Summary Basis of Decision for Bosulif

 

Clinical Pharmacology

The clinical pharmacological data support the use of Bosulif for the specified indication. Appropriate warnings and precautions are in place in the approved Bosulif Product Monograph to address the identified safety concerns.

For further details on the clinical pharmacology, please refer to the Bosulif Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Efficacy

The clinical efficacy of Bosulif was based primarily on the interim analysis of a Phase I/II, multicentre, single-arm, open-label, 2-Part study, namely Study 200-WW (B1871006). A total of 571 patients were enrolled, and 570 received at least one study drug administration.

In Part 1 of the 200-WW study, the objective was to determine the maximum tolerated dose of Bosulif in patients with imatinib-resistant chronic phase (CP) CML. The primary efficacy endpoint was the rate of the major cytogenetic response (MCyR) in CP patients at various dose levels. Major cytogenetic response was defined as having a complete cytogenetic response (CCyR) or a partial cytogenetic response (PCyR). Eighteen patients, including seventeen with CP CML and one with accelerated phase (AP) CML were enrolled at three different dose levels: 400 mg (3 patients), 500 mg (3 patients), and 600 mg (12 patients). The patients were administered a single oral dose on Day 1, followed by continual daily dose administration from Day 3 through Day 28.

Following the completion of Part 1, the identified safety issues included one dose-limiting toxicity (Grade 3 rash) and two 'near' dose-limiting toxicities (Grade 2 rash, Grade 2 diarrhea) observed at the 600-mg level that prompted the stop of the dose escalation in Part 1 and the start of Part 2; prior to defining a maximum tolerated dose. As a result, the recommended starting dose for patients enrolled in Part 2 was that of 500 mg daily.

Part 2 of the study began following the completion of Part 1 and consisted of evaluating the efficacy and safety of 500 mg daily oral Bosulif in patients with CP CML and advanced CML who had received previous treatment with imatinib. The primary efficacy endpoint for this part of the study was the MCyR rate at Week 24 in imatinib-resistant CP CML patients, who had no prior Src, Abl, or Src-Abl inhibitor exposure other than imatinib. The efficacy endpoint used for the sample size calculation in both AP CML and blast phase (BP) CML was confirmed overall hematologic response (OHR), defined as either a major hematologic response (MHR) or return to chronic phase (RCP).

The patients enrolled in Part 1 were included in the analyses of Part 2. Dose escalations to 600 mg were permitted for patients not exhibiting toxicity and who had sub-optimal response to 500 mg. Dose reductions in 100 mg increments to a minimum dose of 300 mg daily were permitted for toxicity. Given Part 2 of the study currently remains ongoing, the efficacy analysis was based on interim efficacy results provided, with a cut-off date of February 15, 2012.

Interim results obtained from Part 2 of the 200-WW study did demonstrate that the primary endpoint of MCyR rate at Week 24 in imatinib-resistant second-line CP CML patients was met with 35.5% [90% Confidence Interval (CI): (29.7, 41.7)] of patients having attained MCyR at Week 24. The results of the two-sided test were also p<0.001, which means that the point estimate of the MCyR at 24 weeks was statistically significantly higher than 23% (null hypothesis). Therefore the primary endpoint was met.

Secondary efficacy objectives in the Part 2 study were to estimate the MCyR rate at Week 24 in patients with imatinib-intolerant CP CML who have imatinib exposure only, estimate time to and duration of MCyR in patients with CP CML who have had imatinib exposure only, estimate the time to and duration of complete hematological response (CHR), estimate the MCyR rate in patients with CP CML previously treated with imatinib who are also resistant to dasatinib or nilotinib or who are intolerant to dasatinib, estimate overall survival (OS) and progression-free survival (PFS) rates at 1 and 2 years, estimate the CHR rate in accelerated phase (AP)/blast phase (BP) CML patients treated with at least previous imatinib and estimate the OHR rate in AP/BP CML patients treated with at least previous imatinib.

Of these, 266 patients with CP CML treated with prior imatinib, 108 patients with CP CML treated with both imatinib and at least 1 additional TKI, and 129 patients with either AP CML or BP CML were evaluable for efficacy. The Demographics and baseline characteristics of the patients in the Bosulif Phase I/II clinical study are represented in Table 10 of the Bosulif Product Monograph, approved by Health Canada and available through the Drug Product Database.

