Preprint
Article
Individual progression of S100 calcium-binding protein beta as a surrogate for epilepsy risk: rationale and design of a meta-analysis project
Altmetrics
Downloads
115
Views
35
Comments
0
This version is not peer-reviewed
Advances in Diagnostics, Brain Delivery Systems and Therapeutics of Neurodegenerative Disease
Submitted:
25 March 2024
Posted:
27 March 2024
You are already at the latest version
Alerts
Abstract
Introduction: This study aims to elucidate the clinical diagnostic value of S100 Calcium Binding Protein beta (S100B) in epilepsy, a topic of ongoing debate among European and Asian epilepsy research centres.Methods and analysis: A structured long-term research plan was developed, comprising three phases. Phase one involves observational studies investigating the correlation between S100B levels in serum or cerebrospinal fluid and epilepsy. Phase two integrates observational and randomised controlled data to establish baseline S100B levels across diverse epilepsy demographics, underpinning the predictive value of S100B for epilepsy. Phase three focuses on randomised controlled trials to evaluate the diagnostic accuracy of S100B as an epilepsy biomarker.Ethics and dissemination: All enrolled patients will provide written consent. The ethics committee of the Second People’s Hospital of Gansu Province, Affiliated Hospital of Northwest Minzu University, China, approved the study (GSSEY2022-KY009-01).Registration: The study is registered with PROSPERO (CRD42023425431).
Keywords:
Subject: Medicine and Pharmacology - Neuroscience and Neurology
Strengths and Limitations of This Study:
Strengths:
- Longitudinal, multi-phase study design enhances the robustness of findings.
- Integration of observational and randomised controlled data provides comprehensive insights.
- A collaborative approach with authors of existing research ensures a rich data pool.
- Limitations:
- Potential selection bias in the initial phase of observational studies.
- Variability in S100B measurement techniques may affect data standardisation.
1. Introduction
The gene encoding S100B is on 21q22.3, which changes gene expression(1) associated with neurological diseases (Figure 1), such as epilepsy (www.genecards.org). The expression level of S100B can now be detected by serum or cerebrospinal fluid. In studying brain injury biomarkers, S100B is often considered a significant factor associated with neurological diseases such as epilepsy(2) and melanoma(3)(www.genecards.org). Among these findings, S100B was considered a potential predictive marker for epilepsy. However, these findings did not reach the endpoint. Is the result that the expression level of S100B is associated with epilepsy reliable? What kind of baseline can S100B establish as a diagnostic marker for epilepsy? How does the S100B level serve as an evaluation for treating epilepsy in one day, week, or month? The debate on the relevance of S100B expression levels to epilepsy will end in this long-term research program.
In most of the research results, S100B has a higher expression level in the serum or cerebrospinal fluid of epileptic patients than in ordinary people. This clear expression level trend provides the basis for establishing baseline standards. In the future, the clinical diagnosis of epilepsy can be based on the level of serum S100B to determine the clinical endpoint, and doctors can also use the above criteria to predict the onset of epilepsy in the future. The current evidence cannot prove that clinical practitioners can use the level of S100B as an evaluation strategy for the diagnosis or prognosis of epilepsy. Before this, two problems must be solved. First, we must determine whether a quantitative association exists between S100B progression and event risk across populations and interventions. Second, the link between treatment effect changes in S100B and changes in clinical endpoints must be significant in the intervention trial. To our knowledge, publications have yet to fill these gaps in evidence.
In clinical practice, it is impossible for a single marker to answer such questions. According to the report, the S100B level before and after the seizure is different within a few hours (1h, 3h, 8h, 24h, 48h, 72h), so the variability of the progression of the S100B level is substantial. Another limitation is the observation conditions. S100B level inspection is currently not a routine detection item for hospital epilepsy. Only clinical research on S100B can undergo S100B detection. To study these problems, we have developed a questionnaire and sent emails to the teams that have made reports on epilepsy related to the S100B level. The couples willing to provide detailed clinical information will become our cooperation group. This mode of investigation provides us with potentially significant and clinically complicated samples for objective evidence.
