A Retrospective Study on Inotuzumab Ozogamicin in Infants and Young Children with Relapsed or Refractory Acute Lymphoblastic Leukemia (ALL)

Background: Inotuzumab ozogamicin (InO) is a CD22-directed monoclonal antibody linked to calicheamicin. It is approved for adults with relapsed/refractory (R/R) CD22+ B-cell precursor acute lymphoblastic leukemia (BCP-ALL) at the dose of 1.8 mg/m2/course fractionated at day 1, 8 and 15 (Kantarjian, NEJM, 2016). The pediatric experience with InO is still limited. Patients (pts) <1 year of age cannot be enrolled in the ongoing Phase 1-2 clinical trial and no data for this age group are published so far. Drug administration in these pts with very low Body Surface Area (BSA), unique body composition and drug disposition is always challenging, especially with new agents with unknown toxicities. Infant ALL is characterized by a dismal prognosis with an EFS of 47% in newly diagnosed pts, and of 20% after relapse (Driessen, Leukemia 2017). There is an unmet need for innovative strategies in these pts, and MLL-rearranged (MLL-r) leukemia, commonly seen in infants (75%), has shown a lower average CD22 antigen density in previous studies (Shah, Ped Blood Cancer 2015).

We collected cases of infants and young children diagnosed with R/R ALL and treated with InO via a compassionate access program.

Methods: Participating international pediatric oncology centers submitted retrospective demographic, outcome, and toxicity data on infants (age <1 year) and young children (age ≥1 and ≤3 years), with a diagnosis of R/R ALL, who received at least 1 dose of InO, after regulatory approval according to local regulation. Study procedures were limited to review of medical records. Survival analysis used the Kaplan-Meier method.

Results: Between 2015 and 2019, 9 infants and 3 children (1.0-2.9 years of age) were diagnosed with ALL and later on received 1-3 cycles of InO due to R/R disease. At the time of InO treatment 1 pt was <1.0 year old, 5 ≥1.0 and <2.0, and 6 ≥2.0 and <3.0. The median age was 5,7 months (range 1,1-27) at the time of ALL diagnosis and 22 months (range - 7,5-35) at the time of InO treatment. Median BSA at InO start was 0.51 m² (range 0.32-0.63), median weight 11,8 Kg (range 5,8-15,9 Kg) and median WBC 7.3 x10⁹/L (range 0.2-36.9). Pts were heavily pretreated: 6 had a ≥2^nd relapse, 4 a refractory 1^st relapse and 2 a 1^st relapse after hematopoietic stem cell transplant (HSCT), median 3 prior treatment regimens (range 2-7). Five pts had been previously transplanted and 7 received prior anti-CD19 directed therapy (Blinatumomab). 9/12 pts were MLL-r.

At the start of InO, 8 (67%) pts had an M3 bone marrow status, 2 (17%) an M2 and 2 an M1 with MRD positivity. CD22 expression was reported in 7/12 pts: in 3 cases >90% of blasts expressed CD22, the others ranging from 52% to 80%. In total 21 courses were administered (60 doses; 3 courses not completed), median administered dose of InO per course 1.74 mg/m² (divided in 3 weekly doses, with the loading dose at D1 of cycle 1): 6 pts received 1.8 mg/m² (adult approved dose), 4 intermediate dosages as 1.5 or 1.6 mg/m², 1 1.4 mg/m2 and 1 1.2 mg/m². The dose was based on BSA; 0.07 mg/Kg/course was the median pro-Kg dose administered (higher than in adults).

6 pts (50%) achieved a complete remission (CR) (3 after 1 course and 3 after 2-3 courses), 1 additional MRD-positive pt. turned MRD-negative (<0.01%) after 1 course. 4 pts reached an MRD <0.1%. Level of CD22 expression seems not to be related to the response. 5 (42%) pts proceeded to HSCT after InO (1 in partial remission (PR) and 1 after additional treatment following reinduction with Ino), after a median time of 34 days (range 21-70) after the last InO dose. Median survival from the start of InO was 160 days. OS at 6 months was 50.0% (95%CI 28.4%- 88.0%).

Only limited data about toxicity were reported in our records, although hematologic toxicity was the most common (thrombocytopenia in 33% of pts). 1 case of neurotoxicity described as toxic leukoencephalopathy was reported after 1 course of InO, causing death in CR. 1 pt. developed severe veno-occlusive disease (VOD) after HSCT performed in PR, which was fatal.

Conclusions: Single agent InO in infants and younger children had a CR rate of 58% in this cohort of heavily pretreated pts. Even with a median dose close to the adult recommended starting dose, InO appeared to be well tolerated, allowing a subsequent HSCT in 5 pts. While taking into account the lower CD22 expression showed by MLL-r pts, further investigation with InO treatment might be valuable in this subgroup of ALL pts with a particular dismal prognosis.