Chronic Phase Chronic Myelogenous Leukemia Previously Treated with Imatinib Only

The efficacy results in the CP CML patients previously treated with imatinib demonstrated that MCyR at Week 24 was achieved in 66 of 186 evaluable patients [35.5%; 95% CI: (28.6, 42.8)] in the primary endpoint analysis cohort (CP CML imatinib-resistant). In the imatinib-intolerant CP CML patients previously treated with imatinib, MCyR at Week 24 was achieved in 24 of 80 evaluable patients [30.0%; 95% CI: (20.3,41.3)]. The median time to MCyR was 32.1 weeks [95% CI: (24.1, 48.0)] for all evaluable patients that were previously treated with imatinib.

Of the 288 patients in the CP CML population that were previously treated with imatinib, 11 patients [3.8%; 95% CI: (1.9, 6.7)] had confirmed disease transformation to AP or BP while on treatment with Bosulif.

Chronic Phase Chronic Myelogenous Leukemia Previously Treated with Imatinib and another Tyrosine Kinase Inhibitor (TKI)

A total of 118 patients with CP CML who were imatinib resistant or intolerant and received at least 1 additional prior TKI therapy [that is (i.e.) dasatinib and/or nilotinib] were enrolled and treated. Patients had a median age of 56 years (range 20 to 79 years), most were <65 years of age (79%), and slightly less than half (45%) of patients were male. Most patients were white (72%) or Asian (14%). Patients had an Eastern Cooperative Oncology Group (ECOG) performance score of 0 (74%) or 1 (26%) at baseline (data was missing for one patient). Slightly more than half of patients (52%) had received prior interferon therapy and 8% had undergone a stem cell transplant. The most common reasons for stopping imatinib treatment were disease progression (64%) and intolerance (31%).

Among the 118 patients receiving Bosulif, all patients received prior therapy with imatinib (resistant or intolerant), 37 patients were dasatinib-resistant, 50 were dasatinib-intolerant, and 27 were nilotinib-resistant; and one patient was nilotinib-intolerant. There were 3 patients who received Bosulif following all previous TKI treatments: 2 were resistant to all three prior TKI therapies (imatinib, dasatinib, and nilotinib) and one was intolerant of all three prior TKI therapies.

The efficacy results of these 118 patients are as follows: In the third line CP CML evaluable population (patients previously treated with imatinib and dasatinib and/or nilotinib), 29 of 108 evaluable patients [26.9%; 95% CI: (18.8, 36.2)] achieved MCyR by Week 24.

Advanced Leukemia : Accelerated Phase Chronic Myelogenous Leukemia & Blast Phase Chronic Myelogenous Leukemia

A total of 164 advanced phase leukemia patients were treated with Bosulif, including 76 patients with AP CML and 64 with BP CML, and 24 with the Philadelphia chromosome positive acute lymphoblastic leukemia (Ph+ ALL).

In the AP CML cohort, the median age was 50.5 years (range 18.0 to 83.0 years), 89% were <65 years of age, and a little more than half (55%) of patients were male. Most patients were white (61%) or Asian (26%). Most patients had an Eastern Cooperative Oncology Group (ECOG) performance score of 0 (54%) or 1 (43%) at baseline. Half of patients (50%) had received prior interferon therapy, 33% had received prior dasatinib therapy, 20% had received prior nilotinib therapy, and 9% had a prior stem cell transplant. The primary reasons for stopping imatinib were disease progression (86%) and adverse events (AEs) (12%).

In the BP CML cohort, the median age was 48.5 years (range 19.0 to 82.0 years), 83% were <65 years of age, and a little more than half (64%) of patients were male. Most patients were white (59%) or Asian (22%). Patients had an ECOG performance score of 0 (34%), 1 (44%), or 2 (22%) at baseline. Thirty percent (30%) had received prior interferon therapy, 36% had received prior dasatinib therapy, 19% had received prior nilotinib therapy, and 6% had a prior stem cell transplant. The primary reasons for stopping imatinib were disease progression (81%) and AEs (19%).

In the Ph+ ALL cohort, the median age was 59.0 years (range 24.0 to 84 years), 54% were <65 years of age, and half (50%) were male. Most patients were white (79%). Patients had an ECOG performance score of 0 (38%), 1 (46%), or 2 (17%) at baseline. Few patients (4%) had received prior interferon therapy, 33% had received prior dasatinib therapy, 1 patient had received prior nilotinib therapy, and 17% had a prior stem cell transplant. The reasons for stopping imatinib were disease progression (92%) and AEs (8%).

