Mutations in the UNC13D gene lead to functional defects in cytotoxic lymphocytes (CL) and cause familial hemophagocytic lymphohistiocytosis type 3 (FHL3).1,2 A proportion of patients with lymphoma have been reported to harbor mutations in the PRF1, UNC13D, STX11, STXBP2 or SH2D1A genes,3 which cause functional defects in CL, and recent studies4,5 have shown that defects in these FHL-related genes may increase the risk of lymphoma and Epstein-Barr virus (EBV)-associated T/NK-cell lymphoproliferative disorders, indicating that genetic defects in CL may increase susceptibility to lymphomagenesis. Data regarding the association between genetic defects and the development of lymphoma in pediatric patients are limited to date. Based on emerging evidence that a certain relationship exists between genetic defects in FHL-related genes and lymphoma pathogenesis, in this retrospective analysis, we comprehensively reported the clinical characteristics of 54 pediatric lymphoma patients as well as their genetic features.
A total of 680 children were diagnosed with lymphoma at Beijing Children’s Hospital from January 2020 to January 2024, 54 of whom carried UNC13D gene mutations. Pathology samples were reviewed and classified based on the World Health Organization (WHO) guidelines.6 Staging was based on clinical evaluation and was defined by the Ann Arbor or St. Jude staging system. Patients were divided into three groups according to the different risk factors and responses to treatment: low-risk, intermediate-risk, and high-risk. Medical records, including clinical features and genetic findings, were retrieved. Hemophagocytic lymphohistiocytosis (HLH) was diagnosed based on HLH-2004 guidelines.7 Genomic DNA was isolated from peripheral blood mononuclear cells for whole-exome sequencing, and exfoliated cells of the oral mucosa were matched to determine the germline origin of the variants. Classification of variants were performed per the American College of Medical Genetics and Genomics (ACMG)8 and the Association for Molecular Pathology (AMP) criteria. Severe cases were defined as patients having been diagnosed with HLH or having manifestations of high tumor burdens, or when patients have superior vena cava compression symptoms or airway obstruction, or symptoms of central nervous system (CNS) invasion or compression. This study has been approved by Institutional Review Board of Beijing Children’s Hospital, and written informed consent was obtained from the patients’ guardians.
Figure 1 and Online Supplementary Table S1 show the clinical characteristics of the patients in our study group. The male-to-female ratio was 1.7:1. The median age at clinical diagnosis was 9 years (range, 1-14 years). Thirteen patients were diagnosed with Hodgkin lymphoma (HL), and 41 patients were diagnosed with non-Hodgkin lymphoma (NHL). A total of 40 patients (74.1%) were diagnosed at stage III and IV, with severe cases accounting for 22.2% of the patients. HLH was diagnosed in nine patients, among which seven cases were concurrent at the time of diagnosis, while another two patients developed secondary HLH during disease progression or relapse. Analysis revealed that patients carrying the UNC13D gene with pathogenic/likely pathogenic variants were correlated with the severity of their clinical features and the occurrence of HLH (x²=6.943; P=0.008; x²=11.76; P=0.001). However, no correlations were found with mortality or clinical disease staging. The follow-up times of the two groups with and without UNC13D mutations (54 patients vs. 626 patients) were 31 (range, 1-60) months and 38 (range, 1-60) months, respectively. The 2-year overall survival (OS) rates of the two groups were 96.3% and 99.5%, respectively, while the 2-year event-free survival (EFS) rates were 83.3% and 96.3%, respectively, with statistically significant differences (P=0.0076; P<0.0001). Furthermore, by further analyzing the pathogenic/likely pathogenic (P/LP) variants group and uncertain significance (VUS) variants group separately against the wild-type (WT) group, we found that the prognosis of the P/LP group was significantly worse than that of the WT group in terms of both EFS and OS (Figure 2).
