Real-world outcomes of ponatinib treatment in 724 patients with CML and Ph+ ALL: a post-marketing surveillance study with a special interest in arterial occlusive events in Japan

Abstract Background In September 2016, ponatinib was approved in Japan for the treatment of patients with chronic myeloid leukemia with resistance/intolerance to prior tyrosine kinase inhibitors and patients with relapsed or refractory Philadelphia chromosome-positive acute lymphoblastic leukemia. Methods We conducted a post-marketing all-case surveillance to study the safety and efficacy of ponatinib in clinical practice, focusing on arterial occlusive events. Results Data from 724 patients were collected for 2 years from the initiation of ponatinib. The arterial occlusive events were reported in 6.49% (47/724) with an exposure-adjusted incidence rate of 6.8/100 person-years. The risks associated with arterial occlusive events were age and comorbidities including hypertension and diabetes. At 104 weeks, the cumulative major molecular response rate in patients with chronic-phase chronic myeloid leukemia was 67.2% and the complete cytogenetic response in patients with Philadelphia chromosome-positive acute lymphoblastic leukemia was 80.0%. Furthermore, the estimated 1-year overall survival rate was 98.5% for chronic-phase chronic myeloid leukemia and 68.6% for Philadelphia chromosome-positive acute lymphoblastic leukemia. Conclusions This surveillance demonstrated that ponatinib has a favorable safety and efficacy profile in Japanese patients and also showed the necessity of closely monitoring arterial occlusive events in older adults and patients with predisposing factors for atherosclerosis.


Introduction
The development of tyrosine kinase inhibitors (TKIs) for the treatment of chronic myeloid leukemia (CML) or Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph + ALL) has led to the introduction of a selective targeted therapy that inhibits the BCR::ABL1 tyrosine kinase and to an improvement in treatment outcomes by suppressing the growth of leukemia cells (1).However, patients may be intolerant to TKIs due to adverse events (AEs) (2) or may develop resistance to TKIs due to mutations in the ABL kinase domain (3).
Ponatinib, a third-generation TKI, was approved in Japan by the Ministry of Health, Labour and Welfare in September 2016 for the treatment of patients with CML with resistance/intolerance to prior TKIs and patients with relapsed and refractory Ph + ALL, based on the results from the phase 2 Ponatinib Ph + ALL and CML Evaluation (PACE) trial conducted outside Japan (4) and a phase 1/2 trial conducted in Japan (5).
Ponatinib has inhibitory activity against BCR::ABL1 kinase, including a variant with threonine-to-isoleucine at position 315 (T315I) (4) but is associated with a higher risk of major arterial events than other TKIs among patients with CML (6).In the phase 2 PACE trial, the incidence of arterial occlusive events (AOEs, i.e. vascular ischemia or thrombosis) was 25%, and the exposureadjusted incidence rate of AOEs per 100 person-years was 13.8 (7).In the phase 1/2 trial of Japanese patients, 14% (5/35) of patients treated with ponatinib experienced AOEs (5).
Since the number of Japanese patients treated with ponatinib in clinical trials so far is quite limited, the safety and efficacy of ponatinib in Japanese patients require further evaluation.In addition, findings from randomized clinical trials, which are conducted in carefully selected patients and follow strictly controlled drug regimens, cannot always be generalized to the whole population (8).Observational studies conducted outside Japan to explore real-life experience with ponatinib have provided valuable information that is sometimes different from that obtained in the clinical trial cohorts (9)(10)(11).Therefore, we conducted a post-marketing surveillance to study the safety and efficacy of ponatinib, focusing on AOEs, to optimize its use in clinical practice in Japan.

