Skip to main content
Download PDF

Univariable and multivariable Mendelian randomization study identified the key role of gut microbiota in immunotherapeutic toxicity

European Journal of Medical Research volume 29, Article number: 161 (2024) Cite this article

Abstract

Background

In cancer patients receiving immune checkpoint inhibitors (ICIs), there is emerging evidence suggesting a correlation between gut microbiota and immune-related adverse events (irAEs). However, the exact roles of gut microbiota and the causal associations are yet to be clarified.

Methods

To investigate this, we first conducted a univariable bi-directional two-sample Mendelian randomization (MR) analysis. Instrumental variables (IVs) for gut microbiota were retrieved from the MiBioGen consortium (18,340 participants). GWAS summary data for irAEs were gathered from an ICIs-treated cohort with 1,751 cancer patients. Various MR analysis methods, including inverse variance weighted (IVW), MR PRESSO, maximum likelihood (ML), weighted median, weighted mode, and cML–MA–BIC, were used. Furthermore, multivariable MR (MVMR) analysis was performed to account for possible influencing instrumental variables.

Results

Our analysis identified fourteen gut bacterial taxa that were causally associated with irAEs. Notably, Lachnospiraceae was strongly associated with an increased risk of both high-grade and all-grade irAEs, even after accounting for the effect of BMI in the MVMR analysis. Akkermansia, Verrucomicrobiaceae, and Anaerostipes were found to exert protective roles in high-grade irAEs. However, Ruminiclostridium6, Coprococcus3, Collinsella, and Eubacterium (fissicatena group) were associated with a higher risk of developing high-grade irAEs. RuminococcaceaeUCG004, and DefluviitaleaceaeUCG011 were protective against all-grade irAEs, whereas Porphyromonadaceae, Roseburia, Eubacterium (brachy group), and Peptococcus were associated with an increased risk of all-grade irAEs.

Conclusions

Our analysis highlights a strong causal association between Lachnospiraceae and irAEs, along with some other gut microbial taxa. These findings provide potential modifiable targets for managing irAEs and warrant further investigation.

Introduction

Applications of immune checkpoint inhibitors (ICIs), especially those targeting CTLA-4 (cytotoxic T-lymphocyte associated protein 4) and PD-1/PD-L1 (programmed cell death protein 1/programmed cell death ligand 1), have revolutionized the treatment of various aggressive cancers [1]. By blocking inhibitory signaling pathways and reinvigorating the natural anti-tumor immune response, these inhibitors have significantly prolonged the lives of numerous cancer patients [2,3,4,5]. However, due to the inhibition of the systemic brake of immune activation, ICIs can cause off-target effects resulting in immune-mediated toxicities to organs and non-malignant tissues. This newly recognized registry of iatrogenic effects, known as immune-related adverse events (irAEs), usually resemble autoimmune disorders, such as colitis, dermatitis, and thyroiditis [6]. Although the majority of irAEs manifest in a mild manner, still, up to 55% of patients develop serious irAEs in combined therapy (anti-CTLA-4 and anti-PD-1) [7]. Notably, serious irAEs pose significant risks to patients’ well-being and may result in morbidity and mortality, not only due to the adverse event itself but also due to the need to suspend or terminate ICIs therapy and the potential impairment of the ICIs-induced immune response while using immunosuppressants (e.g., corticosteroids) [8,9,10]. Therefore, effective management of irAEs is critical to optimize the safety and efficacy of ICIs therapy.

The precise mechanisms underlying irAEs are not fully understood, but emerging evidence indicates that the gut microbiota, a complex and dynamic system of microorganisms colonizing the intestinal tract, may play a crucial role in the regulation of irAEs. Simpson et al. found a reduced alpha-diversity of intestinal microbiota in patients who developed severe irAEs [11]. Furthermore, antibiotics commonly prescribed prophylactically to hospitalized patients have been shown to increase the risk of ICIs therapy-related irAEs that are not limited to the gastrointestinal tract [12,13,14]. The gut microbiota closely interacts with the host immune system and has been implicated in the regulation of various autoimmune and inflammatory disorders [15, 16]. However, consensus on the core microbial drivers or protective microbes of irAEs is still lacking, due to inconsistent findings reported in previous studies [11, 17,18,19]. The discrepancies among previous studies may be attributed to limited sample sizes and susceptibility to confounding factors such as age, diet, and medication usage in observational designs [11, 20].

Mendelian randomization (MR), initially described by Katan in 1986 [21], is a novel method for inferring causal associations between modifiable risk factors and health outcomes using genetic variations as instrumental variables (IVs) [22]. MR effectively addresses the limitations of confounding and measurement errors that often exist in observational studies, as the direction of causation is from the genetic polymorphism to the trait of interest, not vice versa [23]. Therefore, we aim to utilize MR, an increasingly popular method in drug discovery and epidemiology [24, 25], to investigate the potential association between the gut microbiota and irAEs, providing further evidence for the management of irAEs by targeting human gut microbiota.

Methods

Study design and data source

An overview of the study design was illustrated in Fig. 1. In general, we first performed univariable bi-directional two-sample MR analysis, which utilizes single-nucleotide polymorphisms (SNPs) from summary-level data as proxies for the risk factor under investigation. Then, several sensitivity analyses and multivariable MR analysis (MVMR) was conducted to further increase the robustness of our study. To ensure the validity of the MR results, three assumptions needed to be satisfied, as illustrated in Additional file 1: Fig. S1 [22]:

  • (1) Relevance assumption: The genetic variants should demonstrate a strong association with the exposure;

  • (2) Independence assumption: The genetic variants should not be associated with any confounders that could affect the relationship between the exposure and outcome;

  • (3) Exclusion restriction assumption: The genetic variants should not have an independent effect on the outcome aside from their impact through the exposure.

Fig. 1
figure 1

Overview of the study design. Initially, IVs were selected from the summary GWAS data of the gut microbiota and irAEs. Subsequently, by employing thresholds of p value (IVW) < 0.05 and p value (MR PRESSO) < 0.05, the identified gut microbiota that exhibited statistically significant associations were taken into further analysis. irAEs immune-related adverse events, ICIs immune checkpoint inhibitors, BMI body mass index, WC Waist circumference, HP Hip circumference; WHR waist-to-hip ratio. (Created with BioRender.com)

Full size image

This study is based on publicly available GWAS summary statistics and ethical approvals were acquired by the original studies.

Gut microbiota

Genetic variations associated with the composition of gut microbiota were derived from the most comprehensive genome-wide meta-analysis conducted to date by the MiBioGen consortium [26]. This study included a total of 18,340 individuals from 24 cohorts worldwide, mainly of European descent (n = 13,266). Fecal DNA was extracted, and targeted sequencing of variable regions in the bacterial 16S rRNA gene (V4, V3–V4, and V1–V2 regions) was performed to profile the gut bacterial composition. To account for sequencing depth differences across cohorts, all cohorts were rarefied to 10,000 reads per sample. Taxonomic classification was performed using direct taxonomic binning method [26]. Following quality control, imputation, and post-imputation filtering procedures, gut bacterial taxa observed in over 10% of samples were included in the microbiota quantitative trait loci (mbQTL) mapping [26]. This allowed us to identify host genetic variants associated with the relative abundance of bacterial taxa. Further details on microbial data processing can be found in the original study. Genus-level and family-level taxa were included in our analysis, resulting in a total of 131 genus-level and 35 family-level taxa.

