600 mg; ADV, ≥800 mg) or intolerant to any dose of imatinib. Patients in the CP3L cohort were resistant to dasatinib 100 mg per day or nilotinib 800 mg per day and/or intolerant to any dose of dasatinib; patients in the ADV cohort may also have been resistant or intolerant to dasatinib and/or nilotinib. Additional criteria included Eastern Cooperative Oncology Group performance status score of 0 or 1 (2 permitted for the ADV cohort); adequate bone marrow (CP2L and CP3L cohorts), hepatic, and renal function (defined previously13,14); ≥7 days since any antiproliferative treatment (except hydroxyurea and anagrelide); ≥3 months postallogeneic hematopoietic stem cell transplantation; and recovery to grade 0 or 1 or to baseline from any toxicities associated with prior anticancer treatment (except alopecia).

Safety evaluations

The safety population included patients from parts 1 and 2 who received ≥1 bosutinib dose. AEs were assessed continually and graded for severity according to the National Cancer Institute Common Terminology Criteria for Adverse Events, version 3.0. Laboratory tests, electrocardiograms (ECGs), vital signs, and physical examinations were performed; in some instances, abnormalities may have been reported as AEs. Treatment-emergent AEs (TEAEs) were defined as any event increasing in severity from baseline or any new event starting during bosutinib therapy or within 30 days of the last bosutinib dose.



Five hundred seventy patients received bosutinib and were evaluated for safety. Patients were grouped by disease phase and by prior therapy into 3 cohorts (Table 1). The CP2L cohort (n = 286) included patients with CP CML following resistance (n = 196) or intolerance (n = 90) to prior imatinib only (ie, no other TKIs). The CP3L cohort (n = 118) included patients with CP CML following resistance/intolerance to prior imatinib plus dasatinib resistance (n = 38), dasatinib intolerance (n = 50), nilotinib resistance (n = 26), or nilotinib intolerance or exposure to all 3 TKIs (n = 4). The ADV cohort (n = 166) included patients with AP CML (n = 78), BP CML (n = 64), or ALL (n = 24) following resistance/intolerance to prior imatinib or to multiple prior TKIs (ie, imatinib plus dasatinib and/or nilotinib).

View this table:
Table 1

Patient baseline demographic and clinical characteristics

Treatment summary

Median duration of bosutinib treatment across cohorts (analysis cutoff based on an interim unlocked database: May 15, 2013) was 11.1 (range, 0.03-83.4) months (CP2L, 24.8 [0.2-83.4] months; CP3L, 8.5 [0.2-78.1] months; ADV, 4.0 [0.03-77.9] months). The time from the last patient’s first dose to the cutoff was ≥48 months for CP2L and ≥36 months for CP3L and ADV.

Across cohorts, dose reductions and interruptions due to AEs occurred in 45% and 65% of patients, respectively, and were more common in CP2L (49% and 72%) and CP3L (50% and 66%) cohorts vs the ADV cohort (34% and 51%) (Table 2), likely reflecting differences in bosutinib exposure and treatment duration. More CP2L and CP3L patients intolerant vs resistant to prior TKIs had dose reductions and interruptions; intolerant patients were more likely to discontinue bosutinib treatment because of AEs, particularly thrombocytopenia (Table 2). Treatment interruptions due to AEs occurring ≤4 weeks after bosutinib initiation were comparable among the overall CP2L, CP3L, and ADV cohorts but were more common in CP2L and CP3L patients intolerant vs resistant to prior TKIs (Table 2).

View this table:
Table 2

Treatment modifications and discontinuations due to AEs

Median cumulative duration of dose reduction to 400 mg per day was 88 days (range, 1-2323 days) (CP2L: 169 days; CP3L: 63.5 days; ADV: 59.5 days) and 127 days (range, 1-1480 days) to 300 mg per day (CP2L: 115 days; CP3L: 180 days; ADV: 158 days). Among patients with bosutinib dose reduction to 400 mg per day in the CP2L and CP3L cohorts, 18 of 130 (14%) and 4 of 52 (8%), respectively, had a major cytogenetic response (MCyR) before and after reduction, and an additional 56 of 130 (43%) and 15 of 52 (29%) patients first achieved MCyR after reduction; 3 CP2L patients and 2 CP3L patients lost their previously attained MCyR after reduction. Among those with dose reduction to 300 mg per day in the CP2L and CP3L cohorts, 17 of 49 (35%) and 4 of 21 (19%) patients had MCyR both before and after reduction; 8 of 49 (16%) and 4 of 21 (19%) patients first achieved MCyR after reduction; 2 and 0 patients lost their previously attained MCyR after reduction. For ADV patients with bosutinib dose reduction to 400 or 300 mg per day, 11 of 50 (22%) and 4 of 23 (17%) patients had an overall hematologic response (OHR) before and after first reduction; 15 of 50 (30%) and 3 of 23 (13%) first achieved OHR after reduction; 3 and 7 lost their previously attained OHR after reduction. Of 558 patients who received an initial dose of bosutinib 400 or 500 mg per day, 92 (17%) had dose escalation to 600 mg per day due to lack of efficacy; median treatment duration at 600 mg per day was 159 (range, 4-1811) days.