Surveillance design and patients
This was a post-marketing surveillance to obtain real-world data on ponatinib in Japanese patients with CML with resistance/intolerance to prior kinase inhibitors and patients with relapsed/refractory Ph + ALL.The surveillance was conducted in accordance with the Japanese regulatory requirements stipulated in Good Post-Marketing Study Practice.Informed consent was obtained from all patients who were to receive ponatinib in accordance with the package insert of ponatinib, but under these regulations, central Institutional Review Board oversight and patient consent are not required.
The case report forms were collected between 21 November 2016 and 30 June 2018.The observation period was 2 years or until the end of ponatinib administration, and the forms were collected at 3 months, 6 months, 1 year,and 2 years after the initiation of ponatinib.Information collected was as follows: patient characteristics, previous treatment history, administration status of ponatinib, safety (e.g.AEs and laboratory data) and efficacy information (e.g.hematological findings and cytogenetic/molecular analyses).
The recommended starting dose of ponatinib was 45 mg once daily, which might be adjusted to a lower dose (30 or 15 mg) based on the condition of each patient.

Safety analysis
We collected information on AEs to assess the safety of ponatinib.A serious adverse reaction (ADR) is an AE that requires hospitalization or prolongation of existing hospitalization, causes a congenital malformation, results in persistent or significant disability or incapacity, is life-threatening, or results in death (12).
This surveillance focused on the real-world incidence of AOEs in Japanese patients treated with ponatinib.AEs were coded according to the Medical Dictionary for Regulatory Activities/J (MedDRA/J), version 25.0, and AOEs were categorized based on preferred terms related to vascular ischemia or thrombosis.An exposure-adjusted incidence rate of AOEs was calculated as (the number of first events during observation)/(the total exposure during observation, personyears) × 100 (7).We also analyzed the association between AOEs and patient characteristics, such as age, history of ischemic disease and specific comorbidities (diabetes, hypertension and dyslipidemia).

Statistical analysis
Statistical analysis was performed using SAS software, Version 9.3 (SAS Institute Japan Ltd., Japan).The number of patients, the mean and standard deviation (SD), or the median and range were calculated as summary statistics.Univariable logistic regression analysis evaluated the association of patient characteristics (sex, age, performance status, Philadelphia chromosome, BCR::ABL1 transcripts, BCR::ABL1 mutations, smoking, comorbidities, history of ischemic disease, average daily dose of ponatinib, starting daily dose of ponatinib and maximum daily dose of ponatinib) with ADRs and AOEs to obtain the odds ratio (OR) and its 95% confidence interval (CI).The cumulative incidence of AOEs, cumulative response rates and survival rates were shown in Kaplan-Meier plots.

Patient characteristics
A total of 807 patients were enrolled at 334 facilities (Supplementary Fig. 1).Case report forms were collected from 724 patients.The safety analysis was conducted in all 724 patients, and 116 patients were excluded from the efficacy analysis for the following reasons: (i) they were treated for diseases other than CML or Ph + ALL (n = 6), (ii) they were not evaluated for all efficacy endpoints (n = 104) and/or (iii) they had received drugs with the same active ingredient composition as ponatinib before the surveillance (n = 10).
Among those who were tested for the BCR::ABL1 mutation and for whom molecular response rate data was available, the cumulative MMR rate at week 52 was 75.3% in patients with the BCR::ABL1 T315I mutation and 62.0% in patients without the BCR::ABL1 T315I mutation (Supplementary Fig. 2): the cumulative rate at week 52 was 52.5 and 32.8%, respectively, for MR4.0 and 41.2 and 22.3%, respectively, for MR4. 5.
The estimated 1-year overall survival (OS) rate and the estimated 1-year survival rate only considering death due to progressive disease were 98.5 and 100.0%, respectively, in patients with CML-CP (Supplementary Fig. 3A) and 68.6 and 71.2%, respectively, in patients with Ph + ALL (Supplementary Fig. 3B).