IrAEs

Summary-level data of irAEs was obtained from a recent GWAS conducted in the Dana-Farber Cancer Institute (DFCI) cohort [27]. The study included 1,751 cancer patients of European ancestry who underwent ICIs treatments between 2013 and 2020. The majority of patients (approximately 90%) received PD-1/PD-L1 inhibitors, while the remaining 10% received combined immunotherapy (CTLA4 and PD-1/PD-L1 inhibitors). Among the 1,751 cancer patients, 259 cases that experienced high-grade irAEs (grade 3–5 events) were manually curated according to the National Cancer Institute (NCI) Common Terminology Criteria for Adverse Events v.5 guidelines. In addition, algorithm-based autoimmune-like electronic health records were used to identify 339 patients who experienced any grade irAEs (referred to as all-grade irAEs). Most of these cases were grade 2 or higher events [27]. The tumor tissue of these patients was sequenced using the targeted OncoPanel sequencing platform. After quality control steps, germline SNPs were imputed by utilizing ultra-low-coverage off-target reads. Then, the GWAS was conducted in the DFCI cohort to investigate the association of all variants with the time from the start of ICIs treatment to the occurrence of the two phenotypes of irAEs. For more detailed information, please refer to the original publication [27].

Selection of instrumental variables

Several steps were followed in the selection of IVs. First, for the gut microbiome, we selected SNPs associated with bacterial taxa with a p value less than 1 × 10–5 for further analysis [28, 29]. Second, potential SNPs were clumped for independence in the TwoSampleMR package in R software. We used the European 1000 Genomes Project Phase 3 reference panel and set the linkage-disequilibrium threshold (r2) at 0.001 within a 10 Mb window size. Third, we extracted SNPs from the outcome statistics and performed a harmonization procedure. SNPs that were not available in the outcome GWAS data were replaced with proxy SNPs (r2 > 0.8), and palindromic SNPs were removed for further MR analysis. Furthermore, F statistics of selected IVs, which indicate instrument strength, was calculated as [Beta/SE]2. Typically, F statistics > 10 suggest enough IVs strength to avoid weak instrument bias [30]. Finally, All SNPs with positive results were re-examined using PhenoScanner package (version 1.0) in R software to investigate the presence of potential confounders. Bacterial taxa with less than 3 valid SNPs and unknown origin were excluded from the analysis to mitigate potential bias. Consequently, we included a total of 104 genus-level and 28 family-level bacterial taxa (n = 132) for further MR analysis.

Statistical analysis

In this study, we employed several MR analysis methods to explore the potential causal relationship between gut microbiota and irAEs. The methods used included IVW, MR PRESSO, ML, weighted median, weighted mode, and a constrained maximum likelihood and model averaging-based method (cML–MA–BIC). IVW and MR PRESSO were used in the primary analysis. In general, IVW provides maximum statistical power when all instruments are valid [31], while MR PRESSO identifies and removes genetic variants that deviate significantly from the variant-specific causal estimates of other variants, thereby increasing statistical power and addressing potential outliers [32]. The ML method resembles the IVW approach which assumes the absence of both heterogeneity and horizontal pleiotropy. If these assumptions hold true, the ML method yields unbiased results with smaller standard errors compared to the IVW approach [33]. Considering the potential existence of IV pleiotropy, we also performed pleiotropy-robust methods including weighted median, weighted mode, and cML–MA–BIC in the sensitivity analysis. These methods relax the instrumental variable assumptions. Weighted median were introduced when the exclusion restriction assumption was violated (uncorrelated pleiotropy), which typically assume fewer than 50% of genetic variants are invalid [34]. The weighted-mode method clusters genetic variants based on their similarity in causal effect and estimates the overall causal effect based on the cluster with the most number of IVs [35]. The cML–MA–BIC method is a novel approach developed for MR analysis, specifically addressing the issue of invalid IVs exhibiting both uncorrelated and correlated pleiotropy (violation of the independence assumption) [36]. By being robust to such violations, cML–MA–BIC improves the accuracy of MR analysis, reduces Type I error, and increases statistical power [36].

Next, heterogeneity and directional pleiotropy were assessed using Cochra’'s Q statistics and MR Egger intercept. Leave-one-out (LOO) analysis was conducted to identify possible reliance on a specific variant, which involved excluding one SNP at a time for all valid SNPs in the IVW analysis. In addition, reverse MR analysis between irAEs and the identified significant gut bacterial taxa was performed. Moreover, MVMR analysis, which could help us better understand the intricate interplay between the risk factors was further conducted [37]. We considered a Bonferroni-corrected p value of 3.89 × 10–4 (0.05/132) as the significance threshold for gut microbiota. Two-tailed p value < 0.05 was considered suggestive of significance. All analyses were conducted using R packages “TwoSampleMR” (version 0.5.6), “MRPRESSO” (version 1.0), and “MRcML” (version 0.0.0.9) in R software (version 4.2.2).

Results

Genetic instruments and primary MR analysis

A total of 870 single-nucleotide polymorphisms (SNPs) were selected as IVs for the 132 gut bacterial taxa (Additional file 2: Table S1). The F statistics for each SNP ranged from 16.91 to 36.57, with a median value of 21.66. Using the IVW and MR-PRESSO methods, eight gut bacterial taxa associated with high-grade irAEs were identified with p values < 0.05. These taxa include Lachnospiraceae, Verrucomicrobiaceae, Ruminiclostridium6, Coprococcus3, Anaerostipes, Akkermansia, Collinsella, and Eubacterium (fissicatena group). For all-grade irAEs, seven gut bacterial taxa, including Lachnospiraceae, Porphyromonadaceae, Roseburia, RuminococcaceaeUCG004, DefluviitaleaceaeUCG011, Eubacterium (brachy group), and Peptococcus, were identified. Given previous studies suggest that pre-existing autoimmune conditions, such as inflammatory bowel disease, psoriasis, and rheumatoid arthritis, may predispose individuals to irAEs susceptibility [38, 39]. We further examined the SNPs associated with the significant bacterial taxa using PhenoScanner. Only one SNP (rs11597285) for the Collinsella genus was found to be associated with allergic disease (e.g., allergic rhinitis and eczema) (refer to Additional file 2: Table S9). However, the results of Collinsella remained uninfluenced after removing rs11597285 in the LOO analysis (described below). The complete results of the primary MR analysis can be found in Additional file 2: Tables S2 and S3.

