From Undifferentiated Connective Tissue Disease to Common Variable Immunodeficiency: A Novel NFKB1 Mutation in a Pediatric Case

ABSTRACT

Common variable immunodeficiency (CVID) may initially manifest as undifferentiated connective tissue disease (UCTD) in the absence of recurrent infections or hypogammaglobulinemia. In patients presenting with warning signs, neither older age nor normal immunoglobulin levels should preclude consideration of CVID; early genetic testing is recommended.

1. Introduction

CVID is a prevalent inborn error of immunity (IEI) classified as an antibody deficiency disorder, with hypogammaglobulinemia as its hallmark [1]. It is a heterogeneous immunodeficiency syndrome, with an incidence of 1:10,000–1:50,000 abroad, and it is relatively rare in China [2, 3]. In addition to recurrent and severe infections, patients with CVID are also prone to autoimmune and inflammatory complications, and have an increased risk of developing tumors. CVID typically manifests after 2 years of age, although some patients may not develop symptoms until adulthood. Recurrent infections, particularly those involving the respiratory tract, represent the most common initial manifestation; however, a subset of patients first present with autoimmune diseases or lymphoproliferative disorders [4, 5]. Owing to the absence of specific clinical features and insufficient awareness among clinicians, CVID is frequently misdiagnosed, missed, or diagnosed with a significant delay. Herein, we report a case of CVID presenting with UCTD as the initial manifestation, aiming to enhance clinicians' awareness of CVID in atypical presentations.

2. Case History/Examination

2.1. Initial Presentation and Evaluation (September 2016)

2.1.1. History at Initial Presentation

A female patient, aged 11.5 years, was admitted to Shanghai Children's Hospital in September 2016 with a chief complaint of fever, dizziness, and fatigue for 5 days. Five days prior to admission, the patient developed fever, dizziness, and fatigue without any obvious precipitating factors, accompanied by vomiting. She had no cough, nasal discharge, diarrhea, rash, oral ulcers, gross hematuria, foamy urine, or joint swelling. Her past medical history was unremarkable, with no recurrent infections, no history of infectious diseases, no drug or food allergies, no prior surgeries or trauma, and no blood transfusions. She was the first child born at full term via spontaneous vaginal delivery with a birth weight of 3000 g, without perinatal complications. Her growth and developmental milestones were age‐appropriate, and she received all routine childhood vaccinations according to the national immunization schedule, with no reported adverse events. Her parents were non‐consanguineous and healthy, and her paternal half‐brother was also healthy.

2.1.2. Physical Examination at Initial Presentation

Physical examination revealed positive findings including pallor, mild pharyngeal erythema, tachypnea, tachycardia (129 bpm), hepatomegaly (2 cm below the right costal margin), and massive splenomegaly (9 cm below the left costal margin, extending approximately 1.5 cm to the right of the umbilicus) with moderate consistency. All remaining physical examination findings were negative.

2.1.3. Initial Laboratory and Imaging Findings

Initial laboratory investigations revealed pancytopenia with a white blood cell count of 2.51 × 10⁹/L (reference 4.00–10.00), hemoglobin of 40.00 g/L (reference 110–160), and platelet count of 48.00 × 10⁹/L (reference 100–400). Absolute neutrophil count was reduced at 0.88 × 10⁹/L (reference 2–7), while reticulocytes were elevated at 4.60% (reference 0.50%–1.50%). Inflammatory markers including C‐reactive protein and erythrocyte sedimentation rate were within normal limits. Biochemical testing showed hyperbilirubinemia with total bilirubin of 41.09 μmol/L (reference 3.4–17.1) and direct bilirubin of 11.68 μmol/L (reference 0–6.8), with normal transaminases. Renal function tests, including blood urea nitrogen and creatinine, were within normal limits. Serum albumin and globulin levels were normal. Serological testing for viral infections revealed positive hepatitis B surface antibody (HBsAb), while all other hepatitis B markers, as well as antibodies against hepatitis C virus (HCV) and human immunodeficiency virus (HIV), were negative.

