Isatuximab in Combination with Chemotherapy in Pediatric Patients with Relapsed/Refractory Acute Lymphoblastic Leukemia or Acute Myeloid Leukemia (ISAKIDS): Interim Analysis

Introduction: Acute lymphoblastic leukemia (ALL) is the most common pediatric cancer, representing ~25% of cancer diagnoses in children. Approximately 20% of childhood leukemias are of myeloid origin with the majority being acute. Isatuximab (Isa) is a monoclonal antibody that binds to a specific epitope of CD38 and exerts anti-multiple myeloma (MM) effects through several modes of action. Isa is approved for use in MM in adults. CD38 expression is high in some acute leukemia subsets in pediatric patients, making CD38 a potential target for acute leukemia treatment in children.

Methods: ISAKIDS (NCT03860844) is a Phase 2, single-arm, multicenter, open-label study evaluating the antitumor activity, safety, and pharmacokinetics (PK) of Isa in combination with standard salvage chemotherapies in children with relapsed or refractory (R/R) leukemia in first or second relapse; including both T-ALL and B-ALL and acute myeloid leukemia (AML), conducted in 3 separate cohorts. Children from 2 years (yrs) to <18 yrs of age were eligible for inclusion. In all cohorts, participants received 20 mg/kg Isa as a single agent on Day 1 and then weekly for 5 weeks in the ALL groups or weekly for 3 weeks in the AML groups. Combination chemotherapy was added on Day 8 (modified UKALL R3 and fludarabine, cytarabine, G-CSF induction for ALL and AML patients (pts), respectively) unless it was judged necessary to start chemotherapy earlier than Day 8. The primary endpoint was complete response (CR) rate, defined as the proportion of pts with CR or CR with incomplete peripheral recovery (CRi). Secondary endpoints included duration of response, event-free survival, overall survival, and overall response rate. CD38 receptor density was assessed at baseline and CD38 occupancy was assessed at Day 15. Descriptive statistics were carried out on the evaluable population. Confidence intervals were computed using the Clopper-Pearson method. For this first interim analysis, PK parameters for Isa were assessed in the first 20 pts to confirm Isa dose and schedule in very young pts. Safety analyses included treatment-emergent adverse events (TEAEs) of all grades.

Results: 24 pts were enrolled (10, B-ALL; 7, T-ALL; 7, AML) and treated. Baseline characteristics were comparable between cohorts; median age (range) B-ALL 6.5 (3-14) yrs; T-ALL 10.0 (7-16) yrs; AML 7.0 (2-17) yrs. Males (4/10 B-ALL; 6/7 T-ALL; 3/7 AML). Evaluable Lansky scores of 90 - 100 (9/10 B-ALL; 3/4 T-ALL; 4/6 AML). Mean (standard deviation) time from initial diagnosis to first dose of investigational medicinal product (IMP) was 2.6 (1.0) yrs, 1.2 (0.6) yrs, and 1.0 (0.3) yrs for the B-ALL, T-ALL, and AML groups, respectively. Number of prior regimens, median (range) was 6 (1-≥8), 3 (1-≥8), 4 (1-≥8) for the B-ALL, T-ALL, and AML groups, respectively, corresponding to inclusion of pts with very advanced disease. Three pts presented with hyperleukocytosis >20 x 10 ⁹/L. Among 23 pts assessed for CD38 expression, all except 2 (1 B-ALL and 1 AML) were CD38 positive. There were 17 evaluable pts with valid response data collected up to the cutoff date (May 11, 2021). Of the evaluable participants, CR+CRi were observed for: 3/7 (42.9%) in the B-ALL cohort, 2/6 (33.3%) in the T-ALL cohort, and 2/4 (50.0%) in the AML cohort (Table). Safety data showed that Grade ≥3 TEAEs occurred in 17 pts (Table). Infusion reactions (majority Grade 1 or 2) occurred in 9/24 pts (2/10 B-ALL; 4/7 T-ALL; 3/7 AML). Three pts had Grade 5 TEAEs (deaths: 1 B-ALL and 2 AML). One subject in the AML cohort had cytokine-release syndrome (CRS) related to IMP with progressive disease as a confounding factor. The other 2 deaths were not treatment related. One subject in the AML cohort had cellulitis Grade 5 and 1 subject in the B-ALL cohort had septic shock Grade 5 as cause of death. Preliminary PK analysis on 18 pts (>2 yrs) showed that Isa PK parameters including exposure PK parameters in children with ALL or AML are consistent with those observed in adult ALL pts. PK modeling approaches predict a slightly lower exposure (<20%) in pts <2 yrs old (4-12 kg) compared with exposure achieved in adults (51-100 kg) at the same dose.

Conclusions: In a poor prognostic relapsed population, 7/17 (41.2%) pts achieved CR+CRi. Safety profile and exposure is consistent with the available data for adults. Enrollment will be opened for pts <2 yrs of age with R/R ALL or AML.

Figure 1

View largeDownload slide

Figure 1

View largeDownload slide

Close modal