Purpose of Review: Acute myeloid leukemia (AML) in children remains a challenging disease to cure with suboptimal outcomes particularly when compared to the more common lymphoid leukemias. Recent advances in the genetic characterization of AML have enhanced understanding of individualized patient risk, which has also led to the development of new therapeutic strategies. Here, we review key cytogenetic and molecular features of pediatric AML and how new therapies are being used to improve outcomes. Recent Findings: Recent studies have revealed an increasing number of mutations, including WT1, CBFA2T3-GLIS2, and KAT6A fusions, DEK-NUP214 and NUP98 fusions, and specific KMT2A rearrangements, which are associated with poor outcomes. However, outcomes are starting to improve with the addition of therapies such as gemtuzumab ozogamicin and FLT3 inhibitors, initially developed in adult AML. Summary: The combination of advanced risk stratification and ongoing improvements and innovations in treatment strategy will undoubtedly lead to better outcomes for children with AML.

Acute promyelocytic leukemia (APL) is a rare disease accounting for only 5%–10% of pediatric acute myeloid leukemia (AML) and fewer than 1000 cases occur annually in the United States across all age groups. Characterized by t (15; 17), with a resultant PML-RARA gene fusion driving leukemia development, advances in therapy have improved outcomes for APL significantly in the past several decades, now making APL the most curable form of AML in both children and adults. Cure rates in APL are now comparable to pediatric B-lymphoid leukemias. The success of APL treatment is due, in part, to the breadth of understanding of the driver PML-RARA mutation as well as collaborative efforts to quickly introduce and maximize the benefit of new therapies. Here, we review the presentation, clinical features, pathogenesis, and treatment advances in pediatric APL.

Black and Hispanic children with acute myeloid leukemia (AML) have worse outcomes compared with White children. AML is a heterogeneous disease with numerous genetic subtypes in which these disparities have not been specifically investigated. In this study, we used the Therapeutically Applicable Research to Generate Effective Treatments (TARGET) database to examine the association of race-ethnicity with leukemia cytogenetics, clinical features, and survival outcomes within major cytogenetic subgroups of pediatric AML. Compared with White non-Hispanic patients, t(8;21) AML was more prevalent among Black (odds ratio [OR], 2.22; 95% confidence interval [CI], 1.28-3.74) and Hispanic patients (OR, 1.74; 95% CI, 1.05-2.83). The poor prognosis KMT2A rearrangement t(6;11)(q27;q23) was more prevalent among Black patients (OR, 6.12; 95% CI, 1.81-21.59). Among those with KMT2Ar AML, Black race was associated with inferior event-free survival (EFS) (hazard ratio [HR], 2.31; 95% CI, 1.41-3.79) and overall survival (OS) (HR, 2.54; 1.43-4.51). Hispanic patients with KMT2Ar AML also had inferior EFS (HR, 2.20; 95% CI, 1.27-3.80) and OS (HR, 2.07; 95% CI, 1.09-3.93). Similarly, among patients with t(8;21) or inv(16) AML (ie, core-binding factor [CBF] AML), Black patients had inferior outcomes (EFS HR, 1.93; 95% CI, 1.14-3.28 and OS HR, 3.24; 95% CI, 1.60-6.57). This disparity was not detected among patients receiving gemtuzumab ozogamicin (GO). In conclusion, racial-ethnic disparities in survival outcomes among young people with AML are prominent and vary across cytogenetic subclasses. Future studies should explore the socioeconomic and biologic determinants of these disparities.

IKAROS family zinc finger 1 (IKZF1) alterations represent a diverse group of genetic lesions that are associated with an increased risk of relapse in B-cell acute lymphoblastic leukemia. Due to the heterogeneity of concomitant lesions, it remains unclear how IKZF1 abnormalities directly affect cell function and therapy resistance, and whether their consideration as a prognostic indicator is valuable in improving outcome. CRISPR/Cas9 strategies were used to engineer multiple panels of isogeneic lymphoid leukemia cell lines with a spectrum of IKZF1 lesions to measure changes in chemosensitivity, gene expression, cell cycle, and in vivo engraftment that can be linked to loss of IKAROS protein. IKZF1 knockout and heterozygous null cells displayed relative resistance to a number of common therapies for B-cell acute lymphoblastic leukemia, including dexamethasone, asparaginase, and daunorubicin. Transcription profiling revealed a stem/myeloid cell–like phenotype and JAK/STAT upregulation after IKAROS loss. A CRISPR homology-directed repair strategy was also used to knock-in the dominant-negative IK6 isoform into the endogenous locus, and a similar drug resistance profile, with the exception of retained dexamethasone sensitivity, was observed. Interestingly, IKZF1 knockout and IK6 knock-in cells both have significantly increased sensitivity to cytarabine, likely owing to marked downregulation of SAMHD1 after IKZF1 knockout. Both types of IKZF1 lesions decreased the survival time of xenograft mice, with higher numbers of circulating blasts and increased organ infiltration. Given these findings, exact specification of IKZF1 status in patients may be a beneficial addition to risk stratification and could inform therapy.

Cohesin gene mutations occur in many malignancies, including acute myeloid leukemia (AML). Loss-of-function mutations in the four major cohesin complex genes (RAD21, SMC3, SMC1a, and STAG2) occur across most major genetic subtypes of AML but are notably absent in AML harboring CBFB::MYH11, suggesting that cohesin mutations yield distinct biological outcomes dependent on the genetic AML driver. We hypothesized that CBFB::MYH11-expressing leukemias would be dependent on intact cohesin genes given their near-mutual exclusivity. To investigate this, we combined either germline or inducible heterozygous deletions in cohesin genes Smc3 or Rad21, respectively, with an inducible murine model of Cbfb::MYH11 AML. This approach allowed us to evaluate the effects of cohesin haploinsufficiency on leukemia development, chromatin accessibility, and transcriptional output. We demonstrated that intact cohesin function is dispensable for Cbfb::MYH11-driven leukemia. Instead, Cbfb::MYH11 expression is the primary driver of the transcriptional program in transformed leukemic cells. Furthermore, we observed differential effects of Rad21 and Smc3 deletion on leukemia development and secondary engraftment despite only minor differences in gene expression. These results demonstrate that cohesin mutations are not only tolerated in Cbfb::MYH11-expressing cells, but they also likely do not confer a strong selective advantage and are therefore not preferentially selected for during clonal evolution of this leukemia.