Results from an international phase 2 study of the anti-CD22 immunotoxin moxetumomab pasudotox in relapsed or refractory childhood B-lineage acute lymphoblastic leukemia

Abstract

Background:

In a multicenter phase 1 study of children with relapsed/refractory acute lymphoblastic leukemia (ALL), moxetumomab pasudotox, an anti-CD22 immunotoxin, demonstrated a manageable safety profile and preliminary evidence of clinical activity. A phase 2 study further evaluated efficacy.

Procedure:

This international, multicenter, phase 2 study enrolled children with relapsed/refractory B-cell precursor ALL who received moxetumomab pasudotox 40 μg/kg intravenously every other day, for 6 doses per 21-day cycle. The primary objective was to evaluate the complete response (CR) rate. Secondary objectives included safety, pharmacokinetics, and immunogenicity evaluations.

Results:

Thirty-two patients (median age, 10 years) were enrolled at 16 sites; 30 received study drug and were evaluable for safety; 28 were evaluable for response. The objective response rate was 28.6%, with 3 patients (10.7%) achieving morphologic CR, and 5 patients (17.9%) achieving partial response. Disease progression occurred in 11 patients (39.3%). Ten patients (33.3%) experienced at least 1 treatment-related serious adverse event, including capillary leak syndrome (CLS; n=6), hemolytic uremic syndrome (HUS; n=4), and treatment-related death (n=1) from pulmonary edema. No differences were observed in inflammatory markers in patients who did or did not develop CLS or HUS.

Conclusions:

Despite a signal for clinical activity, this phase 2 study was terminated at interim analysis for a CR rate that did not reach the stage 1 target. Preclinical data suggest enhanced efficacy of moxetumomab pasudotox via continuous infusion or in combination regimens; thus, further studies designed to optimize the efficacy and safety of moxetumomab pasudotox in pediatric ALL may be warranted.

INTRODUCTION

Targeting cell-surface antigens in precursor B-cell acute lymphoblastic leukemia (B-ALL) represents a highly effective strategy, as evidenced by the great progress made over the last decade in the development of novel immunotherapeutic approaches to treat relapsed or refractory (r/r) ALL. Chimeric antigen receptor T-cell therapies targeting CD19 have been shown to induce minimal residual disease (MRD)-negative complete response (CR) rates that exceed 70% in early-phase clinical trials,^1–4 which has led to approval by the U.S. Food and Drug Administration (FDA) for children with r/r ALL.⁵ The bispecific CD19/CD3 antibody blinatumomab, also FDA approved for r/r ALL, has been shown to induce high CR rates and to eradicate MRD in a substantial proportion of patients.^6,7 Inotuzumab ozogamicin (IO), a CD22-targeted antibody drug conjugate, has received FDA approval for treatment of adults with r/r ALL.⁸ It is being tested by the Children’s Oncology Group (ClinicalTrials.gov NCT02981628) and by the Innovative Therapies for Children with Cancer Consortium (EU Clinical Trials Register: 2016-000227-71). These therapeutic advances offer new treatment options for patients with B-ALL and relapsed or chemotherapy-refractory disease.

Moxetumomab pasudotox is an anti-CD22 recombinant immunotoxin in development for the treatment of r/r CD22+ lymphoid malignancies. It is a recombinant immunotoxin composed of a disulfide-stabilized murine anti-CD22 immunoglobulin single-chain variable domain genetically fused to a truncated form of Pseudomonas exotoxin A, PE38.⁹ Moxetumomab pasudotox was approved by the FDA for the treatment of hairy cell leukemia (HCL) based on results from a pivotal phase 3 study.^5,10–12 Toxicities unique to moxetumomab pasudotox include capillary leak syndrome (CLS) and hemolytic uremic syndrome (HUS).^5,10–12

With nearly universal expression of CD22 in almost all cases of precursor B-ALL,¹³ moxetumomab pasudotox was tested in a multicenter phase 1 study of pediatric patients with r/r precursor B-ALL. This study led to the determination of a maximum tolerated cumulative dose and established the safety profile of this agent in children with r/r precursor B-ALL.¹² Additionally, evidence for clinical activity was seen by virtue of a 23% CR rate, warranting further study in a larger population. Based on these findings, an international, multicenter, phase 2 study was conducted to further evaluate the activity of this agent in children and adolescents with multiply relapsed and chemotherapy-refractory precursor B-ALL.