At the time of analysis, 147 (26%) patients were receiving treatment. Disease progression was the most common reason for treatment discontinuation (25%), followed by AEs (22%). Rates of discontinuation due to AEs did not substantially differ across patient cohorts (Table 2); median time to bosutinib discontinuation due to AEs was 5.1 (range, 0.2-75.3) months. Of those who discontinued treatment because of an AE, 29% and 28% had their AE managed with bosutinib dose reduction to 300 or 400 mg per day, respectively, before discontinuation; 42% were not reduced to <500 mg per day before discontinuation. Of patients reduced to bosutinib 300 or 400 mg per day because of an AE, 12 of 93 (13%) and 41 of 232 (18%), respectively, later discontinued bosutinib because of progressive disease and 8 of 93 (9%) and 17 of 232 (7%) discontinued due to unsatisfactory response/lack of efficacy. Among 272 patients who had achieved/maintained an MCyR, treatment was discontinued for 139 (51%); more patients discontinued treatment in the ADV (84%) vs CP2L (38%) or CP3L (60%) cohorts. The most common reasons for treatment discontinuation among patients who had achieved/maintained an MCyR were disease progression (CP2L, 13%; CP3L, 15%; ADV, 46%) and AEs (10%, 23%, 18%).

Overall safety

Bosutinib toxicities were generally of mild to moderate severity and managed with concomitant medication, dose interruption and/or dose reduction, or resolved spontaneously. The most frequently reported TEAEs were gastrointestinal events (diarrhea, nausea, and vomiting) and thrombocytopenia; incidences did not substantially differ across cohorts overall (Table 3). Although diarrhea was common (82%), maximum severity was grade 1 (45%) or 2 (28%) in most patients; few experienced maximum grade 3 (8%) or 4 (<1%; 1 ADV patient) diarrhea. Based on differences of ≥10% for TEAEs occurring in ≥10% of patients, rash was more common and pyrexia less common in patients intolerant vs resistant to prior imatinib treatment in the CP2L cohort (Table 3). In the CP3L cohort, vomiting, rash, dyspnea, and pleural effusion were more common and neutropenia was less common in patients intolerant to prior dasatinib therapy vs patients resistant to prior TKI therapy (Table 3). Thrombocytopenia was most common in CP3L resistant to nilotinib. Grade 3/4 TEAEs reported in ≥10% of all patients were hematologic events, including thrombocytopenia (30%), anemia (14%), and neutropenia (14%; Table 3). Among 92 patients who were dose escalated to bosutinib 600 mg per day, generally no increases >5% were observed in the frequency or severity of individual TEAEs following dose escalation.

View this table:
Table 3

TEAEs reported for ≥10% of patients

Serious AEs occurred in 248 (44%) patients, more commonly in the ADV cohort (58%) vs CP2L (39%) and CP3L (34%) cohorts. The most frequently reported serious AEs (≥16 patients) were pneumonia (n = 29), pleural effusion (n = 23), pyrexia (n = 21), and thrombocytopenia (n = 16).

On-treatment grade 3/4 hematologic laboratory abnormalities were reported frequently, particularly in the ADV cohort (Table 4). Grade 3/4 nonhematologic laboratory abnormalities reported on treatment in ≥5% of patients included hypermagnesemia (10%), increased ALT (8%), hypophosphatemia (8%), and increased lipase (7%; Table 4). Grade 3/4 hypermagnesemia was more common in patients intolerant vs resistant to prior TKIs (CP2L or CP3L); hematologic laboratory abnormalities were more common in CP2L patients intolerant vs resistant to prior TKIs.

