Genotyping Offers Insights Into Key Characteristics of ALL Subtypes

Charles G. Mullighan, MBBS (Hons), MSc, MD

Genomic sequencing is a critical step in informing the prognosis of patients with acute lymphoblastic leukemia (ALL), said Charles G. Mullighan, MBBS (Hons), MSc, MD, in a plenary presentation during the 25th Annual International Congress on Hematologic Malignancies®: Focus on Leukemias, Lymphomas, and Myeloma, an event hosted by Physician’s Education Resource®, LCC.¹ He added that more information regarding specific subgroups of ALL, such as lineage ambiguous ALL, could pave the way for more personalized therapies for patients.

“Subtype-defining [alterations] are diverse in leukemia, both in the genes involved and the nature of the alterations,” Mullighan said. “The complexity and diversity really narrow sequencing-based diagnostics and are strongly associated with prognosis and therapeutic opportunities. [The] genotype is the defining feature of subtypes rather than older features like [the] immunophenotype.”

Significant advances have been made to identify and classify subtype-defining alterations in ALL, particularly B-cell ALL (B-ALL). These alterations are among the most powerful predictors of outcomes for this patient population, explained Mullighan, the deputy director of the Comprehensive Cancer Center, co-leader of the Hematological Malignancies Program within the Department of Pathology, medical director of the St. Jude Biorepository, and the William E. Evans Endowed Chair at St. Jude Children’s Research Hospital.

“[Children] generally have good outcomes, although when [ALL] relapses, outcomes are worse [and we have] difficulty in inducing remission. With increasing age, outcomes and prognosis decline. We now know that is due to underlying genomics, as well as treatment-related factors,” said Mullighan.

A key study that was published in Nature Genetics demonstrated the utility of transcriptome sequencing to classify B-ALL subtypes, as well as to inform potential targeted therapies.²

A T-distribution stochastic neighbor embedding model, a statistical method for visualizing data, of 1988 children (n = 1402), adolescent and young adults (AYA; n = 208), and adults (n = 378) with B-ALL revealed 23 distinct and clustered subtypes of ALL, some of which were not previously identified. These subtypes were defined by chromosomal rearrangements, sequencing mutations, or heterogenous genomic alterations.

Alterations in transcription factor genes such as DUX4, ZNF384, and MEF2D were commonly observed among patients. Additionally, rearrangements in BCL-2/MYC were identified and indicative of high-risk disease, Mullighan explained.

Also, single point mutations in genes, such as PAX5, were identified. Moreover, PAX5 p.Pro80Arg mutations were found to impair B-lymphoid development and promote B-ALL development with biallelic PAX5 alterations in vivo, confirming that PAX5 plays a key role as a checkpoint of B-lymphoid maturation and leukemogenesis.

The study also validated that age is key factor in ALL with regard to genomics. For example, children under the age of 15 had a substantially higher incidence of transcription factor rearrangements compared with AYA or adult patients. Conversely, the latter 2 populations had higher proportions of kinase-driven alterations vs children.

Taken collectively, these findings reinforce the importance of genomic testing for patients with B-ALL, said Mullighan. These alterations can be unidentifiable by cytogenetics and polygenic in nature.

As genomic subtyping informs risk, it can also inform prognostic implications for treatment, potential treatment deintensification, and the optimal use of transplant among patients with ALL.

“Although it is relatively early as we scramble experimentally to validate these oncogenes and lesions, we are finding that some of [these alterations] appear to have their own targeted approaches,” Mullighan said.

The Ph-positive or Ph-like subtype of ALL remains the most genetically diverse and an important unmet need, added Mullighan. The subtype encompasses around 60 mutations targeting kinase signaling pathways or genes. Although Ph-like alterations are observed across all age groups, they appear to be most common in AYA and adult populations.

Currently, targeted therapies directed toward kinases such as ABL1/2, PDGFRA, JAK, and others, have demonstrated utility in Ph-like ALL. However, second-generation targeted approaches, as well as venetoclax (Venclexta) may add to the paradigm, said Mullighan.