We report the results from this phase 2 study, which was closed at the interim analysis due to an insufficient benefit:risk profile. Biologic correlative studies that enhance our understanding of the toxicity profile of this agent are also reported.

METHODS

Patient Enrollment

This multicenter, international study (ClinicalTrials.gov NCT02227108) was conducted in compliance with the Declaration of Helsinki, the International Council for Harmonisation Guidance for Good Clinical Practice, and all local laws and requirements, and was approved by the regulatory authorities and/or the institutional review board/independent ethics committee of each study site. All patients or their parents/legal guardians provided written informed consent prior to the conduct of any protocol-specific activity. Before study entry, patients were included in an age-appropriate discussion regarding the study, and assent was obtained from those aged 7 years and older. Patients were enrolled between August 2014 and September 2015.

Pediatric patients aged 6 months to <18 years with relapsed or chemotherapy-refractory precursor B-ALL or B-cell lymphoblastic lymphoma who had received at least 1 front-line and 1 salvage regimen of chemotherapy or a prior allogeneic hematopoietic stem cell transplant (HSCT) were eligible. Patients with a prior HSCT must have been at least 100 days post-HSCT and without active graft-versus-host disease. Prior CD22-targeted therapy was allowed as long as at least 3 half-lives or 30 days had elapsed, whichever was longer. Bone marrow involvement with at least 5% blasts was required. Patients could not receive chemotherapy for at least 14 days before the first moxetumomab pasudotox dose, except for intrathecal or maintenance-like chemotherapy for ALL. Patients with isolated testicular relapse or central nervous system involvement (CNS3) were excluded. Eligibility required aspartate aminotransferase (AST) and alanine aminotransferase (ALT) ≤5 times the upper limit of normal, total bilirubin ≤2 times the upper limit of normal, age-adjusted normal creatinine, and an acceptable performance status, with an Eastern Cooperative Oncology Group performance score of ≤2 or Lansky score of ≥50%. Notable exclusion criteria included uncontrolled infection, KMT2A gene rearrangement due to concerns for partial CD22 expression,¹³ and history of thrombotic microangiopathy or HUS.

Study Design

In this global, multicenter, open-label, single-arm, phase 2 study, patients received intravenous (IV) moxetumomab pasudotox 40 μg/kg administered over 30 minutes every other day with 6 total doses per 21-day cycle until progressive disease, up to 6 cycles, or until they otherwise became ineligible (Figure 1). The maximum-tolerated cumulative dose was determined to be 50 μg/kg × 6 doses in the phase I trial.¹² However, an interim change in the manufacturing process led to greater purity of the new material (“Process 3”), thus a dose adjustment was made to ensure the same amount of biologically equipotent active product. The 40-µg/kg/dose utilized in the phase 2 study was deemed to be biologically equivalent to the 50-µg/kg/dose utilized in the phase 1 study.¹² The primary objective of this phase 2 study was to evaluate the efficacy of moxetumomab pasudotox as measured by the composite complete response (CRc) rate based on an efficacy-evaluable population/intent-to-treat population analysis. CRc was defined as achieving a complete response (CR; M1 marrow with absence of extramedullary sites of disease and a neutrophil count ≥1000/μL and platelet count ≥100,000/μL) or a CR with incomplete blood count recovery (CRi, defined as meeting CR criteria but without the aforementioned specified blood count parameters). Secondary objectives included determination of MRD-negativity rates (by flow cytometry) in those who attained a CRc, along with evaluation of safety, pharmacokinetics (PK), and immunogenicity of moxetumomab pasudotox. Exploratory objectives included the determination of baseline CD22 expression by flow cytometry as a biomarker of response, and the evaluation of cytokines to assess for possible association with the development of HUS or CLS.