2. Methods
2.1. Study Design and Study Population
The ethics committee of the Second People's Hospital of Gansu Province, Affiliated Hospital of Northwest Minzu University, China, approved the study (GSSEY2022-KY009-01). PROG-S100B is a multinational, multi-centre project (registration: https://www.crd.york.ac.uk/PROSPERO/, CRD-ID: CRD42023425431), aiming to integrate future research from large data centre S100B The individual data from the study were combined to answer a series of questions and to resolve the debate about whether S100B levels have clinical value in epilepsy. We tried to contact the heads of epilepsy research centres worldwide researching S100B and epilepsy-related epilepsy to cooperate more efficiently to complete this research. Investigators are collaborating to build an extensive S100B database. With the dynamic operation of bibliometrics, new reports on S100B and epilepsy-related research will gradually appear, so this project will incorporate more data sources.
2.2. Patient Enrollment and Study Conduction
The first phase mainly consists of observational studies focusing on resolving the debate on the correlation between S100B levels in serum or cerebrospinal fluid and epilepsy. The second phase uses observational and randomised controlled research data, focusing on solving the baseline standard of S100B level and epilepsy population with different characteristics and providing a research basis for the predictive value of S100B level on epilepsy. The third phase is mainly based on randomised controlled studies, focusing on the accuracy of S100B as a biomarker for diagnosing epilepsy. Regarding the three-stage tasks of studying epilepsy and S100B, the series of problems we need to solve are as follows:
- (1)
- Are S100B concentration levels in serum or CSF associated with epilepsy (significantly)?
- (2)
- What is the range of baseline levels of S100B concentrations in serum or cerebrospinal fluid in different types of healthy people?
- (3)
- What is the range of baseline levels of S100B concentration levels in serum or cerebrospinal fluid in epilepsy patients with different characteristics or high-risk epilepsy groups?
- (4)
- How can we transform (2) and (3) into a set of feasible solutions to provide predictive value for clinical epilepsy patients?
- (5)
- Do S100B concentration levels in serum or CSF serve (significantly) as a diagnostic biomarker for epilepsy?
- (6)
- Can S100B concentration in serum or CSF be quantified as a prognostic marker for epilepsy treatment?
To solve the above six problems in the three-stage tasks, we mainly searched the original research on epilepsy-related S100B in PubMed (Supplementary Table S1). We searched for the contact information of the person in charge of the epilepsy research centre of epilepsy-related S100B worldwide through the search results. As well as the company that produces the S100B detection kit, if necessary, it is essential to search the reference list of the above search results manually.
For the multi-centre, multi-country, large-sample meta-analysis, the best strategy for collecting data is to find the heads of epilepsy research centres that study S100B and then convince these heads to cooperate with us, which can avoid extracting data from published articles. It can also become a research model for meta-analysis and systematic review. In addition, the epilepsy research team that studies S100B-related epilepsy has advantages over recruiting epilepsy research teams. These advantages are mainly reflected in the fact that it is easier to recruit epilepsy patients and has more weight in academic research. Once the person in charge is willing to cooperate with us, we will nominate the person in the direction to enter the program-S100B project after signing documents such as informed consent.
The study had been run continuously for one year before the publication of this study protocol. We have developed inclusion criteria (Table 1), but slight changes may be made when the project is run. After the PubMed search, we made the search results of epilepsy and biomarkers into a visualised chart(4), which is convenient for understanding the latest research trends in the world, and the 37 research results that meet our project have been compiled into the list of research that planned to be recruited (Supplementary Data).
2.3. Analysis Strategy for Stage 1
We have found that etiologies of epilepsy include but are not limited to stroke, tumour, and genetics. Subjects suffering from epilepsy must meet the diagnostic criteria of the World Health Organization (WHO) or the International Society Against Epilepsy or be excluded from the first stage analysis. The clinical endpoint event was seizure or non-seizure, and the starting point event was a diagnosis of epilepsy, evaluation of treatment effect and prediction of epilepsy recurrence. The studies that were included used different definitions of clinical thresholds. We will select variables with clinical starting points as consistently as possible to reduce heterogeneity.
