Neuropathy in X-linked agammaglobulinemia: A systematic review and meta-analysis

Literature search

A systematic review and meta-analysis were conducted according to the updated 2020 Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines [Figure 1]. The protocol of the study had been registered in the International Prospective Register of Systematic Reviews (PROSPERO) database (Registration number: CRD42024614529). Medical Subject Headings (Mesh) terms and keywords in combination using Boolean operators like “OR” or “AND” were used to search relevant studies in the PubMed database. Of the 436 articles identified through the literature search using (((X linked agammaglobulinemia) AND (Case)) OR (Bruton’s agammaglobulinemia)) AND (Case), we excluded 383 articles because they did not involve nervous system. Out of the remaining 53 articles, one was a case of hyper IgM syndrome, one was a study performed on animals, and one was a review article. These three articles were also excluded, and finally, 50 articles were included in this study, which consisted of 45 case reports, three case series, and two original studies of XLA with neuropathies.

Close

Figure 1:

Preferred reporting items for systematic reviews and meta-analyses flow diagram for neuropathies in X-linked agammaglobulinemia (XLA).

Export to PPT

Inclusion criteria for articles were (1) case reports or case series involving patients who were new cases diagnosed to be XLA with involvement of nervous system, (2) case reports or case series involving patients who were known cases of XLA with involvement of nervous system, and (3) original articles with cohort or cross-sectional study for a case of XLA with neuropathies. We excluded articles that (1) had no cases with neuropathies, (2) had no cases with XLA, (3) were review articles or systematic reviews, and (4) were non-human studies.

Grading and assessment

For assessing the risk of selection and measurement bias for each included article and categorizing each article as having a low, moderate, serious, or critical risk of bias, we used a tool proposed by Murad et al. Different questions are employed in this tool to assess the selection, determination, follow-up, and level of detail in case reports and case series.^[5,6] To determine the certainty of evidence in each article, categorizing as high, moderate, low, or very low, we used GRADE assessment. Case reports, case series, cohort, and cross-sectional studies could be assessed by these methods. There was no control for bias or certainty in the data analysis.^[6,7]

Data collection and analysis

For case reports, case series, and original studies, the data were extracted, listed together, and merged to form a single combined dataset to determine the frequency of each variable of interest after omitting cases with absent data for that variable. Details elicited encompassed total number of cases of XLA with neuropathy in each article, patient’s demographics (age at presentation, sex), age at onset of symptoms, age at diagnosis of XLA, which neuropathy was present, type of genetic mutation identified, information about pedigree analysis if done, laboratory parameters, cerebrospinal fluid (CSF) findings, treatment given and outcome. For the meta-analysis of reported parameters across individual case studies, we implemented the following statistical approach. Since primary studies often lacked reported measures of dispersion, we calculated a pooled standard deviation (SD) from all available data points to estimate variability. For studies where individual SDs were unavailable, we assumed the standard error to equal the pooled SD, consistent with methodologies for single-case estimates (n = 1 per study). The lower confidence interval (CI) bounds were truncated at zero to maintain clinical plausibility wherever applicable. This adjustment preserved statistical validity while preventing biologically impossible negative values.

Disclosure of ethical statements

This systematic review and meta-analysis have been registered in the PROSPERO database and registration number: CRD42024614529. The authors confirm that we have adhered to the ethical policies of the journal. No identification details of the patients are shared.

Neural manifestations

Multiple neurological diseases may be identified in a case of XLA. The most common among them is meningoencephalitis. Bacterial meningitis most commonly pneumococcal meningitis or encephalitis, can sometimes be the first sign of XLA, during which clinicians might easily miss any underlying immunodeficiency.^[14-17] Encephalitis may be caused due to viruses, commonly herpes simplex, varicella-zoster, Epstein Barr, mumps, measles, astrovirus, enteroviruses such as polioviruses, echoviruses, and coxsackieviruses A and B; bacteria; fungi; and parasites.^[18]