Figure 1.

Study flow diagram. Patients continued treatment until progressive disease, for up to 6 cycles, or until they otherwise became ineligible.

QOD, every other day.

Patients were considered evaluable for response if they received at least 1 dose of study drug and had both baseline and post-therapy disease evaluation. Bone marrow aspirate/biopsy were used for morphologic disease assessments and were also sent to a central lab for flow cytometry for disease evaluation prior to cycle 2 and all subsequent cycles, at end of treatment, and during the follow-up period. Hematologic activity was defined as a response that did not meet the criteria for a CR or a partial remission (PR) but that met at least 1 of the following criteria: (1) 50% reduction in marrow or peripheral blood blasts; (2) improvement of absolute neutrophil count >1,000/μL in patients with pretreatment neutropenia; or (3) platelet count >100,000/μL in patients with pretreatment thrombocytopenia.

A Simon 2-stage design was used for statistical analysis, with a desirable CRc rate of ≥35% and an undesirable CRc rate of ≤20%. An interim analysis was planned after enrollment of the first 33 evaluable patients, and study termination was planned if there were 7 or fewer patients with CRc. Adverse events (AEs) were collected from the time that written informed consent was obtained through day 30 after the last dose of study drug. Adverse events of special interest included hepatic function abnormalities, CLS, HUS, and ocular events. No dose reductions were permitted.

Supportive Care

Hydration at a rate of ≥90 mL/m²/hour began 3 hours before and continued for 2 hours after each dose, with daily fluid intake approximating 1,440 mL/m²/day until 18 hours after the last dose of each cycle. Premedication consisted of acetaminophen, diphenhydramine, and ranitidine. Dexamethasone 2.5 mg/m²/dose was given 30 minutes prior to each moxetumomab pasudotox infusion and again at 12 hours after the start of each moxetumomab pasudotox infusion, for prevention of CLS for all doses during cycle 1. Concomitant intrathecal chemotherapy at restaging time points was permitted for central nervous system prophylaxis but could not be given on days of moxetumomab pasudotox administration.

Pharmacokinetics

Plasma levels of moxetumomab pasudotox were determined using a sandwich enzyme-linked immunosorbent assay (ELISA; see Supporting Information for details).¹¹ Samples were obtained before, immediately after, and at 1, 3, and 6 hours after the day 1 dose, and before and immediately after dose 6 during cycles 1, 2, 3, and every subsequent odd-numbered cycle until the end of treatment. Additional samples were obtained within 24 hours of diagnosis, during the AE of interest, and after resolution of any grade HUS, CLS, or ocular toxicity. Pharmacokinetic parameters were estimated by noncompartmental analysis using Phoenix® WinNonlin® software (version 6.3, Certara; Princeton, NJ).

Immunogenicity Assays

Patients were assessed for immunogenicity to moxetumomab pasudotox using a tiered approach and were considered evaluable if they had at least 1 post-baseline sample. Clinical samples were first tested in the validated antidrug antibody (ADA) screening electrochemiluminescence (ECL)-based immunoassay.¹⁴ ADA-positive samples were then evaluated for neutralizing activity using a validated Raji cell cytotoxicity assay.¹⁴ Neutralizing antibody–positive samples were characterized for titer levels.¹⁴ Samples were obtained prior to the start of each cycle for cycles 1, 2, and 3, for subsequent odd-numbered cycles, at the end of treatment, and at the 30-day follow-up visit. Additional samples were also obtained within 24 hours of diagnosis and after resolution of any grade HUS, CLS, or ocular toxicity.