Our research topic is whether there is a correlation between epilepsy and the concentration of S100B. Seizures are more accessible to judge because of convulsions, and the most critical observation index is the S100B level (serum or cerebrospinal fluid), so the S100B level will be used as the primary target for data extraction. Due to the small number of variables, explaining the differences in S100B levels in different populations is challenging. Age and region may be used as correction variables to avoid S100B measurement errors, and we can use linear regression to estimate the progress of individual S100B levels. We then reanalysed the comparisons based on estimates of the mean difference between the epileptic and non-epileptic groups, correcting for population bias.
2.4. Analysis Strategy for Stage 2
The project's second phase continues the observational study of the first phase and adds high-risk groups for epilepsy. If the survey recruits a large enough sample, the baseline criteria for S100B in different high-risk groups will be confirmed. Inclusion criteria may be lowered if fewer people are recruited (Table 1).
2.5. Analysis Strategy for Stage 3
Clinical trials measuring S100B have more EEG, and S100B examinations are performed more frequently. The regression model was used to test the confounding factors, and the correlation of S100B in different risk groups was analysed. The extensive sample data is based on the previously determined S100B concentration baseline interval to determine the breakpoint value. Taking the clinician's diagnosis as the gold standard and the breakpoint value of S100B concentration as the pseudo-diagnosis, the true positive, false positive, true negative and false negative of epilepsy were deduced. The sample sizes of the four characteristics were stratified according to the studies, and joint tabulations were made to calculate the sensitivity and 95% confidence interval of S100B for diagnosing epilepsy.
For the treatment experiments, we used the model of van Houwelingen (5,6). For diagnostic investigations, we use a model that balances accuracy and two-dimensionality (7,8). Regardless of treatment experiments or diagnostic tests, it is easier to obtain routine data processing results such as mean logit sensitivity, mean logit specificity, standard error, and 95% confidence interval. It is also necessary to calculate variability and covariance to produce visual effects. After hard work, the total receiver operating curve (ROC) of the S100B stratification study is presented in an elliptical contour diagram.
Heterogeneity across study centres was assessed using I². We used funnel plots and asymmetric regression tests to rule out publication bias. Only one author was assessed to assess methodological bias.
2.6. Statistical Calculations
Obtaining quantitative baseline data and confirming associations between S100B and subgroups in large high-risk epilepsy populations is complicated. The following planned data analysis strategies are necessary, but unplanned ones may appear in the actual process. Due to the study design in the enacted recruitment strategy as observational and randomised controlled studies, the data we obtained may be two-arm or single-arm.
We will use Stata's "meta" command package to perform random effects meta-analysis using inverse variance weighting. Categorical data were used to calculate risk ratios (RRs) using dichotomous or dummy variable methods, and the rest of the data was transformed into continuous data to calculate the mean and standard deviation.
We then employed a Cox regression model, utilising the method of moments approach of DerSimonian and Laird (9), to study the influence of epileptic disorder characteristics on S100B levels. To eliminate the potential effect of time on S100B levels, period variables such as years of epilepsy diagnosis, seizure frequency, and the interval between seizures and detection of S100B were used as covariates. Measuring principles and kit companies for S100B levels were considered potential sources of heterogeneity, and these variables would be appropriately added to the model. In recruiting patients in each research centre, we subjectively believe that age and gender are manipulated and may not be the primary source of heterogeneity at the S100B population level. However, considering the characteristics of high-risk groups of epilepsy, we are worried that age and gender factors are the epidemiological factors of epilepsy, so we have to add them to the model. There may be more potential confounding factors that have yet to be considered, and we can only add them if we have accurate data from each centre.