View this table:
Table 4

Maximum on-treatment grade 3 and 4 laboratory abnormalities reported for ≥5% of patients

Forty-four deaths (including 32 deaths in the ADV cohort) occurred within 30 days after the last bosutinib dose (supplemental Table A1, available on the Blood Web site). Common reasons included CML disease progression (n = 25 [4%]) and AEs considered unrelated to bosutinib by the investigators (n = 16 [3%]), most frequently pneumonia/pneumonitis (n = 5) and cardiac events (n = 6). Three deaths were attributed to an AE considered bosutinib-related by the investigator (lower gastrointestinal hemorrhage with thrombocytopenia [CP3L cohort], myocardial infarction [ADV], and acidosis [with respiratory failure complicated by sepsis; ADV]). All deaths in the CP2L cohort occurred in patients with resistance (not intolerance) to prior imatinib treatment; all deaths in the CP3L cohort occurred in patients with resistance or intolerance to prior dasatinib (not nilotinib) treatment (supplemental Table A1).

Characteristics and management of individual toxicities

Diarrhea occurred in 82% of all patients with little variability across patient cohorts. Diarrhea TEAEs were noted soon after bosutinib initiation, with a median time to first event of 2.0 days (Table 5). Diarrhea incidence decreased over time on treatment (supplemental Figure A1). Diarrhea TEAEs were predominantly of low severity and were typically transient, with a median duration of 1 to 2 days per event across cohorts. Diarrhea TEAEs were considered bosutinib-related in 455 (80%) patients, with <2% of bosutinib-related diarrhea AEs considered serious (n = 8). Concomitant antidiarrheal medication was used in 65% of patients who experienced diarrhea, representing the primary management strategy. Loperamide (58%; Imodium) was used most frequently. Less commonly, diarrhea was managed by bosutinib interruption (14%) and reduction (6%), allowing most affected patients to maintain their bosutinib dose. Few patients (1%) in the safety population discontinued treatment because of diarrhea, emphasizing the manageability of this toxicity in the majority of patients. Nearly all (97%) patients whose diarrhea was managed with temporary interruption of bosutinib dosing and who were subsequently rechallenged did not discontinue bosutinib because of diarrhea. Diarrhea characteristics were generally consistent across the 3 cohorts.

View this table:
Table 5

Characteristics and management of diarrhea TEAEs

Pleural effusion TEAEs occurred in 57 patients (10%) in the safety population, including 17 (3%) with maximum toxicity grade of 3 (n = 15), 4 (n = 1), or 5 (n = 1; CP2L, death on study day 2290 due to congestive heart failure and pleural effusion, both unrelated to bosutinib). Twenty-one patients (4%) experienced a serious pleural effusion TEAE. Among the 3 cohorts, pleural effusions were more common in the CP3L (all grades, n = 18 of 118 [15%]; grade 3/4/5, 4 of 118 [3%]) vs CP2L (n = 23 of 286 [8%] and n = 6 of 286 [2%], respectively) and ADV (n = 16 of 166 [10%] and 7 of 166 [4%]) cohorts. Median time to onset of the first pleural effusion was 541 (range, 3-1993) days; median event duration was 21 (range, 1-421) days. Of patients who experienced pleural effusion, 26% had dose reductions, 47% had dose interruptions, and 58% were managed with concurrent medications. Four (1%) patients discontinued bosutinib because of pleural effusion.

Of 57 patients with pleural effusion during bosutinib treatment, 22 (39%) had a history of pleural effusion, of whom 16 attributed prior TKI discontinuation to pleural effusion (dasatinib, n = 14; imatinib and nilotinib, n = 1 each). Of 17 CP3L patients with prior dasatinib treatment who experienced pleural effusion during bosutinib treatment, 11 (65%) indicated pleural effusion as the reason for dasatinib intolerance. Most patients (n = 33 of 57 [58%]) with pleural effusions during bosutinib treatment had never been exposed to dasatinib.

The collective incidence of cardiac-related TEAEs (combined analysis of MedDRA system organ class Cardiac Disorders and Investigations [cardiac and vascular terms]) was relatively low (18% of patients experienced any cardiac event, of whom most presented with prior or ongoing cardiac disorders at screening), with most patients (10%) experiencing maximum grade 1 or 2 events, 5% maximum grade 3 events, 2% maximum grade 4 events, and 1% grade 5 events (including death due to myocardial infarction [n = 3], congestive cardiac failure [n = 2], and cardiac failure and coronary artery disease [n = 1 each]). The most common cardiac events were pericardial effusion (3%), atrial fibrillation, congestive cardiac failure, tachycardia, and palpitations (2% each). In the overall population, 33 of 570 (6%) patients experienced cardiac events considered bosutinib-related; 13 patients (2%) had treatment-related grade 3/4 events. Management of cardiac TEAEs included bosutinib dose interruption (24%) or reduction (7%); 40% received concomitant medication. Few patients (n = 12 [2%]) in the safety population discontinued treatment due to cardiac events. Of patients whose cardiac TEAE was managed with dose interruption, 19 of 24 (79%) were rechallenged with bosutinib; only 4 rechallenged patients subsequently discontinued bosutinib due to a cardiac event.