Immunological evaluation revealed normal immunoglobulin levels: IgG 4.69 g/L (reference 3.82–14.04), IgA 0.60 g/L (reference 0.14–1.79), and IgM 0.85 g/L (reference 0.29–1.44). Cellular immunity assessment showed normal counts of CD3+ T cells, CD4+ T cells, and CD8+ T cells; however, the CD19+ B cell count was decreased at 0.235 × 10⁹/L (reference 0.45–0.82 × 10⁹/L). However, due to the limitations of medical resources at the local institution a decade ago, more detailed lymphocyte subtyping, including switched memory B cells and other advanced immunological phenotyping, was not performed at the initial presentation. Complement levels were markedly decreased, including total complement (CH50 15.68 U/mL, reference 31–89), C1q (94.40 mg/L, reference 159–233), C3 (0.50 g/L, reference 0.79–1.52), and C4 (0.02 g/L, reference 0.1–0.4). A comprehensive autoantibody panel, including antinuclear antibodies (by indirect immunofluorescence on HEp‐2 cells), anti‐dsDNA, anti‐Sm, anti‐RNP, anti‐SSA/Ro, anti‐SSB/La, anti‐Jo‐1, anti‐Scl‐70, anti‐ribosomal P, anti‐neutrophil cytoplasmic antibodies, and anticardiolipin antibodies (IgG/IgM), was negative. The direct antiglobulin test (Coombs) was positive for IgG and C3d.

Bone marrow aspiration showed no evidence of malignancy or hemophagocytosis. Infectious workup for viral, bacterial, and tuberculosis etiologies was negative. Abdominal computed tomography confirmed splenomegaly, while electrocardiography and echocardiography were normal.

2.1.4. Initial Diagnosis and Management

The initial presentation—characterized by fever, pancytopenia, hypocomplementemia, splenomegaly, and a positive direct antiglobulin test—was consistent with an autoimmune/connective tissue disease. However, the patient did not fulfill the classification criteria for any definite connective tissue disease, including systemic lupus erythematosus, Sjögren's syndrome, or rheumatoid arthritis. Other potential etiologies, including hematological disorders, malignancies, and viral infections, were systematically excluded through bone marrow aspiration, imaging studies, and comprehensive serological testing.

According to the European Society for Immunodeficiencies (ESID) 2019 diagnostic criteria for CVID [6], a diagnosis requires the following: (1) at least one clinical manifestation, including increased susceptibility to infections, autoimmune manifestations, granulomatous disease, unexplained polyclonal lymphoproliferation, or a family history of antibody deficiency; (2) at least two measurements showing markedly decreased IgG and IgA, with or without low IgM (≥ 2 SD below the normal mean for age); (3) at least one of the following: poor antibody response to vaccines and/or lack of isohemagglutinins, or low switched memory B cells (< 70% of age‐related normal value); (4) exclusion of secondary causes of hypogammaglobulinemia; (5) diagnosis after 4 years of age (symptoms may appear earlier); and (6) no evidence of precursor T‐cell deficiency.

At the time of initial presentation, our patient manifested autoimmune cytopenias and polyclonal lymphoproliferation (splenomegaly), which met one of the clinical criteria. However, she had no history of recurrent infections, no family history of antibody deficiency, and normal immunoglobulin levels (IgG, IgA, and IgM). Cellular immunity assessment revealed normal T‐cell counts, and her antibody response to hepatitis B vaccination was adequate. Therefore, she did not fulfill the laboratory requirements of the ESID criteria, and a diagnosis of CVID could not be established at that time. Consequently, a working diagnosis of UCTD with possible early lupus was made. After excluding contraindications, the patient received intravenous corticosteroids and intravenous immunoglobulin (IVIG) upon admission, which led to rapid normalization of blood cell counts and improvement in complement levels. Subsequently, the patient was started on oral prednisone at an initial dose of 2 mg/kg/day (60 mg/day).

2.2. Follow‐Up and Treatment Adjustments (September 2016–August 2020)

After discharge in September 2016, the patient was followed up monthly at the Department of Nephrology, Rheumatology, and Immunology of our hospital. During the initial 6 months of follow‐up, she remained clinically stable with no disease flares, and the prednisone dose was gradually tapered. However, when the prednisone dose was reduced to 7.5 mg/day at 6 months, the patient experienced a disease flare characterized by fever, splenomegaly, pancytopenia, and hypocomplementemia. Re‐escalation of the corticosteroid dose led to clinical improvement, but the prednisone could not be completely withdrawn, and the minimum effective dose required to maintain disease control progressively increased over time. Approximately one and a half years after initiating corticosteroid therapy, mycophenolate mofetil (MMF) was added. After the addition of MMF, the patient's condition stabilized, allowing for gradual tapering of corticosteroids. Approximately 6 months after initiating MMF, prednisone was successfully discontinued. However, the patient subsequently did not adhere to regular follow‐up and discontinued MMF on her own accord, after which she experienced multiple disease flares characterized by fever, anemia, hypocomplementemia, and splenomegaly. These flares responded to corticosteroid re‐administration, but the patient declined further use of MMF due to financial constraints. From then until August 2020, she was maintained on varying doses of prednisone. During this period, she visited multiple hospitals and frequently adjusted or discontinued prednisone on her own, resulting in incomplete medical records; therefore, the specific dosages and administration schedules of corticosteroids could not be ascertained. Throughout the follow‐up period from September 2016 to August 2020, during her visits to our hospital, autoantibody profiles, immunoglobulin levels, and cellular immunity were repeatedly assessed (summarized in Tables 1, 2, 3, 4).