3. Discussion
3.1. Annual Trend Analysis of Research Trends on Biomarkers of Epilepsy
1767 works of literature were included in this survey of epilepsy-related biomarker research trends (Supplementary Table S1). From Supplementary Figure S1, we can see that the overall number of published papers is on the rise, and relevant research articles have risen sharply since 2017, reaching a peak (177) in 2020. In the past few years, from 2017 to 2022, the number of published articles has remained relatively high.
3.2. Spatial-Temporal Distribution and Research Strength of Research Trends on Biomarkers of Epilepsy
After statistical analysis, the top ten authors and their affiliated research institutions are in Supplementary Table S2. Among them, Dr Orrin Devinsky, whose institution is New York University, has the highest number of publications in the literature, with 24 journals.
According to the analysis results of Citespace (Figure 2), the author's graphs of literature are scattered, and there are few connections. The number of nodes for authors and research institutions is 1046, the number of connections is 1665, and the density is only 0.003. In the author analysis, the centrality of each node is 0, indicating that the author is not closely related. They have mainly formed a team led by Orrin Devinsky.
3.3. Hotspots and Frontiers of Research on Biomarkers of Epilepsy
In this study, the keywords in the pieces of literature were statistically analysed and sorted according to frequency. The final top ten keywords are shown in Supplementary Table S3. The top ten keywords with frequency include "temporal lobe epilepsy", "epilepsy surgery", "high-frequency oscillation", "antiepileptic drug", "drug-resistant epilepsy", "oxidative stress", "hippocampal sclerosis", "high-frequency oscillation", "magnetic resonance imaging", "machine learning".
The research trend map of biomarkers of epilepsy diseases drawn through keyword co-occurrence analysis (Figure 3) shows that the larger the size of the node, the greater the weight of the keyword and the more times it appears. The top 10 keywords of document centrality are "temporal lobe epilepsy", "antiepileptic drug", "drug-resistant epilepsy", "epilepsy surgery", "magnetic resonance imaging", "hippocampal sclerosis", "oxidative stress", "machine learning", "autoimmune encephalitis", "heart rate variability". The figure shows that the connections between keyword nodes are crisscrossed, indicating that these are closely related.
We can find research hotspots in the current field through keyword clustering network analysis. CiteSpace software can carry out K-means clustering analysis on literature keywords to obtain clustering analysis, respectively (as shown in Figure 4). Among them, the number size of the cluster is inversely proportional to the range of the bunch. It can be seen from the figure that the research topics related to treatment since 1981, for #0 epileptogenic zone, #1 blood serum, #2 hippocampal sclerosis, #3 cerebrospinal fluid marker, #4 antiepileptic drug, #5 oxidative stress, #6 methylation status. From the clustering, we can notice that the current common attributes are mainly concentrated in cerebrospinal fluid markers.
Supplementary Materials
The following supporting information can be downloaded at the website of this paper posted on Preprints.org.
Author Contributions
All authors should have made substantial contributions to all the following: Jianxun He has the conception and design of the study and is responsible for writing the legal and ethical documents such as the informed consent form. Beilei Li is responsible for managing the revenue and expenditure of this study to provide sound financial decisions and maintain the regular operation of the study. Chuhan Miao has drafted, and Jianxun He reviewed the article.
Funding
Jianxun supports this research.
Institutional Review Board Statement
All enrolled patients will provide written consent. The ethics committee of the Second People’s Hospital of Gansu Province, Affiliated Hospital of Northwest Minzu University, China, approved the study (GSSEY2022-KY009-01). The meta-analysis is registered at https://www.crd.york.ac.uk/PROSPERO/ (CRD42023425431).
Acknowledgments
The authors acknowledge the data processing support of Professor Ma Li Yang and Dr Zhao Fei from the Medicine of Northwest Minzu University in Lanzhou, China. Meanwhile, we thank the Guangming District People's Court of Shenzhen, China, for its legal and regulatory support. Moreover, Dr Jianxun He supports this research. Finally, we sincerely thank the future epilepsy research team that will join our recruitment program.
