AUTHORS
Wencke Walter, PhD, Sophia Prohaska2, Torsten Haferlach, MD, PhD3, Anna Stengel, PhD4
AFFILIATIONS
1 MLL Munich Leukemia Laboratory, Munich, Germany
2 MLL Munich Leukemia Laboratory, Munich, Germany
3 MLL Munich Leukemia Laboratory, Munich, Germany
4 MLL Munich Leukemia Laboratory, Munich, Germany
ABSTRACT
Background
TET2 and IDH1/2 mutations are among the most frequent alterations in myeloid malignancies and are widely considered mutually exclusive due to their convergent roles in epigenetic regulation. Co-occurrence detected in bulk sequencing is typically attributed to independent clones. However, isolated reports and emerging datasets challenge this dogma, raising the possibility that both mutations may co-exist within the same malignant clone.
Aim
To determine the prevalence, clonal architecture, and clinical impact of TET2/IDH1/2 co-mutations across myeloid malignancies and to validate clonal relationships using single-cell proteogenomics.
Methods
All patients were analyzed by targeted panel sequencing (in-house panel, median coverage 1,500x, variant calling threshold 3% VAF). Cases with co-occurring TET2 and IDH1/2 mutations were classified into three clonal architecture categories based on bulk VAF patterns: (i) multiclonal architecture, defined by both VAFs <20%; (ii) putative single-clone co-occurrence, defined by both VAFs ≥20% with a VAF difference ≤15%; and (iii) sequential clonal evolution, comprising all remaining cases, with the higher-VAF mutation designated as the presumed first hit. Selected cases underwent additional single-cell proteogenomic analysis (Takahashi, MissionBio; TotalSeq-D Heme Oncology Panel, BioLegend).
Results
We identified 5,036 patients diagnosed with myeloid neoplasms between 2022 and 2025 who were positive either for TET2 (n=3,826, 76%) or IDH1/2 (n=1,210, 24%). Patients had a median age of 76 years [20-98y] and 62.8% of patients were male. 63.9% of IDH1/2-positive cases were diagnosed with AML, whereas TET2-positive cases were distributed across MDS (31.3%), CMML (20.7%), CCUS/MDS (20.6%), and AML (18.9%), with a smaller proportion of MDS/MPN (6.8%) and MPN (1.8%) cases. In addition, we identified in the same time frame 250 patients with co-occurring TET2 and IDH1/2 mutations: AML (n=88, 34.8%), MDS (n=89, 36.0%), CMML (n=20, 8.1%), CCUS/MDS (n=28, 11.3%), MDS/MPN (n=9, 3.6%), and MPN (n=16, 6.5%). Median age was 75 years [20–92y], and 62.7% of patients were male. Nearly all AML cases were classified as AML-MR (n=61, 69.3%) or AML-NPM1 (n=21, 23.9%), with only 6 cases of other AML subtypes (6.8%) and none harboring additional defining genetic alterations.
Multiclonal architecture represented the most frequent overall pattern (n=91, 36.8%), particularly in CCUS/MDS (71.4%, 20/28) and MDS (43.8%, 39/89), reflecting high clonal diversity at an early disease stage. In line with this, putative single-clone co-occurrence was most frequent in AML (31.8%, 28/88), CMML (35.0%, 7/20), and MDS/MPN (33.3%, 3/9), indicating the outgrowth of this clone due to Darwinian selection in the more advanced stages. Sequential clonal evolution with IDH1/2 as the presumed first hit accounted for 20.2% of cases (n=50) and was most prevalent in AML (28.4%), whereas TET2-first evolution accounted for 19.4% (n=48) and was enriched in CMML (35.0%) and MDS (20.2%).
Single-cell proteogenomic analysis in 11 cases confirmed intra-clonal co-occurrence of TET2 and IDH1/2 mutations in 5 cases (1 CCUS/MDS, 2 AML-MR, 2 AML-NPM1), directly demonstrating that these mutations can reside within the same malignant clone. In the remaining 6 cases, single-cell analysis identified independent clones. Most of these cases showed one dominant mutation (VAF >20%) accompanied by a low-level second mutation (VAF <10%), supporting the concept that many TET2/IDH1/2 co-mutations detected by bulk sequencing reflect independent subclones. Notably, one MDS-IB2 case with comparable VAFs (IDH2 5.1%, TET2 7.2%) still demonstrated independent clonal architecture by single-cell analysis, supporting the validity of the applied <20% VAF threshold for multiclonal classification.
Conclusions
The distribution of clonal architectures reflects a progressive reduction in clonal diversity along the CCUS/MDS–MDS–AML continuum. In early disease, multiclonal co-existence of TET2- and IDH1/2-mutant clones may foster active clonal competition and extinction, maintaining a relatively stable pre-malignant equilibrium. Selective elimination of the IDH1/2-mutant clone by early IDH-inhibitor intervention risks disrupting this equilibrium, potentially providing a selective advantage to co-existing, more aggressive clones that would otherwise have been outcompeted. These findings caution against early intervention without prior clonal architecture assessment.
DISCLOSURES
Wencke Walter, PhD: MLL Munich Leukemia Laboratory - Current Employment