AFFILIATIONS
1 MDS Research Group, Josep Carreras Leukaemia Research Institute, ICO-Hospital Germans Trias i Pujol, Badalona, Barcelona, Spain, Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain
2 MDS Research Group, Josep Carreras Leukaemia Research Institute, ICO-Hospital Germans Trias i Pujol, Badalona, Barcelona, Spain
3 MDS Research Group, Josep Carreras Leukaemia Research Institute, ICO-Hospital Germans Trias i Pujol, Badalona, Barcelona, Spain, Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain
4 Hematology and Oncology Program, CIMA Universidad de Navarra, Instituto de Investigaciones Sanitarias de Navarra (IdiSNA), Cancer Center Clinica Universidad de Navarra (CCUN), Pamplona, Navarra, Spain
5 Hematology and Oncology Program, CIMA Universidad de Navarra, Instituto de Investigaciones Sanitarias de Navarra (IdiSNA), Cancer Center Clinica Universidad de Navarra (CCUN), Pamplona, Navarra, Spain
6 MDS Research Group, Josep Carreras Leukaemia Research Institute, ICO-Hospital Germans Trias i Pujol, Badalona, Barcelona, Spain
7 MDS Research Group, Josep Carreras Leukaemia Research Institute, ICO-Hospital Germans Trias i Pujol, Badalona, Barcelona, Spain
8 Hematology Department, Hospital Germans Trias i Pujol-Catalan Institute of Oncology, Badalona, Spain
9 Hematology Department, Hospital Germans Trias i Pujol-Catalan Institute of Oncology, Badalona, Spain
10 Hospital Universitario la Paz, Madrid, Madrid, Madrid, Spain
11 Hospital Universitario la Paz, Madrid, Madrid, Madrid, Spain
12 Hematology Research Group, Instituto de Investigación Sanitaria La Fe, Valencia, Spain
13 Haematology Service and Genetics Unit, Hospital Universitario y Politécnico La Fe, Valencia, Spain
14 Centro de Investigación Biomédica en Red de Cáncer (CIBERONC), Madrid, Spain, Haematology Service and Genetics Unit, Hospital Universitario y Politécnico La Fe, Valencia, Spain
15 Institut d'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Barcelona, Spain
16 Experimental Hematology, Vall d’Hebrón Institute of Oncology (VHIO), Barcelona, Barcelona, Spain
17 Department of Hematology, Vall d'Hebron Institute of Oncology, University Hospital Vall d'Hebron, Barcelona, Spain, Experimental Hematology, Vall d’Hebrón Institute of Oncology (VHIO), Barcelona, Barcelona, Spain
18 Department of Hematology, Vall d'Hebron Institute of Oncology, University Hospital Vall d'Hebron, Barcelona, Spain, Experimental Hematology, Vall d’Hebrón Institute of Oncology (VHIO), Barcelona, Barcelona, Spain
19 MDS Research Group, Josep Carreras Leukaemia Research Institute, ICO-Hospital Germans Trias i Pujol, Badalona, Barcelona, Spain, Microarrays Unit, Josep Carreras Leukaemia Research Institute, Badalona, Spain
20 Immunohematology and Glycobiology Group, Josep Carreras Leukaemia Research Institute, Badalona, Barcelona, Spain, MDS Research Group, Josep Carreras Leukaemia Research Institute, ICO-Hospital Germans Trias i Pujol, Badalona, Barcelona, Spain
21 MDS Research Group, Josep Carreras Leukaemia Research Institute, ICO-Hospital Germans Trias i Pujol, Badalona, Barcelona, Spain
22 Hematology and Oncology Program, CIMA Universidad de Navarra, Instituto de Investigaciones Sanitarias de Navarra (IdiSNA), Cancer Center Clinica Universidad de Navarra (CCUN), Pamplona, Navarra, Spain
23 Department of Hematology, Vall d'Hebron Institute of Oncology, University Hospital Vall d'Hebron, Barcelona, Spain, Experimental Hematology, Vall d’Hebrón Institute of Oncology (VHIO), Barcelona, Barcelona, Spain
24 IBSAL, Instituto de Investigación Biomédica de Salamanca, Salamanca, Salamanca, Spain
25 Department of Hematology, Hospital Clinic, Barcelona, Barcelona, Spain, Institut d'Investigacions Biomèdiques August Pi i Sunyer (IDIBAPS), Barcelona, Barcelona, Spain
26 MDS Research Group, Josep Carreras Leukaemia Research Institute, ICO-Hospital Germans Trias i Pujol, Badalona, Barcelona, Spain, Universitat Autònoma de Barcelona, Bellaterra, Barcelona, Spain
ABSTRACT
Germline predisposition is a recognized contributor to myelodysplastic syndromes (MDS), particularly in patients with early-onset disease or a suggestive family history. Accurate identification of these cases is relevant for prognosis, treatment decisions, donor selection for allogeneic hematopoietic stem cell transplantation, and genetic counselling. However, current diagnostic strategies based on targeted sequencing restricted to coding regions of predefined predisposition genes show limited diagnostic yield, potentially missing novel or non-canonical germline variation and contributing to heterogeneous diagnostic practices. To address these limitations, we characterized the germline mutational landscape of patients with clinical features suggestive of germline predisposition using whole-exome sequencing (WES) and investigated whether transcriptomic profiling by RNA sequencing (RNA-seq) could identify molecular signatures associated with germline predisposition and reveal biologically relevant alterations in unexplained cases.