Our study revealed that approximately 7.9% of patients carried the UNC13D gene mutations. All mutations were confirmed to be germline-derived. A total of 34 different UNC13D (NM_199242.2) mutations were identified in 54 unrelated patients (Online Supplementary Table S1), including 44 with monoallelic mutations, three with homozygous mutations and seven with compound heterozygous mutations. The UNC13D gene contains 32 exons and 31 introns.1 In this group of patients, 77.7% and 11.1% of the UNC13D gene mutations occurred in exons and introns, respectively; 9.3% of the patients had mutations in both exons and introns, and one patient (1.9%) had an unclear mutation region. The most prevalent UNC13D mutations were missense mutations (57.4%), followed by synonymous mutations (16.7%), splicing errors (11.1%) and deletion mutations (5.5%), and 7.4% of patients had both a missense mutation and a splicing mutation. The Munc13-4 protein has four domains (C2A, MHD1, MHD2, and C2B) and a region that interacts with RAB27A. Protein domain analysis revealed that the main affected domain of the Munc13-4 protein in lymphoma patients was the RAB27A domain (29.6%), followed by non-domain involvement (22.4%); C2B and MHD2 each accounted for 16.6%, and C2A and MHD1 each accounted for 7.4% (Figure 3). Among the 12 patients with severe cases, two had homozygous mutations (c.2588G>A, c.118-308C>T), three had compound heterozygous mutations, and the remaining patients had heterozygous mutations, including three deletion mutations involving the same site (c.3229_3235del/p.R1077Sfs*48), two splicing mutations in the same intron region (c.2553+5C>G), and two mutations in the same exon region. Data analysis revealed that the biallelic mutation rate of UNC13D in severe cases was 41.7%, which was significantly higher than that in mild cases (11.9%), with a statistically significant difference (P=0.019), which may suggest that UNC13D biallelic mutations in lymphoma patients are associated with severe clinical manifestations.
Figure 1.Distribution of UNC13D mutations and clinical characteristics in pediatric lymphoma patients. The upper part of the graph shows the frequencies of each mutation site in the present study (green), the Lower part displays the main clinical characteristics of patients carrying UNC13D mutations (gray). Each column shows the clinical information of a patient and their type of mutation site. HPS: hemophagocytic syndrome; HSCT: hematopoietic stem cell transplantation; HL: Hodgkin Lymphoma; NHL: non-Hodgkin lymphoma; 0/1 represents a heterozygous genotype, where 1 reference allele and 1 alternate allele are present; 1/1 represents a homozygous genotype for the alternate allele, where both alleles at that locus are the alternate form. Freq.: frequency; EBV: Epstein-Barr virus; CNS: central nervous system; inv: involvement; P: pathogenic; LP: likely pathogenic; VUS: variants of uncertain significance.
In addition, we also identified concomitant mutated genes in pediatric lymphoma patients (Online Supplementary Figure S1), most of which were associated with immunodeficient phenotypes. The most common co-mutated gene in this study was the PRF1 gene (12.9%), followed by BRCA2 (11.1%). Notably, in two patients with the same intronic mutation (c.2553+5C>G) in the UNC13D gene, one experienced clinical recurrence, and the other presented with severe manifestations. This finding also suggests that intronic variants are significant contributors to human disease.
In this study, we first reported the clinical characteristics of pediatric lymphoma patients with different UNC13D gene mutations. A total of 34 different mutations were identified in 54 unrelated patients (13 mutations have been reported, and 21 mutations have not been reported). The majority of mutations detected in the current study were heterozygous missense mutations, which was consistent with previous reports;4,9 this may explain why these patients developed lymphoma later in life rather than having an outbreak of fatal FHL during infancy, which suggests that such monoallelic mutations may contribute to the pathogenesis of the disease. Notably, one patient carried a homozygous disease-causing mutation in intron one of the UNC13D gene (NM_199242:c.118-308C>T), although the initial presentation was an outbreak of fatal hemophagocytic syndrome-like manifestations, the final biopsy pathology revealed EBV-positive diffuse large B-cell lymphoma. Due to the rapid progression of the disease, the patient lost the opportunity for further treatment. Deep intronic variants in UNC13D have also been identified in HLH patients from Europe and North America.10,11 These variants frequently involve the first intron of UNC13D, which appears to serve as a key regulatory region, and one variant, c.118-308C>T, has been shown to disrupt transcription factor binding to a cytotoxic lymphocyte-specific alternative promotor, thereby selectively diminishing UNC13D expression in cytolytic cells.12 It is speculated that the genetic defects in CL may increase susceptibility to lymphomagenesis.
