Safety and effectiveness of clofarabine in Japanese patients with relapsed/refractory acute lymphoblastic leukaemia: a post-marketing surveillance study

Abstract Objective Clofarabine is used to treat acute lymphoblastic leukaemia, but evidence of its safety and effectiveness in Japanese patients is limited. We evaluated the safety and effectiveness of clofarabine in patients with relapsed/refractory acute lymphoblastic leukaemia in real-world clinical practice in Japan. Methods An observational, multicenter, post-marketing, all-case surveillance was conducted for safety. Effectiveness analyses were conducted in patients aged ≤21 years and those treated with clofarabine monotherapy and combination therapy (clofarabine plus etoposide and cyclophosphamide). Results In the all-case survey, 260 of 264 registered patients were eligible for safety analysis. Among the 225 patients eligible for effectiveness analysis, 139 were aged ≤21 years. For monotherapy and combination therapy, 20/31 and 34/88 patients were eligible, respectively. In the all-case survey, the median age was 16.0 years, and 47.7% of patients were <15 years old. Adverse drug reaction incidence was 83.5% and the most common were hematologic toxicities. The best overall response rates in the population aged ≤21 years were complete remission, 29.7%; complete remission without platelet recovery, 7.3% and partial remission, 10.9%. The rest (52.2%) were classified as ineffective. The sum of complete remission, complete remission without platelet recovery and partial remission rates (effectiveness rate) was 47.8% (66/138 patients). The effectiveness rates in the monotherapy and combination therapy surveys were 10.0% (2/20 patients) and 58.8% (20/34 patients), respectively. Conclusions These post-marketing surveys provide real-world evidence of the safety and effectiveness of clofarabine regimens, including monotherapy and combination therapy in Japanese patients with relapsed/refractory acute lymphoblastic leukaemia. The safety and effectiveness profiles were comparable with those of previous prospective studies.


Introduction
Acute lymphoblastic leukaemia (ALL) is a malignant haematological disease that affects both children and adults, with most cases occurring in healthy individuals (1).It has been estimated that despite chemotherapy, 10-15% of paediatric patients with ALL and 40-50% of adult patients with ALL experience relapse (2)(3)(4).Disease relapse is the leading cause of death among patients with ALL; after relapse, the median survival ranges from 3 to 8.4 months (4)(5)(6).Although overall survival has improved in the last decade with advances in treatment, outcomes of patients with refractory or relapsed ALL remain poor after conventional chemotherapy.
Clofarabine is a second-generation purine nucleoside analogue.Its cytotoxic effect depends on intracellular phosphorylation to the triphosphate metabolite (7).Clofarabine exerts its antitumour effect by reducing the synthesis of deoxynucleotide triphosphates, inhibiting DNA polymerase, increasing incorporation of clofarabine triphosphate into DNA and eventually inducing mitochondria cytochrome c release through DNA damage, which induces apoptosis (8).
Previous Phase 1 and 2 studies have evaluated clofarabine administered as monotherapy or combined with etoposide and cyclophosphamide in patients with ALL (9)(10)(11)(12)(13)(14)(15)(16).Most of these studies showed encouraging activity of clofarabine in patients with acute leukaemia and refractory or relapsed ALL.However, a Phase 1 study of clofarabine monotherapy in Japanese paediatric patients with ALL showed that there were no responders (patients evaluated as having complete remission [CR], CR without platelet recovery [CRp] or partial remission [PR]) because only seven paediatric patients were enrolled for tolerability in that clinical trial (13).In addition, the safety profile of clofarabine has yet to be fully investigated in Japanese patients.
The present report summarizes a post-marketing surveillance conducted to evaluate the safety and effectiveness of clofarabine in real-world clinical practice in Japan.The main study (all-case survey) aimed to confirm the real-world safety and effectiveness of clofarabine in patients with relapsed/refractory ALL and identify the following: unexpected adverse drug reactions (ADRs), the occurrence of ADRs in real-world clinical practice and the occurrence of adverse events (AEs) of special interest.The additional studies aimed to evaluate the effectiveness of clofarabine monotherapy and clofarabine in combination with etoposide and cyclophosphamide in real-world clinical practice in Japanese patients with relapsed/refractory ALL.

