Impact of IL-6 receptor inhibitors on cerebrospinal fluid metabolites in type 2 diabetes patients

Abstract

This study aimed to investigate whether interleukin-6 (IL-6) receptor inhibitors affect cerebrospinal fluid (CSF) metabolites in type 2 diabetes (T2D) patients. The initial phase employed drug-targeted Mendelian randomization to investigate the association between IL-6 receptor inhibitors and T2D across diverse populations. Subsequently, variables related to T2D were utilized as mediators in the relationship between IL-6 receptor inhibitors and CSF metabolites. Finally, the mediating effect of T2D was evaluated using mediation analysis. In the European cohort, the use of IL-6 receptor inhibitors was associated with a 25.3% reduction in the risk of T2D, as indicated by an odds ratio of 0.747 with a 95% confidence interval ranging from 0.556 to 0.938 (P = .003). In the East Asian cohort, IL-6 receptor inhibitors were associated with a 1.188-fold increase in the risk of elevated fasting insulin levels, with an odds ratio of 1.188 and a 95% confidence interval of 1.034 to 1.343 (P = .029). Mediation analysis revealed that IL-6 receptor inhibitors significantly elevated the levels of galacto-glycero-lipid (GG) in CSF, with 3.06% of the effect attributable to the reduced risk of T2D, and 96.94% directly resulting from the action of the IL-6 receptor inhibitors. The findings preliminarily indicated that race constituted a variable potentially influencing the differential effects of IL-6 receptor inhibitors among patients with type 2 diabetes (T2D) across various regions. While the administration of IL-6 receptor inhibitors might impact glucose concentrations in CSF, the role of T2D as a mediator in CSF metabolites was weak. The predominant determinant appeared to be the administration of IL-6 receptor inhibitors.

1. Introduction

Type 2 diabetes (T2D) represents a pervasive chronic health issue globally, characterized primarily by insulin resistance and impaired insulin secretion.^[1] Recent studies have demonstrated that chronic subclinical inflammation constitutes a significant contribution to the pathogenesis of T2D. Adipocytes, for instance, contribute to elevated levels of C-reactive protein (CRP) through the secretion of pro-inflammatory cytokines such as tumor necrosis factor alpha (TNF-α) and interleukin-6 (IL-6).^[2] CRP is implicated in the pathophysiology of the vascular system and the development of metabolic syndrome by influencing physiological processes, including endothelial function, vasodilation, and vascular remodeling.^[3,4] These findings underscore the potential benefits of further investigating the interplay between inflammation and T2D.

With the progression of molecular biology, immune-targeted therapies have shown significant efficacy in the treatment of tumors and autoimmune diseases.^[5,6] Compared to traditional immunosuppressive drugs, immune-targeted agents provide the advantages of enhanced efficacy and diminished toxicity. Immune-targeted therapies are categorized into 5 categories based on the specific antibody targets: TNF inhibitors, IL inhibitors, integrin inhibitors, B cell depletion/inhibition agents, and T cell depletion agents.^[7] IL-6 is a pleiotropic cytokine involved in the regulation of a wide array of physiological and pathological processes, notably contributing to inflammation, immune response, metabolism, among other functions.^[8,9] Tocilizumab (TCZ), a humanized monoclonal antibody that targets the interleukin-6 (IL-6) receptor, inhibits the IL-6 mediated inflammatory cascade implicated in the pathogenesis of rheumatoid arthritis (RA).^[10] Clinical trials have demonstrated that TCZ monotherapy exhibits superior efficacy compared to monotherapy with TNF-α inhibitors (TNFi) and methotrexate monotherapy.^[11,12] Exploring the relationship between immune-targeted therapies and T2D, is of significant importance for the development of novel therapeutic agents and the optimization of clinical treatment strategies.