TABLE 1.

Longitudinal autoantibody profile during follow‐up (2016–2025).

Date	Follow‐up
	Antinuclear antibodies	Anti‐dsDNA antibodies	Anti‐Sm, anti‐RNP, anti‐SSA/Ro, anti‐SSB/La	Anti‐Jo‐1, anti‐Scl‐70, anti‐ribosomal P	Anti‐neutrophil cytoplasmic antibodies (ANCA)	Anticardiolipin antibodies (IgG/IgM)
2016.09	Negative	Negative	Negative	Negative	Negative	Negative
2016.10	Negative	Negative	Negative	Negative	Negative	Negative
2017.08	Negative	Negative	Negative	Negative	Negative	Negative
2018.11	Negative	Negative	Negative	Negative	Negative	Negative
2019.09	Negative	Negative	Negative	Negative	Negative	Negative
2020.04	Negative	Negative	Negative	Negative	Negative	Negative
2020.08	Negative	Negative	Negative	Negative	Negative	Negative
2025.05	Negative	Negative	Negative	Negative	Negative	Negative

Note: ANA was detected by indirect immunofluorescence (IIF) on HEp‐2 cells; anti‐dsDNA, anti‐Sm, anti‐RNP, anti‐SSA/Ro, anti‐SSB/La, anti‐Jo‐1, anti‐Scl‐70, anti‐ribosomal P, ANCA, and anticardiolipin antibodies (IgG/IgM) were detected by enzyme‐linked immunosorbent assay (ELISA) and/or immunoblotting. All results remained negative throughout the 9‐year follow‐up period.

TABLE 2.

Longitudinal serum immunoglobulin levels during follow‐up (2016–2025).

Date	Follow‐up
	IgG (g/L)	IgA (g/L)	IgM (g/L)
2016.09	4.69	0.6	0.85
2017.08	8.5	< 0.0663↓	0.45
2018.04	3.88↓	< 0.0625↓	0.58
2018.11	4.41↓	< 0.0663↓	0.45
2019.09	3.69↓	< 0.0694↓	0.8
2020.04	3.87↓	< 0.0694↓	1.3
2020.08	2.14↓	< 0.0694↓	0.26↓
2020.09	4.61↓	< 0.06↓	0.14↓
2021.01	3.03↓	< 0.0691↓	0.11↓
2022.02	0.74↓	< 0.0663↓	0.12↓
2022.07	1.17↓	< 0.0672↓	0.16↓
2022.11	2.59↓	< 0.0672↓	0.06↓
2025.05	< 0.082↓	< 0.0566↓	< 0.0606↓

Note: Reference ranges (age‐adjusted): IgG: neonates 6.6–17.5 g/L, 1 month–1 year 2.0–9.5 g/L, 1–12 years 3.6–15.5 g/L, 12–18 years 7.0–15.5 g/L; IgA: neonates < 0.06 g/L, 1 month–1 year 0.06–0.91 g/L, 1–12 years 0.14–2.91 g/L, 12–18 years 0.58–3.21 g/L; IgM: neonates 0.06–0.20 g/L, 1 month–1 year 0.17–1.50 g/L, 1–12 years 0.37–2.40 g/L, 12–18 years 0.49–2.61 g/L. The downward arrow (↓) indicates values below the lower limit of normal. The patient started regular intravenous immunoglobulin (IVIG) replacement therapy in August 2020. Measurements obtained after August 2020 may be influenced by exogenous immunoglobulin; levels recorded from February 2022 onward were measured at least 2–3 months after the most recent IVIG infusion, thus reflecting endogenous immunoglobulin production. Dates are presented in year and month format (e.g., 2016.09 = September 2016).

TABLE 3.

Longitudinal T‐ and B‐cell subset kinetics during follow‐up (2016–2025).