Minimal Residual Disease Evaluation

Measurements of MRD in bone marrow aspirates were performed at the time of response evaluation after a cycle of moxetumomab pasudotox administration. The MRD was evaluated using a 6-color flow cytometry precursor B-ALL MRD panel, which has a demonstrated sensitivity of <0.01%.¹⁵ The following antibody combinations were included in the panel: CD20-FITC/CD10-PE/CD38-PerCPCy5.5/CD58-APC/CD19-PECy7/CD45-APCH7, CD9-FITC/CD13+33-PE/CD34-PercPCy5.5/CD10-APC/CD19-PECy7/CD45-APCH7, and a combination containing Syto-16 to identify nucleated cells.¹⁵ The MRD value was expressed as a percentage of mononuclear cells. Samples were tested at Johns Hopkins University (Baltimore, Maryland) or at the University of Washington (Seattle, Washington), depending on the location of the clinical site.

Cytokine and CD22 Assessment

An exploratory analysis evaluated inflammatory biomarkers (Table S1) in patients treated with moxetumomab pasudotox who experienced CLS or HUS (n = 8) versus those who did not develop CLS or HUS (n = 20); 2 patients without CLS or HUS did not have samples sent for biomarker analysis. Notably, for patients who developed CLS or HUS, PK samples were collected within 24 hours of diagnosis, during the event as clinically indicated, and after resolution of the event. Baseline CD22 expression by flow cytometry was evaluated as an exploratory biomarker (Figure S1).

RESULTS

Patients

A total of 32 patients, all with a primary diagnosis of precursor B-ALL, were enrolled in the study. The median age was 10 years (range, 4–17 years). Nine patients (28%) were primarily refractory to first-line therapy, and 23 patients (72%) had relapsed disease. Ten patients (31.3%) had relapsed after a prior allogeneic HSCT. Only 1 subject had received prior CD22-targeted therapy (CAR T-cells). Five subjects had received prior blinatumomab and 9 had received CAR T-cells (5 CD19-directed, 4 unspecified). The majority of patients (87.5%) had an M3 marrow (≥25% marrow involvement) at baseline (Table 1), with a median blast percentage of 69.5% (range, 5.0%–99.0%). Among the 32 patients enrolled, 2 did not receive study drug because of complications related to progressive disease that occurred between the time of enrollment and initiation of study drug; thus, only 30 patients were evaluable for safety. Twenty-eight patients were evaluable for efficacy, which was limited to those patients who received any study drug and had at least 1 post-baseline disease assessment. Two of the treated patients were excluded from the efficacy evaluable population owing to missing disease assessments related to AE development.

TABLE 1.

Patient characteristics

Parameter	Patients
Total enrolled	32
Age, years, median, range	4–17
>6 months and ≤3 years, n (%)	0
>3 and ≤12 years, n (%)	22 (68.8)
>12 and <18 years, n (%)	10 (31.3)
Sex, n (%)
Female	13 (40.6)
Male	19 (59.4)
Time from diagnosis to study entry, months; median (range)	45.7 (5–179)
Diagnosis of precursor B-ALL, n (%)	32 (100)
Children’s Oncology Group risk criteria28 (at diagnosis), n (%)
Low	0
Standard	9 (28.1)
High	18 (56.3)
Very high	2 (6.3)
Unknown	2 (6.3)
Prior treatment response, n (%)
Refractory disease	9 (28.1)
Relapsed disease	23 (71.9)
Number of prior treatments, median (range)	2 (1–7)
Prior allogeneic hematopoietic stem cell transplantation, n (%)	10 (31.3)
Marrow status at enrollment, n (%)
M1	1 (3.1)a
M2	3 (9.4)
M3	28 (87.5)

^aOne patient who met the inclusion criteria without deviation was enrolled, but subsequently had M1 status at screening.

Among the 30 patients who received moxetumomab pasudotox, the median number of cycles on therapy was 1.5 (range, 1–4), with 7 doses as the median number of doses received (range, 2–24). Fifteen patients (50%) received 1 cycle, 12 (40%) received 2 cycles, 1 (3%) received 3 cycles, and 2 (7%) received 4 cycles of therapy.