Study design, treatment and patients
Three observational, multicenter, post-marketing surveillances were conducted from 18 September 2013 to 19 September 2018 (all-case survey), 26 June 2014 to 8 April 2018 (monotherapy survey) and 19 May 2014 to 9 March 2020 (combination therapy survey) in Japan.Of the patients enrolled in the all-case survey, those administered clofarabine monotherapy or combined with etoposide and cyclophosphamide were simultaneously enrolled in each outcome surveillance.As the number of patients who received the combination regimen was insufficient when the registration of the all-case survey was completed, only the combination regimen cases continued to be registered after the completion of the all-case survey.
Clofarabine 52 mg/m 2 (body surface area) was administered by intravenous infusion over 2 h daily for five consecutive days and then discontinued for at least 9 days (Cycle 1) until the next cycle (for a maximum of six cycles).Among patients who received combination therapy, clofarabine was administered at a dose of 52 mg/m 2 , and etoposide and cyclophosphamide were administered at a dose of 150 and 400 mg/m 2 , respectively.The intravenous infusion was administered over at least 2 h daily for five consecutive days, then discontinued for at least 9 days (Cycle 1) until the next cycle.The dose of clofarabine could be decreased depending on the patient's condition.
The observation period was defined as the period between the start of the study drug to the end of the final cycle (at least 9 days from the end of the 5-day treatment period).The observation period was set to a maximum of 6 cycles.Patient data were collected from case report forms (CRFs).
The institutional review board approved the protocol at each study site.The study was conducted in accordance with the Declaration of Helsinki and adhered to the guidelines for Good Postmarketing Study Practice in Japan.Informed consent was exempted under Good Post-marketing Study Practice.Agreements were made with the study sites to publish anonymized data.

Safety assessments
ADRs, serious ADRs, AEs of special interest (including hematotoxicity, infection, renal disorders, hepatobiliary disorders, capillary leak syndrome and systemic inflammatory response syndrome [SIRS], tumour lysis syndrome and cardiovascular events) and abnormal changes in laboratory parameters were evaluated.AEs were coded using the Medical Dictionary for Regulatory Activities, version 23.1, and graded according to the Common Terminology Criteria for Adverse Events, version 4.0.

Effectiveness assessments
Response to treatment was evaluated by the best overall response (CR, CRp or PR), the remission rate was evaluated by CR + CRp and the effectiveness rate was evaluated by CR + CRp + PR.The criteria for CR were defined as meeting all of the following: no leukaemic cells in peripheral blood and no extramedullary infiltration, <5% of leukaemia cells in bone marrow and platelet count and neutrophil count in peripheral blood recovered to ≥100 000 and ≥1000/mm 3 , respectively.The criteria for CRp were the same as for CR criteria except platelet count recovery (≥100 000/mm 3 ).The definition of PR was: no leukaemic cells found in peripheral blood and leukaemic cells in bone marrow >5% and <25% with normal blood cell precursors observed (or if <5% of leukaemic cells in bone marrow, but not meeting the criteria for CR and CRp).Ineffective was defined as not meeting the CR, CRp or PR criteria.
The effectiveness analysis was performed for patients with relapsed/refractory ALL aged ≤21 years because the target population of clinical trials in Japan and overseas at the New Drug application submission in Japan was ≤21 years of age.In addition, the effectiveness was also summarized for patients aged ≥22 years.

Statistical methods
The minimum number of cases required to detect AEs occurring at a frequency of 3% with a probability of ≥95% was 100 cases.Thus, 120 patients were set as the target sample size.Baseline demographic and clinical characteristics were evaluated in the safety analysis population and are summarized as mean ± standard deviation (SD)

Patients
The patient disposition is shown in Supplementary Fig. S1.In the allcase survey, 264 patients from 115 study sites were enrolled, and 262 CRFs were collected.The safety and effectiveness analysis populations included 260 and 225 patients, respectively.In the monotherapy survey, 31 patients from 24 study sites were enrolled, and 30 CRFs were collected.The safety and effectiveness analysis populations included 27 and 20 patients, respectively.In the combination therapy survey, 88 patients from 44 study sites were enrolled, and 87 CRFs were collected.The safety and effectiveness analysis populations included 87 and 34 patients, respectively.Among the 262, 30 and 87 patients evaluated in the all-case, monotherapy and combination therapy surveys, 152 (58.0%), 23 (76.7%) and 44 (50.6%) discontinued treatment, respectively (Supplementary Table S1).The most common reasons for discontinuation in all three surveys were insufficient effectiveness and primary disease progression.

