Favourable Outcomes in Newly Diagnosed Paediatric AML with Gemtuzumab Ozogamicin and Risk-Stratified Therapy: Results from the International Phase III Myechild 01 Trial

Introduction

Adding one dose of 3mg/m² of Gemtuzumab ozogamicin (GO), an anti CD33 antibody drug conjugate, showed an event-free survival (EFS) benefit in children with acute myeloid leukaemia (AML) in the AAML0531 trial (Gamis et al, JCO, 2014) whilst the adult ALFA-0701 trial reported a survival benefit of 3 fractionated doses with standard induction (Castaigne et al, Lancet, 2012). MyeChild 01, an international trial (UK, France, Australia, New Zealand, Ireland, Switzerland) for paediatric AML, high risk myelodysplastic syndrome (MDS > 10% blasts) or isolated myeloid sarcoma (IMS), embedded a GO dose finding study which established the safety of combining up to 3 doses of 3mg/m² with intensive induction chemotherapy. 515 patients were randomised to 1 vs 3 doses of GO during course 1 (given on day 4 or days 4, 7, and 10 respectively), from Jan-2019 to Jun-2022. Due to regulatory commitments comparative results by randomisation are not available at the time of writing, but these data will be presented at the ASH annual meeting. Overall results are presented here.

Methods

Randomisation of GO doses was stratified by age; diagnosis (AML, MDS, IMS); disease type (de novo, secondary); white cell count (WCC) (<100, ≥100x10⁹/L). Statistical analyses are by intention to treat with a primary outcome of EFS.

All patients were allocated mitoxantrone and cytarabine (MA) with GO in course 1 and thereafter stratified by cyto/molecular genetics, remission status after course 1 and measurable residual disease (MRD). Patients with resistant disease post course 1 or poor risk (PR) cyto/molecular genetics were stratified high risk (HR) and received fludarabine, cytarabine and idarubicin (FLA-Ida) for course 2; all other patients received a second course of MA. Data is available on 472 patients; 142 received FLA-Ida and 330 MA. HR patients proceeded to allogeneic stem cell transplantation (HSCT). Non-HR patients were subsequently allocated treatment based on their cyto/molecular genetics and MRD response.

Results

Patient characteristics were: males 55%; median age 10yr; median WCC 14 x10⁹/L; AML 96%, MDS 2.5%, isolated MS 1.9%; de novo 98%; CNS2 16%, CNS3 9%; non CNS extramedullary disease 16%.

Of 515 patients randomised to 1 vs 3 doses, all but 16 received GO (6 ineligible, 5 prior toxicity, 5 other). 179 patients (35%) had a confirmed HSCT. Median follow-up is 3 years.

The overall 2 yr EFS is 70% (95% CI: 66-74%) and overall survival (OS) 88% (85-91%). 94% achieved complete remission (CR) or CR with incomplete count recovery (CRi) post course 1 or 2; 6% failed to achieve CR/CRi (resistant disease 2.5%, non-evaluable 1%, unknown response 2.3%). 127 of 485 patients who achieved CR/CRi have relapsed giving a 2 yr cumulative incidence of relapse (CIR) of 25% (21-29%). Overall, there were 79 deaths but only 11 (2%) in first remission. 63 deaths were disease-related, 12 transplant-related, 3 off-trial treatment related and 1 other non-cancer.

A cyto/molecular risk group was available for 485 patients: 192 GR (40%), 162 IR (33%) and 131 PR (27%). GR patients had a CR/CRi of 100%, 2 yr EFS 81% (76-87%), CIR 18% (13-24%) and OS 96% (93-99%); IR patients had a CR/CRi of 97.5%, 2 yr EFS 64% (57-72%), CIR 33% (26-41%) and OS 87% (82-93%); PR patients had a CR/CRi of 93.9%, 2 yr EFS 68%(61-77%), CIR 24%(16-32%) and OS 78% (71- 85%).

Time to count recovery was defined as time from day 1 of a course to date of neutrophils >0.75 x10⁹/L and platelets >75 x 10⁹/L. Median count recovery was 41days post course 1, 44 days post course 2 MA and 45 days post course 2 FLA-Ida.

59% of patients had at least one grade ≥3 adverse event or any grade serious adverse event, evaluated from the start of treatment until 30 days after end of induction. Most events were consistent with AML chemotherapy. Only 9 patients (1.7%) had grade ≥3 hyperbilirubinemia and 2 patients confirmed veno-occlusive disease (VOD) in induction.

Conclusion

At least one dose of GO in combination with mitoxantrone and cytarabine in induction followed by risk-adapted therapy has produced excellent results. Despite intensive treatment the death in first remission rate is remarkably low at 2% and VOD was rare. Although this abstract reports only pooled data, outcomes for the 1 vs 3 dose GO randomisation will be presented at the ASH annual meeting.