In Pediatric AML, CD33 Targeting Agent Improved EFS, Risk of Relapse

Myeloblasts with Auer rods seen in AML; source: Paulo Henrique Orlandi Mourao

Treatment with gemtuzumab ozogamicin (GO) improved the event-free survival in children and adolescents with acute myeloid leukemia (AML) by reducing the risk of relapse among those able to achieve remission, according to results of a phase III Children’s Oncology Group trial presented at the 2013 American Society of Hematology (ASH) Annual Meeting and Exposition.

Based on these results, study author Alan S. Gamis, MD, MPH, of Children’s Mercy Hospital and Clinics said that GO could be “added to standard therapy at the time of induction for low-risk patients” and that its use may also “enhance the benefits of stem cell transplant for high-risk patients” as well.  

In recent years the survival for pediatric AML has gradually improved due to both the use of chemotherapy intensification as well as the incorporation of allogeneic stem cell transplant and better supportive care. However, according to Gamis, researchers have hit a ceiling with their success due to limits in the escalation of therapy, as well as toxic mortality and late cardiac sequelae, leading them to search for new targeted therapies.

Approximately 80% to 85% of patients with AML have high CD33 expression levels. GO is a humanized anti-CD33 monoclonal antibody. According to Gamis, GO was previously removed from the market when earlier clinical trials failed to show clinical value of the drug. However, trial data released subsequent to its market withdrawal indicated that GO had a benefit when researchers looked at disease-free survival and reduced relapse risk endpoints, and that benefit seemed to occur in subgroups of patients with low- or intermediate-risk AML.

Gamis and colleagues conducted this phase III trial to determine if GO improved event-free survival in pediatric patients with AML and to see if improvements might vary by risk group and method of intensification.

The researchers enrolled 1,022 patients aged younger than 30 years. The median age of patients was 10 years. Patients were randomly assigned to standard therapy or two doses of GO 3 mg/m² per dose on day 6 of induction I and on day 7 on intensification II in a five cycle chemotherapy backbone. Additional chemotherapy was given to low-risk patients and allogeneic donor stem cell transplants to high-risk patients. Patients were followed for a median of 3.6 years.

Patients assigned to the investigational drug experienced improved event-free survival (HR = 0.83; 95% CI, 0.70-0.99; P = .04) and relapse-free survival (HR = 0.74; 95% CI, 0.6-0.93; P = .01) compared with patients who underwent standard therapy. Specifically, at 3 years the event-free survival was 46.9% for patients on standard treatment compared with 53.1% for GO (P = .05). Event-free survival with GO remained significantly longer upon multivariate analysis adjusting for significant adverse risk factors. The researchers found no significant difference in overall survival.

Treatment with GO conferred no better induction complete remission compared to standard therapy (88% vs 85%). Therefore, the researchers looked at outcomes from the end of induction and found that patients assigned GO had a reduced risk of relapse and an improved disease-free survival. Data also showed that patients assigned GO had an increased risk for non-leukemic mortality (P = .08).

When looking at the data stratified by risk groups, patients in the low-risk groups had relapse rates that trended lower when assigned GO, but toxic mortality was significantly worse for these patients compared with patients on standard therapy (7.5% vs 1.8%).

Additionally, event-free survival, relapse rate, and overall survival trended toward improvement in intermediate-risk patients assigned GO, but when the researchers censored patients at the time of transplant no significant difference in outcomes was found. 

Finally, when looking at high-risk patients who underwent stem cell transplant, the researchers found that patients assigned GO had a reduced relapse rate (27% vs 44.8%), an improved disease-free survival (55.9% vs 40.3%) and an improved overall survival (67.5% vs 48.5%) compared with the standard treatment arm.

“CD33 targeting does reduce relapse risk,” Gamis said. “We believe that this merits further investigation.”