Response

Among the 28 patients who were evaluable for response, 23 received 6 doses in cycle 1 and 27 had formal restaging evaluations (Table 2; Figure S2). Three patients (10.7%) attained an MRD-positive morphologic CR after 1 (n = 1) or 2 (n = 2) cycles of therapy. The duration of CR was approximately 2 months in 1 patient, who subsequently received chemotherapy and blinatumomab. In the other 2 patients, there was no subsequent response assessment after the first documentation of CR; one of these patients proceeded directly to HSCT and achieved MRD negativity post-transplant. Five patients (17.9%) attained a partial response leading to an overall objective response in 8 patients (28.6%). Two patients had a best overall response of hematologic activity, with both patients having more than a 50% reduction in marrow blasts; 1 of these patients had grade 4 neutropenia at baseline that improved to a normal neutrophil count, and the other patient had normal neutrophil values at baseline and throughout treatment. Six patients (21.4%) had stable disease and 11 patients (39.3%) had progressive disease.

TABLE 2.

Best overall response

Parameter	Patients, n (%)
Total evaluable for efficacya	28b
Complete response, composite	3 (10.7%)c
Partial response	5 (17.9%)
Hematologic activity	2 (7.1%)
Stable disease	6 (21.4%)
Progressive disease	11 (39.3%)

^a2 patients who received treatment were not included in the efficacy evaluable population: 1 patient was excluded due to missing baseline evaluations and 1 patient discontinued treatment due to an AE before post-baseline disease assessment.

^bIn 1 patient evaluable for efficacy, a formal response assessment was not reported after the patient discontinued therapy due to HUS after 4 doses in cycle 1, and was subsequently referred to their local hospital.

^cAll patients who attained composite complete response remained positive for minimal residual disease.

Adverse Events

Among the 30 patients who received moxetumomab pasudotox, all reported at least 1 AE (Table 3). Treatment-related AEs greater than or equal to grade 3 were observed in 10 patients (33%). The most common treatment-related AEs were transiently increased ALT, increased AST, increased weight, CLS, and blurred vision (Table S2). Veno-occlusive disease (VOD)/sinusoidal obstructive syndrome (SOS) was not seen. Six patients (20%) experienced at least 1 treatment-related AE that led to treatment discontinuation. The AEs of special interest included ocular dysfunction (periorbital edema or blurred vision) in 11 patients (36. 7%), CLS in 6 patients (20%), and HUS in 4 patients (13.3%). CLS was grade 2 in 3 patients, grade 3 in 2 patients, and grade 4 in 1 patient who developed associated acute respiratory distress syndrome. Four patients experienced HUS, which was grade 3 in 3 patients and grade 2 in the other patient. In 5 patients, at least 1 AE of special interest led to permanent discontinuation of study drug. The onset of all CLS or HUS events occurred early during cycle 1 (dose 2 or 3), except for 1 patient each with CLS or HUS in cycle 2 (dose 9). One treatment-related death was reported in a patient who received 6 doses of treatment and experienced fatal pulmonary edema in the setting of disease progression, with the cause of death attributable to both moxetumomab pasudotox and underlying ALL, and was independent of the other reported CLS events.

TABLE 3.