The main function of cell-mediated immunity (CMI) in protecting the host against the majority of viral infections is widely acknowledged. Enteroviruses are well known for targeting these patients because of their weakened mucosal immune response, and notably, the primary defense mechanism against these viruses appears to be neutralizing antibodies. In addition, the inability of BTK-deficient dendritic cells to effectively recognize viral single stranded ribo-nucleic acid (ssRNA) through toll-like receptor 8 is thought to contribute to the vulnerability of XLA patients to enteroviral infections. Given that individuals with XLA do not produce secretory Igs, one might speculate that enteroviruses could persist for a longer duration and potentially spread to the central nervous system (CNS). While the occurrence of enteroviral meningoencephalitis in XLA ranges from 1% to 5%, the risk of death is considerably elevated. Specifically, individuals with XLA have an increased susceptibility to vaccine-associated paralytic poliomyelitis (VAPP) and chronic enteroviral meningoencephalitis (CEME).^[15,19-24]

The dermatomyositis-like syndrome (DLS), which seems to be linked with the spread of enteroviruses, typically echovirus and occasionally coxsackievirus, is a characteristic feature associated with CEME.^[8,9,25] DLS is marked by peripheral edema, erythematous rashes, and evidence of inflammation in skin and muscle biopsy specimens. Some patients with viral dissemination have only edema or myositis, and fasciitis.^[8,9] Echovirus encephalitis/myositis can also be associated with XLA.^[26,27] Atypical presentations such as associations with fibromatosis, traumatic neuromas, or space-occupying lesions extending intracranially may also be seen.^[15]

To confirm CEME, it is essential to isolate an enterovirus, ideally from the CSF. The virus might not be found in the CSF until symptoms begin to appear and abnormalities in the CSF are noted.^[8,28] The examination of the brain biopsy showed an inflammatory reaction affecting both the cortex and white matter, accompanied by gliosis, progressive loss of neurons, neuronophagia, and an increase in microglial cells. The cortical neurons might remain fairly intact, but prominent lymphocytic infiltration accompanied by astrocytosis is typically observed. Lymphocytic infiltration around the blood vessels is seen in the white matter; however, there is no evident primary degeneration of the myelin sheaths. Immunohistochemistry (IHC) is required for the detection of viral antigens, including enterovirus 71, coxsackieviruses B1 and 3, poliovirus, John Cunningham virus (JCV), and lymphocytic choriomeningitis virus.^[19,29] High-dose IVIG is the first line of treatment for CEME, also in cases of regular IVIG replacement therapy, with intraventricular Ig being reserved for those who deteriorate while on IVIG.^[21,30,31] Fluoxetine is a potential therapy for CEME, but further research is necessary.^[32] Ribavirin has shown promising activity in vitro against enteroviruses and crosses the blood–brain barrier after oral administration.^[21] The failure to raise IgG levels in CSF results in the virus remaining in the ventricular system or being trapped in the brain tissue beneath the cortical surface, which can lead to deadly echovirus encephalitis while we wait for the development of new antiviral treatments.^[33] In instances where all these treatments have failed, intrathecal and intraventricular IFNventricular for CEME resulting from echovirus type 6 has been employed as well.^[34]

Isolates from iVDPV showed predominance of types 2 (64%), followed by type 1 (21%) and type 3 (15%) serotype distribution. By contrast, VAPP in immunocompetent oral poliovirus vaccine (OPV) recipients and household contacts is most frequently associated with type 3 (71%), followed by type 2 (26%).^[4,12,35] VAPP in a male patient receiving OPV was triggered by a recombinant strain of Sabin type 3 and type 1, who was subsequently discovered to have a novel mutation in the BTK.^[22] Since OPV is given at birth, when many PIDs are often difficult to detect, nearly all individuals with immunodeficiencies have received the OPV vaccination. Individuals with immunodeficiency who receive OPV may be at risk of becoming long-term excretors of iVDPV. This presents a potential danger of both experiencing paralysis and reintroducing reverted virulent poliovirus into the community once polio vaccination has ceased. Transitioning to a strategy that exclusively uses inactivated poliovirus vaccine (IPV) or employing a combination of IPV and OPV will diminish the likelihood of VAPP in the future.^[22] In exceptional circumstances, a patient with XLA might endure a paralytic enteroviral infection (most probably wild polio) with complications.^[36] Viral shedding in feces can persist for several months and may stop after the administration of IVIG.^[37] Pocapavir ought to be regarded as a potentially safe and effective treatment option for individuals with iVDPV infection.^[38]