Discussion
In this 2-year post-marketing surveillance of 724 patients with CML or Ph + ALL, ponatinib demonstrated favorable efficacy and tolerability compared with previous clinical trials and also showed the necessity of closely monitoring arterial occlusive events in older adults and patients with predisposing factors for atherosclerosis.As demonstrated in the PACE trial (7) and a meta-analysis (6), AOEs are considered notable safety risks of ponatinib that require further investigation.In the current surveillance, the incidence of AOEs was 6.49% (47/724), and the exposure-adjusted incidence rate of AOEs per 100 person-years was 6.8, which did not exceed 25% and 13.8 in the PACE trial (7).Among patients with CML-CP in the current surveillance, the incidence of AOEs was 6.22% (12/193) and the exposure-adjusted incidence rate of AOEs per 100 person-years was 4.5, which again did not exceed 31% (84/270) and 14.1 at year 2 in the PACE trial (7).
The most common AOEs in this surveillance were cerebrovascular events in CML-CP and cardiovascular events in Ph + ALL.Although we assume that patients with Ph + ALL were at risk for cardiovascular events because of previous treatment (e.g.anthracycline-based chemotherapy, cyclophosphamide and concomitant glucocorticoids), it is unclear whether AOEs in this   Values are given as n (%). a Four patients died from myocardial infarction (CML-AP, n = 1; CML-BC, n = 1; Ph + ALL, n = 2).b Two patients died from disseminated intravascular coagulation (Ph + ALL, n = 2).c Two patients died from thrombotic microangiopathy (CML-BC, n = 1; Ph + ALL, n = 1).AP, accelerated-phase; BC, blast crisis; CML, chronic myeloid leukemia; CP, chronic-phase; Ph + ALL, Philadelphia chromosome-positive acute lymphoblastic leukemia surveillance were treatment-related or were due to pre-existing vascular disease.Further studies are necessary to determine whether the type of AOE was affected by the patient's leukemia type.
The multivariate analyses of pooled data from three clinical trials showed that ponatinib dose intensity, history of ischemic disease and age were strong predictors of increased risk of an AOE, suggesting a 33% reduction in the risk of an AOE or for each 15 mg/day decrease in ponatinib dose intensity (15).The question is whether the dose escalation or reduction method can balance safety and efficacy in patients with highly refractory CML-CP.The phase 2 Optimizing Ponatinib Treatment in CP-CML (OPTIC) trial was conducted to assess the benefit/risk ratio across three ponatinib starting doses (45, 30 and 15 mg/day) in patients with highly resistant CML-CP (16).The results showed that optimal benefit/risk outcomes occurred with a regimen consisting of a 45 mg/day starting dose reduced to 15 mg upon response (BCR::ABL1 transcripts [IS] ≤ 1%) (16).Furthermore, the 5-year results of the PACE trial, in which patients received an initial dose of 45 mg of ponatinib, showed the exposureadjusted incidence rate of new AOEs per 100 person-years decreased each year along with median dose intensity among patients with CML-CP, from 15.8/100 person-years at 32.1 mg/day during year 1 to 4.9/100 person-years at 20.4 mg/day during year 4 (7).In the current surveillance, while the maximum daily dose was 45 mg in 36.3% (263/724) of patients, only 19.8% (143/724) of patients received an initial daily dose of 45 mg.Further studies are necessary to ascertain the optimal starting dose to obtain maximum safety and effectiveness in Japanese patients.
Our surveillance showed a significant association between AOEs and age (OR 1.039, 95% CI 1.017-1.062),hypertension (OR 2.137, 95% CI 1.169-3.907)and diabetes (OR 1.950, 95% CI 1.026-3.707).Considering that age and a history of ischemic disease were previously reported as prognostic factors for AOEs (15), a benefit-risk evaluation and close monitoring are especially crucial when treating older patients and those with predisposing factors for atherosclerosis.
The MMR (BCR::ABL1 transcripts [IS] ≤ 0.1%) rate at month 12 and month 24 was 28 and 34%, respectively, in the PACE trial and 20% and 34%, respectively, in the OPTIC trial (in the group on a 45 mg/day starting dose) (17).In the current surveillance, the cumulative MMR rate in patients with CML-CP who had not achieved MMR before treatment with ponatinib was 58.8% at week 52 and 67.2% at week 104; the cumulative molecular rates  (MMR, MR4.0 and MR4.5) increased over time regardless of T315I mutation status, as in the PACE and OPTIC trials.While we cannot directly compare those trials and the current surveillance because of different study designs, the results showed that ponatinib has realworld effectiveness among Japanese patients who were refractory to prior TKIs.In this surveillance, ponatinib showed a good proportion of complete hematologic remission (73.6%, 237/322) and CCyR (79.7%, 141/177) at any time in Japanese patients with Ph + ALL (data not shown).Furthermore, the estimated 1-year OS rate was 68.6% for Ph + ALL patients and 98.5% for CML patients, which are favorable results and similar to those obtained in the OPTIC and PACE trials (7,16).Taken together, ponatinib has a favorable safety and efficacy profile in Japanese patients with Ph + ALL as well as CML-CP.
This surveillance has some limitations.First, the incidence of AEs in post-marketing surveillance studies may be underreported, especially for lower-grade events.Therefore, we did not statistically compare the incidence of AEs and ADRs in our surveillance to that of previous clinical trials.Second, the prognosis beyond 2 years in patients treated with ponatinib remains unknown.Further studies are necessary to evaluate the long-term safety and efficacy, especially among patients with CML-CP.In conclusion, this 2-year post-marketing surveillance of patients with CML with resistance/intolerance to prior kinase inhibitors or patients with relapsed/refractory Ph + ALL has demonstrated favorable safety and efficacy of ponatinib in clinical practice in Japan.Older patients and those with predisposing factors for AOEs should be closely monitored for possible vascular adverse events.
or contracts from any entity from KM Biologics and Sysmex; consulting fee from Sysmex and LSI medicine; honoraria from Otsuka Pharmaceutical, Novartis and Bristol Meyer Squib.Akira Masunari, Seiji Nishibayashi, Akiko Kageyama and Yasuhiko Fukuta are employees of Otsuka Pharmaceutical.

