Phase II Trial of Inotuzumab Ozogamicin in Children and Adolescents With Relapsed or Refractory B-Cell Acute Lymphoblastic Leukemia: Children's Oncology Group Protocol AALL1621

Abstract

Purpose: Children's Oncology Group trial AALL1621 was conducted to prospectively determine the safety and efficacy of inotuzumab ozogamicin (InO) in pediatric and adolescent patients with relapsed or refractory (R/R) B-cell acute lymphoblastic leukemia (B-ALL).

Patients and methods: This single-arm phase II trial enrolled patients age 1-21 years with R/R CD22-positive B-ALL. In cycle 1, InO dosing was 0.8 mg/m² intravenously on day 1 and 0.5 mg/m² on days 8 and 15 of a 28-day cycle with response evaluation at day 28. Using a two-stage design, the trial was continuously monitored for dose-limiting toxicities and sinusoidal obstruction syndrome (SOS). CD22 expression was retrospectively evaluated by central flow cytometry.

Results: Forty-eight patients were evaluable for response and toxicity; 19 had complete response (CR) and nine CR with incomplete count recovery (CRi) after cycle 1 (CR/CRi rate: 58.3%; two-sided 90% CI, 46.5 to 69.3). Twenty-seven of 28 patients with CR or CRi had minimal residual disease measured by flow cytometry; 18 (66.7%) had minimal residual disease < 0.01%. Seven of 28 patients (25%) with CR or CRi had delayed count recovery past day 42 in cycle 1. Three (6.3%) patients had grade 3 ALT elevation and one patient had grade 3 hyperbilirubinemia in cycle 1. Of 21 patients undergoing hematopoietic stem-cell transplantation after InO, 6 (28.6%) developed grade 3 SOS. Partial CD22 expression and lower CD22 site density were associated with lower likelihood of response to InO.

Conclusion: InO is effective and well tolerated in heavily pretreated children and adolescents with R/R CD22-positive B-ALL. SOS after hematopoietic stem-cell transplantation and prolonged cytopenias were notable. CD22 modulation was identified as a mechanism of resistance. Expanded study of InO combined with chemotherapy is underway.

Trial registration: ClinicalTrials.gov NCT02981628.

FIG 1.

Swimmer's plot depicting individual patient response, subsequent therapy, and outcomes for 48 patients. Data are censored as of December 31, 2020. Patients with Down syndrome are denoted as DS. Dark bars indicate InO given on study. Initial response to cycle 1 is indicated by CR, CRi, PR, SD, or PD. For patients who received cycle 2, response is indicated as CCR or CCRi if previously achieved CR or CRi with cycle 1. Duration of initial response is indicated by color (CR or CRi = blue; PR or SD = red, PD = gray). HSCT is indicated by star, and CAR T-cell therapy is indicated by triangle. Patient 9 received six cycles of therapy on study and continued to receive commercial InO for an additional year before relapse. Patient 19 with KMT2A rearrangement received CD19 CAR T-cell therapy following three cycles of InO and subsequently developed lineage switch to acute myeloid leukemia (denoted as SMN). C, censored; CAR, chimeric antigen receptor; CCR, continuous complete response; CCRi, continuous complete response with incomplete count recovery; CR, complete response with or without count recovery; CRi, complete response with incomplete count recovery; D, death; DS, Down syndrome; HSCT, hematopoietic stem-cell transplantation; InO, inotuzumab ozogamicin; PD, progressive disease; PR, partial response; R, relapse; SD, stable disease; SMN, second malignant neoplasm.

FIG 2.

Flow cytometry evaluation of CD22 expression in three patients pretreatment (left column) and post-treatment (right column) with InO. The CD22 site density shown reflects that of the entire blast population, irrespective of percentage of CD22 positivity. (A) Partial CD22 expression pre-InO, which is further decreased post-InO. (B) Uniformly positive expression of CD22 pre-InO, with partial/bimodal expression of CD22 post-InO. (C) CD22 expression ranging from moderate to dim pre-InO with diminished CD22 expression post-InO (KMT2A-rearrangment with subsequent lineage switch to acute myeloid leukemia after CD19-directed therapy). InO, inotuzumab ozogamicin.

FIG 3.

Outcomes after InO monotherapy. (A) EFS and OS of the 48 patients treated on study. The estimated EFS rate was 28.6% (95% CI, 15.9 to 42.8) at 2 years and 19.6% (95% CI, 7.9 to 35.1) at 3 years. The estimated OS rate was 36.0% (95% CI, 22.3 to 49.9) at 2 years or 3 years. (B) EFS and OS of patients with morphologic CR or CRi and MRD < 0.01% by flow cytometry with one or two cycles of treatment. Among patients with best response of CR or CRi and MRD < 0.01%, EFS was 71.4% (95% CI, 47.2 to 86) at 1 year and 57.7% (95% CI, 31.9 to 76.8) at 2 year, and the 2-year OS rate was 68.6% (95% CI, 41.8 to 84.9). (C) EFS and OS of patients who received HSCT or CART without other bridging therapy following a best response of CR or CRi after up to two cycles of InO. Of the 30 patients who achieved CR or CRi after up to two cycles of InO, four patients had an event at the end of InO therapy or shortly after being off protocol, two patients did not have HSCT or CART within 6 months without other therapy, and the remaining 24 patients had HSCT or CART following CR or CRi. Of the 24 patients who received HSCT or CART following CR or CRi, 2-year EFS and OS rates were 62.0%% (95% CI, 39.6 to 78.1) and 66.2% (95% CI, 43.6 to 81.5), respectively. CART, chimeric antigen receptor T-cell therapy; CR, complete response; CRi, complete response with incomplete count recovery; EFS, event-free survival; HSCT, hematopoietic stem-cell transplantation; InO, inotuzumab ozogamicin; MRD, minimal residual disease; OS, overall survival.

FIG 4.

CD22 site density. (A) Surface CD22 site density measured using ABC at baseline and after cycle 1 in 13 patients with residual leukemia after cycle 1. (B) Surface CD22 expression (CD22%) at baseline and after cycle 1 in 13 patients with residual leukemia after cycle 1, demonstrating decrease from baseline in eight of 13 patients. ABC, antibody bound per cell.