Drug-targeted Mendelian randomization (MR) has been widely employed to identify therapeutic agents for a range of diseases. There is considerable interest in exploring the effects of lipid-lowering drugs on non-cardiovascular diseases.^[13,14] In contrast to traditional 2-sample MR, drug-targeted MR focuses exposure factors on specific target genes, improving the accuracy of confounding variable elimination. However, the causal relationship between IL6 inhibitors and T2D remains unresolved. It is also crucial to acknowledge that T2D is associated with changes in various metabolite levels, which can serve as indicators of physiological alterations. Consequently, this study considered T2D as a mediating factor to explore whether cerebrospinal fluid (CSF) metabolites were influenced in patients with T2D who were treated with IL-6 receptor inhibitors. This research enhanced our understanding of the interplay between IL-6 receptor inhibitors, T2D and CSF metabolites, offering valuable insights for clinicians.

2. Materials and methods

2.1. Research process

This study was structured into 3 phases. The initial phase aimed to examine the potential association between IL-6 receptor inhibitors and outcomes related to T2D. In the subsequent phase, the study investigated whether T2D-related variables could induce alterations in cerebrospinal fluid metabolites. In the final phase, mediation analysis was employed to assess the mediating role of T2D-related variables on CSF metabolites. Unlike previous research, this study considered the possibility that IL-6 receptor inhibitors could exert differential effects across various racial groups. Based on data size and availability, data from European and East Asian populations were selected for drug-targeted MR analysis, as illustrated in Figure 1.

Figure 1.

Study flow chart. This study was divided into drug-targeted MR analysis and mediation analysis. Drug-targeted MR analysis was used to assess whether IL-6 inhibitors based on CRP level had a causal relationship with T2D. Mediation analysis was used to assess whether T2D performed a mediating effect between IL-6 inhibitors and CSF metabolites. CSF = cerebrospinal fluid, CRP = C-reactive protein, IL = interleukin, IVW = inverse variance weighted, MR = mendelian randomization, SNP = single nucleotide polymorphism, T2D = type 2 diabetes.

2.2. Selection of IL6 related variables

Building on existing literature, we identified the serum CRP dataset (ebi-a-GCST90029070) as an indicator of systemic inflammation to serve as the exposure.^[15] This dataset encompassed 5,75,531 individuals across 2 cohorts. The first cohort comprised 4,27,367 participants of European descent from the UK Biobank. The second cohort, part of the Heart and Aging Research in Genomic Epidemiology (CHARGE) Consortium, included 1,48,164 individuals of European ancestry. This was the largest dataset of genome-wide association studies on CRP to date. Initially, we conducted a screening of this dataset to identify single nucleotide polymorphisms (SNPs) that exhibited genome-wide significant associations (P < 5 × 10⁻⁸) and demonstrated no substantial linkage disequilibrium (r² < 0.3). The information regarding IL-6 genes was based on the Genome Reference Consortium Human Genome Build 37. We selected 14 SNPs located within ± 100 kb of the IL-6 gene from the CRP dataset to serve as instrumental variables.

2.3. Selection of T2D related variables

Rheumatoid arthritis (RA) was selected as the positive control based on existing literature.^[16] To achieve a more comprehensive evaluation of T2D, we incorporated T2D, its complications, and laboratory indicators as variables. We selected datasets with the largest sample sizes from European and East Asian populations for drug-targeted MR analysis. The European dataset comprised 4,90,089 participants and 2,41,67,560 SNPs. The dataset from East Asia included 4,33,540 participants and 1,12,22,507 SNPs. Complications associated with diabetes were diabetic nephropathy, diabetic neuropathy, and proliferative diabetic retinopathy. Laboratory indicators included glycated hemoglobin (HbA1c), fasting glucose, fasting insulin, and 2-hour glucose levels. SNPs were selected based on genome-wide significant associations (P < 5 × 10⁻⁸) and demonstrated no substantial linkage disequilibrium (r² < 0.3). Table 1 presented a detailed overview of the characteristics of T2D related variables.

Table 1.

Features of T2D related variables.