Main MR results and sensitivity analysis for high-grade irAEs

As shown in Fig. 2, the IVW estimate suggested the abundance of Lachnospiraceae family was associated with a shortened time to high-grade irAEs (Beta = − 1.22, 95% CI − 1.99 to − 0.44, p = 2.17 × 10–3), indicating that Lachnospiraceae serves as a risk factor for the development of high-grade irAEs. The deleterious effect remained significant in pleiotropy-robust cML–MA–BIC estimation (Beta = − 1.24, 95% CI − 2.45 to − 0.02, p = 4.62 × 10–2). Surprisingly, Ruminiclostridium6 genus was significantly associated with an increased risk of high-grade irAEs in all MR approaches, including IVW (Beta = − 2.11, 95% CI − 2.98 to − 1.23, p = 2.47 × 10–6), cML–MA–BIC (Beta = − 2.17, 95% CI − 3.49 to − 0.86, p = 1.19 × 10–3), Weighted median (Beta = -2.43, 95% CI − 4.05 to − 0.81, p = 3.29 × 10–3), and other methods. In addition, the IVW estimate indicated a protective effect of the Akkermansia genus on high-grade irAEs (Beta = 1.27, 95% CI 0.28–2.25, p = 0.01), and this finding was confirmed by the IVW estimate of its paternal taxon Verrucomicrobiaceae (Beta = 1.27, 95% CI 0.29–2.25, p = 0.01). The IVW estimate of Anaerostipes genus also indicated a suggestive protective effect against high-grade irAEs (Beta = 2.1, 95% CI 0.85–3.35, p = 1.02 × 10–3), as well as cML–MA–BIC (Beta = 2.17, 95% CI 0.57–3.77, p = 7.88 × 10–3). Moreover, significant effects of the Coprococcus3 (Beta = − 2.04, 95% CI − 2.7 to − 1.39, p = 8.93 × 10–10), Collinsella (Beta = − 1.12, 95% CI − 1.7 to − 0.53, p = 1.7 × 10–4), and Eubacterium (fissicatena group) genus (Beta = -0.73, 95% CI − 1.01 to − 0.46, p = 1.93 × 10–7) were all revealed by IVW estimate, indicating an increased risk of high-grade irAEs. Scatter plots reflecting the effect size of IVs on both bacterial taxa and high-grade irAEs are shown in Fig. 3.

Fig. 2
figure 2

Forest plots of MR results for the causal association between the identified eight gut microbial taxa and high-grade irAEs (grade 3–5 events). NSNP number of SNPs; Beta = effect size from the exposure to the outcome; CI  confidence interval

Full size image
Fig. 3
figure 3

Scatter plots of MR analysis between the gut microbial taxa and high-grade irAEs (grade 3–5 events)

Full size image

In the subsequent analysis of heterogeneity and horizontal pleiotropy, Cochra’'s Q statistics revealed no significant heterogeneity (p value > 0.05) among the IVs for the gut bacterial taxa in high-grade irAEs analysis (see Table 1). No significant evidence for directional horizontal pleiotropy was found in the MR–Egger regression intercept analysis and MR PRESSO global test (Table 1 and Additional file 2: Table S4). In addition, the LOO analysis identified no predominant SNP that influenced the results (Fig. 4). We further performed reverse MR analysis and demonstrated no reverse causation exists between high-grade irAEs and the abundance of gut bacterial taxa (Additional file 2: Table S6).

Table 1 Results of heterogeneity and horizontal pleiotropy analysis
Full size table
Fig. 4
figure 4

Leave-one-out plots of MR analysis between the gut microbial taxa and high-grade irAEs (grade 3–5 events)

Full size image

Main MR results and sensitivity analysis for all-grade irAEs

Figure 5 reveals the association between bacterial taxa and all-grade irAEs. It is noteworthy that the deleterious impact of the Lachnospiraceae family was also detected in all-grade irAEs, as revealed by IVW (Beta = − 2.05, 95% CI − 3.27 to − 0.82, p = 1.06 × 10–3), cML–MA–BIC (Beta = − 2.21, 95% CI − 3.32 to − 1.1, p = 9.16 × 10–5), and Weighted median estimates (Beta = − 1.72, 95% CI − 3.2 to − 0.24, p = 2.25 × 10–2). The IVW estimate of Roseburia genus also showed an increased risk of all-grade irAEs (Beta = − 2.07, 95% CI − 3.21 to -0.93, p = 3.76 × 10–4). Consistent results were observed in ML (Beta = − 2.07, 95% CI − 3.67 to − 0.47, p = 1.14 × 10–2) and cML–MA–BIC (Beta = − 2.08, 95% CI − 3.71 to − 0.45, p = 1.23 × 10–2) estimates. On the contrary, RuminococcaceaeUCG004 (Beta = 1.07, 95% CI 0.43–1.72, p = 1.05 × 10–3) and DefluviitaleaceaeUCG011 (Beta = 0.82, 95% CI 0.38–1.25, p = 2.15 × 10–4) were identified to decrease the risk of all-grade irAEs according to the IVW approach. Subsequently, the results for RuminococcaceaeUCG004 were consistent with the cML–MA–BIC (Beta = 1.1, 95% CI 0.04–2.15, p = 4.12 × 10–2) and ML method (Beta = 1.09, 95% CI 0.04–2.14, p = 4.16 × 10–2). Moreover, the IVW estimates suggested that Porphyromonadaceae (Beta = − 1.11, 95% CI − 1.83 to − 0.39, p = 2.46 × 10–3), Eubacterium (brachy group) (Beta = − 0.71, 95% CI − 1.14 to − 0.27, p = 1.38 × 10–3), and Peptococcus (Beta = − 0.71, 95% CI − 1.24 to − 0.18, p = 8.82 × 10–3) may increase the risk of all-grade irAEs. Scatter plots reflecting the effect size of each IV on both bacterial taxa and all-grade irAEs are shown in Fig. 6.

Fig. 5
figure 5

Forest plots of MR results for the causal association between the identified seven gut microbial taxa and all-grade irAEs (grade 1–5 events). NSNP number of SNPs; Beta = effect size from the exposure to the outcome; CI = confidence interval

Full size image
Fig. 6
figure 6

Scatter plots of MR analysis between the gut microbial taxa and all-grade irAEs (grade 1–5 events)

Full size image

Similarly, Cochran’s Q statistics indicated an absence of notable heterogeneity in the IVs of gut bacterial taxa (refer to Table 1). In addition, the results from the MR–Egger regression intercept analysis and the MR PRESSO global test demonstrated no significant evidence of directional horizontal pleiotropy (Table 1 and Additional file 2: Table S4). Next, no significant single SNP was identified in the LOO analysis that influenced the results (Fig. 7). Overall, these results provide evidence for the association between specific gut bacterial taxa and the development of high-grade and all-grade irAEs, and highlight the potential role of the gut microbiota in irAEs.