Effectiveness
In the all-case survey, of the 225 patients in the effectiveness analysis population, 139 were patients with relapsed/refractory ALL aged younger than 22 years.After excluding the one non-evaluable case, among the 138 patients evaluated, 29.7% (n = 41) had CR, 7.3% (n = 10) had CRp, 10.9% (n = 15) had PR and 52.2% (n = 72) were classified as ineffective.The CR + CRp rate (remission rate) was 37.0% (n = 51), and the CR + CRp + PR rate (effectiveness rate) was 47.8% (n = 66).The best overall response rates in the population aged ≥22 years are shown in Supplementary Table S3.
The results of tumour response in all the patients (regardless of age) in the all-case, monotherapy and combination therapy surveys are summarized in Table 4. Effectiveness was confirmed after more than one cycle of clofarabine was administered.

Discussion
This study evaluated the safety and effectiveness of clofarabine in Japanese patients with relapsed/refractory ALL in real-world clinical practice.This study is important because to date, evidence of the real-world safety and effectiveness of clofarabine in Japanese patients with relapsed/refractory ALL has been limited.
Regarding the safety findings, in the all-case survey, ADRs were observed in 83.5% of the patients.The most common ADRs were hematotoxicity-related events, similar to ADRs reported in previous clinical trials (13)(14)(15)(16).
The effectiveness rate (CR + CRp + PR rate) was 47.8% in the patients with relapsed/refractory ALL aged ≤21 years from the all-case survey, 10.0% in the monotherapy survey and 58.8% in the combination therapy survey.Of note, the effectiveness rate of clofarabine monotherapy was notably lower than the effectiveness rate in previous studies of paediatric patients with relapsed/refractory ALL who were treated with clofarabine monotherapy (30%) (11) and combination therapy (56-64%) (17,18).This trend showing less effectiveness with monotherapy in this study may be related to differences between the studies regarding patient age (unlike previous studies, which included only paediatric patients, the present monotherapy survey included both paediatric and adult patients) and other background characteristics.In addition, the number of patients included in the monotherapy survey (n = 27) was very limited.The discontinuation rate in the monotherapy survey was higher than that in the combination therapy survey, as was the proportion of patients with 'Other AEs (including worsening of complications)'.Discontinuations occurred more frequently in the first cycle in the monotherapy survey than in the combination therapy survey.In a recent Phase 3 study of clofarabine added to intensive treatment in adult patients with newly diagnosed ALL, the addition of clofarabine failed to improve outcomes in this patient population (19).This result in adult patients with ALL, along with the observation that variations in the prevalence of chromosomal alterations, which can affect treatment responsiveness, exist between patients of different ages (20), may partly explain the relatively low effectiveness shown in this study.Thus, the risk and benefit of treatment with clofarabine in adult patients with ALL should be carefully considered.Furthermore, in a previous study of paediatric patients with multiple relapsed or refractory ALL who received clofarabine in combination with cyclophosphamide and etoposide, the probability of overall survival was significantly higher in patients with B-lineage vs T-lineage (33% N, number of patients; n, number of analysed patients.CI, confidence interval; CR, complete remission; CRp, complete remission without platelet recovery; PR, partial remission vs 0%; P < 0.001) (18).The immunophenotype may have affected the results in the real world as well.This study has some limitations, including those inherent to postmarketing surveys (e.g.non-interventional, non-controlled, observational study design).Additionally, the sample size of the monotherapy survey was very small, and the generalizability of the study findings is limited to the Japanese population.
In conclusion, this study provides real-world evidence of the safety and effectiveness of clofarabine administered as monotherapy and in combination with etoposide and cyclophosphamide in Japanese patients with relapsed/refractory ALL.These results indicate that clofarabine monotherapy was effective in Japanese patients, but its effectiveness was considerably lower than with combination therapy.No new safety concerns were raised.

Table 1 .
Causes of death

Table 2 .
Background patient characteristics (safety analysis population) a After enrollment, it was revealed that this patient had not received monotherapy.b Any drugs other than antitumour drugs may be included.c This includes patients whose reason for using clofarabine was having a diagnosis of relapsed/refractory acute lymphoblastic leukaemia (all-case survey, n = 226; monotherapy, n = 25; combination therapy, n = 85).SD, standard deviation.

Table 3 .
Adverse events, adverse drug reactions and serious adverse drug reactions