Clinical and genetic characterization of X-linked agammaglobulinemia in a Colombian cohort☆ Authors: Manuela Olaya Hernández, Jacobo Triviño Arias, Oriana Arias Valderrama, Vanessa Cabrera Guarín, Jaime Alberto Patiño, Harry Mauricio Pachajoa, Diego Medina Valencia, Alexis Antonio Franco, Lina M Castaño Jaramillo, Natalia Vélez Tirado, Andrés Felipe Zea-Vera, María Claudia Ortega-López, Carlos Olmos-Olmos, Yaicith Arévalo-Montenegro, Laura Rincón-Arenas, José Luis Franco, Julio Cesar Orrego, Estefanía Vásquez, Catalina Obando-Gil, Isabel Moribe, Maria Lucía Valderrama, Catalina Gómez-Parada, Marta Álvarez, Andrés Felipe Arias, Paola Marsela Pérez, Oscar Correa Jiménez Published: 2026 ABSTRACT AbstractIntroductionX-linked Agammaglobulinemia (XLA) is an inborn error of immunity (IEI) caused by mutations in the BTK gene, resulting in the absence of mature B-cells and altered serum immunoglobulin levels. In Colombia, unified epidemiological and genetic data are lacking. This study aims to characterize XLA patients in the country demographically, phenotypically, and genotypically.MethodsA multicenter, observational, cross-sectional study based on clinical records. Thirty-eight patients with B-cell counts <2% and hypogammaglobulinemia were included. Clinical, immunological, and genetic variables and outcomes were analyzed using descriptive statistics in R-Studio.ResultsThirty-eight males were analyzed, with a median age at diagnosis of 16 months (IQR 7.2–21.8). Genetic testing was available for 84.2% (n = 32), identifying 15 distinct variants, 53% of which were novel. Sinopulmonary infections were the primary manifestation (92.1%), notably pneumonia (42.1%) and otitis media (39.5%). Median baseline IgG levels were 187 mg/dL. Bronchiectasis was documented in 34.2% of cases. The eight-year survival rate was 71.1%.DiscussionThe Colombian cohort shows a relatively early diagnosis but a high burden of structural sequelae (bronchiectasis), suggesting a need to optimize immunoglobulin replacement therapy. The high proportion of novel genetic variants underscores the importance of regional studies. This study represents the most robust characterization of XLA in Colombia, providing key data to improve clinical suspicion and prognosis.Keywords: Bruton's agammaglobulinemia, X-linked agammaglobulinemia, BTK, Inborn errors of immunity, Genetics Introduction X-linked agammaglobulinemia (XLA) is an inborn error of immunity caused by pathogenic variants in the Bruton tyrosine kinase (BTK) gene.1 Mutations in BTK result in an early block in B-cell development, leading to the absence of mature B lymphocytes and plasma cells and, consequently, an inability to produce immunoglobulins of all isotypes.2 The BTK gene is located on the long arm of the X chromosome at Xq21.3–q22, and XLA predominantly affects male patients.3,4 Nevertheless, rare cases of XLA in females have been reported, mainly attributed to skewed X-chromosome inactivation.5 Among congenital defects of antibody production, XLA is the most common genetic disorder, accounting for approximately 85% of cases,6 with an estimated incidence ranging from 1:100,000 to 1:200,000 live births.6 According to the European Society for Immunodeficiencias (ESID) agammaglobulinemia is defined by recurrent infections before 5 years of age, serum IgG levels <500 mg/dL, IgA and IgM levels below −2 standard deviations, and <2% circulating B cells.7 Clinical manifestations typically begin after 6 months of age, coinciding with the physiological decline of maternally derived immunoglobulin G (IgG).6,8 Patients commonly present with recurrent or severe infections, predominantly involving the respiratory tract, musculoskeletal system, and gastrointestinal tract,9 with otitis media being the most frequent infection.10 The pathogens most commonly implicated include Streptococcus pneumoniae, Staphylococcus spp., Pseudomonas spp., and Haemophilus influenzae.9 Infections may also be caused by other bacterial and viral agents and can be challenging to treat.1,11 Early and regular treatment with standard IgG replacement therapy (intravenous or subcutaneous) constitutes the cornerstone of management, in conjunction with prompt