Fig. 7
figure 7

Leave-one-out plots of MR analysis between the gut microbial taxa and all-grade irAEs (grade 1–5 events)

Full size image

Multivariable MR analysis

Given the strong correlation between body mass index (BMI) and gut microbiota [40, 41], we incorporated BMI and weight-associated anthropometric traits, as well as several lifestyle risk factors associated with irAEs [42], into the univariable MR analysis. Detailed GWAS data sets information of these anthropometric traits and lifestyle risk factors can be found in Additional file 2: Table S10. The univariable MR analysis suggested a correlation between BMI and an increased risk of all-grade irAEs (Table 2). Thus, we performed subsequent MVMR analysis by including BMI as an additional exposure alongside previously identified gut microbial taxa (Additional file 2: Table S11). Notably, after accounting for the effect of BMI in the MVMR analysis, Lachnospiraceae still remained significantly associated with the risk of developing all-grade irAEs (Beta = − 1.07, 95% CI − 1.96 to − 0.18, p = 0.02), and high-grade irAEs (Beta = − 0.91, 95% CI − 1.84 to 0.01, p = 0.05) with a marginal p value (Table 3).

Table 2 Univariate MR estimates between common risk factors and irAEs
Full size table
Table 3 Multivariable MR estimates between Lachnospiraceae, BMI and irAEs
Full size table

Discussion

In this study, univariable bi-directional MR and MVMR analyses were conducted to infer the causal association between gut microbiota and irAEs. We identified fourteen gut bacterial taxa that were causally associated with high-grade and all-grade irAEs. Surprisingly, Lachnospiraceae strongly associated with an increased risk of both irAEs phenotypes, even after accounting for the effect of BMI in the MVMR analysis. In addition, we found robust evidence indicating that Ruminiclostridium6 predisposes ICIs receivers to developing high-grade irAEs. Coprococcus3, Collinsella, and Eubacterium (fissicatena group) were also associated with an increased risk of high-grade irAEs, while Akkermansia, Verrucomicrobiaceae, and Anaerostipes exhibited protective roles in high-grade irAEs. For all-grade irAEs, Porphyromonadaceae, Roseburia, Eubacterium (brachy group), and Peptococcus were associated with an elevated risk, while RuminococcaceaeUCG004 and DefluviitaleaceaeUCG011 were associated with a reduced risk.

Several observational studies have demonstrated associations between gut microbiota and irAEs [11, 17,18,19, 43, 44]. Lachnospiraceae species (such as Coprococcus and Roseburia), which are obligately anaerobic, variably spore-forming bacteria, were found to correlate with increased risk of various types of irAEs [17, 18, 45]. In a more recent study, two species of the Lachnospiraceae family were specifically enriched in irAEs that occur in endocrine organs [46]. Consistent with these findings, we also identified the Lachnospiraceae family, and its two genera (i.e., Coprococcus3 and Roseburia) associated with an increased risk of irAEs. Importantly, the harmful influence of the Lachnospiraceae family was observed in high-grade irAEs and all-grade irAEs, even after adjusting the effect of BMI, which provides further validity and robustness to our study. Ruminiclostridium6 has been less studied in irAEs, but some studies have observed its accumulation in a mouse model of DSS-induced colitis. While treated with phloretin (a dihydrochalcone flavonoid) or sodium butyrate (one of the short chain fatty acids [SCFAs]), both of which alleviates DSS-induced colitis, the abundance of Ruminiclostridium6 was reduced [47, 48]. In addition, the Ruminiclostridium genus has also been associated with autoimmune-related diseases, such as the experimental multiple sclerosis model and Alzheimer’s disease [49, 50]. Based on our strong association of Ruminiclostridium6 with increased risk of high-grade irAEs, it is suggested that Ruminiclostridium6 may play a pivotal role in the development of autoimmune conditions and could be a potential target for relieving irAEs symptoms, although more evidence is needed.

Akkermansia muciniphila, an anaerobic gram-negative species that belongs to Akkermansia genus, and Verrucomicrobiaceae family, has gained much attention in immunotherapy due to their association with a favorable response in ICIs therapy [51,52,53]. Akkermansia muciniphila has also been shown to exhibit a protective role in ICIs-associated colitis [54]. Mechanically, Wang et al. demonstrated that Akkermansia muciniphila and its purified membrane protein mitigated colitis by regulating macrophages and CD8 + T cells in the colon tissue [55]. Our study further supported the protective role of Akkermansia in high-grade irAEs. Ruminococcaceae, a key family of bacteria producing short-chain fatty acids (SCFAs), has been observed to be enriched in ICIs responders without severe irAEs [11]. Previous studies have found that Faecalibacterium prausnitzii, a species belonging to the Ruminococcaceae family, was decreased in non-responders and those experiencing severe irAEs [11, 17, 45]. These findings suggest a potential role of Ruminococcaceae as protective bacteria, possibly through the facilitation of SCFAs accumulation, in the mitigation of irAEs. Collinsella and Anaerostipes have limited evidence in irAEs, but the Collinsella genus has been reported to increase the production of IL-17A and enhance rheumatoid arthritis severity [56]. In contrast, Anaerostipes, which belongs to the Lachnospiraceae family, was conversely associated with the risk of high-grade irAEs in our study [11, 18]. These discrepancies observed in previous clinical studies might be attributed to several factors, including limited sample sizes in previous observational studies, heterogeneity among the samples, and inadequate exploration of the taxonomic classification at the genus level of the gut microbiota. Therefore, a more detailed taxonomy for gut microbiota is crucial in dissecting the underlying mechanisms.

Gut microbiota plays a pivotal role in modulating human immune homeostasis, and an imbalance in gut microbial composition, known as gut dysbiosis, has been implicated in several autoimmune diseases [15, 16, 57]. IrAEs resemble autoimmune diseases in many aspects [38, 58, 59]. Thus, despite the underlying mechanisms by which gut microbiota manipulates the development of irAEs remain poorly understood, we hypothesize that there might be some shared etiology of microbiota in autoimmune diseases and irAEs. These mechanisms include: (1) “Molecular mimicry”: Evidence has shown that exposure to homologous amino acid sequences or epitopes of microbiota and aberrant activation of autoreactive B or T cells leads to multiple autoimmune diseases, such as multiple sclerosis [60], Guillain–Barré syndrome [61], Type 1 diabetes [62], Rheumatoid arthritis [63], and primary biliary cholangitis [64], which is referred to as “molecular mimicry” [65]. It is believed that the systematic activation of the immune system during ICIs treatment triggers irAEs by bypassing self-tolerance in normal organs. One intriguing fact is that most irAEs occur in barrier organs (e.g., the intestinal tract, skin, and lungs) [58, 66]. This implies that the activated immune response might target the commensal microbiome as antigenic targets, although this hypothesis has not been fully demonstrated. (2) Decreased accumulation of SCFAs: SCFAs, including acetate, propionate, and butyrate, are a group of organic compounds primarily produced by the gut microbiota during the fermentation of dietary fibers. These metabolites were shown to improve the anti-cancer function of effector T cells, but they also seem to exhibit anti-inflammatory characteristics [67, 68]. Butyrate, one of the well-studied SCFAs, was shown to inhibit the activation of NF-κB and its downstream pathway [69], thereby reducing the production of pro-inflammatory cytokines, such as IL-8 [70], while increasing the levels of anti-inflammatory factors, such as IL-10 [71]. Moreover, SCFAs serve as a key energy source for colonocytes and maintain intestinal barrier integrity [72]. Thus, the reduced abundance of SCFAs-producing bacteria along with its metabolites may participate in the development of irAEs. [3] Other mechanisms: stimulation of the immune response by microbial-associated molecular patterns (e.g., include lipopolysaccharides (LPS), lipoproteins, flagellin and bacterial DNA) [73] and compromised vitamin B and polyamine metabolism that associated with gut dysbiosis [19] may also contribute to irAEs.