The efficacy results in the advanced leukemia patients treated with at least imatinib showed that for AP CML patients, 38 of 69 evaluable patients [55.1%; 95% CI: (42.6, 67.1)] maintained or attained confirmed overall hematologic response (OHR). The median duration of OHR was not reached. In BP CML patients, 17 of 60 evaluable patients [28.3%; 95% CI: (17.5, 41.4)] maintained or attained confirmed OHR. The median duration of OHR was 31.5 weeks [95% CI: (28.9, 48.0)].

Issuance of a Notice of Deficiency and Response

During review of the New Drug Submission (NDS) for Bosulif, a Notice of Deficiency (NOD) was issued to the sponsor in November 2012 for two main reasons as follows:

1. Serious design flaws, limitations and equivocal electrophysiologic data were noted in QT Study 3160A-105-US which rendered the study inconclusive. In addition, discordant electrocardiogram (ECG) results and the lack of a placebo- or active comparator group in pivotal Study 200-WW (B1871006) further complicated the interpretation of these data.
2. Due to a lack of efficacy data for all patient cohorts evaluated in Study 200-WW (B1871006), the evidence to support the proposed indication for Bosulif was lacking. In addition, the length of follow-up for Study 200-WW (B1871006) was insufficient to demonstrate long-term durability of the primary efficacy endpoint (MCyR at Week 24) for the main cohort, CML-CP (2^nd line). Longer follow-up (minimum 2 years) was therefore recommended by Health Canada.

In the Response to Notice of Deficiency (R-NOD), the sponsor did satisfactorily address the two main concerns identified above, as well as other issues contained in the NOD letter. In addressing the first main concern, although there remains existing concerns with the dedicated QT Study 3160A-105-US design, the Product Monograph labelling was revised and now provides several cautionary statements throughout the document in regards to the study conclusions. In addition, the revised labelling has been strengthened with insertion of a box warning and also contraindications which notify prescribers and patients of the potential for QT prolongation with Bosulif, especially in patients with hepatic impairment; as well as ECG findings based on longer follow-up for Study 200-WW (B1871006) and Study 3000-WW (B1871008) [see Clinical Safety for further information on Study 3000-WW (B1871008)].

In reference to the second major concern, based on the more mature efficacy data from Study 200-WW (B1871006), there is promising evidence of a continued trend toward improved and durable long-term response rates for Bosulif for the proposed indication. Of note is the revised labelling which includes a distinction between the primary endpoint analysis cohort, identified as previously treated imatinib-resistant CML-CP patients, with Major Cytogenetic Response (MCyR) rate at Week 24 of 35.5% (95% CI, 28.6, 42.8 based on 186 evaluable patients) and the remaining cohorts which are considered exploratory. Response rates for the primary cohort and exploratory cohorts show promising evidence of efficacy of bosutinib: previously treated imatinib-intolerant CML-CP patients, with a MCyR rate at Week 24 of 30.0% (95% CI, 20.3, 41.3 based on 88 evaluable patients).

During the original filing of this New Drug Submission (NDS), the proposed indication by the sponsor was the following: Bosulif (bosutinib) is indicated for the treatment of chronic, accelerated, or blast phase Ph+ CML in adult patients with resistance or intolerance to prior therapy.

The indication which has been approved by Health Canada is as follows:Bosulif (bosutinib) is indicated for the treatment of chronic, accelerated, or blast phase Ph+ CML in adult patients with resistance or intolerance to prior TKI therapy, and for whom subsequent treatment with imatinib, nilotinib, and dasatinib is not clinically appropriate. This indication accurately reflects the use of Bosulif to fulfill an unmet medical need in patients with mutation-mediated resistance or with co-morbidities that would make treatment with other TKIs sub-optimal choices due to increased risk of developing side-effects or an inability to tolerate the side-effects of the currently market authorized TKIs.

Based on these promising results from the Phase II Study 200-WW, it was recognized that Bosulif does fulfil an unmet medical need as an alternate treatment option for CML patients who do not respond to or who are intolerant of currently market authorized TKIs in Canada, and in whom Bosulif is clinically appropriate. Therefore, it was decided to issue the sponsor market authorization under the Notice of Compliance with Conditions (NOC/c) Guidance Document, in recognition of the promising but unconfirmed evidence of clinical efficacy of Bosulif in the submission.

For more information, refer to the Bosulif Product Monograph, approved by Health Canada and available through the Drug Product Database.

