Inflammatory bowel disease in a young female patient with a novel de novo TRAF3 frameshift variant responsive to ustekinumab: a case report
Article information
Abstract
Tumor necrosis factor receptor-associated factor 3 (TRAF3) is an anti-inflammatory molecule that negatively regulates the non-canonical nuclear factor-κB pathway. Although TRAF3 haploinsufficiency (TRAF3 HI) can influence innate and adaptive immune cells, its effect on inflammatory bowel disease (IBD) development remains unclear. Here, we report the first case of severe early-onset IBD with a novel TRAF3 variant leading to HI, successfully treated with ustekinumab. A 6-year-old girl with a recurrent parotitis, otitis media, tonsilitis, and atopic dermatitis developed IBD involving the stomach, small intestine, and colon. At diagnosis, the immunoglobulin (Ig)G and IgA levels were relatively high, and lymphocyte subsets showed increased counts of plasmablasts, class-switch recombination B cells, and circulating T-follicular helper cells. Treatment with azathioprine and infliximab failed to maintain remission marked by several relapses accompanied by erythema nodosum and arthritis; however, ustekinumab, an anti-interleukin (IL)-12/23p40 antibody, led to long-term clinical remission, normalizing the Ig level and reducing abnormal lymphocyte counts. Whole-exome sequencing revealed a novel heterozygous mutation in TRAF3 [p.(Pro487Leufs*8)], resulting in TRAF3 under-expression. Our case may highlight the contribution of TRAF3 HI to the development of IBD and provide insights into IBD pathophysiology, suggesting the involvement of the IL-12/23-T-follicular helper cell pathway affected by genetic mutations.
INTRODUCTION
Inflammatory bowel disease (IBD), mainly comprising Crohn’s disease (CD) and ulcerative colitis, is a chronic and idiopathic disorder affecting the gastrointestinal tract, leading to progressive gastrointestinal damage and disability [1]. Recent advances in genetics have revealed monogenic disorders with inborn errors of immunity and congenital epithelial barrier defects, resulting in the development of enterocolitis. These conditions are referred to as monogenic IBD and more than 75 genes have been identified as causatives for monogenic IBD. However, there are still many undiagnosed patients because of mutations in genes that have not yet been reported [2]. Identifying these novel genetic mutations is crucial for advancing our understanding of the pathogenesis of IBD and for developing targeted therapeutic strategies. Moreover, extensive analyses of these monogenic disorders may reveal critical roles of the mutated molecules in the pathogenesis of inflammatory diseases, including IBD [3].
Among tumor necrosis factor receptor-associated factor (TRAF) family members, TRAF3 performs unique functions as signaling adaptor proteins for the tumor necrosis factor (TNF) receptor or Toll-like receptors (TLRs). It is also a constitutive negative regulator of the non-canonical nuclear factor-κB pathway activated by various molecules, including CD40L and TLRs, thus influencing innate and adaptive immune cells [4]. Germline TRAF3 haploinsufficiency (TRAF3 HI) results in recurrent bacterial infections, autoimmune disorders such as Sjogren’s syndrome, B cell lymphoproliferation, and hypergammaglobulinemia [5]. While some cases with TRAF3 loss-of-function (LoF) variants reportedly involved gastrointestinal symptoms suggestive of IBD [5,6], its effect on IBD development remains unclear due to the lack of detailed information on clinical features.
This is the first report of severe early-onset IBD in a girl with a novel heterozygous pathogenic variant of TRAF3 [p.(Pro487-Leufs*8)] who was successfully treated with the anti-interleukin (IL)-12/23p40 monoclonal antibody ustekinumab.