Based on ECG laboratory data, only 1 patient experienced on-treatment grade 3 Fridericia corrected QT interval prolongation (patient had eligibility violation of grade 2 prolongation at baseline). Based on echocardiogram/multiple gated acquisition scan analysis, 1 patient had a shift in left ventricular ejection fraction (LVEF) decrease from normal (baseline) to grade 3 (on treatment); 3 patients had grade 3 LVEF at baseline that continued on therapy; 1 patient had on-treatment grade 3 LVEF with no available baseline results.

Overall, cardiac-related toxicities on bosutinib were infrequent, occurred mostly in patients with preexisting cardiac conditions reported at baseline, and were manageable with dose interruption and/or reduction and concomitant medication.

Relatively few vascular disorders (MedDRA system organ class) were observed (overall, 76 [13%]; treatment related, 18 [3%]). Hypertension was the most frequent TEAE (any grade, 37 [6%]; grade 3/4, 11 [2%]); for all other vascular events, frequency was ≤2%. Notably, 1 event of peripheral arterial occlusive disease (grade 2) was reported in a CP3L patient with prior nilotinib exposure; the event was considered serious, unrelated to bosutinib, and resolved within 10 days.

Myelosuppression (ie, anemia, neutropenia, and thrombocytopenia) was reported frequently; thrombocytopenia was the most commonly observed grade 3/4 TEAE (Table 3) and grade 3/4 laboratory abnormality (Table 4). Several patients with grade 3/4 laboratory abnormalities on treatment had baseline grade 3/4 hematologic laboratory abnormalities (thrombocytopenia, n = 56 of 202 [28%]; neutropenia, n = 30 of 140 [21%]; anemia, n = 15 of 108 [14%]; supplemental Table A2).

Median time to first myelosuppression TEAE was 22.0 days across cohorts (Table 6). Myelosuppression TEAEs were generally transient; median event duration was 14.0 days (range, 1-1373 days). Myelosuppression events were managed primarily with bosutinib treatment modifications; 46% of affected patients temporarily stopped treatment, 32% underwent dose reduction. Across cohorts, 10% of affected patients received growth factor support; 1% underwent transfusion for management of ≥1 event. Although these events represented the most severe toxicities, only 7% of patients in the safety population discontinued treatment because of a myelosuppression TEAE, suggesting overall manageability and tolerability of these events.

View this table:
Table 6

Characteristics and management of myelosuppression TEAEs

Across cohorts, ALT and aspartate aminotransferase (AST) elevation TEAEs were reported in 99 patients (17%) and 81 patients (14%), respectively; 112 patients (20%) had either or both types of elevation. Maximum grade 3 ALT/AST TEAEs were reported in 39 patients (7%); maximum grade 4 TEAEs were reported in 2 patients (<1%). Median time to first ALT/AST TEAE was 33.5 days; median event duration was 21.0 days. Despite the relative frequency of ALT and AST elevations, relatively few patients (13 [2%]) discontinued treatment because of these events (Table 7). The incidence of ALT events decreased over time on treatment (supplemental Figure A2). ALT and AST TEAEs were primarily managed with bosutinib dose reduction or temporary interruption (18% and 35%, respectively, of affected patients). Among patients who required a temporary dose interruption for ALT and/or AST elevation and were subsequently rechallenged with bosutinib, 9 of 35 (26%) discontinued bosutinib due to ALT/AST elevation. ALT and AST elevation TEAEs were managed with concomitant medication in 13% of affected patients.