Outcome	Population	Sample size	Case	Control	Year	PMID
Rheumatoid arthritis	European	4,84,598	5427	4,79,171	2021	33959723
Type 2 diabetes	European	4,90,089	38,841	4,51,248	2021	34594039
	East Asian	4,33,540	77,418	3,56,122	2020	32499647
Diabetic nephropathy	European	4,52,280	1032	4,51,248	2021	34594039
	East Asian	1,32,984	220	1,32,764	2021	34594039
PDR	European	2,12,889	8681	2,04,208	2021	–
Diabetic neuropathy	European	1,63,616	1415	1,62,201	2021	–
HbA1c	European	3,89,889	–	–	2018	34017140
	East Asian	2566	2566	0	2020	–
Fasting glucose	European	2,00,622	–	–	2021	34059833
	East Asian	2566	2566	0	2020	–
Fasting insulin	European	1,51,013	–	–	2021	34059833
	East Asian	29,792	–	–	2021	34059833
Two-hour glucose	European	63,396	–	–	2021	34059833
	East Asian	8509	–	–	2021	34059833

HbA1c = glycated hemoglobin, PDR = proliferative diabetic retinopathy, T2D = type 2 diabetes.

2.4. Selection of CSF metabolites related variables

The dataset of CSF metabolites was obtained from 2 distinct longitudinal cohort studies focused on Alzheimer’s disease: the Wisconsin Alzheimer’s Disease Research Center (WADRC) and the Wisconsin Alzheimer’s Disease Prevention Registry (WRAP).^[17,18] To enhance the study’s applicability, only CSF metabolomics from healthy participants were included. The average age of participants was approximately 60 years, with a specific average of 64.7 years in the WADRC cohort and 62.0 years in the WRAP cohort. The gender distribution was comparable across both cohorts, with approximately two-thirds of participants being female (63.2% in WADRC and 66.2% in WRAP). A total of 338 CSF metabolites were collected from 291 unrelated individuals of European descent.^[19] SNPs exhibited genome-wide significant associations (P < 1 × 10⁻⁵) and demonstrated no substantial linkage disequilibrium (r² < 0.3).

2.5. Evaluation of MR results

This study employed R (version 4.3.1) and the “MRCIEU/TwoSampleMR” package for statistical analysis. For the MR analysis, the MR-Egger, weighted median, weighted mode and inverse variance weighted (IVW) methods were used. The results from the IVW method were used as the primary reference.^[20] A relationship between exposure and outcome was inferred when P_IVW was >0.05, or when the beta values from the other methods were in the same direction as the IVW method.^[21] To ensure the robustness of the IVW results, Cochran’s Q test was used to evaluate heterogeneity. A P-value below .05 indicated significant heterogeneity, suggesting potential unreliability. Horizontal pleiotropy was assessed using the MR-Egger intercept, MR-PRESSO, and funnel plot models. Horizontal pleiotropy was considered negligible when the intercept was approximately 0, the P-value from MR-PRESSO exceeded .05, and the SNPs formed a roughly symmetric inverted funnel plot. Under these conditions, it was assumed that the results were unbiased.^[22]

The robustness of MR was assessed using the “leave-one-out” method, which involved sequentially excluding each SNP and calculating the meta-effects of the remaining SNPs. The MR results were considered robust if all SNPs demonstrated effects on the same side of 0. Furthermore, to determine the directionality of causation, reverse MR analysis was conducted in both drug-targeted MR and mediation analysis. The criteria included a P-value from the IVW method of less than .05 at each step, and the P_IVW of a reverse MR needed to be greater than .05. Results exhibiting bidirectional MR (P_IVW < 0.05) were excluded to prevent interference with the mediation effect.

3. Results

3.1. Causal relationship between IL-6 receptor inhibitors and T2D

Initially, we conducted an MR analysis to evaluate IL-6 receptor inhibitors and RA in Europeans. The results indicated that IL-6 receptor inhibitors were associated with a reduced risk of RA, corroborating findings from previous clinical studies.^[23] Subsequently, we investigated the relationship between IL-6 receptor inhibitors and T2D, along with its complications and laboratory indicators, in both European and East Asian populations. Results exhibiting excessively large or small odds ratios (ORs) were excluded from the analysis. Additionally, statistical results with a 95% confidence interval (CI) exceeding 1 were excluded. The MR analysis indicated that, among the European cohort, IL6 inhibitors were associated with a 25.3% reduction in the risk of T2D compared to nonusers (OR: 0.747, 95% CI: 0.556–0.938, P = .003). In the East Asian cohort, IL6 inhibitors were associated with a 1.188-fold risk of elevated fasting insulin (OR: 1.188, 95% CI: 1.034–1.343, P = 0.029), as shown in Table 2.