CASE REPORT
A 6-year-old girl with a medical history of recurrent otitis media since the age of 1, parotitis since the age of 4, tonsilitis, and atopic dermatitis was admitted to our hospital because of bloody diarrhea and fever that had persisted for 1 month. The patient had no family history of autoinflammatory or autoimmune diseases. Gastrointestinal endoscopy revealed multiple small ulcerations on the edematous mucosa in her stomach, small intestine, and colon as well as pathological findings of cryptitis, and crypt abscess (Fig. 1A-C), leading to a diagnosis of CD. The patient’s condition was complicated by aseptic osteomyelitis at the right distal tibia and calcaneus, lip erosion, and blisters on the back (Fig. 1D-F). Total and differential leukocyte quantification, immunoglobulin (Ig) assay, lymphocyte analysis, and dihydrorhodamine test showed no findings suggesting primary immunodeficiency. However, immune profiling indicated increased levels of IgG and IgA and decreased counts of CD3+ and CD4+ cells at diagnosis compared with their reference ranges (Table 1). Moreover, the counts of plasmablasts, class-switch recombination B cells, and circulating T-follicular helper (cTfh) cells were relatively high (Table 1).

Clinical characteristics and a novel heterozygous TRAF3 mutation in a patient with early-onset inflammatory bowel disease (IBD). (A) Endoscopic findings of the patient with IBD. Multiple small ulcers on the edematous mucosa in the stomach, small intestine, and colon at diagnosis (upper panels) improved after ustekinumab treatment (lower panels). (B) Findings of small bowel capsule endoscopy. Multiple erosions and small ulcers from the jejunum to the ileum (arrows). (C) Histopathological findings with cryptitis in the colon at diagnosis (hematoxylin and eosin staining, ×200). (D) Oral ulcers and (E) blisters on the back at diagnosis. (F) Positron emission tomography/computed tomography findings of the right osteomyelitis. 18F-FDG PET uptake was observed at the distal tibia and calcaneus. (G) Magnetic resonance imaging finding of parotitis with IBD recurrence. An apple tree pattern compatible with Sjogren’s syndrome manifestation was observed in the right parotid gland. (H) Family pedigree of the patient. Circles indicate female individuals, squares indicate male individuals, and the filled circle indicates our patient. The other family members did not carry the TRAF3 variant. (I) Schematic representation of the TRAF3 protein. (J) Western blot analysis of TRAF3 expression in peripheral blood mononuclear cells and T cells from the patient and control groups (n=2). 18F-FDG PET, 18F-fluorodeoxyglucose positron emission tomography; TRAF3, tumor necrosis factor receptor-associated factor 3; PBMC, peripheral blood mononuclear cell; ACTB, β actin.
Initial treatment with corticosteroids was effective, but subsequent therapies with azathioprine and infliximab, an anti-TNF drug, failed to maintain remission, marked by several relapses accompanied by erythema nodosum and arthritis. Upon IBD relapse, parotitis also recurred. Magnetic resonance imaging findings revealed an apple tree pattern in the right parotid gland, consistent with Sjogren’s syndrome presentation (Fig. 1G). For anti-TNF failure in CD, ustekinumab was introduced at 7 years of age. Although temporary escalation of corticosteroids for recurrence of arthritis was required at 8 years of age, ustekinumab treatment was effective, allowing for the tapering of corticosteroids to 0.02 mg/kg and maintaining the remission of gastrointestinal manifestations and other comorbidities, including aseptic osteomyelitis and arthritis, over 4 years (Fig. 1A lower panel, Table 2). Moreover, the IgG and IgA levels improved to the normal range and the counts of plasmablasts, class-switch recombination B cells, and cTfh cells decreased after ustekinumab treatment (Table 1).

Clinical Features of the Patient in Our Study Compared with Those of Individuals with TRAF3 Haploinsufficiency Reported in a Previous Study
Given that the patient’s early-onset intractable IBD was complicated by aseptic osteomyelitis, lip erosion, and blisters, familial-based whole-exome sequencing was performed to identify the genetic etiology of the disease [9]. For causal variants of the narrowed results, the American College of Medical Genetics/Association for Molecular Pathology (ACMG/AMP) criteria were used [10]. A de novo heterozygous variant was identified on TRAF3 (NM_145725.3, c.1457del [p.(Pro487Leufs*8)], ClinVar accession number: SCV003843246) (Fig. 1H and I).