Figure 2.Kaplan-Meier curves for overall survival and event-free survival. (A) Overall survival (OS) of the whole study population without UNC13D mutations/wide-type group (WT, N=626) and the target group (N=54). (B) Event-free survival (EFS) of the WT group and the target group. (C) OS of the WT group and patients carrying pathogenic/likely pathogenic (P/LP) variants (N=9). (D) OS of the WT group and patients carrying variants of uncertain significance (VUS) (N=46). (E) EFS of the WT group and patients carrying P/LP variants. (F) EFS of the WT group and patients carrying VUS.
This study revealed that a subset of pediatric lymphoma patients carried monoallelic missense mutations in the UNC13D gene. Unlike the study by Chen et al.,4 which reported that the variant UNC13D c.2588G>A was the founder mutation for lymphoma in the Chinese population, our study demonstrated a higher prevalence of c.1228A>C/p. I410L compared with c.2588G>A/p. G863D in pediatric lymphoma patients (22.2% vs. 14.8%). Twelve c.1228A>C/p. I410L mutations included one compound heterozygous mutation, 11 monoallelic mutations, and eight c.2588G>A/p. G863D mutations included one homozygous mutation, one compound heterozygous mutation and six monoallelic mutations. This single amino acid substitution occurred in an evolutionarily conserved position and was predicted to be pathogenic via the sorting-intolerant form tolerant (SIFT) sorting algorithm. The mutation site of c.1228A>C/p. I410L occurred in the interaction region with RAB27A, and SIFT predicted that its pathogenicity was tolerable. Previous studies13 have shown that the most prevalent UNC13D mutations in FHL3 patients are splice errors (35%) and missense mutations (20.5%), and the rarest mutations are insertions and deletions (1%). In our study, we found that the most common UNC13D mutation in pediatric lymphoma patients was a missense mutation (57.4%). Protein domain analysis revealed that the main affected domains of the Munc13-4 protein in FHL3 patients were the C2A and C2B domains,13 whereas in pediatric lymphoma patients, the RAB27A interaction domain was the primary domain involved. These findings suggest that the type of mutation in the UNC13D gene and the functional domains of the affected protein may influence the clinical phenotype and disease severity. In this study, biallelic mutations in UNC13D were found in five of 12 severe cases, all of which were associated with hemophagocytic manifestations, which is consistent with previous reports that compound heterozygous mutations in the UNC13D gene may be detrimental to FHL.14,15
UNC13D c.3229_3235del occurred in three children from unrelated families, all of whom exhibited hemophagocytic manifestations. UNC13D c.3229_3235del was classified as likely pathogenic for lymphoma according to the ACMG guidelines. The deletion mutation results in a frame-shift mutation starting from amino acid 1,077 and leads to a premature termination codon in the terminal region. All three patients were co-morbid with HLH, presenting with severe clinical manifestations and poor clinical treatment outcomes. To the best of our knowledge, the recurrent monoallelic deletion mutation c.3229_3235del/p. R1077Sfs*48 inducing hemophagocytosis and developing into lymphoma has been reported for the first time, and its underlying pathogenesis remains to be further explored.
Figure 3.Frequency and type of variants in the UNC13D gene. (A) Type of mutations in pediatric lymphoma patients within exons and introns of the UNC13D gene. (B) Distribution of mutations in pediatric lymphoma patients within exons and introns of the UNC13D gene. (C) Proportion of UNC13D genotype in pediatric lymphoma patients. (D) Proportion of affected domain of UNC13D protein in pediatric lymphoma patients.
Yang et al.16 described a case of peripheral T-cell lymphoma and HLH complicated by multiple germline heterozygous mutations, including HLH-specific variants (UNC13D and CD27), which emphasized that a thorough understanding of both the functional and genetic facets of genetic defects is essential.
In summary, UNC13D gene mutations are relatively common among pediatric patients with lymphoma. We found that the prevalence of c.1228A>C/p. I410L was the highest in pediatric lymphoma patients. Lymphoma patients with the c.3229_3235del/p.R1077Sfs*48 monoallelic deletion mutation all presented with HLH and showed poor clinical treatment outcomes. Patients carrying the UNC13D gene with P/LP variants were correlated with the severity of their clinical features and the occurrence of HLH, furthermore, the prognosis of the P/LP group was significantly worse than that of the WT group in terms of both EFS and OS, and we recommend screening for UNC13D gene mutations in newly diagnosed pediatric lymphoma cases. The findings of this study hopefully will prompt more thorough investigations into the relationship between germline defects in cytotoxic lymphocytes and lymphoma pathogenesis.