Clinical Safety

The safety profile of Bosulif was based primarily on interim analysis of the ongoing Phase I/II previously described in the Clinical Efficacy section. In addition, further safety data was also obtained from a Phase III study, namely the 3000-WW study. The 3000-WW study was a multicentre, multinational, randomized, open-label study which compared the efficacy and safety of Bosulif alone to that of imatinib alone in patients with newly diagnosed chronic phase CML where patients were randomized 1:1 to either treatment or control. The primary efficacy endpoint of this study was the CCyR rate at one year in the intent-to-treat (ITT) population. Although this study failed to meet its primary efficacy endpoint, the safety data obtained from the study was taken into consideration to further evaluate the overall safety of Bosulif.

In the Phase I/II safety population, adverse reactions of any toxicity grade reported for ≥20% of patients were diarrhea (80% of which 7.7% were Grade 3/4), nausea (42% of which 1.1% were Grade 3/4), vomiting (34% of which 2.5% were Grade 3/4), thrombocytopenia (29% of which 20.5% were Grade 3/4), and rash (27% of which 6.0% were Grade 3/4). In the single-arm Phase I/II clinical study, the median time of onset for diarrhea (all grades) was 2 days and the median duration per event was 2 days.

In Study 200-WW, approximately 33% of patients experienced nonfatal SAEs including infections and infestations, gastrointestinal, general disorders and administration site conditions, cardiac disorders, respiratory/thoracic/mediastinal disorders, blood and lymphatic system disorders, and nervous system disorders. Approximately 20% of CP CML patients and 25% of AP CML patients discontinued treatment with Bosulif due to AEs, including mostly thrombocytopenia, elevation liver enzymes, and neutropenia. Fewer patients in the advanced cancer groups discontinued due to AEs since in most of these patients, disease progression was the reason for discontinuation. In Study 3000-WW, the most frequently occurring SAEs in patients receiving Bosulif included diarrhea, ALT increased, pneumonia, AST increased, and pyrexia.

Safety Topics of Special Interest

QT Prolongation

A dedicated QT Study (3160A-105-US) with single-dose 500 mg Bosulif and the CYP3A4 inhibitor (ketoconazole) rendered inconclusive results based on sub-therapeutic exposure to Bosulif (see Clinical Efficacy : Issuance of a Notice of Deficiency and Response); however, the available results were consistent with statistically significant QTc (corrected QT by the Fridericia method) interval prolongation associated with Bosulif treatment alone and concomitantly with ketoconazole. Given these results, a revision to the labelling was therefore instigated in the Bosulif Product Monograph with insertion of cautionary statements throughout in response to the study conclusions. Furthermore, the revised labelling was strengthened by a box warning and contraindications to notify prescribers and patients of the potential for QT prolongation with Bosulif, especially in patients with hepatic impairment (see What follow-up measures will the company take?).

Bosulif should therefore be administered with caution to patients who have a history of or predisposition for QTc prolongation, who have uncontrolled or significant cardiac disease including recent myocardial infarction, congestive heart failure, unstable angina or clinically significant bradycardia, or who are taking medicinal products that are known to prolong the QT interval (e.g. anti-arrhythmic medicinal productsand other substances that may prolong QT). The presence of hypokalemia and hypomagnesaemia may further increase this effect.

Patients with hepatic impairment who are receiving treatment with Bosulif are also at higher risk of developing QT interval prolongation.

Monitoring for an effect on the QTc interval is advisable and a baseline ECG is recommended prior to initiating therapy with Bosulif and as clinically indicated. Hypokalemia or hypomagnesaemia must be corrected prior to Bosulif administration and should be monitored periodically during therapy.

Cardiac Toxicity

Cardiac-related treatment-emergent adverse events were reported in 10.2% of patients in the Phase I/II study, with Grade 3 or 4 cardiac-related events reported in 3.5% of patients. Atrial fibrillation (2.1%) (Grade 3 or 4 in 0.9%), cardiac failure congestive (1.8%), and tachycardia (1.8%) were most commonly reported. Grade 3 or 4 events of acute myocardial infarction, cardiac failure, and coronary artery disease were reported in 0.5% of patients each, as well as coronary artery stenosis (0.4%), left ventricular dysfunction (0.4%), and pulmonary edema (0.2%). In the Phase III study, 3.6% of patients in the Bosulif arm experienced cardiac events (0.8% with Grade 3 or 4) versus 1.6% of patients in the imatinib arm (none with Grade 3 or 4).

Caution should therefore be exercised in patients with a history of or predisposition to relevant cardiac disorders including recent myocardial infarction, congestive heart failure, unstable angina or clinically significant bradycardia.