As this variant is absent in the Genome Aggregation Database, we considered it to be pathogenic based on ACMG/AMP criteria.
To clarify the effect of this variant on TRAF3 protein expression, we analyzed TRAF3 expression in the patient’s peripheral blood mononuclear cells (PBMCs) and T cells using Western blotting. TRAF3 protein expression in PBMCs and T cells was markedly reduced compared with that in controls (Fig. 1J), suggesting TRAF3 HI is related to its variant.
This study was approved by the Institutional Review Board of the National Center for Child Health and Development, Tokyo, Japan (#378, #926). Written informed consent was obtained from the patient’s parents, and assent was obtained from the patient. The detailed methods are presented in the Supplementary Materials.
DISCUSSION
In total, 12 patients with TRAF3 LoF variants in 7 families have been reported (Supplementary Table 1) [5,6]. The manifestations in them were considerably similar to those in our patient, including recurrent parotitis, otitis media, tonsilitis, atopic dermatitis, arthritis, increased levels of Igs, decreased counts of CD3+ and CD4+ cells, and increased count of cTfh cells [5]. Among the 4 patients with gastrointestinal involvement, 1 patient with common variable immunodeficiency, harboring p.Gln407* in TRAF3, developed pancolitis and ileitis that was steroid-dependent and treated with biologics. However, detailed clinical information, including endoscopic and histopathological images, the type of biologic agent used, and treatment responsiveness, was not provided [5,6]. Therefore, the involvement of TRAF3 LoF in the development of IBD remains unclear. Myeloid cell-conditional Traf3-knockout mice exhibit increased inflammatory cytokine levels following TLR stimulation in macrophages and increased T-cell-dependent antibody responses [11]. These mice also develop exacerbated colitis induced by dextran sulfate sodium [12]. Moreover, the TRAF3 level in the plasma and colonic mucosa is higher than that in healthy individuals [13]. Together with the critical roles of TRAF3 in immune regulation, these findings suggest the involvement of TRAF3 in the pathogenesis of IBD. Among the patients with TRAF3 LoF exhibiting gastrointestinal symptoms, a truncated protein was detected in patients harboring p.Gln407*. This finding indicates the potential involvement of a dominant-negative effect in disease pathogenesis; however, a detailed functional analysis has not been performed. In contrast, truncated protein was not detected in our case (data not shown). Further studies are needed to clarify the phenotype-genotype correlation. Given that less than half of the patients with monogenic IBD with a major causative variant develop IBD [14], environmental factors, such as dysbiosis of the gut microbiota, could be crucial for the development of CD, even in monogenic disorders.
Another important finding of this study was the difference in responsiveness to anti-cytokine therapy between anti-TNF and ustekinumab, as well as the dynamics of cTfh cells before and after treatment in our patient with TRAF3 HI. CD is considered a heterogeneous disease, and the effectiveness of specific treatments varies significantly among patients. Although approximately 10%–30% of patients show no response to anti-TNF and 40%–50% lose their response over time, anti-IL-12/23p40 therapy can be beneficial to some patients who do not respond to anti-TNF [15]. Therefore, clarifying each patient’s detailed disease pathophysiology is important for the selection of a suitable treatment. Although exact mechanisms of action of anti-TNF in CD remain unclear, 1 mechanism involves neutralization of soluble and membrane-bound TNF, causing inhibition of TNF-induced signaling through its 2 receptors, TNFR1 and TNFR2 [16]. The failure of anti-TNF therapy in our patient may partially be attributed to the fact that TRAF3 is not directly involved in TNFR1 and TNFR2 signaling. In contrast, TRAF3 is vital for TLR signaling; its deficiency reportedly increases IL-12 and IL-23 production in murine macrophages [12], both of which are crucial in Tfh cell differentiation [17]. Tfh cells help B cells, supporting the formation of germinal centers that allow affinity maturation of antibody responses; they have been suggested to be involved in the pathophysiology of a wide range of inflammatory diseases, including CD [18]. Our patient with TRAF3 HI exhibited increased cTfh cell counts before ustekinumab treatment. Moreover, she was responsive to ustekinumab and the cTfh cell count reduced after treatment. These findings raise the hypothesis that the IL-12/23-Tfh pathway could play a role in the pathophysiology of CD in our patient with TRAF3 HI. Similar findings have been reported in some patients with non-monogenic CD. Peripheral blood cTfh cell frequencies are higher in patients with active CD than in those in remission, and their frequencies significantly decrease in ustekinumab responders after treatment. However, the cTfh cell counts in ustekinumab non-responders and anti-TNF responders/non-responders remain unchanged [19]. This pattern may suggest a potential involvement of the IL-12/23-Tfh pathway, as well as TRAF3, in the pathophysiology of some patients with CD.