Footnotes
- Received March 17, 2025
- Accepted February 20, 2026
Correspondence
Disclosures
No conflicts of interests to disclose.
Contributions
Funding
Funding support was provided by the Special Funding Project for Research on Children’s Clinical Nutrition (2022E17-B4).
Acknowledgments
Thanks goes to all patients for their consent to participate and to our colleagues for data collection. Revised manuscript language editing was provided by Bethune Pediatric Young and the Middle-aged Doctors’ Capacity Enhancement Program.
References
- Stepp SE, Dufourcq-Lagelouse R, Deist FL. Perforin gene defects in familial hemophagocytic lymphohistiocytosis. Science. 1999; 286(5446):1957-1959. Google Scholar
- Feldmann J, Callebaut I, Raposo G. Munc13-4 is essential for cytolytic granules fusion and is mutated in a form of familial hemophagocytic lymphohistiocytosis (FHL3). Cell. 2003; 115(4):461-473. Google Scholar
- MachaczkaM Klimkowska M, Chiang SCC. Development of classical Hodgkin’s lymphoma in an adult with biallelic STXBP2 mutations. Haematologica. 2013; 98(5):760-764. Google Scholar
- Chen X, Zhang Y, Wang F. Germline cytotoxic lymphocytes defective mutations in Chinese patients with lymphoma. Oncol Lett. 2017; 14(5):5249-5256. Google Scholar
- Guan YQ, Shen KF, Yang L. Inherited genetic susceptibility to nonimmunosuppressed Epstein-Barr virus-associated T/NK-cell lymphoproliferative diseases in Chinese patients. Curr Med Sci. 2021; 41(3):482-490. Google Scholar
- World Health Organization (WHO): WHO classification of tumours of haematopoietic and lymphoid tissues. 2008. Google Scholar
- Henter JI, Horne A, Aricó M. HLH-2004: diagnostic and therapeutic guidelines for hemophagocytic lymphohistiocytosis. Pediatr Blood Cancer. 2007; 48(2):124-131. Google Scholar
- Richards S, Aziz N, Bale S. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015; 17(5):405-424. Google Scholar
- Clementi R, Locatelli F, Dupré L. A proportion of patients with lymphoma may harbor mutations of the perforin gene. Blood. 2005; 105(11):4424-4428. Google Scholar
- Qian Y, Johnson JA, Connor JA. The 253-kb inversion and deep intronic mutations in UNC13D are present in North American patients with familial hemophagocytic lymphohistiocytosis 3. Pediatr Blood Cancer. 2014; 61(6):1034-1040. Google Scholar
- Entesarian M, Chiang SCC, Schlums H. Novel deep intronic and missense UNC13D mutations in familial haemophagocytic lymphohistiocytosis type 3. Br J Haematol. 2013; 162(3):415-418. Google Scholar
- Cichocki F, Schlums H, Li HC. Transcriptional regulation of Munc13-4 expression in cytotoxic lymphocytes is disrupted by an intronic mutation associated with a primary immunodeficiency. J Exp Med. 2014; 211(6):1079-1091. Google Scholar
- Amirifar P, Ranjouri MR, Abolhassani H. Clinical, immunological and genetic findings in patients with UNC13D deficiency (FHL3): a systematic review. Pediatr Allergy Immunol. 2021; 32(1):186-197. Google Scholar
- Yoon HS, Kim HJ, Yoo KH. UNC13D is the predominant causative gene with recurrent splicing mutations in Korean patients with familial hemophagocytic lymphohistiocytosis. Haematologica. 2010; 95(4):622-626. Google Scholar
- Meeths M, Chiang SCC, Wood SM. Familial hemophagocytic lymphohistiocytosis type 3 (FHL3) caused by deep intronic mutation and inversion in UNC13D. Blood. 2011; 118(22):5783-5793. Google Scholar
- Yang TT, Chen RR, Zhang MM. Relapsed/refractory peripheral T-cell lymphoma-associated hemophagocytic lymphohistiocytosis with UNC13D and CD27 germline mutations. Cell Transplant. 2024; 33:9636897231221887. Google Scholar
Data Supplements
Figures & Tables
Article Information
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.