Background
The mutations in KMT2D gene are well known to be associated with Kabuki syndrome. Up to date, 1,027 different germline KMT2D variants have been reported in ClinVar (https://www.ncbi.nlm.nih.gov/clinvar/?term=KMT2D%5Bgene%5D), and among them, 365 variants (35.5%) are classified as uncertain significance and 53 (5.2%) are assigned to conflicting interpretations. This means about 40% of KMTD2 variants still need further evaluation to figure out whether they are 'pathogenic' or 'likely pathogenic'. Currently, the problem with the respect to the interpretation of variants is that they are prone to be reported as variants of unknown significance (VUS) without further study if the KMT2D variants have not been clinically described and reported previously. And in the clinical field, the KMT2D variants interpreted as VUS most likely do not get paid attention by the clinicians and can be easily ignored without further testing. Recently, Hadjadj et al.1 brought up the criteria 'probably pathogenic' variants which include the minor allele frequency less than 0.01 in the general population, and in Silico algorithms predicts the pathogenicity of missense variants. The authors found these 'probably pathogenic' variants were not significantly different from 'pathogenic' mutations in Evans syndrome in terms of clinical manifestations.
Hypothesis and objective
We hypothesized that there could be 'probably pathogenic' KMT2D variants which have been reported as VUS in patients with hematologic cytopenias and primary immune deficiency and we sought to find these 'probably pathogenic' variants for further evaluation to figure out their pathogenicity.
Methods
We analyzed the next generation sequencing data performed on patients with hematologic cytopenias and primary immune deficiency at the Ann and Robert H. Lurie Children's Hospital from March 2017 to May 2019. This primary immune deficiency (PID) panel analyzes 290 genes and the clinical manifestations of the patients were compared with the previous reports published regarding the KMT2D pathogenic mutations. We reviewed not only the online PubMed resource, but the websites well known for gene analysis were also used as references (https://www.ncbi.nlm.nih.gov/clinvar/, http://exac.broadinstitute.org/, http://genetics.bwh.harvard.edu/pph2/, http://provean.jcvi.org/index.php, https://sift.bii.a-star.edu.sg/, etc).
Results
Among 52 patients who had a PID panel ordered, 10 patients (19.2%) were found to have KMT2D variants. Among them, 4 variants (7.6%) were polymorphisms (c.7705G>A, p.Gly2569Ser; c.11849A>G, p.Gln3950Arg; c.1408C>T, p.Pro470Ser; c.2506C>A, p.Ile238Val) and 6 (11.5%) 'probably pathogenic' KMT2D variants were newly identified. Novel heterozygous KMT2D variants, previously reported as VUS: c.15341A>C, p.His5114Pro; c.10640G>A, p.Arg3547; c.6902C>T, p.Pro2301Leu; c.7328G>T, p.Arg2443Leu; c.15694A>G, p.I5232V; c.7001G>A, p.R2334Q were identified in patients who presented with various clinical manifestations known to be associated with KMT2D mutations which are neutropenia, anemia, thrombocytopenia, cardiac anomaly, urogenital anomaly, primary immune deficiency, variable degrees of mental retardation and distinctive facial features that include arched eyebrows, long eyelashes, elongated eyelids with lower lids that turn out, prominent ears, a flat tip of the nose, and a downward slant to the mouth. Further parental testing revealed one of them was a novel de novo mutation which was reclassified as 'likely pathogenic', 4 variants were inherited from mother or father and parental testing was not done for 1 patient.
Conclusion
We concluded that the mutational gene analysis should be correlated with patients' clinical manifestations. The second look analysis of the 'probably pathogenic' variants reported as VUS to figure out the clinical meaning would be important in terms of diagnosis and treatment of patients in the future. Further evaluation regarding family history and gene analysis would be necessary as well as further downstream pathway studies to confirm the pathogenicity of the 'probably pathogenic' KMT2D variants.
References
1. Jérôme Hadjadj, Nathalie Aladjidi, Helder Fernandes, et al. Pediatric Evans syndrome is associated with a high frequency of potentially damaging variants in immune genes, Blood 2019 134:9-21.
Thompson:Baxalta: Research Funding; Novartis: Consultancy, Research Funding; Celgene: Consultancy, Research Funding; bluebird bio, Inc.: Consultancy, Research Funding.
Author notes
Asterisk with author names denotes non-ASH members.