treatment of infections and, in selected cases, antibiotic prophylaxis.4 The aim of this study was to characterize the clinical and paraclinical phenotype of patients with XLA in Colombia during the period from 2013 to 2024. Materials and methods This was a multicenter, registry-based study. Patients diagnosed with XLA and treated at multiple healthcare institutions in Colombia—including Fundación Valle del Lili, HOMI Fundación Hospital Pediátrico La Misericordia, Hospital Infantil Universitario de San José, Fundación Neumológica Colombiana, Fundación Cardioinfantil, Hospital Infantil Club Noel, and Centro Médico La Sabana—as well as academic centers (Primary Immunodeficiency Group of Universidad de Antioquia, Universidad Icesi, and Universidad del Valle) were included between 2013 and 2024. Patients with a CD19+ B-cell count <2% and hypogammaglobulinemia were eligible for inclusion, 1 part of the group had clinical suspicion, hypogammaglobulinemia, and genetic testing with an alteration compatible with XLA. Hypogammaglobulinemia was assessed by measuring only IgG levels below 500 mg/dL, given that the vast majority of patients were older than 12 months, and with decreased or absent IgA and IgM.Genetic studies were performed using Sanger sequencing or single-exome sequencing in other cases. We collected information about family history, prior infections (were considered characteristic of this type of humoral deficiency: septic arthritis, meningitis, pneumonia, recurrent otitis media, sinusitis, skin infections, sepsis of gastrointestinal origin, acute lymphadenitis, osteomyelitis, and post-vaccination reaction to poliomyelitis. These same infections were analyzed at the time of diagnosis and after diagnosis with immunoglobulin replacement), neutrophil count, quantitative immunoglobulin levels, and treatment-related complications. Bronchiectasis, malignancy, and respiratory allergy were considered complications. We included vital status. We excluded registries with absence of follow-up at least 6 months in the same center. Data collection was conducted ensuring confidentiality using the REDCap (Research Electronic Data Capture) platform version 15.0.2 © 2025 Vanderbilt University. Statistical analysis Categorical variables are presented as frequencies and percentages. Quantitative variables were assessed for normality using the Shapiro–Wilk test. Depending on distribution, data are presented as medians with interquartile ranges (IQRs) or means with standard deviation (SD). Statistical analyses were performed using RStudio software version 4.3.3. Ethics approval The study was reviewed and approved by the Biomedical Research Ethics Committee of Fundación Valle del Lili (approval number 2021.1847). Approval was also obtained from the ethics committees of all participating institutions. Results A total of 38 patients with a diagnosis of XLA were included. The median age at diagnosis was 16.0 months (IQR 7.3–21.8), and all patients (100%) were male. Genetic testing confirming BTK mutations was available in 32 patients (84.2%) (Table 1). Regarding genetic epidemiology, 20 patients (52.6%) were classified as sporadic cases, while family history data were unavailable for 10 patients (26.3%). Among the 32 patients with genetic testing, detailed variant characterization was possible in 20 individuals, identifying 15 distinct BTK variants, predominantly located within exons encoding the Protein Kinase, Src homology 3, and Tec homology domains. Missense variants were most common (8/15), followed by nonsense (4/15), splicing (3/15), and indel mutations (1/15). According to ACMG criteria, variants were mainly classified as pathogenic (7/15) or likely pathogenic (5/15), while 3 were variants of uncertain significance. Notably, 53% of the identified variants had not been previously reported. Some variants, including c.1116G > C and c.1786 A > G, were observed in multiple patients, suggesting possible recurrent or founder effects. Table 1Genetic distributionTable 1LocalizationDomainNucleotide ChangeAa ChangeMutation TypePatientsClassificationACMG CriteriaPrevious reportCountry of previously reportExon 16Protein kinasec.1630 A > Gp. (Arg544Gly)Missense1PathogenicPM1, PM5, PP3, PP5, PM2YesSpain, MexicoExon 15Protein kinasec.1558C > Gp. (Arg502Gly)Missense1PathogenicPM1, PP3, PM5, PP5, PM2YesArgentina, USAIntron 9Src homology 3c.839+5G > Ap.?Splicing1Likely PathogenicPP3, PP5, PM2NoIntron 11Src homology 2c.1076+4 A > Gp.?Splicing1Uncertain SignificancePP3, PM2NoExon 8Src homology 3c.763C > Tp. (Arg255Ter)Nonsense1PathogenicPVS1, PP5, PM2YesUSA, EuropeIntron 9Src homology 3c.840-3C > Gp.?Splicing1Uncertain SignificancePP3, PM2NoExon 13Src homology 2c.1116G > Cp. (Arg372Ser)Missense5Likely PathogenicPM1, PM5, PP3, PP5, PM2NoExon 17Protein kinasec.1684C > Tp. (Arg562Trp)Missense2PathogenicPM1, PM5, PP3, PP5, PS3, PM2YesMexico, diferenets partesExon 16Protein kinasec.1606 A > Gp. (Lys536Glu)Missense1Likely Pathogenic (Franklin) - Uncertain Sginificance (Clinvar)PM1, PP2, PM2, PP3NoExon 16Protein kinasec.1573C > Tp. (Arg525Ter)Nonsense1PathogenicPM3, PVS1, PM2,YesEuropa, USAExon 6Tec homologyc.461G > Ap. (Cys154Tyr)Missense1Likely PathogenicPP3, PM2, PM5, PP2NoExon 6Pleckstrin homologyc.364_367delp. (Lys122GlyfsTer9)Indel1Likely PathogenicPVS1, PM2NoIntron 16Protein kinaseBTK c.1566+2 T > Gp,?Splicing1Likely PathogenicPVS1, PM2YesItaliaExon 18Protein kinasec.1763G > Ap. (Trp588Ter)Nonsense1PathogenicPVS1, PP5, PM2YesEuropa,USAExon 9Src homology 3c.778C > Tp. (Gln260Ter)Nonsense1PathogenicPVS1, PP5, PM2YesEuropa,USA Genetic distribution Baseline IgG levels were available in 34 patients, with a median of 187.0 mg/dL (IQR 28.5–320.0); 4 patients lacked baseline immunoglobulin measurements and only had post-replacement values. IgM levels were documented in 33 patients, with a median of 15.0 mg/dL (IQR 3.0–25.0). IgA levels were available in 34 patients, with a median of 4.5 mg/dL (IQR 4.0–21.0). B-cell quantification was documented in 25 patients, showing a median of <2% circulating B cells (IQR 0.1–2.0). Neutropenia at diagnosis was observed in 4 patients (13.3%). Prior to diagnosis, 35 patients (92.1%) experienced documented infections (Table 1). Sinopulmonary infections, including pneumonia and otitis media, decreased following diagnosis but continued to occur (Fig. 1). The most frequently isolated microorganisms in our cohort, were Aspergillus fumigatus, Pseudomonas aeruginosa, Enterococcus spp., Escherichia coli, rhinovirus/enterovirus, cytomegalovirus, Streptococcus pneumoniae, and rotavirus. Prior to diagnosis, no patient developed acute lymphadenitis or vaccine-associated poliomyelitis, despite the majority of patients having received live oral poliovirus vaccine (POV). Frequency of infections in X-linked Agammaglobulinemia: comparison pre and post diagnosis. During follow-up, 26 patients (68.4%) developed infectious, including pneumonia (42.1%), recurrent otitis media (39.5%), sinusitis (27.3%), sepsis (13.2%), skin or gastrointestinal infections (10.5%), and septic arthritis (2.6%). No cases of meningitis, acute lymphadenitis, or osteomyelitis were observed. Non-infectious complications were reported in 21 patients (55.3%), most commonly bronchiectasis (34.2%) and respiratory allergy (31.6%); 1 patient developed malignancy (2.6%) (Fig. 1). Overall survival at 8 years of follow-up was 71.1%. Among the eleven deceased patients (28.9%), 1 death (0.11%) was attributed to complications of major depressive disorder; causes of death for the remaining patients were not recorded. Discussion X-linked agammaglobulinemia (XLA) remains a diagnostic challenge and requires a high index of clinical suspicion, particularly during early infancy. A careful integration of family history, clinical course, and physical examination is essential, especially in male infants presenting with recurrent upper and lower respiratory tract infections, in association with absent or hypoplastic tonsils, lingual nodules, and poorly developed or absent peripheral lymphoid tissue. These findings constitute well-established warning signs that should prompt early immunological evaluation.12 In the present cohort, the median age at diagnosis was 16 months, reflecting a substantial diagnostic delay. This age is higher than that reported in several international series, including cohorts from Italy and other Latin American