Background Checkpoint kinase 2 (CHEK2) is a critical mediator of the DNA damage response and plays a role in DNA double strand break-induced apoptosis and cell cycle arrest (Bartek, 2003). Deleterious germline CHEK2 variants (CHEK2germ) have been associated with an increased risk for breast, colon, prostate, thyroid and kidney cancers (Cybulski, 2004). Early studies suggested that germline CHEK2 variants also predispose to myeloid malignancies, including myelodysplastic syndromes (MDS) (Janiszewska, 2018). Recently, CHEK2 was identified as the most frequent germline mutated gene in a cohort of 391 adult patients with AML (Yang, 2022). However, more data is warranted to clarify the role of germline CHEK2 mutations in the development of myeloid malignancies. Here we analyzed the prevalence of germline CHEK2 variants in a large series of unselected patients diagnosed with myeloid malignancies and describe their clinical features. Methods Between February 2018 and June 2022, 2322 consecutive adult patients diagnosed at Saint-Louis hospital, Paris with a myeloid malignancy were included in the study. DNA obtained from bone marrow or peripheral blood cells was subjected to capture-based next-generation sequencing of genes recurrently mutated in myeloid malignancies, including CHEK2. In selected patients, germline mutation testing was performed using Sanger sequencing of DNA obtained from skin fibroblasts or sorted CD3-positive cells from peripheral blood. Results We identified a CHEK2 variant in 38 out of 2322 (1.6%) unrelated patients. We focused on the 24 patients with a variant allele frequency (VAF) equal or higher than 40%, highly suggestive of a germline origin, which was subsequently confirmed in all patients with available germline DNA. Of these, 16 variants (Fig1 A) were considered as pathogenic/likely pathogenic (P/LP) according to the American College of Medical Genetics as well as 3 missense variants which were semi-functional in an in-vivo, yeast-based, functional assay (Delimitsou, 2019). The 5 remaining variants were considered of unknown significance (VUS). Finally, 19 CHEK2germ variants were considered causal for the myeloid malignancy representing a frequency of 0.8% (19/2322). Noteworthy, two out of them (10.5%) had a second somatic hit in CHEK2. (Fig1 B) The median age of the 19 CHEK2germ patients was 75 years (42-87) at diagnosis of the hematological disease. According to the revised WHO classification, there were 7 (37%) MDS-MLD, 2 (11% ) MDS-RS, 4 (21%) MDS-EB, 3 (16%) MPN, 2 (11%) AML, and 1 (5%) ICUS. Karyotype was normal in 7 patients (39%), 8 patients had an isolated abnormality, 3 patients had a complex karyotype and one had cytogenetic failure. Additional driver mutations (Fig 1 B) were identified in most patients (18/19, 95%), however no particular association was noticed. Seven patients (37%) had a personal history of cancer (prostate n=4, breast n=1, thyroid n=1, colon n=1). Five patients (26%) had a family history of hematologic disorders (MPN, n=2, multiple myeloma n=2, ICUS n=1) and 7 patients (37%) of non-hematological cancer (breast n=2, brain n=2, lung n=1, ovary n=1, intestinal n=1). Eleven patients were classified as low-risk MDS according to the IPSS classification with a median survival of 5.6 years. They mainly received ESA (N=2) or lenalidomide (N=2). Among the 3 high-risk MDS patients 2 received hypomethylating agent with a survival of 5 and 1.7 years, the remaining one is still alive after sequential bone marrow transplant. One AML patient is still receiving hypomethylating agent with BCL2-inhibitor in complete remission with a follow up of 1.1 year, while the other AML patient died of relapse after 2 sequential bone marrow transplants. Regarding the 3 MPN patients, 1 is still alive after 1.3 year and the 2 others died 11.5 and 3.8 years after diagnosis. Conclusion This is the largest study addressing the frequency (0.8%) and characteristics of carriers of germline CHEK2 pathogenic variants within patients with myeloid disorders. Their presentation is rather heterogenous with a majority of low-risk MDS. They did not appear to have a more complex karyotype or a poorer prognosis and seems to be less frequent than previously suggested. Compared to more frequent genetic predisposition to myeloid malignancies like DDX41, CHEK2germ hematological malignancies did not seem to represent a specific entity, in line with recent debate on CHEK2germ solid cancer. Figure 1View largeDownload PPTFigure 1View largeDownload PPT Close modal
Read full abstract