Outcome of Pediatric Patients with Acute Myeloid Leukemia and -5/5q-Abnormalities Enrolled On Five Children's Oncology Group Acute Myeloid Leukemia Treatment Protocols

Studies of acute myeloid leukemia (AML) in adults have documented that abnormalities of chromosome 5q (-5/5q-), primarily deletions, confer a poor prognosis. However, there are no large studies that specifically focus on -5/5q- in pediatric AML.

To elucidate the disease correlates of this group, we retrospectively analyzed cytogenetic data from 5 studies of childhood AML: Children's Oncology Group (COG-AAML03P1), Children's Cancer Group (CCG-2961 and 2891) and Pediatric Oncology Group (POG-8821 and 9421). Data from all patients whose cytogenetic clones included -5/5q-, with the exception of those with acute promyelocytic leukemia (M3) or Down syndrome, were included. A total of 2035 patients from these 5 studies had cytogenetic data available for review.

Twenty-two (1.1%) of the 2035 patients had −5 or 5q-. The majority of these patients were male (63.6%). The median age was 12.9 years (range 0.3–20.7 years) with a significant number of patients in the 11–21 year age range (63.6%, p=0.032). The median white blood cell count was 17.4 ×10³/μL (range 1.4–98 ×10³/μL) and the median bone marrow blast percentage was 77% (range 15–99%). Patients with -5/5q- had a significantly higher median platelet count at diagnosis than those without this abnormality (88×10³/μL versus 53×10³/μL, p=0.015). Of the 22 patients with -5/5q-, their FAB classification showed that a significant number of patients had M0 morphology (28.6%, p<0.001) versus patients without -5/5q- (2.6%). The remaining patients had M1 morphology (23.8%), M2 (19%), M4 (9.5%), M5 (9.5%), M6 (4.8%) and M7 (4.8%) morphology. Eighteen of the 22 patients (81.8%) had a complete response (CR) to induction chemotherapy. The 5-year event free survival (EFS) from the time of diagnosis for these 22 patients was 27% (±19%) and the 5-year overall survival (OS) was 32% (±20%). The 5-year EFS and OS for the patients on these studies without -5/5q- were 41% (±2%) and 51% (±2%) respectively. The 5-year EFS and OS rates were not significantly different between the two groups (p=0.182 and 0.120 respectively). For the 18 patients who achieved a CR with induction chemotherapy, from time of CR the 5-year disease free survival (DFS) was 33% (±22%), the 5-year OS was 39% (±25%), the 5-year relapse risk (RR) was 61% (±23%) and the 5-year treatment related mortality (TRM) was 6% (±11%). For the 1674 patients without -5/5q- who obtained a CR after induction, from the time of CR the 5-year DFS was 47% (±2%), OS was 57% (±2%), RR was 44% (±3%) and TRM was 9% (±5%). None of these values were significantly different between the groups (p>0.1).

In this, the largest retrospective study of pediatric patients with AML and -5/5q- to date, this cytogenetic subgroup was found to have a poor outcome. The median 5-year overall survival across studies was 32%, and the median 5-year event free survival was 27%. These findings support the use of more aggressive therapy for the treatment of children with -5/5q- AML, as has been previously supported based on data from adults with -5/5q- AML. This subset of patients may also benefit from treatment with innovative agents.

No relevant conflicts of interest to declare.