countries, such as Peru and Argentina.13,14 Delayed diagnosis in our population likely reflects limited awareness of inborn errors of immunity among primary care providers, as well as restricted access to specialized clinical immunology services in many regions of Colombia. These findings underscore the urgent need to strengthen educational strategies for frontline healthcare professionals and to expand access to immunology referral networks nationwide. Most patients in our cohort fulfilled classical diagnostic criteria for XLA and exhibited the expected infectious phenotype, dominated by recurrent respiratory tract infections, pneumonia, otitis media, septic arthritis, and sepsis, in line with previous reports.15 The predominance of encapsulated bacterial pathogens and respiratory viruses at diagnosis is consistent with the fundamental defect in humoral immunity characteristic of XLA. Interestingly, cutaneous and gastrointestinal manifestations were relatively uncommon in our cohort, contrasting with higher frequencies reported in other series.1 This discrepancy may reflect differences in environmental exposure, microbiological surveillance, or referral patterns and warrants further investigation. Following initiation of regular immunoglobulin replacement therapy, a marked reduction in the frequency and severity of infections was observed, confirming its central role in disease management. Nevertheless, persistent sino-respiratory infections of lower severity continued to occur in some patients, as described in other long-term follow-up studies.15 Importantly, a significant proportion of patients developed bronchiectasis, a complication most plausibly attributable to delayed diagnosis and delayed initiation of immunoglobulin replacement, resulting in cumulative inflammatory and infectious lung damage. These findings highlight the critical importance of early diagnosis and prompt initiation of therapy to prevent irreversible organ damage. The low incidence of malignancy observed in our cohort aligns with previously reported rates ranging from 1.5% to 6%.16 Notably, no patient experienced adverse events related to live oral poliovirus vaccination prior to diagnosis, despite widespread administration in this cohort. This observation is particularly relevant in the Colombian context, where the national immunization program historically relied on oral poliovirus vaccines. The gradual transition to inactivated poliovirus vaccine (IPV), officially implemented in 2022, represents an important public health advance for patients with underlying immunodeficiencies. Overall survival in our cohort was 71% at a median follow-up of 8 years, which, while comparable to historical data, remains suboptimal. Loss to follow-up due to administrative barriers within the Colombian healthcare system represents a significant limitation and likely contributes to underestimation of morbidity and mortality. Beyond diagnostic delay, additional factors influencing outcomes include delayed or suboptimal immunoglobulin replacement and the development of immune dysregulation or infectious sequelae despite adequate therapy.17,18 Although immunoglobulin replacement remains the cornerstone of treatment, emerging therapeutic strategies continue to evolve. Hematopoietic stem cell transplantation has been considered in selected cases, particularly in patients with severe, refractory infections or associated malignancy, and is most effective when performed before irreversible organ damage occurs.18 Gene therapy for XLA is currently under investigation and represents a promising future approach.16,19 From a genetic perspective, our findings reinforce the marked mutational heterogeneity of the BTK gene. Consistent with previous studies, missense variants predominated, particularly within functionally critical domains such as Protein Kinase and Src homology 3, underscoring how subtle structural alterations can profoundly impair BTK function.20 Notably, over half of the variants identified in our cohort had not been previously reported, suggesting the presence of population-specific variants or potential founder effects within Latin America. Furthermore, as described by Chin