Table 2.

Association between IL-6 inhibitors and T2D by IVW method.

Outcome	Population	NSNP	OR (95% CI)	P-value
Rheumatoid arthritis	European	14	0.991 (0.985–0.998)	.010*
Type 2 diabetes	European	14	0.747 (0.556–0.938)	.003*
	East Asian	11	1.009 (0.812–1.206)	.929
Diabetic nephropathy	European	14	0.139 (−1.446 to 1.725)	.015*
	East Asian	10	79.098 (76.307–81.889)	.002*
PDR	European	14	0.489 (−0.108 to 1.087)	.019*
Diabetic neuropathy	European	14	0.085 (−1.617 to 1.788)	.005*
HbA1c	European	14	0.949 (0.882–1.015)	.120
	East Asian	12	1.217 (0.480–1.952)	.602
Fasting glucose	European	14	0.964 (0.902–1.025)	.235
	East Asian	12	1.217 (0.480–1.952)	.602
Fasting insulin	European	14	0.941 (0.875–1.007)	.072
	East Asian	14	1.188 (1.034–1.343)	.029*
Two-hour glucose	European	14	1.096 (0.819–1.373)	.516
	East Asian	10	2.019 (0.985–3.053)	.183

CI = confidence interval, HbA1c = glycated hemoglobin, IL = interleukin, OR = odds ratio, PDR = proliferative diabetic retinopathy, SNP = single nucleotide polymorphism site, T2D = type 2 diabetes.

^*P < .05.

We assessed the robustness of the drug-targeted MR results. The Q test yielded a result of P > .05, suggesting no heterogeneity. Subsequently, the MR-Egger intercept was calculated to be 0.0031 for the relationship between IL-6 receptor inhibitors and T2D, and −0.0035 for the association between IL-6 receptor inhibitors and fasting insulin, as shown in Table S1, Supplemental Digital Content 1. The funnel plot exhibited a symmetric distribution of the SNPs, indicating reliability, as illustrated in Figure 2A, B. The sensitivity analysis supported the robustness results, as presented in Figure 2C, D. The findings indicated that IL-6 receptor inhibitors decreased the risk of T2D within the European population while potentially elevating the risk of fasting insulin in the East Asian population.

Figure 2.

The robustness of the drug-targeted MR results. (A) The funnel plot of IL-6 inhibitors and T2D in Europeans. (B) The funnel plot of IL-6 inhibitors and fasting insulin in the East Asians. (C) The sensitivity analysis of IL-6 inhibitors and T2D in Europeans. (D) The sensitivity analysis of IL-6 inhibitors and fasting insulin in the East Asians. IL = interleukin-6, MR = mendelian randomization, T2D = type 2 diabetes.

3.2. Causal relationship between T2D and CSF metabolites

In light of the observed effects of IL-6 receptor inhibitors on T2D and fasting insulin levels, we conducted further investigations to explore potential associations with 338 CSF metabolites. The results indicated 7 metabolites that demonstrated a significant association with T2D. Reverse MR analysis revealed that two of these metabolites were not enriched with SNPs, while the P values for the remaining 5 metabolites exceeded .05, as detailed in Table 3. Fasting insulin was utilized as the exposure variable in the reverse MR analysis. However, due to the enrichment of only 1 SNP, MR analysis was not feasible. We selected 5 metabolites to serve as the outcomes. We intended to conduct further mediation analysis based on these datasets.

Table 3.

Causal relationship between T2D/IL-6 inhibitors and CSF metabolites.