Taken together, the gut microbiota and the human immune system maintain a delicate balance under normal physiological conditions. Once the balance has been disturbed (e.g., ICIs treatment), the dysregulated microbiota might lead to the development of undesirable irAEs. The primary management strategy for irAEs (> grade 2) involves the suspension of ICIs and/or utilizing immunosuppressive therapy [74]. Nevertheless, one concern is that discontinuing ICIs or using immunosuppressants may compromise treatment efficacy. Ideally, approaches to boost ICIs efficacy while reducing the accompanied irAEs are to be expected in the future. Fecal Microbiota Transplantation (FMT), an approach to modulate gut microbiota, has been shown to increase ICIs efficacy in melanoma patients [75, 76], and emerging evidence has demonstrated the mitigation of ICIs-related colitis through FMT in clinical practice [77]. Interestingly, while irAEs and ICIs efficacy are often considered two sides of the same coin, certain gut bacteria, such as Akkermansia muciniphila and Faecalibacterium prausnitzii, have been shown to ameliorate irAEs and reinforce ICIs efficacy at the same time [78]. This suggests that targeting gut microbiota could be an ideal approach to relieve irAEs symptoms and maintain ICIs efficacy, but further real-world evidence is needed to support this hypothesis.

Our study possesses several strengths. First, we applied the MR approach to infer the causal associations between gut microbiota and irAEs, which effectively mitigates the influence of confounding factors and provides robust causal inference. Second, we conducted reverse MR analyses, confirming the absence of reverse causation, thereby enhancing the validity of our study. Third, we applied MVMR in our analysis which further strengthened the validity and robustness of our study. Moreover, we incorporated several pleiotropy-robust methods such as MR PRESSO and cML–MA–BIC, further strengthening the robustness of our study. However, there are also certain inherent limitations in our study that should be considered when interpreting the findings. First, our analysis is based on European-derived GWAS summary statistics, which might confine the generalization of the findings to other populations. Second, due to the utilization of summary statistics instead of raw data in the analysis, subgroup analyses based on ICIs regimes (e.g., PD-1/PD-L1 group, CTLA-4 group, and combined therapy group) or different cancer subtypes (e.g., gastrointestinal cancer and genitourinary cancer) could not be performed. Third, the gut microbiota is shaped by multiple environmental factors, which confines the number of the identified significant gene loci in the GWAS [26]. Thus, we relaxed the significant threshold of IVs to 1 × 10–5 [29] and employed Bonferroni correction to mitigate potential false positive results. Dietary patterns may influence the composition of gut microbiota and the changes of gut microbial taxa may also contribute to the development of irAEs. However, due to the limitations of summary statistics, detailed analysis among diet, gut microbiota and irAEs was not available. Further study is needed in the future. In addition, future studies may focus on the role of different microbial patterns residing on different barrier interfaces of human organs in the development of irAEs.

Conclusion

Our univariable and multivariable MR analysis identified a strong causal association between Lachnospiraceae and irAEs, along with several other gut microbial taxa, such as Akkermansia and Ruminiclostridium6 etc.. However, whether the FMT or probiotics could be used as interventional approaches to mitigate irAEs while reserving ICIs efficacy, additional randomized clinical trials (RCTs) are warranted. Furthermore, in-depth investigations are needed to elucidate the precise mechanisms through which the gut microbiota influences the development of irAEs.

Data availability

This study was conducted using public-available data. The GWAS summary statistics for gut microbiota are available at www.mibiogen.org, and GWAS summary statistics for irAEs are available at https://zenodo.org/record/6800429. The summary statistics for irAEs were downloaded from https://zenodo.org/record/6800429.

References

  1. Ribas A, Wolchok JD. Cancer immunotherapy using checkpoint blockade. Science. 2018;359(6382):1350–5.

    Article  PubMed  PubMed Central  ADS  CAS  Google Scholar 

  2. Hodi FS, O’Day SJ, McDermott DF, Weber RW, Sosman JA, Haanen JB, et al. Improved survival with ipilimumab in patients with metastatic melanoma. N Engl J Med. 2010;363(8):711–23.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  3. Motzer RJ, Escudier B, McDermott DF, George S, Hammers HJ, Srinivas S, et al. Nivolumab versus everolimus in advanced renal-cell carcinoma. N Engl J Med. 2015;373(19):1803–13.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  4. Overman MJ, McDermott R, Leach JL, Lonardi S, Lenz HJ, Morse MA, et al. Nivolumab in patients with metastatic DNA mismatch repair-deficient or microsatellite instability-high colorectal cancer (CheckMate 142): an open-label, multicentre, phase 2 study. Lancet Oncol. 2017. https://doi.org/10.1016/S1470-2045(17)30422-9.

    Article  PubMed  PubMed Central  Google Scholar 

  5. Gandhi L, Rodríguez-Abreu D, Gadgeel S, Esteban E, Felip E, De Angelis F, et al. Pembrolizumab plus chemotherapy in metastatic non-small-cell lung cancer. N Engl J Med. 2018;378(22):2078–92.

    Article  PubMed  CAS  Google Scholar 

  6. Robert C. A decade of immune-checkpoint inhibitors in cancer therapy. Nat Commun. 2020;11(1):3801.

    Article  PubMed  PubMed Central  ADS  CAS  Google Scholar 

  7. Postow MA, Chesney J, Pavlick AC, Robert C, Grossmann K, McDermott D, et al. Nivolumab and Ipilimumab versus Ipilimumab in Untreated Melanoma. N Engl J Med. 2015;372(21):2006–17.

    Article  PubMed  PubMed Central  Google Scholar 

  8. Suresh K, Naidoo J. Lower survival in patients who develop pneumonitis following immunotherapy for lung cancer. Clin Lung Cancer. 2020;21(3):e169–70.

    Article  PubMed  Google Scholar 

  9. Schadendorf D, Wolchok JD, Hodi FS, Chiarion-Sileni V, Gonzalez R, Rutkowski P, et al. Efficacy and safety outcomes in patients with advanced melanoma who discontinued treatment with nivolumab and ipilimumab because of adverse events: a pooled analysis of randomized phase ii and iii trials. J Clin Oncol Off J Am Soc Clin Oncol. 2017;35(34):3807–14.

    Article  CAS  Google Scholar 

  10. Brahmer JR, Lacchetti C, Schneider BJ, Atkins MB, Brassil KJ, Caterino JM, et al. Management of immune-related adverse events in patients treated with immune checkpoint inhibitor therapy: American society of clinical oncology clinical practice guideline. J Clin Oncol Off J Am Soc Clin Oncol. 2018;36(17):1714–68.