View this table:
Table 7

Characteristics and management of ALT and AST TEAEs


This analysis comprehensively assessed toxicities and toxicity management associated with long-term bosutinib therapy (CP2L, 4-year minimum follow-up [unless discontinued earlier]: CP3L and ADV, 3-year minimum follow-up [unless discontinued earlier]) in patients with Ph+ leukemia resistant or intolerant to prior TKI therapy. Overall, bosutinib tolerability did not differ substantially among Ph+ leukemia patients, regardless of disease phase and treatment line. Within second-line and third-/fourth-line CP CML cohorts, dose reductions/interruptions and treatment discontinuation due to AEs, particularly thrombocytopenia, were more common in patients intolerant vs resistant to prior TKIs; in the CP2L cohort, patients intolerant to prior TKI therapy had 0 deaths in 90 patients within 30 days of last dose vs 7 in 196 for patients resistant to prior TKI therapy. Among all cohorts, the percentage of patients discontinuing bosutinib because of AEs in this analysis (CP2L, 22%; CP3L, 25%; ADV, 18%) was similar to that reported for first-line bosutinib therapy in the phase 3 Bosutinib Efficacy and Safety in Newly Diagnosed Chronic Myeloid Leukemia (BELA) trial (19%) with a median bosutinib treatment duration of 13.8 months.19

Gastrointestinal events, particularly diarrhea, were the most commonly reported TEAEs (all grades, grade 3/4). Although diarrhea occurred in most patients (CP2L, 86%; CP3L, 83%; ADV, 74%), events were typically transient and of mild to moderate severity, with most patients experiencing diarrhea early after treatment initiation (median time to first event, 2.0 days) and few treatment-related events reported as serious. Gastrointestinal AEs (diarrhea/nausea/vomiting) were less frequently reported in patients with imatinib-resistant/intolerant CP CML previously treated with dasatinib 100 mg per day (25%/18%/7%, respectively)19 or nilotinib (12%/25%/13%).20 These gastrointestinal AEs occurred more frequently in the CP2L (86%/46%/37%) and CP3L (83%/48%/38%) CML cohorts in the present analysis than in patients receiving first-line bosutinib treatment (68%/31%/32%).19 Diarrhea AEs associated with bosutinib in all lines of therapy were manageable with concurrent medications and dose interruptions and/or reductions. No patients discontinued therapy primarily because of diarrhea in the first-line setting,19 and only 1% of patients receiving bosutinib second-line (or later) discontinued due to diarrhea. Among patients who had dose interruption due to diarrhea and were subsequently redosed, 97% were successfully rechallenged with bosutinib. These data suggest bosutinib remains well tolerated despite the relatively high incidence of gastrointestinal events. With proper management, these events are transient and mild for most patients.

Treatment-related pleural effusion is associated with dasatinib (all grades, 14%) in patients with imatinib-resistant/intolerant CP CML,20 but not nilotinib (all grades, 1%).21 In our study, pleural effusion TEAEs occurred in relatively few CP CML patients with imatinib resistance/intolerance (CP2L cohort, 8%) and overall (10%) during bosutinib treatment. The proportion of patients experiencing pleural effusion was highest in the CP3L cohort (n = 18 [15%]), and the majority (17 of 18) of affected patients received prior dasatinib. Furthermore, 11 of these 17 patients reported pleural effusion as the reason for dasatinib intolerance. Notably, pleural effusions rarely led to bosutinib discontinuation (1%).

Cardiac and vascular events were observed at relatively low frequency in this study (any cause, 18% and 13%, respectively; treatment-related, 6% and 3%), with low incidences of individual events (≤3% and ≤6%). The majority of patients requiring dose interruption for cardiac toxicity were successfully rechallenged; 2% of patients discontinued bosutinib for cardiac-related events. Notably, the incidence of drug-related cardiac AEs in first-line CP CML patients was also low (4%).19

In our study, the most common grade 3/4 TEAEs and laboratory abnormalities were myelosuppression events, which is unsurprising because myelosuppression is commonly associated with other BCR-ABL TKIs (dasatinib and nilotinib) as second-line CML treatment.16-18 In a phase 3 trial of dasatinib in patients with imatinib-resistant/intolerant CP CML, reported rates of grade 3/4 treatment-related myelosuppression laboratory abnormalities were 35% for neutropenia, 23% for thrombocytopenia, and 13% for anemia after a median treatment duration of 22 months.20 Similarly, a phase 2 trial of nilotinib in imatinib-resistant/intolerant CP CML reported incidences of grade 3/4 laboratory abnormalities of 31% for neutropenia, 30% for thrombocytopenia, and 11% for anemia.21 Rates of grade 3/4 laboratory abnormalities of neutropenia were lower with bosutinib in both the CP2L (17%) and CP3L (20%) cohorts vs dasatinib and nilotinib, whereas rates of thrombocytopenia (CP2L, 25%; CP3L, 26%) and anemia (CP2L, 14%; CP3L, 8%) were fairly similar. The incidences of grade 3/4 myelosuppression laboratory abnormalities with bosutinib were higher in ADV patients (thrombocytopenia, 60%; neutropenia, 41%; anemia, 35%) than in CP CML patients. Myelosuppression was managed with bosutinib dose modifications and less frequently with transfusions and growth factors; myelosuppression events were the primary reason for treatment discontinuation in 7% of patients. Myelosuppression grade 3/4 laboratory abnormalities with bosutinib were lower (14% for thrombocytopenia, 11% for neutropenia, 6% for anemia) with first-line therapy.19