Date	Follow‐up
	CD3+ cell (109/L)	CD8+ cell (109/L)	CD4+ cell (109/L)	CD19+ cell (109/L)
2016.09	1.533	0.547	0.867	0.235↓
2017.08	1.404↓	0.634	0.671↓	0.16↓
2018.04	3.252	1.474	1.634	0.33↓
2019.09	1.449	0.474↓	0.703↓	0.088↓
2020.08	1.766	0.66	0.916	0.082↓
2020.09	3.349	1.304	1.924	0.246↓
2021.02	2.588	1.078	1.34	0.162↓
2022.02	2.005	0.743	1.037	0.1↓
2022.07	1.466	0.441↓	0.857	0.07↓
2022.11	1.664	0.663	0.914	0.076↓
2025.05	1.305↓	0.689	0.594↓	0.024↓

Note: Reference ranges (age‐adjusted for pediatric patients): CD3+ T cells 1.41–3.38 × 10⁹/L; CD8+ T cells 0.54–1.36 × 10⁹/L; CD4+ T cells 0.71–1.84 × 10⁹/L; CD19+ B cells 0.45–0.82 × 10⁹/L. The downward arrow (↓) indicates values below the lower limit of normal. Lymphocyte subset analysis was performed by flow cytometry. The patient started regular intravenous immunoglobulin (IVIG) replacement therapy in August 2020. CD19+ B cells were consistently decreased throughout the entire follow‐up period (from initial presentation in September 2016 to the most recent assessment in May 2025), even when serum immunoglobulin levels were still within normal ranges, suggesting an early and persistent B‐cell defect. CD4+ T‐cell counts showed intermittent decreases, while CD3+ and CD8+ T‐cell counts were generally within or near normal ranges. Dates are presented in year.month format (e.g., 2016.09 = September 2016).

TABLE 4.

Longitudinal detailed B‐cell subset immunophenotyping during follow‐up (2020–2025).

Date	Follow‐up
	Naive B (cells/μL)	Memory B (cells/μL)	Transitional B (cells/μL)	Plasmablast B (cells/μL)
2020.08	130	13↓	3	1
2022.02	88↓	2↓	2↓	0↓
2022.07	34↓	2↓	1↓	0↓
2022.11	93↓	3↓	1↓	0↓
2025.05	15↓	0↓	1↓	0↓

Note: Reference ranges for B‐cell subpopulations: naive B cells, 98–296 cells/μL; memory B cells, 27.00–70.00 cells/μL; transitional B cells, 3–19 cells/μL; plasmablasts, 1–15 cells/μL. The downward arrow (↓) indicates values below the lower limit of normal. B‐cell subset analysis was performed by flow cytometry. The patient started regular intravenous immunoglobulin (IVIG) replacement therapy in August 2020; the August 2020 measurement was obtained prior to the first IVIG infusion. Follow‐up measurements from February 2022 onward demonstrate progressive decline in memory B cells, transitional B cells, and plasmablasts, reflecting persistent humoral immune defect. Dates are presented in year.month format (e.g., 2020.08 = August 2020).

The patient first showed a marked decrease in IgA in August 2017, with IgA levels dropping below 0.0663 g/L, while IgG levels remained within normal limits at that time. During subsequent follow‐up, IgA remained persistently low, and IgG gradually declined. By August 2020, IgG had fallen significantly below the normal range, accompanied by a concurrent decrease in IgM. At this point, an underlying IEI was strongly suspected, prompting further immunological evaluation, including detailed lymphocyte phenotyping and whole‐exome sequencing. The absolute count of memory B cells was markedly decreased at 13.00 cells/μL (reference 27.00–70.00 cells/μL). Whole‐exome sequencing revealed a heterozygous de novo frameshift mutation in the NFKB1 gene (NM_001165412: c.902delC, p.Ser301fs), which results in a frameshift starting at codon 301. Sanger sequencing confirmed the variant in the proband, while her parents and paternal half‐brother were all wild‐type at this locus (Figure 1). According to the American College of Medical Genetics and Genomics (ACMG) guidelines [7], this variant was classified as pathogenic (criteria: PVS1 + PS2 + PM2).

FIGURE 1.

Sanger sequencing chromatograms of the proband and his family members. The proband (panel a) harbors a heterozygous variant in NFKB1 (NM_001165412): C.902delC (p.S301fs); this variant was not detected in his father (panel b), mother (panel c), or half‐brother (panel d).