Growing evidence suggests that CD is a heterogeneous disease, which needs stratification considering the different management strategies with various biologics, such as anti-TNF, anti-integrins, and anti-IL-12/23 or IL-23 antibodies, in addition to corticosteroids, immunomodulators, and Janus kinase inhibitors [20]. Although further studies will be required to elucidate the role of TRAF3 in the pathogenesis of IBD and more case studies are required to provide evidence for relevant targeted therapies, understanding the specific molecules/pathways affected by genetic mutations, such as TRAF3 mutations, might lead to critical insights into disease mechanisms and further understanding of IBD pathophysiology.
Notes
Funding Source
This study was supported in part by grants from the National Center for Child Health and Development (grant numbers 2019A-3 to KA, and 2020B-10 and 2024B-24 to IT) and the Japan Agency for Medical Research and Development (AMED: grant numbers JP21ek0109489 to Hayashi K and JP24ek0410 106 to Morita H). The funders had no role in the study design, data collection and interpretation, or decision to submit the work for publication.
Conflict of Interest
Arai K received honoraria from Mitsubishi Tanabe Pharma Co., Ltd., Janssen Pharmaceutical K.K., Nippon Kayaku Co., Ltd., and AbbVie GK and research grants from Janssen Pharmaceutical K.K. and AbbVie GK. The remaining authors disclose no conflict of interest.
Data Availability Statement
Data are available upon reasonable request.
Author Contributions
Conceptualization: Takeuchi I, Taniguchi K, Arai K, Uchiyama T, Morita H. Formal analysis: Takeuchi I, Taniguchi K, Hori A. Funding acquisition: Takeuchi I, Arai K, Morita H, Hata K. Investigation: Takeuchi I, Taniguchi K, Arai K, Uchiyama T, Terao M, Kawai T, Yoshioka T, Fujita S, Motomura K, Okazaki Y, Ishikawa T, Matsumoto K, Takada S, Onodera M, Morita H, Hata K. Resources: Takeuchi I, Arai K, Kyodo R, Shimizu H, Ogura M, Hayashi K. Supervision: Arai K, Matsumoto K, Onodera M, Morita H, Hata K. Writing - original draft: Takeuchi I, Taniguchi K, Uchiyama T, Morita H. Writing - review & editing: Arai K, Terao M, Hori A, Kawai T, Yoshioka T, Kyodo R, Shimizu H, Fujita S, Motomura K, Okazaki Y, Ishikawa T, Ogura M, Hayashi K, Matsumoto K, Takada S, Onodera M, Hata K. Approval of final manuscript: all authors.
Additional Contributions
We thank the patient and her family for their sincere efforts and cooperation during this study. We also want to express our sincere gratitude to all members of the National Center for Child Health and Development.
Supplementary Material
Supplementary materials are available at the Intestinal Research website (https://www.irjournal.org).
Supplementary Materials.
Supplementary Table 1.
Summary of the Patient in Our Study and Individuals with TRAF3 LoF Variants Reported in Previous Studies