Background: Clinical heterogeneity, variable latency, and incomplete penetrance characteristic of inherited myeloid malignancies syndromes (HMMSs) suggest that the acquisition of additional alterations are potential drivers of clonality and are required for myelodysplastic syndrome (MDS)/acute myeloid leukemia (AML) onset. However, despite advances in the identification of germline variants, predicting the clinical implications of clonal abnormalities in these syndromes is still challenging.
Aims: We aim at exploring the putative associations between germline and somatic molecular signatures found in a cohort of consecutive patients diagnosed with HMMS at Hospital La Fe, in Valencia (Spain).
Methods: We examined paired germline/somatic samples from 43 MDS and 75 AML consecutive patients along with 5 patients selected from our historical cohort (4 MDS and 1 GATA2-deficiency). The median age was 56.5 (range 12-81 years). Thirteen cases were considered therapy-related. The germinal samples included 89 fibroblasts cultured from a skin biopsy, 29 CD3+ lymphocytes isolated from peripheral blood (PB), 1 bone marrow in remission, and 5 PB. We analyzed them, through a targeted next-generation sequencing (NGS) using a capture-based custom gene panel of 180 genes to assess the diagnosis of HMMSs, and two commercial panels to investigate somatic alterations. Conventional cytogenetics was available in 118 (96%) patients and NGS-based somatic profiling in 115 (93.5%).
Results: Sixteen (13.5%) patients from the consecutive cohort and 5 patients from the historical cohort presented (likely) pathogenic germline variants in genes responsible for HMMSs. Findings derived from cytogenetics and molecular characterization in these patients are shown in Figure1. High risk karyotype was higher in patients with solid cancer predisposition compared to the rest of patients of the series (71% vs 29%; p=0.02). Somatic NGS analysis found 42 clinically significant variants in genes commonly altered in myeloid neoplasms. The most recurrently somatic mutated genes were RUNX1 (5 variants) and TP53 (4 variants). The mean number of somatic mutations per patient in group associated with platelet disorders and group associated with cancer solid predisposition (3.3 and 3 respectively) was similar to patients with a sporadic neoplasm of the same cohort (2.9). However, DDX41 group and group associated with organ dysfunction presented a lower number of somatic mutations per patient (1). In addition, 30.7% of therapy-related neoplasms in our series presented (likely) pathogenic germline variants in contrast with 15,4% in sporadic neoplasms (p=ns). They showed a 2.25 mean number of somatic changes and a high-risk karyotype in 3 out of 4 cases.
Figure 1. Somatic acquired alterations in patients with inherited myeloid malignancies syndromes. AML/SMD-t: AML/SMD therapy-related. NK: normal karyotype. CK: complex karyotype. Chr: chromosome. Blue boxes represent variants with allele frequency (VAF) >10%. Yellow boxes represent variants with VAF <10%.
Image:
Summary/Conclusion: In our series, 13.5 % of patients had a potential HMMS, being the incidence higher in those with a therapy-related neoplasm (30.7%). Patients with solid cancer predisposition presented a higher risk karyotype. Patients with DDX41 mutations and those HMMSs associated with organ dysfunction presented a lower number of somatic mutations per patient. It is necessary to expand this series to find early detection biomarkers of progression in high-risk families.