Treatment-emergent adverse events in ≥10% of patients

Adverse event	Patients, n (%)
Pyrexia	16 (53.3)
Headache	13 (43.3)
Alanine aminotransferase increased	11 (36.7)
Anemia	11 (36.7)
Weight increased	10 (33.3)
Platelet count decreased	9 (30.0)
Aspartate aminotransferase increased	8 (26.7)
Hypertension	8 (26.7)
Nausea	8 (26.7)
Febrile neutropenia	7 (23.3)
Back pain	6 (20.0)
Capillary leak syndrome	6 (20.0)
Dyspnea	6 (20.0)
Hypoalbuminemia	6 (20.0)
Hypocalcemia	6 (20.0)
Neutrophil count decreased	6 (20.0)
Pain in extremity	6 (20.0)
Vision blurred	6 (20.0)
Edema peripheral	5 (16.7)
Hypokalemia	5 (16.7)
Hyponatremia	5 (16.7)
White blood cell count decreased	5 (16.7)
Anxiety	4 (13.3)
Bone pain	4 (13.3)
Constipation	4 (13.3)
Face edema	4 (13.3)
Gamma-glutamyl transferase increased	4 (13.3)
Hemolytic uremic syndrome	4 (13.3)
Hyperglycemia	4 (13.3)
Hypotension	4 (13.3)
Hypoxia	4 (13.3)
Vomiting	4 (13.3)
Abdominal pain	3 (10.0)
Cough	3 (10.0)
Diarrhea	3 (10.0)
Fatigue	3 (10.0)
Hypophosphatemia	3 (10.0)
Periorbital edema	3 (10.0)
Pulmonary edema	3 (10.0)
Sinus tachycardia	3 (10.0)
Tachycardia	3 (10.0)

Pharmacokinetics

Of the 30 patients who received treatment, PK samples were available for 28 patients. Mean plasma concentration-time profiles and the mean peak plasma concentration of moxetumomab pasudotox following the first dose of cycle 1 are shown in Figure 2. Moxetumomab pasudotox exhibited a monophasic disposition profile, with peak concentrations attained immediately after IV infusion. The mean elimination half-life of moxetumomab pasudotox was approximately 1.07 hours (range, 0.82–1.44 hours), demonstrating fast plasma clearance (range, 24.4–41.8 mL/h/kg). Consistent with a short half-life, no drug accumulation occurred with repeated dosing. Mean PK exposure, as measured by peak plasma concentration (C_max), was similar across cycles (Figure 2B). Apart from the last cycle, the coefficient of variation for C_max was between 28% and 34%, suggesting moderate inter-individual variability.

Figure 2.

(A) Mean plasma concentration-time profiles of moxetumomab pasudotox in pediatric patients with B-cell progenitor acute lymphoblastic leukemia following the first dose of cycle 1. (B) Mean peak concentration levels of moxetumomab pasudotox in pediatric patients with B-cell progenitor acute lymphoblastic leukemia following the first dose of each cycle. Error bars represent standard deviations of the mean. Standard deviation was only calculated if sample size was >3. Arrows indicate dosing events. (C) Overlay of concentration-time profiles of moxetumomab pasudotox in patients with or without adverse events of special interest. Those with (colored) and without (gray) capillary leak syndrome are compared. (D) Those with (colored) and without (gray) hemolytic uremic syndrome are compared.

IV, intravenous; LLOQ, lower limit of quantification.

There was no obvious correlation of PK exposure with safety (CLS or HUS) or response (Table S3, Figure S3, Figure S4). The overall exposures from patients who experienced CLS or HUS were within the range of non-CLS/non-HUS patients (Figure 2C and D), as were mean C_max values (Figure S3). Because of treatment discontinuation for most patients after the occurrence of AEs, PK data after the first dose were too limited to adequately assess for potential correlation of PK and safety. Limited PK data collected during the onset, 24 hours after diagnosis, or after the resolution of the AE showed no detectable moxetumomab pasudotox concentration (Table S3).

CD22 Expression

Flow cytometry analysis of the baseline blood samples (n=24) showed a broad range of CD22 expression (Figure S1). There was no correlation of CD22 expression with the clinical response to moxetumomab pasudotox (Figure S1).

Immunogenicity

Samples for immunogenicity (ADA) assessment were available for 28 of the 30 treated patients at baseline, of whom 23 patients also had post-treatment assessments. Therefore, 23 patients were considered evaluable for an ADA response to moxetumomab pasudotox, of whom 18 (78%) were ADA positive at baseline. After treatment, 13 (57%) of 23 patients were ADA positive and were positive for the presence of neutralizing antibodies. Most ADA-positive patients had low titers and persistent profile during treatment. Three patients who tested ADA negative at baseline but became positive after treatment had ADA titers ranging from 160 to 20,480 (Figure S5).