    Article  CAS  Google Scholar 

  11. Simpson RC, Shanahan ER, Batten M, Reijers ILM, Read M, Silva IP, et al. Diet-driven microbial ecology underpins associations between cancer immunotherapy outcomes and the gut microbiome. Nat Med. 2022;28(11):2344–52.

    Article  PubMed  CAS  Google Scholar 

  12. Jing Y, Chen X, Li K, Liu Y, Zhang Z, Chen Y, et al. Association of antibiotic treatment with immune-related adverse events in patients with cancer receiving immunotherapy. J Immunother Cancer. 2022;10(1): e003779.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Abu-Sbeih H, Herrera LN, Tang T, Altan M, Chaftari AP, Okhuysen PC, et al. Impact of antibiotic therapy on the development and response to treatment of immune checkpoint inhibitor-mediated diarrhea and colitis. J Immunother Cancer. 2019;7(1):242.

    Article  PubMed  PubMed Central  Google Scholar 

  14. Mohiuddin JJ, Chu B, Facciabene A, Poirier K, Wang X, Doucette A, et al. Association of antibiotic exposure with survival and toxicity in patients with melanoma receiving immunotherapy. J Natl Cancer Inst. 2021;113(2):162–70.

    Article  PubMed  Google Scholar 

  15. Yoo JY, Groer M, Dutra SVO, Sarkar A, McSkimming DI. Gut microbiota and immune system interactions. Microorganisms. 2020;8(10):1587.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  16. Belkaid Y, Hand T. Role of the microbiota in immunity and inflammation. Cell. 2014;157(1):121–41.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  17. Chaput N, Lepage P, Coutzac C, Soularue E, Le Roux K, Monot C, et al. Baseline gut microbiota predicts clinical response and colitis in metastatic melanoma patients treated with ipilimumab. Ann Oncol Off J Eur Soc Med Oncol. 2017;28(6):1368–79.

    Article  CAS  Google Scholar 

  18. McCulloch JA, Davar D, Rodrigues RR, Badger JH, Fang JR, Cole AM, et al. Intestinal microbiota signatures of clinical response and immune-related adverse events in melanoma patients treated with anti-PD-1. Nat Med. 2022;28(3):545–56.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  19. Dubin K, Callahan MK, Ren B, Khanin R, Viale A, Ling L, et al. Intestinal microbiome analyses identify melanoma patients at risk for checkpoint-blockade-induced colitis. Nat Commun. 2016;2(7):10391.

    Article  Google Scholar 

  20. Bajinka O, Tan Y, Abdelhalim KA, Özdemir G, Qiu X. Extrinsic factors influencing gut microbes, the immediate consequences and restoring eubiosis. AMB Express. 2020;10(1):130.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  21. Katan MB. Apolipoprotein E isoforms, serum cholesterol, and cancer. Lancet Lond Engl. 1986;1(8479):507–8.

    Article  CAS  Google Scholar 

  22. Davies NM, Holmes MV, Davey SG. Reading Mendelian randomisation studies: a guide, glossary, and checklist for clinicians. BMJ. 2018;12(362): k601.

    Article  Google Scholar 

  23. Davey Smith G, Hemani G. Mendelian randomization: genetic anchors for causal inference in epidemiological studies. Hum Mol Genet. 2014;23(R1):R89-98.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  24. Chen L, Peters JE, Prins B, Persyn E, Traylor M, Surendran P, et al. Systematic Mendelian randomization using the human plasma proteome to discover potential therapeutic targets for stroke. Nat Commun. 2022;17(13):6143.

    Article  ADS  Google Scholar 

  25. Zhang X, Theodoratou E, Li X, Farrington SM, Law PJ, Broderick P, et al. Genetically predicted physical activity levels are associated with lower colorectal cancer risk: a Mendelian randomisation study. Br J Cancer. 2021;124(7):1330–8.

    Article  PubMed  PubMed Central  Google Scholar 

  26. Kurilshikov A, Medina-Gomez C, Bacigalupe R, Radjabzadeh D, Wang J, Demirkan A, et al. Large-scale association analyses identify host factors influencing human gut microbiome composition. Nat Genet. 2021;53(2):156–65.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  27. Groha S, Alaiwi SA, Xu W, Naranbhai V, Nassar AH, Bakouny Z, et al. Germline variants associated with toxicity to immune checkpoint blockade. Nat Med. 2022;28(12):2584–91.

    Article  PubMed  CAS  Google Scholar 

  28. Liu X, Tong X, Zou Y, Lin X, Zhao H, Tian L, et al. Mendelian randomization analyses support causal relationships between blood metabolites and the gut microbiome. Nat Genet. 2022;54(1):52–61.

    Article  PubMed  CAS  Google Scholar 

  29. Sanna S, van Zuydam NR, Mahajan A, Kurilshikov A, Vich Vila A, Võsa U, et al. Causal relationships among the gut microbiome, short-chain fatty acids and metabolic diseases. Nat Genet. 2019;51(4):600–5.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  30. Zheng J, Baird D, Borges MC, Bowden J, Hemani G, Haycock P, et al. Recent Developments in Mendelian Randomization Studies. Curr Epidemiol Rep. 2017;4(4):330–45.

    Article  PubMed  PubMed Central  Google Scholar 

  31. Burgess S, Davey Smith G, Davies NM, Dudbridge F, Gill D, Glymour MM, et al. Guidelines for performing Mendelian randomization investigations. Wellcome Open Res. 2019;4:186.

    Article  PubMed  Google Scholar 

  32. Verbanck M, Chen CY, Neale B, Do R. Detection of widespread horizontal pleiotropy in causal relationships inferred from Mendelian randomization between complex traits and diseases. Nat Genet. 2018;50(5):693–8.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  33. Burgess S, Dudbridge F, Thompson SG. Combining information on multiple instrumental variables in Mendelian randomization: comparison of allele score and summarized data methods. Stat Med. 2016;35(11):1880–906.

    Article  MathSciNet  PubMed  Google Scholar 

  34. Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in mendelian randomization with some invalid instruments using a weighted median estimator. Genet Epidemiol. 2016;40(4):304–14.

    Article  PubMed  PubMed Central  Google Scholar 

  35. Hartwig FP, Davey Smith G, Bowden J. Robust inference in summary data Mendelian randomization via the zero modal pleiotropy assumption. Int J Epidemiol. 2017;46(6):1985–98.

    Article  PubMed  PubMed Central  Google Scholar 

  36. Xue H, Shen X, Pan W. Constrained maximum likelihood-based Mendelian randomization robust to both correlated and uncorrelated pleiotropic effects. Am J Hum Genet. 2021;108(7):1251–69.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  37. Sanderson E, Spiller W, Bowden J. Testing and correcting for weak and pleiotropic instruments in two-sample multivariable Mendelian randomization. Stat Med. 2021;40(25):5434–52.

    Article  MathSciNet  PubMed  PubMed Central  Google Scholar 

  38. Champiat S, Lambotte O, Barreau E, Belkhir R, Berdelou A, Carbonnel F, et al. Management of immune checkpoint blockade dysimmune toxicities: a collaborative position paper. Ann Oncol Off J Eur Soc Med Oncol. 2016;27(4):559–74.