et al. we observed considerable phenotypic variability associated with identical pathogenic variants, highlighting the importance of integrating genetic findings with clinical and immunological data to achieve accurate diagnosis and prognostication.21 The recognition of atypical XLA phenotypes, characterized by residual or aberrant BTK expression and near-normal B-cell counts, further complicates diagnostic evaluation and may delay genetic confirmation.14,15,18 This underscores the need for comprehensive diagnostic algorithms and broader access to molecular testing. In summary, this study provides novel and clinically relevant data from a historically underrepresented population. Our findings emphasize the need for regional epidemiological and genetic studies to refine diagnostic strategies, optimize patient outcomes, and contribute to a more comprehensive global understanding of X-linked agammaglobulinemia. This study has several key limitations. Its cross-sectional, registry-based design precludes causal inference, and the small sample size limits generalizability. Missing and heterogeneous data introduce potential information bias, while recruitment from referral centers may overrepresent more severe cases. Incomplete follow-up data, including undocumented causes of death, affect the accuracy of survival estimates, and the lack of genotype–phenotype correlation and functional validation limits interpretation of genetic findings. Despite these limitations, the study has important strengths, including being one of the largest multicenter characterizations of X-linked agammaglobulinemia in Colombia, providing valuable data from a previously underrepresented population. The inclusion of multiple referral and academic centers enhances cohort diversity and supports a broader national perspective, while the integration of clinical, immunological, and genetic data allows for a comprehensive disease characterization. Furthermore, the high rate of genetic testing and the identification of numerous novel BTK variants contribute meaningful insights to the global understanding of XLA genetic heterogeneity, and the use of standardized data collection strengthens the internal consistency of the findings. Confirmation of unpublished work This manuscript is original, it has not been published and is not under consideration for publication elsewhere. Disclosure statement regarding the use of artificial intelligence (AI) and AI-assisted technologies Nothing to disclose. Author contributions All authors listed in this manuscript have contributed significantly to the work and meet the ICMJE criteria for authorship. Specifically, M.O·H., J.T.A., O.A.V., and V·C.G. were involved in the conception and design of the study. M.O.H., J.T.A., O.A.V., V·C.G., J.A.P., H.M.P., D.M.V., A.A.F., L.M.C.J., N·V.T., A.F.Z.V., M.C·O.L., C·O., Y.A.M., L.R.A., J.L.F., J.C·O., E.V., C·O.G., I.M., M.L.V., C.G., C·O., M.A., A.F.A., P.M.P., and O·C.J. contributed to the acquisition of clinical and genetic data. M.O.H., J.T.A., O.A.V., and V·C.G contributed to the analysis and interpretation of the results. The manuscript was drafted by M.O·H., J.T.A., O.A.V., and V·C.G and was critically revised for important intellectual content by all co-authors. Finally, all authors have given their final approval of the version to be submitted and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Ethics approval and consent to participate Approved by the Biomedical Research Ethics Committee of Fundación Valle del Lili (approval number 2021.1847). Consent for publication The authors consent to the publication of this article in World Allergy Organization Journal, as above. Data availability The datasets used during the current study are available from the corresponding author on reasonable request. Funding This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. Conflicts of interest The authors declare that they have no competing interest. Footnotes Full list of author information is available at the end of the article References Associated Data Data Availability Statement The datasets used during the current study are available from the corresponding author on reasonable request.