Pharmacokinetic exposures (C_max and area under the concentration-time curve [AUC]) were not found to be statistically different (t test, p>0.05) between ADA-positive and ADA-negative patients. Moreover, after the first dose, PK profiles of pre-treatment ADA-positive patients (n = 18) were within the range of ADA-negative patients, and no impact of preexisting ADA on PK parameters was evident. There was also no apparent correlation between ADA status and toxicity in this limited number of patients.

Inflammatory Biomarkers

No apparent differences were observed in baseline levels of the inflammatory biomarker panel (Table S1) of patients treated with moxetumomab pasudotox who developed adverse responses of CLS or HUS compared with patients who did not develop CLS or HUS. In addition, no apparent differences were observed in changes in the inflammatory biomarker panel post-treatment with moxetumomab pasudotox in patients who developed CLS or HUS versus those who did not.

DISCUSSION

Effective targeting of CD22 in precursor B-ALL remains an important therapeutic strategy, particularly in the evolving field of CD19 negative disease following CD19 directed therapies.^16–20 Current CD22-directed interventions include IO,⁸ an anti-CD22 conjugated immunotoxin, which has received FDA approval in adults with ongoing development in children²¹ and CD22-directed chimeric antigen receptor T cells, which are still in early phase trials.^22,23 Moxetumomab pasudotox, as an anti-CD22–targeted recombinant immunotoxin that was recently approved by the FDA for treatment of HCL, represents another strategy in precursor B-ALL. Based on the experience in HCL,^10,11 which included the elimination of MRD,⁵ and on the early experience in pediatric precursor B-ALL,¹² with an established toxicity profile and the determination of a phase 2 dose, further testing in pediatric ALL was undertaken.

This multicenter phase 2 study tested the hypothesis that moxetumomab pasudotox may provide clinical benefit, with an end point based on CR induction rates, in patients with r/r precursor B-ALL. Although the objective response rate (ORR) of 28.6% in the current phase 2 trial was comparable to that observed in the phase 1 trial,¹² the proportion of patients achieving CR versus PR was lower. Specifically, 3 of 28 evaluable patients (10.7%) attained CR, none of whom were MRD negative. In comparison, in the prior phase 1 trial,¹² among 47 evaluable patients, the ORR was 32%, with 11 subjects (23%) having a CR, of whom 5 were MRD negative. No new toxicities were identified compared with the phase 1 study, although rates of CLS and HUS were higher in the current study. The most frequent AEs related to protocol therapy were increased ALT, increased AST, increased weight, CLS, and blurred vision. Adverse events of special interest included ocular dysfunction in 11 patients (36.7%), CLS in 6 patients (20%), and HUS in 4 patients (13.3%). The etiology for ocular dysfunction remains unclear but may potentially be related to underlying capillary leak; ocular dysfunction also was seen in the corresponding phase 1 study and is included in the FDA package insert for this drug. Notably, moxetumomab pasudotox has not been associated with VOD/SOS, differentiating it from an established toxicity of IO.^18,24 Ultimately, the study was terminated upon the interim analysis because the CRc rate did not reach the stage 1 target.