2.3. Final Diagnosis, Treatment Modification, and Follow‐Up (August 2020–May 2025)

2.3.1. Final Diagnosis, Treatment Modification

Retrospectively, the patient fulfilled all the diagnostic criteria for CVID according to the ESID 2019 criteria: (1) age at onset of 11.5 years; (2) presence of autoimmune manifestations (autoimmune cytopenias); (3) repeatedly decreased serum levels of IgG, IgA, and IgM on multiple measurements; (4) decreased switched memory B cells; (5) exclusion of secondary causes of hypogammaglobulinemia; and (6) absence of precursor T‐cell deficiency. Furthermore, the identification of a pathogenic heterozygous de novo frameshift mutation in the NFKB1 gene (NM_001165412: c.902delC, p.Ser301fs) provided genetic confirmation of the diagnosis, as NFKB1 mutations are well‐established monogenic causes of CVID. Taken together, the combination of clinical, immunological, and genetic evidence definitively established the diagnosis of CVID in this patient.

Given the definitive diagnosis of CVID, which was clinically characterized by fever, autoimmune cytopenias, hypocomplementemia, splenomegaly, and hypogammaglobulinemia, a comprehensive treatment strategy was formulated for this patient. The regimen consisted of corticosteroid therapy in combination with regular scheduled infusions of intravenous immunoglobulin (IVIG).

2.3.2. Outcome and Follow‐Up

The patient has been maintained on a combination of corticosteroids and regular IVIG infusions for over 5 years. During this period, corticosteroids were successfully tapered and eventually discontinued on multiple occasions, although intermittent corticosteroid courses were occasionally required for recurrent cytopenias. Notably, when needed, the corticosteroid doses were substantially lower than those used during the initial treatment and could be tapered off more rapidly. The frequency of IVIG infusions was initially once monthly and was subsequently extended to every 2–4 months based on clinical response and the patient's financial constraints. Throughout this period, the patient did not experience recurrent infections. Although intermittent cytopenias occurred occasionally, their severity was markedly milder compared to that at initial presentation, and complement levels were consistently maintained within normal ranges. The patient received most of her medical care at local hospitals near her home, with intermittent assessments at our institution, where autoantibody profiles, immunoglobulin levels, and cellular immunity were repeatedly evaluated (summarized in Tables 1, 2, 3, 4). Since February 2022, all immunoglobulin measurements were obtained at least 2–3 months after the most recent IVIG infusion, revealing that the patient's endogenous immunoglobulin levels (IgG, IgA, and IgM) remained persistently very low. The most recent immunological evaluation in May 2025 demonstrated that IgG, IgA, and IgM levels were all approaching zero, confirming the sustained nature of the patient's humoral immunodeficiency.

3. Discussion

The present case exemplifies a diagnostically challenging scenario: a pediatric patient presenting with autoimmune cytopenias, hypocomplementemia, and massive splenomegaly at 11.5 years of age—without recurrent infections or hypogammaglobulinemia—who was ultimately found to carry a novel NFKB1 mutation. This case highlights two clinically important aspects: the diagnostic pitfalls of CVID presenting as an UCTD‐like phenotype, and the expanding genetic spectrum of NFKB1‐associated disease.

At initial presentation, the patient's clinical and laboratory features—fever, pancytopenia, hypocomplementemia, splenomegaly, and positive direct antiglobulin test—were strongly suggestive of an autoimmune/connective tissue disease. However, she did not fulfill criteria for any definite connective tissue disease, leading to a working diagnosis of UCTD. CVID was not initially suspected because of normal immunoglobulin levels, adequate vaccine response, and no recurrent infections or family history. This diagnostic scenario carries two important lessons. First, autoimmune cytopenias and splenomegaly are well‐recognized presenting features of CVID that can precede the diagnosis by years. In our patient, the interval from initial presentation to CVID diagnosis was 4 years. In a 10‐year single‐center study of 85 CVID patients, those with autoimmune diseases had a diagnostic delay of 84 months, with cytopenias being the most common autoimmune manifestation and splenomegaly the most common organ manifestation [2]. Second, immunoglobulin levels may remain normal early in the disease course—our patient maintained normal IgG, IgA, and IgM at onset, with IgA not dropping significantly until 1 year later, followed by gradual IgG decline over the subsequent 3 years. Therefore, neither normal immunoglobulin levels nor older age at onset should be used as sole exclusion criteria for CVID in patients presenting with otherwise unexplained autoimmune or lymphoproliferative manifestations. For such patients, early genetic testing is recommended to facilitate timely diagnosis and appropriate therapeutic intervention.

