Targeted interventions and new risk stratification approaches for the improvement of outcomes in B cell precursor acute lymphoblastic leukemia (BCP-ALL) require an understanding of high-risk driver subtypes, their clinical phenotypes and actionable targets. Currently, over 20 BCP-ALL subtypes have been identified based on genomic driver alterations and corresponding gene expression signatures, in particular in pediatric ALL. This complex molecular landscape is so far only partially defined in adult BCP-ALL patients.
To characterize driver subtypes and their clinical phenotypes we performed transcriptome sequencing (Illumina) on Ph-negative adult BCP-ALL patients (n=376) treated on subsequent pediatric inspired protocols of the German Acute Lymphoblastic Leukemia Study Group (GMALL). Using random forest analyses of a well-defined patient subset with available WES/gene panel and DNA-methylation data, we established a 300-gene classifier which together with fusion calling and hotspot mutation calling reliably allocated samples to driver subtypes.
Subgroup allocation was feasible for n=267 / 308 Ph-negative patients (86.7%) with Ph-like ALL (27.7%) being the most frequently observed subtype followed by DUX4 (14.6%), KMT2A (10.1%), PAX5-plus (9.7%), ZNF384 (9.0%), Near haploid - High hyperdiploid (NH-HeH, 6.0%), Low haploid - Near triploid (LH-NT, 6.0%) and TCF3-PBX1 (5.6%). With frequencies below 5% were observed: PAX5-alt, MEF2D, BCL2/MYC, ETV6-RUNX1, CEBP family member fusions and TCF3-HLF.
The age distribution of these driver subtypes varied across our cohort (median age: 37 years, range: 16 - 81; left panel). Adult patients harboring ETV6-RUNX1 fusions (n=4) were 25 years or younger. NH-HeH, PAX5-plus and DUX4 ALL were also observed predominantly in younger patients (median ages: 24, 24, 30 years respectively; p<0.03) Conversely, LH-NT and KMT2A rearranged ALL frequencies increased in older patients (median ages: 47, 46 years respectively; p<0.04).
As expected, male patients were slightly more represented in the cohort (55% male). However, significant differences were observed in the sex-distribution of molecular subgroups with a predominance of male patients in Ph-like (65.4%; p=0.045) and NH-HeH (82.4% p=0.024) ALL, while female patients were more frequent in DUX4 (60%, p=0.046) or KMT2A (75.5%; p=0.031) ALL. These data implicate that sex-specific host factors favor the selection of ALL driver alterations in adult ALL patients.
We measured MRD by qRT-PCR quantification of patient specific immune gene rearrangement markers in our GMALL central reference lab. Data on MRD status after consolidation I which is the most relevant time-point for identification of high-risk patients in the GMALL protocol, were available in 132 patients (aged 18-55 years; right panel). Ph-like and KMT2A patients showed an unfavorable response to intensive induction with Molecular CR achieved in 60.6% and 61.1% respectively. In ZNF384 ALL patients, only n=5/13 achieved Molecular CR, while n=4 each were MRD-positive or had MRD below the quantitative detection range, suggesting that ZNF384 ALL might be a novel high-risk subgroup in adult BCP-ALL. Gene set enrichment analyses showed comparable patterns of JAK/STAT activation in ZNF384 and Ph-like cases as possible functional underpinning of a shared poor-response phenotype. Actionable genomic alterations were identified in n=13/74 Ph-like patients (17.6%). Out of these, ABL-class fusions as targets of ABL-TKI were identified in n=14 patients (PDGFRB: n=5; ABL1: n=4; CSF1R n=2) and one patient harbored an ETV6-NTRK3 fusion amenable to specific inhibitors.
Our data extent the in-depth characterization of BCP-ALL molecular landscape in adult patients. We observe that sex-specific host factors favor the selection of leukemogenic drivers (male predominance: Ph-like, NH-HeH, female predominance: DUX4, KMT2A). ZNF384 ALL seems to be associated with unfavorable therapy response in adults and shares JAK/STAT activation patterns with Ph-like ALL.
Fiedler:Jazz Pharmaceuticals: Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: travel accomodations; Novartis: Membership on an entity's Board of Directors or advisory committees; Celgene: Membership on an entity's Board of Directors or advisory committees; Pfizer: Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: support in medical writing; Amgen: Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: travel accomodations, support in medical writing, Research Funding; Daiichi Sankyo Oncology: Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: travel accomodations; BerGenBio ASA: Research Funding; AbbVie: Honoraria, Membership on an entity's Board of Directors or advisory committees, Other: support in medical writing; BMS: Honoraria; Gilead: Honoraria; Ariad/Incyte: Membership on an entity's Board of Directors or advisory committees; Servier: Honoraria, Other; Morphosys: Membership on an entity's Board of Directors or advisory committees. Viardot:Novartis: Honoraria, Other: advisory board; Kite/Gilead: Honoraria, Other: advisory board; Roche: Honoraria, Other: advisory board; Amgen: Honoraria, Other: advisory board. Topp:Amgen, Boehringer Ingelheim, KITE, Regeneron, Roche: Research Funding; Amgen, KITE, Novartis, Regeneron, Roche: Consultancy. Goekbuget:Servier: Consultancy, Research Funding; Jazz: Consultancy, Research Funding; Novartis: Consultancy, Research Funding; Amgen: Consultancy, Research Funding; Pfizer: Consultancy, Research Funding; Erytech: Consultancy; Kite: Consultancy; Gilead: Consultancy. Brüggemann:Affimed: Research Funding; Incyte: Consultancy; Celgene: Consultancy; Amgen: Consultancy, Honoraria, Research Funding; Janssen: Consultancy, Honoraria; Roche: Consultancy; Regeneron: Research Funding.
Asterisk with author names denotes non-ASH members.