Exposure	Outcome	Forward MR analysis		Reverse MR analysis
		P1	OR1 (95% CI)	P2
T2D	GCST90026327	.014*	0.959 (0.928–0.991)	–
T2D	GCST90026086	.018*	0.951 (0.912–0.991)	.752
T2D	GCST90026107	.009*	0.956 (0.924–0.989)	–
T2D	GCST90026111	.009*	1.097 (1.023–1.177)	.623
T2D	GCST90026314	.011*	0.963 (0.936–0.991)	.768
T2D	GCST90026237	.045*	0.968 (0.938–0.999)	.817
T2D	GCST90026161	.010*	0.961 (0.933–0.990)	.166
IL-6 inhibitors	GCST90026086	.667	–	–
IL-6 inhibitors	GCST90026111	.067	–	–
IL-6 inhibitors	GCST90026314	.751	–	–
IL-6 inhibitors	GCST90026237	.060	–	–
IL-6 inhibitors	GCST90026161	.022*	1.480 (1.144–1.817)	.381

CI = confidence interval, CSF = cerebrospinal fluid, IL = interleukin, MR = mendelian randomization, OR = odds ratio, T2D = type 2 diabetes.

^*P < .05.

3.3. Causal relationship between IL-6 receptor inhibitors and CSF metabolites

Following the determination that T2D could influence CSF metabolites, we used the drug-targeted MR analysis to examine the relationship between IL-6 receptor inhibitors and CSF metabolites. In this analysis, IL-6 was utilized as the exposure factor, while 5 CSF metabolites were used as outcomes. The drug-targeted MR analysis revealed that, within the European cohort, individuals treated with IL-6 receptor inhibitors exhibited a 48.0% risk of elevated galacto-glycero-lipid (GG) levels in CSF (OR: 1.480, 95% CI: 1.144–1.817, P = 0.022). The reverse MR analysis showed no causal association between GG levels and IL-6 receptor inhibitors. IL-6 receptor inhibitors did not exhibit a significant effect on the other 4 metabolites, as shown in Table 3.

3.4. Analysis of mediating effects of type 2 diabetes

Causal relationships were identified among IL-6 receptor inhibitors, T2D and GG levels. We calculated the mediation analysis to assess whether T2D served as a mediator for GG levels during treatment with IL-6 receptor inhibitors. The analysis revealed that the mediation effect was consistent with the direction of the total effect (mediation effect = −0.012, total effect = −0.392), with the mediation effect contributing to 3.06% of the total effect. The data presented in Table 4 indicated that IL-6 receptor inhibitors partially mediated the increase in GG levels. The remaining 96.94% was directly associated with the administration of IL-6 receptor inhibitors.

Table 4.

Mediating effects of T2D.

Exposure	Outcome	Beta	Mediating/direct effect	Effect size	P
IL-6 inhibitors	GCST90026161	−0.392b	−0.012/−0.38	3.06%	.022*
T2D	GCST90026161	−0.040b2	–	–	.009*
IL-6 inhibitors	T2D	0.292b1	–	–	.003*

T2D = type 2 diabetes, IL = interleukin, mediating effects = b1 × b2, direct effect = b − b1 × b2, effect size = b1 × b2/b.

^*P < .05.

4. Discussion

Previous research has identified a significant correlation between chronic inflammation and T2D.^[24] Individuals with T2D exhibited an excessive production of inflammatory stimuli, including acids, reactive oxygen species, and cellular debris, which arose from metabolic dysregulation.^[25] These detrimental stimuli elicited inflammatory responses from immune cells. Additionally, the compromise of the intestinal mucosa and other barriers resulted in an imbalance of the gut microbiota, which subsequently released pro-inflammatory factors into the systemic circulation.^[26] These factors played a crucial role in sustaining chronic inflammation and contributed to the maintenance of chronic inflammation.