Conflicts of Interest
Our recruitment program is based on informed consent from both parties, so the authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
References
- Moreira, G.G.; Cantrelle, F.-X.; Quezada, A.; Carvalho, F.S.; Cristóvão, J.S.; Sengupta, U.; Puangmalai, N.; Carapeto, A.P.; Rodrigues, M.S.; Cardoso, I.; et al. Dynamic interactions and Ca2+-binding modulate the holdase-type chaperone activity of S100B preventing tau aggregation and seeding. Nat. Commun. 2021, 12, 1–16. [Google Scholar] [CrossRef] [PubMed]
- Maiti, R.; Mishra, B.R.; Jena, M.; Mishra, A.; Nath, S.; Srinivasan, A. Effect of anti-seizure drugs on serum S100B in patients with focal seizure: a randomised controlled trial. J Neurol. 2018, 265, 2594–2601. [Google Scholar] [CrossRef] [PubMed]
- Harpio, R.; Einarsson, R. S100 proteins as cancer biomarkers with focus on S100B in malignant melanoma. Clin. Biochem. 2004, 37, 512–518. [Google Scholar] [CrossRef]
- Chen, C.; Song, M. Visualising a field of research: A methodology of systematic scientometric reviews. PLoS One 2019, 14, e0223994. [Google Scholar]
- Van Houwelingen, H.C.; Zwinderman, K.H.; Stijnen, T. A bivariate approach to meta-analysis. Statistics in Medicine 1993, 12, 2273–2284. [Google Scholar] [CrossRef]
- van Houwelingen, H.C.; Arends, L.R.; Stijnen, T. Advanced methods in meta-analysis: multivariate approach and meta-regression. Stat. Med. 2002, 21, 589–624. [Google Scholar] [CrossRef] [PubMed]
- Reitsma, J.B.; Glas, A.S.; Rutjes, A.W.; Scholten, R.J.; Bossuyt, P.M.; Zwinderman, A.H. Bivariate analysis of sensitivity and specificity produces informative summary measures in diagnostic reviews. J. Clin. Epidemiology 2005, 58, 982–990. [Google Scholar] [CrossRef] [PubMed]
- Riley, R.D.; Abrams, K.R.; Sutton, A.J.; Lambert, P.C.; Thompson, J.R. Bivariate random-effects meta-analysis and the estimation of between-study correlation. BMC Med Res. Methodol. 2007, 7, 3–3. [Google Scholar] [CrossRef]
- DerSimonian, R.; Laird, N. Meta-analysis in clinical trials. Controlled Clinical Trials 1986, 7, 177–188. [Google Scholar] [CrossRef] [PubMed]
Figure 1.
The top 150 predicted target genes of S100b by using the Multi Experiment Matrix (MEM, http://biit.cs.ut.ee/mem/) website.
Figure 1.
The top 150 predicted target genes of S100b by using the Multi Experiment Matrix (MEM, http://biit.cs.ut.ee/mem/) website.
Figure 2.
Literature author publishing cooperation network.
Figure 3.
Keyword co-occurrence map of research trends in biomarkers of epilepsy.
Figure 4.
Research Trends of Epilepsy Biomarkers Keyword Cluster Analysis.
Table 1.
Inclusion criteria for stages 1 to 3 of the project.
| Stage 1 | Stage 2 | Stage 3 | |
| Prospective longitudinal study design | |||
| Observational study | Randomised controlled trial | ||
|
Investigation of one of the following at-risk populations:
|
||
| Alternatively, population-based samples with a relevant subcohort (at least ten events) in at least one of the above-named at-risk categories | |||
| Well-defined and disclosed inclusion criteria and recruitment strategy | |||
| At least two EEG and S100B serum tests were performed to determine the type of epilepsy and S100B level. | |||
| A minimum of 10 events per endpoint | No minimum rate of endpoint events | ||
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content. |
© 2024 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.
MDPI Initiatives
Important Links
© 2024 MDPI (Basel, Switzerland) unless otherwise stated