Figure 1 .
Figure 1.Cumulative incidence rates of arterial occlusive events in patients with CML-CP and Ph + ALL.The cumulative incidence rates of cerebrovascular (Solid line), cardiovascular (Dotted line), and peripheral arterial events (Dashed line) are shown in patients with CML-CP (A) and Ph + ALL (B).Incidence rates are labeled at years 1 and 2. CML, chronic myeloid leukemia; CP, chronic-phase; Ph + ALL, Philadelphia chromosome-positive acute lymphoblastic leukemia.

Figure 3 .
Figure 3. Complete and partial cytogenetic response rates in patients with Ph + ALL.The complete cytogenetic response rate (Solid) and partial cytogenetic response rate (Dotted) before and after initiating ponatinib treatment in patients with Ph + ALL are shown.CCyR, complete cytogenetic response; PCyR, partial cytogenetic response; Ph + ALL, Philadelphia chromosomepositive acute lymphoblastic leukemia.

Table 1 .
Baseline characteristics of patients Patients with both TKI resistance and intolerance are included in TKI intolerance; patients with TKI resistance and other reasons are included in TKI resistance; patients with TKI intolerance and other reasons are included in TKI intolerance; and patients with TKI resistance, intolerance, and other reasons are included in TKI intolerance.c Ischemic disease includes coronary artery disease, cerebrovascular disease, retinal artery occlusion, peripheral arterial occlusive disease, and venous thromboembolism.

Table 2 .
Summary of adverse drug reactions

Table 3 .
Summary of arterial occlusive events (adverse drug reactions)

Table 4 .
Risks associated with arterial occlusive events, univariable logistic analysis Ischemic disease includes coronary artery disease, cerebrovascular disease, retinal artery occlusion, peripheral arterial occlusive disease, and venous thromboembolism.b Average daily dose was calculated by dividing the total mass of ponatinib prescribed over the period by the study duration.c 95% CI of the corresponding odds ratio does not include 1. a