A total of 81 patients were analyzed, including 68 diagnosed before the age of 60 years (yMDS) and 13 patients with a family history of hematologic malignancies (fMDS). Paired tumor/germline WES was performed in 81 patients using bone marrow and matched CD3+ T cells or hair follicle-derived DNA. WES was performed at 300× and 100× coverage in tumor and control tissue, respectively, using the Illumina platform. Only pathogenic or likely pathogenic variants were considered for further analysis. RNA-seq was performed on 34 CD34+ samples, using an Illumina stranded total RNA protocol with 40 million median reads/sample. Differential expression analysis was conducted using DESeq2.
Pathogenic or likely pathogenic germline variants in genes associated with genetic predisposition to myeloid neoplasms (gMDS) or bone marrow failure syndromes (BMF) were identified in 30.9% (21/68) of yMDS patients and 30.7% (4/13) of fMDS patients. DDX41 was the most frequently mutated gMDS gene, while BMF-related variants were mainly found in RTEL1 and SBDS. An additional 13.2% (9/68) of yMDS and 30.7% (4/13) of fMDS patients carried germline pathogenic mutations in SEC23B, JAK2, TET2 and CHEK2, which are myeloid-related genes with emerging or less-established links to germline myeloid predisposition (egMDS). Across both cohorts, 53.1% (43/81) of patients carried no germline pathogenic mutation detectable by WES.
Stratification by germline background revealed differences in co-occurring somatic alterations, somatic mutation burden and karyotype complexity. gMDS patients showed recurrent somatic alterations in DDX41, BMF patients in TP53 and egMDS patients in ASXL1 and RUNX1. Germline-negative patients displayed a more heterogeneous somatic profile with scattered alterations in SF3B1, DNMT3A, and SETBP1. Compared with germline-positive patients, they exhibited a lower somatic mutation burden (2.03 vs. 1.40 mutations per patient) and were overrepresented in the low-risk IPSS-R category and among cases with non-complex karyotypes.
To further characterize the germline-negative patients, total RNA-seq was performed and sex at birth included as a covariate after being identified as a major driver of transcriptional variance. Unsupervised clustering did not segregate patients by germline status: germline-negative patients were distributed across all transcriptional clusters, intermixing with BMF, gMDS, and egMDS samples. Each cluster nonetheless displayed distinct, biologically coherent expression programs. Notably, nearest-neighbor analysis of the sex-corrected expression profile showed that 30.4% (7/23) of germline-negative patients clustered transcriptionally closest to a confirmed germline case. This suggests that a subset of genetically unexplained cases share expression features with confirmed germline predisposition, consistent with either convergent disease biology or unrecognized germline predisposition.
Integrative genomic and transcriptomic profiling in yMDS and fMDS patients reveals a heterogeneous germline landscape, with distinct molecular and clinical features associated with each predisposition group. The transcriptional proximity between a subset of germline-negative patients and confirmed germline cases suggests that transcriptomic profiling may help refine classification in patients where targeted germline sequencing fails to identify a causal variant. Funding: FI-1 00200, PI 23/00007, PI23/00992, PMP 24/00025