    Article  CAS  Google Scholar 

  39. Plaçais L, Dalle S, Dereure O, Trabelsi S, Dalac S, Legoupil D, et al. Risk of irAEs in patients with autoimmune diseases treated by immune checkpoint inhibitors for stage III or IV melanoma: results from a matched case–control study. Ann Rheum Dis. 2022;81(10):1445–52.

    Article  PubMed  Google Scholar 

  40. Stanislawski MA, Dabelea D, Lange LA, Wagner BD, Lozupone CA. Gut microbiota phenotypes of obesity. NPJ Biofilms Microbiomes. 2019;5(1):18.

    Article  PubMed  PubMed Central  Google Scholar 

  41. Deng K, Shuai M, Zhang Z, Jiang Z, Fu Y, Shen L, et al. Temporal relationship among adiposity, gut microbiota, and insulin resistance in a longitudinal human cohort. BMC Med. 2022;20(1):171.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  42. Chennamadhavuni A, Abushahin L, Jin N, Presley CJ, Manne A. Risk factors and biomarkers for immune-related adverse events: a practical guide to identifying high-risk patients and rechallenging immune checkpoint inhibitors. Front Immunol. 2022. https://doi.org/10.3389/fimmu.2022.779691.

    Article  PubMed  PubMed Central  Google Scholar 

  43. Liu T, Xiong Q, Li L, Hu Y. Intestinal microbiota predicts lung cancer patients at risk of immune-related diarrhea. Immunotherapy. 2019;11(5):385–96.

    Article  PubMed  CAS  Google Scholar 

  44. Chau J, Yadav M, Liu B, Furqan M, Dai Q, Shahi S, et al. Prospective correlation between the patient microbiome with response to and development of immune-mediated adverse effects to immunotherapy in lung cancer. BMC Cancer. 2021;21(1):808.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  45. Liu W, Ma F, Sun B, Liu Y, Tang H, Luo J, et al. Intestinal microbiome associated with immune-related adverse events for patients treated with anti-PD-1 inhibitors, a real-world study. Front Immunol. 2021;16(12): 756872.

    Article  Google Scholar 

  46. Zhang Y, Cheng S, Zou H, Han Z, Xie T, Zhang B, et al. Correlation of the gut microbiome and immune-related adverse events in gastrointestinal cancer patients treated with immune checkpoint inhibitors. Front Cell Infect Microbiol. 2023. https://doi.org/10.3389/fcimb.2023.1099063.

    Article  PubMed  PubMed Central  Google Scholar 

  47. Dou X, Gao N, Yan D, Shan A. Sodium butyrate alleviates mouse colitis by regulating gut microbiota dysbiosis. Anim Open Access J MDPI. 2020;10(7):1154.

    Google Scholar 

  48. Wu M, Li P, An Y, Ren J, Yan D, Cui J, et al. Phloretin ameliorates dextran sulfate sodium-induced ulcerative colitis in mice by regulating the gut microbiota. Pharmacol Res. 2019;150: 104489.

    Article  PubMed  CAS  Google Scholar 

  49. Johanson DM, Goertz JE, Marin IA, Costello J, Overall CC, Gaultier A. Experimental autoimmune encephalomyelitis is associated with changes of the microbiota composition in the gastrointestinal tract. Sci Rep. 2020;10(1):15183.

    Article  PubMed  PubMed Central  Google Scholar 

  50. Ning J, Huang SY, Chen SD, Zhang YR, Huang YY, Yu JT. Investigating casual associations among gut microbiota, metabolites, and neurodegenerative diseases: a mendelian randomization study. J Alzheimers Dis JAD. 2022;87(1):211–22.

    Article  PubMed  CAS  Google Scholar 

  51. Routy B, Le Chatelier E, Derosa L, Duong CPM, Alou MT, Daillère R, et al. Gut microbiome influences efficacy of PD-1–based immunotherapy against epithelial tumors. Science. 2018;359(6371):91–7.

    Article  PubMed  ADS  CAS  Google Scholar 

  52. Lee KA, Thomas AM, Bolte LA, Björk JR, de Ruijter LK, Armanini F, et al. Cross-cohort gut microbiome associations with immune checkpoint inhibitor response in advanced melanoma. Nat Med. 2022;28(3):535–44.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  53. Derosa L, Routy B, Thomas AM, Iebba V, Zalcman G, Friard S, et al. Intestinal Akkermansia muciniphila predicts clinical response to PD-1 blockade in patients with advanced non-small-cell lung cancer. Nat Med. 2022;28(2):315–24.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  54. Wang Y, Wiesnoski DH, Helmink BA, Gopalakrishnan V, Choi K, DuPont HL, et al. Fecal microbiota transplantation for refractory immune checkpoint inhibitor-associated colitis. Nat Med. 2018;24(12):1804–8.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  55. Wang L, Tang L, Feng Y, Zhao S, Han M, Zhang C, et al. A purified membrane protein from Akkermansia muciniphila or the pasteurised bacterium blunts colitis associated tumourigenesis by modulation of CD8+ T cells in mice. Gut. 2020;69(11):1988–97.

    Article  PubMed  CAS  Google Scholar 

  56. Chen J, Wright K, Davis JM, Jeraldo P, Marietta EV, Murray J, et al. An expansion of rare lineage intestinal microbes characterizes rheumatoid arthritis. Genome Med. 2016;8(1):43.

    Article  PubMed  PubMed Central  Google Scholar 

  57. Miyauchi E, Shimokawa C, Steimle A, Desai MS, Ohno H. The impact of the gut microbiome on extra-intestinal autoimmune diseases. Nat Rev Immunol. 2023;23(1):9–23.

    Article  PubMed  CAS  Google Scholar 

  58. Martins F, Sofiya L, Sykiotis GP, Lamine F, Maillard M, Fraga M, et al. Adverse effects of immune-checkpoint inhibitors: epidemiology, management and surveillance. Nat Rev Clin Oncol. 2019;16(9):563–80.

    Article  PubMed  CAS  Google Scholar 

  59. Postow MA, Sidlow R, Hellmann MD. Immune-related adverse events associated with immune checkpoint blockade. N Engl J Med. 2018;378(2):158–68.

    Article  PubMed  CAS  Google Scholar 

  60. Du C, Yao SY, Ljunggren-Rose Å, Sriram S. Chlamydia pneumoniae infection of the central nervous system worsens experimental allergic encephalitis. J Exp Med. 2002;196(12):1639–44.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  61. Rodríguez Y, Rojas M, Pacheco Y, Acosta-Ampudia Y, Ramírez-Santana C, Monsalve DM, et al. Guillain-Barré syndrome, transverse myelitis and infectious diseases. Cell Mol Immunol. 2018;15(6):547–62.