IL-6 is a multifunctional cytokine that plays a crucial role in maintaining the homeostasis of the immune microenvironment. Empirical studies have demonstrated elevated serum IL-6 levels in individuals with T2D, which correlated with disease progression.^[27,28] Further research demonstrated that, even after controlling for variables such as insulin resistance and obesity, there were significant elevations in plasma IL-6 levels among T2D patients.^[29] These findings underscored a strong association between IL-6 and T2D. Interestingly, the relationship between IL-6 and T2D was complex and multifaceted. Recent research indicated that IL-6 exerted both pathogenic and protective influences on T2D.^[30] The function of IL-6 was tissue-specific. In adipose tissue, IL-6 exhibited pro-inflammatory properties and promoted insulin resistance.^[31] In skeletal muscle, IL-6 has anti-inflammatory effects, enhancing insulin-mediated glucose uptake and promoting fatty acid oxidation.^[32] Simultaneously, the relationship between IL-6 gene variants and T2D has been reported with conflicting results. The human IL-6 gene is located on chromosome 7p21 and shares a high degree of sequence homology with mouse.^[8] Research exploring the genetic associations between IL-6 polymorphisms and diseases, including T2D, insulin resistance, and metabolic syndrome, has primarily focused on 3 IL-6 promoters: G(-597)A, G(−572)C, G(−174)C.^[33] The IL6 G(−174)C polymorphism demonstrated a significant association with the risk of T2D in the Native American population.^[34] In contrast, studies conducted in Germany and Finland did not reveal any association between the G(−174)C and T2D.^[35] The Asp358 variant has been positively correlated with T2D in the European population (OR:1.30), whereas no such association was observed in the African American population.^[36] These findings suggest that the relationship between IL-6 and T2D is worthy of further study. Additionally, the differences between different races, as well as between humans and laboratory animals, are important factors that require consideration in future research.

This study was characterized by several key elements: the investigation of the relationship between IL-6 receptor inhibitors and T2D across diverse demographic groups. Previous research indicated that the impact of site-specific alterations in T2D varied among ethnicities. Consequently, datasets with the largest sample sizes from European and East Asian populations were selected for drug-targeted MR analysis. Our findings supported the hypothesis that IL-6 receptor inhibitors have different effects on European and Asian populations. The use of IL-6 receptor inhibitors could potentially reduce the risk of T2D in European populations. Conversely, it was associated with increased fasting insulin levels in East Asian populations. Given the current uncertainty regarding the relationship between IL-6 and T2D, RA was used as a positive control. This made the results more convincing. Following the elucidation of the relationship between IL-6 receptor inhibitors and T2D, we conducted a mediation analysis to investigate the potential impact of T2D on the interaction between IL-6 receptor inhibitors and CSF metabolites.

Studies have suggested a positive correlation between plasma IL-6 levels and glioblastoma cell invasion.^[37,38] Plasma IL-6 levels were elevated in patients with bipolar disorder compared with those in remission.^[39] In fibroblasts derived from patients with major depressive disorder, transcription levels of the IL-6 receptor gene were significantly upregulated in response to IL-6 stimulation.^[40,41] These findings indicated a potential association between plasma IL-6 and CSF metabolites. Given the impact of IL-6 receptor inhibitors on T2D, we conducted an analysis to assess whether these therapies further influence CSF metabolites in the context of T2D. Our findings revealed that IL-6 receptor inhibitors were associated with elevated levels of GG in the CSF, with T2D serving as a partial mediator. These results have significant implications for informing pharmacological strategies. In particular, the inclusion of all CSF metabolites from healthy individuals enhanced the applicability of this study. We found that T2D was associated with a reduction in GG levels (OR: 0.961), whereas the administration of IL-6 receptor inhibitors resulted in an increase in GG levels (OR: 1.480). This was an important addition to the existing research. A recent analysis has indicated that T2D was correlated with accelerated disease progression in Parkinson’s.^[42] These alterations included a reduction in the expression of peroxisome proliferator-activated receptor-γ coactivator-1 (PGC-1α), alongside an upregulation of phosphoprotein enriched in astrocyte 15 (α/PGC-1) and phosphoprotein (PGC-1).^[43] These molecular changes have been implicated in the pathogenesis of PD. Therefore, it was necessary to explore the association between T2D and CSF metabolites. Although we did not further explore the relationship between IL-6 receptor inhibitors and PD, it was suggested that IL-6 receptor inhibitors could facilitate an increase in GG levels and potentially reduce the risk of T2D in the European population. This finding provides novel insights into the use of IL-6 receptor inhibitors.