Hemorrhage

In the Phase I/II population, treatment-emergent adverse events of any severity grade related to bleeding events were commonly reported in 9.6% of patients. Hemorrhage events included duodenal ulcer hemorrhage, eye hemorrhage, gastrointestinal hemorrhage, operative hemorrhage, pericardial hemorrhage, rectal hemorrhage, retroperitoneal hemorrhage, subarachnoid hemorrhage, vaginal hemorrhage, and cerebral hemorrhage.

Patients with coagulation dysfunction/low platelet counts should be closely monitoring during treatment with Bosulif.

Gastrointestinal Toxicity

Patients with recent or ongoing clinically significant gastrointestinal disorder should use Bosulif with caution and only after a careful benefit-risk assessment, as patients with recent or ongoing clinically significant gastrointestinal disorder [for example (e.g.) severe vomiting and/or diarrhea] were excluded from CML clinical studies. Treatment with Bosulif is associated with events of diarrhea and vomiting. In the single-arm Phase I/II clinical study, the median time of onset for diarrhea (all grades) was 2 days and the median duration per event was 2 days. Eighty-one percent (81%) of patients treated with Bosulif experienced diarrhea, with 8.1% being either Grades 3 or 4.

In newly diagnosed CML CP patients, a higher rate of drug-related vomiting was reported in the Bosulif-treated group (31.5%) relative to the imatinib-treated group (13.5%). In these CML CP patients, a higher rate of drug-related diarrhea was also observed in the Bosulif-treated group (65.7%) relative to imatinib-treated group (17.9%). Bosutinib patients who reported a treatment-emergent diarrhea, 45.8% have experienced an individual episode of diarrhea for more than 28 consecutive days. Patients with these events should be managed using standard of care treatment, including anti-diarrheal medication,and/or fluid replacement. Since some anti-emetics and anti-diarrheals are associated with a risk of increased QT interval prolongation with the potential to induce "torsade de pointes", concomitant treatment with these agents should be carefully considered. In addition, these events can also be managed by withholding Bosulif temporarily, dose reduction, and/or discontinuation of Bosulif.

Fluid Retention

Treatment with Bosulif is associated with fluid retention (pericardial effusion, pleural effusion, pulmonary edema, and/or peripheral edema).

In the single-arm Phase I/II clinical study in 570 patients with Ph+ leukemias treated with prior therapy, severe (Grade 3 and 4) fluid retention was reported in 18 patients (3.2%). Sixteen patients had a Grade 3 or 4 effusion (13 patients had Grade 3 or 4 pleural effusions [2.3%] and 3 patients [0.5%] had a Grade 3 or 4 pericardial effusion). In the Phase III study, 3 patients (1.2%) experienced acute pulmonary edema or pulmonary edema (all grades) in the setting of either a pleural effusion or a pericardial effusion.

Patients should be weighed regularly and monitored for signs and symptoms of fluid retention, and managed using standard of care treatment, such as diuretics. In addition, these events can also be managed by withholding Bosulif temporarily, dose reduction, and/or discontinuation of Bosulif.

Safety Conclusion

Although Health Canada clearly recognizes the limitations of establishing a drug's safety profile primarily on a single-arm study; based on the promising results from the Phase II study 200-WW and the current unmet medical need for an alternate treatment option for CML patients who do not respond to or who are intolerant to currently marketed TKIs, Health Canada is satisfied that through rigorous safety revisions to the Bosulif Product Monograph appropriate risk mitigation measures are in place for the approval of this product. Measures taken include the insertion of a box warning in the Bosulif Product Monograph describing the serious warnings and precautions. Furthermore, submission of a Risk Management Plan (RMP) for Bosulif was also provided by Pfizer Canada to Health Canada. Upon review, the RMP was considered to be acceptable. The RMP is designed to describe known and potential safety issues, to present the monitoring scheme and when needed, to describe measures that will be put in place to minimize risks associated with the product.

Additionally, given the submission was evaluated under the Notice of Compliance with Conditions (NOC/c) Guidance, as part of the marketing authorization of Bosulif, Health Canada has requested the sponsor agree to several commitments to be addressed post-market. These commitments include submission of final study reports for studies 200-WW and 3000-WW. Furthermore, the sponsor shall also provide post-market safety monitoring studies [for example (e.g.) report(s) of all serious adverse drug reactions that occurred in Canada, semi-annual/annual safety summary reports current with the Guidance Document: Notice of Compliance with Conditions].

For more information, refer to the Bosulif Product Monograph, approved by Health Canada and available through the Drug Product Database.