Another area of significant need and ongoing research is lineage ambiguous ALL, particularly within the subgroups of ALL that appears to arise from a hematopoietic stem cell instead of a committed progenitor.

Currently, the classification of these subgroups remains uncertain, but includes early T-cell precursor leukemia (ETP-ALL) and acute leukemias of ambiguous lineage, such as mixed phenotype acute leukemia (MPAL) and acute undifferentiated leukemia (AUL). Often, these subtypes have varying classification based on immunophenotypic criteria rather than genomic criteria, and are distinguished by a single marker, such as the enzyme myeloperoxidase (MPO).

“[MPO] can be one of the most robustly difficult markers to [detect] in the lab, and it is just the single [marker] that is distinguishing these groups. However, it has profound therapeutic implications,” said Mullighan.

Moreover, the majority of these cases do not have defined genotypes, such as Ph-positive rearrangements, and the lineage may shift over time, Mullighan added.

However, data from the study published in Nature Genetics reported that patients with ALL who harbor ZNF384 alterations can have conventional B-ALL or B-myeloid MPAL.² Moreover, the genetic landscapes of these subtypes appear similar, suggesting that all ZNF384-altered ALL is essentially the same disease, Mullighan explained.

During the 2020 ASH Annual Meeting and Exposition, findings from a large-scale analysis of transcriptome sequencing samples of 2573 pediatric and adult patients with B-ALL (n = 1411), acute myeloid leukemia (n = 262), MPAL (n = 126), and T-ALL (n = 774) were presented virtually.³ The study identified a new subset of patients with a distinct gene expression profile and immunophenotype (typically, cCD3+ CD7+ CD1a- CD2+ CD5- CD8- cMPO+/- and myeloid/stem cell marker positive). Of the 55 patients who had available data, 25 (45.5%) had T/myeloid-MPAL, 20 (36.4%) had ETP-ALL, 8 (14.5%) had AML, and 2 (3.6%) had AUL.

Notably, all patients with available data exhibited monoallelic expression of BCL11B with noncoding structural variants involving BCL11B locus and 80% harbored FLT3 alterations.

These findings, as well as those from a validation cohort of 91 adult patients with T-ALL, confirmed that this BCL11B immunophenotypic subgroup account for about 30% of all T/myeloid-MPAL and ETP-ALL.¹

Moreover, it was revealed that the structural rearrangement breakpoints occur upstream or downstream of BCL11B and involve mutations on at least 8 different chromosomes, including ARID1B on chromosome 6 (37.7%), tandem amplification on chromosome 14 (21.3%), CCDC26 on chromosome 8 (13.1%), CDK6 on chromosome 7 (4.9%), ETV6 on chromosome 12 (1.6%), SATB1 on chromosome 3 (1.6%), RUNX1-BCL11B on chromosome 21 (3.3%), and ZEB2-BCL11B on chromosome 2 (6.6%).^1,3

Additionally, chromatin topology analysis demonstrated enhancer hijacking of ectopic BCL11B activation in a primitive stem cell/progenitor cell. As such, a hematopoietic stem cell represents the cell or origin for T/myeloid-ALL.

Taken collectively, these findings suggest that minimal residual disease–adapted therapies, TKIs, FLT3 inhibitors, immunotherapy, BH3 mimetics, and MLL-directed therapies could provide new therapeutic options for patients with lineage ambiguous leukemia, concluded Mullighan.

References

1. Mullighan CG. New concepts in resistance to therapy in acute lymphoblastic leukemia. Presented at: 25th Annual International Congress on Hematologic Malignancies^®: Focus on Leukemias, Lymphomas, and Myeloma; February 25-28, 2021; Virtual.
2. Gu Z, Churchman ML, Roberts KG, et al. PAX5-driven subtypes of B-progenitor acute lymphoblastic leukemia. Nat Genet. 2019;51(2):296-307. doi:10.1038/s41588-018-0315-5
3. Montefiori L, Seliger S, Gu Z, et al. Enhancer hijacking of BCL11B defines a subtype of lineage ambiguous acute leukemia. Blood. 2020;136(suppl 2):LBA-3. doi:10.1182/blood-2020-143881