ALT and AST elevation TEAEs in our study were typically well tolerated; few patients (2%) discontinued treatment because of these TEAEs. The percentage of affected patients with bosutinib dose interruptions (35%) or reductions (18%) due to ALT or AST elevation was lower in our analysis vs first-line therapy (57% and 36%, respectively).19 In our analysis, the majority (74%) of patients subsequently rechallenged with bosutinib following dose interruption due to ALT and/or AST elevation did not discontinue treatment because of ALT/AST elevation, again suggesting that these toxicities, while notable and requiring monitoring, were manageable.

Generally, bosutinib demonstrated acceptable tolerability in patients with Ph+ CML and ALL, with an AE profile distinct from that of other TKIs. Toxicities observed with bosutinib treatment were generally manageable with treatment modification and/or concomitant medication. Additional experience with bosutinib treatment may further improve the management of observed toxicities. Our findings support the continued clinical development of bosutinib as monotherapy for the treatment of Ph+ CML patients.


Contribution: J.E.C., H.J.K., and C.G.-P. contributed to the study design; J.E.C., D.-W.K., T.H.B., G.M., N.B., and K.T. collected and assembled the data; H.M.K., J.E.C., D.-W.K., H.J.K., T.H.B., K.P., G.M., S.D., E.L., V.K., K.T., N.B., and C.G.-P. contributed to the analysis and/or interpretation of data; H.M.K., J.E.C., D.-W.K., H.J.K., T.H.B., K.P., G.M., S.D., and C.G.-P. provided study materials and/or enrolled patients in the study; E.L. performed statistical analyses; and all authors assisted in the writing and/or critical review of the manuscript and approved the final version of the manuscript for submission.

Conflict-of-interest disclosure: The study was sponsored by Pfizer Inc. H.M.K. received research funding from Pfizer. J.E.C. has served as an advisor/consultant for and received research funding from Ariad, Bristol-Myers Squibb, Chemgenex, Novartis, and Pfizer. D.-W.K. has served as an advisor/consultant for and received research funding and honoraria from Bristol-Myers Squibb, Novartis, and Pfizer. H.J.K. has served as an advisor/consultant for Pfizer. T.H.B. has served as an advisor/consultant for and received honoraria from Ariad, Bristol-Myers Squibb, Novartis, and Pfizer, and has also received research funding from Novartis. K.P. has served as an advisor/consultant for and received research funding and honoraria from Bristol-Myers Squibb and Novartis. G.M. has served as an advisor/consultant for and received honoraria from Bristol-Myers Squibb and Novartis. S.D. has served as an advisor/consultant for and received research funding from Pfizer. E.L. is a current employee of Pfizer, and V.K. and N.B. are former employees of Pfizer. K.T. was an employee of Inventiv, who was a paid consultant to Pfizer in the support of this study and development of this manuscript. K.T. is currently an employee of Pfizer. C.G.-P. has served as an advisor/consultant for and received research funding from Pfizer.

Correspondence: Hagop M. Kantarjian, University of Texas MD Anderson Cancer Center, 1515 Holcombe Blvd, Unit 428, Houston, TX 77030; e-mail: hkantarjian{at}


We acknowledge all of the participating patients and their families as well as the global network of investigators, research nurses, study coordinators, and operations staff.

This study was sponsored by Wyeth Research, which was acquired by Pfizer Inc. in October 2009. Programming support was provided by Gaurav Rathi of Pfizer Inc. Editorial/medical writing support was provided by Kimberly Brooks of SciFluent, Cynthia Gobbel of Complete Healthcare Communications, and Tiffany Brake of Complete Healthcare Communications, and was funded by Pfizer Inc.


  • The online version of this article contains a data supplement.

  • There is an Inside Blood commentary on this article in this issue.

  • The publication costs of this article were defrayed in part by page charge payment. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.

  • Submitted July 9, 2013.
  • Accepted December 4, 2013.


View Abstract