Reasons for the lower-than-expected response rate are unclear. Disease burden was similar between the phase 1 and 2 studies; however, more patients in this trial had relapsed after prior immunotherapy, raising the possibility that this study population may have been more refractory. As noted, different manufacturing processes were utilized during the development of moxetumomab pasudotox. It is possible that the change in manufacturing could have contributed to differences between the results of the phase 1 and phase 2 trials. However, the limited data with Process 3 material in the phase I study did not reveal any obvious difference in activity or toxicity.¹² Importantly, this manufacturing change also occurred during the development of moxetumomab pasudotox for HCL without any apparent difference in outcomes, and Process 3 material was approved by the FDA for the treatment of adults with HCL.^5,10

Pharmacokinetics, immunogenicity, and cytokine assays were performed as part of this study. No apparent correlation was observed between PK profiles and the development of CLS or HUS. Immunogenicity did not appear to influence drug exposure or correlate with treatment outcomes or AEs. The high rates of baseline immunogenicity and ADAs were similar to those in adults with HCL, most likely due to previous exposure of study patients to Pseudomonas exotoxin A, which is abundantly present in the environment and shares a PE38 domain with moxetumomab pasudotox.¹⁴ Cytokine profiling demonstrated no apparent relationship between baseline or post-treatment levels of the inflammatory biomarker panel with onset of CLS or HUS. Predictive markers for the development of CLS and HUS are lacking and further studies to elucidate the mechanism of these toxicities are needed.

In a recent pivotal study in patients with r/r HCL, moxetumomab pasudotox resulted in durable complete response and the eradication of MRD ^5,10with acceptable tolerability; this study was the basis of the US approval of the drug for third-line treatment of r/r HCL. For reasons that are not entirely established, the agent has been less effective in pediatric precursor B-ALL.¹² This may be attributable in part to differential proliferation rates, which is rapid in ALL. Response rates might also be reduced because of the relatively lower CD22 expression in ALL compared with HCL.^13,25 In light of the particularly short half-life of this agent, but with some evidence of clinical activity, additional preclinical models are exploring the feasibility and tolerability of adjusted dosing schemas to enhance clinical activity and to reduce toxicity. Data suggest that responses in ALL might be improved by replacing bolus dosing with smaller doses at more frequent intervals or with continuous infusion administration to increase exposure time.²⁶ Notably, in comparison to the short elimination half-life of moxetumomab pasudotox (1.07 hours), IO has a half-life of 12.3 days. Thus, prolonged exposure might account for the higher response rates observed after IO for ALL.^8,21 Notably, the mechanisms of action of IO and moxetumomab pasudotox differ (ie, direct DNA damage vs. inhibition of protein synthesis, respectively); thus, there are likely different potential mechanisms of resistance. Future clinical trials could evaluate moxetumomab pasudotox administered via continuous infusion, which has been associated with improved activity in preclinical models. Combining moxetumomab pasudotox with other agents to potentially enhance activity is another strategy. Possibilities include combination with azacytidine, which may improve its antitumor effect,²⁷ or with bryostatin1, which might increase the efficacy of moxetumomab pasudotox by upregulation of CD22 expression.²⁸

In conclusion, while the risk-benefit profile did not support progressing to phase 3 evaluation, these data demonstrate that moxetumomab pasudotox is an agent with some activity against precursor B-ALL. Given the FDA approval of this agent in adults with HCL and an emerging population of patients in need of effective CD22-targeted therapies, further study of this agent to optimize response rates and tolerability in precursor B-ALL could be considered.

Abbreviations Key

ADA

Antidrug antibodies

AE

Adverse event

ALL

acute lymphoblastic leukemia

ALT

alanine aminotransferase

AST

aspartate aminotransferase

AUC

area under the concentration-time curve

B-ALL

B-cell acute lymphoblastic leukemia

BCP

B-cell precursor

CLS

capillary leak syndrome

C_max

peak plasma concentration

COG

Children’s Oncology Group

CR

complete response

CRc

composite complete response

ECL

electrochemiluminescence

ELISA

enzyme-linked immunosorbent assay

FDA

Food and Drug Administration

HCL

hairy cell leukemia

HSCT

hematopoietic stem cell transplantation

HUS

hemolytic uremic syndrome

IO

inotuzumab ozogamicin

IV

intravenous

MRD

minimal residual disease

PK

pharmacokinetic

PR

partial response

SOS

sinusoidal obstructive syndrome

VOD

veno-occlusive disease