The NFKB1 gene encodes the precursor protein p105, which is processed to generate the NF‐κB transcription factor subunit p50. NF‐κB signaling plays a pivotal role in adaptive immunity, cell proliferation, and inflammation [8]. Loss‐of‐function NFKB1 variants are the most common monogenic cause of CVID in European populations, accounting for approximately 4% of cases [9]. The pathogenic mechanism underlying NFKB1‐related CVID is haploinsufficiency rather than negative dominance [10]. Our patient was found to carry a heterozygous de novo frameshift mutation in NFKB1 (NM_001165412: c.902delC, p.Ser301fs). To our knowledge, this variant has not been previously reported in public databases (gnomAD, ExAC, 1000 Genomes) or in the literature. According to ACMG/AMP guidelines, the variant was classified as pathogenic based on: PVS1 (frameshift leading to premature stop codon and loss of function), PS2 (confirmed de novo occurrence, as parents and paternal half‐brother were wild‐type), and PM2 (absent from population databases). The combination of these criteria provides strong evidence for pathogenicity, consistent with the autosomal dominant inheritance pattern of NFKB1‐associated CVID. The clinical phenotype of our patient aligns well with the established NFKB1‐related disease spectrum: reduced switched memory B cells, progressive hypogammaglobulinemia, autoimmune cytopenias, and splenomegaly. In a large cohort study of 157 individuals with heterozygous NFKB1 mutations, the phenotype included hypogammaglobulinemia (88.9%), reduced switched memory B cells (60.3%), autoimmunity (57.4%), and lymphoproliferation (52.4%) [11]. Notably, patients with frameshift or nonsense NFKB1 variants—such as the c.902delC mutation in our patient—have been shown to have more pronounced noninfectious complications compared to those with missense variants [12], which may explain the prominent autoimmune and lymphoproliferative phenotype in this case.

The rationale and timing for genetic testing merit further explanation. At initial presentation, although the patient's clinical and laboratory features were suggestive of an autoimmune/connective tissue disease that could not be classified into any definite category, the absence of recurrent infections, normal immunoglobulin levels, adequate vaccine response, and no family history of immunodeficiency did not support a clinical suspicion of IEI. Genetic testing was discussed with the family at that time; however, it was declined due to financial constraints. Therefore, genetic testing was not pursued initially. During 4 years of follow‐up, the patient developed progressive hypogammaglobulinemia and markedly decreased memory B cells, prompting whole‐exome sequencing in August 2020, which was ultimately performed with family consent. Retrospective evaluation confirmed that the patient fulfilled all ESID 2019 diagnostic criteria for CVID, and the genetic finding provided definitive molecular confirmation.

In conclusion, this case expands the clinical and genetic spectrum of NFKB1‐associated CVID and underscores the importance of considering underlying IEI in patients with unclassifiable autoimmune/connective tissue disease phenotypes accompanied by warning signs of immune dysregulation. Genetic testing should be considered in such selected cases regardless of age or immunoglobulin levels. Functional validation of this novel variant is warranted in future studies.

4. Conclusion

We report a pediatric case of CVID initially presenting as UCTD without infections or hypogammaglobulinemia, carrying a novel pathogenic NFKB1 mutation (c.902delC). This case expands the genetic spectrum of NFKB1‐associated CVID.

Clinically, it demonstrates that CVID may present solely with autoimmune features years before typical immunodeficiency emerges. For patients with unclassifiable or treatment‐resistant connective tissue diseases—especially those with cytopenias, splenomegaly, or lymphoproliferation—CVID should be considered regardless of age or immunoglobulin levels. Early genetic testing facilitates definitive diagnosis, timely IVIG therapy, and avoidance of unnecessary immunosuppression.

In selected patients with unclassifiable autoimmune phenotypes accompanied by warning signs of immune dysregulation (e.g., cytopenias, splenomegaly, lymphoproliferation, or a family history of autoimmunity/immunodeficiency), underlying IEI should be suspected and genetic testing is recommended, irrespective of age or immunoglobulin levels.

Author Contributions

Yu‐Jie Hu: writing – original draft, data curation. Sheng Hao: validation. Dan Feng: data curation. Bing‐Xue Huang: investigation. Wen‐Yan Huang: funding acquisition, writing – review and editing.

Funding

This study was supported by the National Natural Science Foundation of China (82170683, Wen‐Yan Huang) and Shanghai Science and Technology Commission Natural Foundation Project (23JC1401202, Wen‐Yan Huang).

Consent

Written informed consent from the patient was obtained according to journal guidelines.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are available from the corresponding author upon reasonable request.