Fasting insulin is an endogenous hypoglycemic hormone secreted by pancreatic islet beta cells, which play a vital role in inhibiting glycogenolysis and maintaining stable blood glucose levels. Elevated levels of fasting insulin indicate hyperinsulinemia. Early lifestyle modification and intervention are essential to prevent the progression of elevated fasting insulin and the subsequent development of insulin resistance. A cross-sectional study in Asia suggested that hyperinsulinemia served as a precursor to insulin resistance and represented the earliest subclinical metabolic abnormality observed in obese children.^[44] A comparative analysis of obese males in Europe indicated that fasting insulin served as a marker for identifying insulin resistance in obese individuals in both research and clinical contexts.^[45] This study highlighted a strong correlation between fasting insulin and insulin resistance across diverse populations and age groups. Research underscored the significance of insulin resistance, not only as a robust predictor of T2D but also as a critical therapeutic target for managing hyperglycemia.^[46]

We identified a potential association between IL-6 receptor inhibitors and a risk of fasting insulin among East Asian populations. However, these pharmacological agents did not appear to elevate the risk of T2D in the same demographic. The question was whether the use of targeted drugs constituted a risk factor for diabetes. Previous research indicated that certain medications, such as calcineurin inhibitors, antipsychotic drugs, hormones, and antihypertensive agents, contributed to inadequate insulin secretion or insulin resistance, potentially leading to drug-induced diabetes. In European patients undergoing tacrolimus treatment, the IL-6 (−174 G/C) gene polymorphism with the GG genotype was associated with a slightly elevated risk of developing new-onset diabetes after transplantation compared to individuals with the CC genotype (28.3% vs 13.3%; P = .12).^[47] In a randomized controlled trial with T2D and healthy individuals in Africa, carriers of the IL-6 G allele (−174) were found to have a 2.82-fold higher likelihood of developing T2D than those carrying the C allele (OR: 2.81, 95% CI: 1.78–4.5).^[48] These studies suggested a positive correlation between IL-6 gene polymorphism and T2D, with racial differences influencing this relationship. This study elucidated the impact of racial variations on the association between IL-6 receptor inhibitors and T2D. In Asian populations, IL-6 inhibitors did not exhibit a direct correlation with T2D. However, they were associated with increased fasting insulin in tissues. This elevation was correlated with insulin resistance, a crucial factor in the pathogenesis of T2D. Therefore, we speculated that IL-6 receptor inhibitors could promote the change of IL-6 gene polymorphism in the East Asian population, potentially contributing to increased fasting insulin levels. This hypothesis warranted further investigation to confirm its validity.

The study is subject to several limitations. First, while we accounted for potential racial differences in MR results, our analysis was restricted to data from European and East Asian populations, thereby excluding data pertinent to diabetes in South Asian, African, and other populations. Second, this study examined the association between IL-6 receptor inhibitors and CSF metabolites without concurrently analyzing blood metabolites. In future research, we intend to conduct a comprehensive search of the database for more detailed studies. Concurrently, while drug-targeted MR analysis presents distinct advantages, it remains imperative to validate these findings through large-scale, multi-center prospective studies and clinical drug trials. These represent critical areas for further investigation and development.

5. Conclusion

The findings preliminarily indicated that race constituted a variable potentially influencing the differential effects of IL-6 receptor inhibitors among patients with T2D across various regions. While the administration of IL-6 receptor inhibitors might impact glucose concentrations in CSF, the role of T2D as a mediator in CSF metabolites was weak. The predominant determinant appeared to be the administration of IL-6 receptor inhibitors.

Author contributions

Conceptualization: Naidan Zhang.

Data curation: Naidan Zhang.

Formal analysis: Naidan Zhang, Chaixia Ji.

Funding acquisition: Naidan Zhang.

Investigation: Naidan Zhang, Qi Xin.

Methodology: Baibing Xie, Chengliang Yuan.

Project administration: Naidan Zhang.

Resources: Naidan Zhang.

Software: Naidan Zhang.

Supervision: Chaixia Ji.

Validation: Chengliang Yuan.

Visualization: Naidan Zhang.

Writing – original draft: Naidan Zhang, Chaixia Ji.

Writing – review & editing: Naidan Zhang, Chengliang Yuan.