The range of causative agents linked to encephalitis is broadening, emphasizing the significance of unbiased molecular technology as an essential resource for the differential diagnosis of unexplained CNS diseases in XLA. Even with the application of different diagnostic techniques such as serology, culture, IHC, or molecular methods, a specific causative factor remains undetermined in many cases of encephalitis. This diagnostic failure could indicate that a known agent or its molecular signature was not present at the time of sampling, that the handling of the specimen was not optimal, that the assay lacks sensitivity, or that there is an unforeseen or new agent that was not accounted for in standard assays. A deeper comprehension of new and reemerging pathogens associated with encephalitis outbreaks, such as West Nile virus, Hendra virus, Nipah virus, Australian bat lyssavirus, and enterovirus 71, highlights an immediate requirement for innovative tools aimed at quick differential diagnostic testing and monitoring. Often, it can be difficult to differentiate between infectious diseases and progressive encephalitis of uncertain origin. The histological results will align with viral infection, and since molecular, serological, and morphological techniques do not detect an infectious agent, it is essential to conduct an unbiased pyrosequencing of RNA, polymerase chain reaction amplification, next-generation sequencing, and antigen detection from the brain biopsy sample which can help in identifying rare pathogens such as astrovirus and Cache Valley virus.^[18,39-41] There are instances when no organisms are identified even after all the investigations, resulting in progressive non-infectious encephalitis.^[29]

Patients with XLA may also exhibit PML, which is caused by the human polyomavirus, JCV.^[40] It is commonly observed in individuals experiencing depression of CMI, but there have been few reports of its occurrence in situations involving unusual humoral immunity, such as in XLA. Medical treatments such as IFN-α and cytarabine have been tried for PML, but none have demonstrated effectiveness, while we wait for a new medical treatment to be developed.^[42] The prognosis of CNS infections in XLA is not solely dependent on Ig levels, and these infections are not necessarily progressive or lethal. The outcome of meningoencephalitis in XLA may be influenced by the interplay between the host’s immune response and the virulence of the responsible virus.^[43,44] Progressive neurodegenerative disorders with no causative microorganisms are a rare CNS complication in XLA.^[45]

Uncommon instances of XLA linked to isolated GH deficiency exhibited proper linear growth while receiving GH replacement therapy, yet there was no indication of correction of the humoral immune defect.^[9] A recently identified mutation in the myeloid elf-1–like factor could be the cause of the disease. These individuals might experience growth and puberty delays and need to be differentiated from those with XLA who have growth issues due to malnutrition.^[10] Some cases manifested either a worsening myelopathy or an encephalopathy, and occasionally, a combination of encephalomyelopathic symptoms. This could indicate the emergence of a progressively fatal neurodegenerative disorder or a progressive, lethal DPS.^[46] The cause of progressive neurodegeneration in individuals with XLA remains uncertain, though there has been speculation about a connection to long-term use of IVIG.^[47-49] It is also suggested that the tyrosine kinase domain at the end of BTK may influence the role of BTK protein in the CNS and that alterations in this region might be associated with these cases. In upcoming research, it will be crucial to explore the possible correlation between genotype and phenotype in a larger cohort of XLA patients receiving IVIG, including those with and without neurodegenerative conditions.^[47]

A limited number of XLA cases have shown characteristics similar to MTS, which is also referred to as deafness-dystonia-optic atrophy neuronopathy or MTS/XLA contiguous gene syndrome. This condition results from mutations in the DDP1/TIMM8A gene, which plays a role in the mitochondrial transport of metabolites. The close proximity of BTK and TIMM8A, at 770 base pairs, accounts for the likelihood that substantial deletions in this area can lead to this syndrome. Nonetheless, the majority of BTK mutations consist of missense, non-sense, splice site changes, and small insertion or deletion events, with large deletions representing only about 3.5% of XLA.^[11,50,51] One of the case reports outlines the occurrence of von Recklinghausen disease, also referred to as neurofibromatosis type 1, alongside XLA. Since there was no known family history of tumors, it is possible that mutations in this gene arose de novo rather than being inherited from the parents. CMI plays a crucial role in preventing tumor growth, and the impact of B-cell deficiency on tumor immunity requires further investigation. It remains unclear whether the simultaneous presence of these two factors is merely coincidental or if there is an underlying cause-and-effect relationship.^[52] There was a case presenting with autism spectrum disorder, which is a complicated neurodevelopmental condition with increasing evidence pointing to genetic factors and immune system irregularities. It is linked to dysregulation of the immune system, autoimmune reactions, and lower serum concentrations of Igs.^[53]