    Article  PubMed  PubMed Central  Google Scholar 

  62. Gülden E, Wong FS, Wen L. The gut microbiota and type 1 diabetes. Clin Immunol. 2015;159(2):143–53.

    Article  PubMed  PubMed Central  Google Scholar 

  63. Lundberg K, Kinloch A, Fisher BA, Wegner N, Wait R, Charles P, et al. Antibodies to citrullinated α-enolase peptide 1 are specific for rheumatoid arthritis and cross-react with bacterial enolase. Arthritis Rheum. 2008;58(10):3009–19.

    Article  PubMed  CAS  Google Scholar 

  64. Tanaka A, Leung PSC, Gershwin ME. Pathogen infections and primary biliary cholangitis. Clin Exp Immunol. 2019;195(1):25–34.

    Article  PubMed  CAS  Google Scholar 

  65. Rojas M, Restrepo-Jiménez P, Monsalve DM, Pacheco Y, Acosta-Ampudia Y, Ramírez-Santana C, et al. Molecular mimicry and autoimmunity. J Autoimmun. 2018;95:100–23.

    Article  PubMed  CAS  Google Scholar 

  66. Wang SJ, Dougan SK, Dougan M. Immune mechanisms of toxicity from checkpoint inhibitors. Trends Cancer. 2023;9(7):543–53.

    Article  PubMed  CAS  Google Scholar 

  67. Park J, Kim M, Kang SG, Jannasch AH, Cooper B, Patterson J, et al. Short-chain fatty acids induce both effector and regulatory T cells by suppression of histone deacetylases and regulation of the mTOR–S6K pathway. Mucosal Immunol. 2015;8(1):80–93.

    Article  PubMed  CAS  Google Scholar 

  68. Luu M, Riester Z, Baldrich A, Reichardt N, Yuille S, Busetti A, et al. Microbial short-chain fatty acids modulate CD8+ T cell responses and improve adoptive immunotherapy for cancer. Nat Commun. 2021;12(1):4077.

    Article  PubMed  PubMed Central  ADS  CAS  Google Scholar 

  69. Zhou G, Zhang N, Meng K, Pan F. Interaction between gut microbiota and immune checkpoint inhibitor-related colitis. Front Immunol. 2022;27(13):1001623.

    Article  Google Scholar 

  70. Böcker U, Nebe T, Herweck F, Holt L, Panja A, Jobin C, et al. Butyrate modulates intestinal epithelial cell-mediated neutrophil migration. Clin Exp Immunol. 2003;131(1):53–60.

    Article  PubMed  PubMed Central  Google Scholar 

  71. Lee C, Kim BG, Kim JH, Chun J, Im JP, Kim JS. Sodium butyrate inhibits the NF-kappa B signaling pathway and histone deacetylation, and attenuates experimental colitis in an IL-10 independent manner. Int Immunopharmacol. 2017;51:47–56.

    Article  PubMed  CAS  Google Scholar 

  72. den Besten G, van Eunen K, Groen AK, Venema K, Reijngoud DJ, Bakker BM. The role of short-chain fatty acids in the interplay between diet, gut microbiota, and host energy metabolism. J Lipid Res. 2013;54(9):2325–40.

    Article  Google Scholar 

  73. Wang Y, Jenq RR, Wargo JA, Watowich SS. Microbiome influencers of checkpoint blockade-associated toxicity. J Exp Med. 2023;220(3): e20220948.

    Article  PubMed  PubMed Central  CAS  Google Scholar 

  74. Schneider BJ, Naidoo J, Santomasso BD, Lacchetti C, Adkins S, Anadkat M, et al. Management of immune-related adverse events in patients treated with immune checkpoint inhibitor therapy: ASCO guideline update. J Clin Oncol. 2021;39(36):4073–126.

    Article  PubMed  CAS  Google Scholar 

  75. Davar D, Dzutsev AK, McCulloch JA, Rodrigues RR, Chauvin JM, Morrison RM, et al. Fecal microbiota transplant overcomes resistance to anti-PD-1 therapy in melanoma patients. Science. 2021;371(6529):595–602.

    Article  PubMed  PubMed Central  ADS  CAS  Google Scholar 

  76. Baruch EN, Youngster I, Ben-Betzalel G, Ortenberg R, Lahat A, Katz L, et al. Fecal microbiota transplant promotes response in immunotherapy-refractory melanoma patients. Science. 2021;371(6529):602–9.

    Article  PubMed  ADS  CAS  Google Scholar 

  77. Chen M, Liu M, Li C, Peng S, Li Y, Xu X, et al. Fecal microbiota transplantation effectively cures a patient with severe bleeding immune checkpoint inhibitor-associated colitis and a short review. Front Oncol. 2022;10(12): 913217.

    Article  Google Scholar 

  78. Soularue E, Lepage P, Colombel JF, Coutzac C, Faleck D, Marthey L, et al. Enterocolitis due to immune checkpoint inhibitors: a systematic review. Gut. 2018;67(11):2056–67.

    Article  PubMed  CAS  Google Scholar 

Download references

Acknowledgements

We acknowledge all the investigators who have made their GWAS data available.

Funding

This study was supported by the National Nature Science Foundations of China (Grant number 82203108), China Postdoctoral Science Foundation (Grant number 2022M722275), and the Key R&D Program of Sichuan Province, China (Grant number 2023YFS0278).

Author information

Authors and Affiliations

  1. Department of General Surgery, West China Hospital, Sichuan University, Chengdu, 610041, Sichuan, People’s Republic of China

    Baike Liu, Tianxiang Jiang, Xiaonan Yin, Zhaolun Cai, Xiaoqiao Zou & Bo Zhang

  2. Gastric Cancer Center, West China Hospital, Sichuan University, Chengdu, 610041, Sichuan, People’s Republic of China

    Baike Liu, Tianxiang Jiang, Xiaonan Yin, Zhaolun Cai, Xiaoqiao Zou & Bo Zhang

  3. Department of Biotherapy and National Clinical Research Center for Geriatrics, Cancer Center, West China Hospital, Sichuan University, Chengdu, Sichuan, People’s Republic of China

    Zheran Liu

  4. State Key Laboratory of Biotherapy and Cancer Center, West China Hospital, Sichuan University and Collaborative Innovation Center for Biotherapy, Chengdu, 610041, Sichuan, People’s Republic of China

    Xiangshuai Gu & Lei Dai

Authors
  1. Baike Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Zheran Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Tianxiang Jiang
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Xiangshuai Gu
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Xiaonan Yin
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Zhaolun Cai
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Xiaoqiao Zou
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Lei Dai
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Bo Zhang
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

BL conceptualized this study. BL, ZL, and TJ were involved in the analyses and manuscript drafting of this study. XG and XZ were involved in the data curation. BL, TJ, XY, ZC, and LD were involved in the interpretation of the methodology and results. BL and ZL were involved in the visualization of the results. BZ was involved in obtaining funding and critical revision of the manuscript.

Corresponding authors

Correspondence to Lei Dai or Bo Zhang.

Ethics declarations

Ethics approval and consent to participate

Not applicable.

Competing interests

All authors declare no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Additional file 1: Figure S1.

Illustration of MR assumptions

Additional file 2: Table S1.

All Instrumental variables used for gut microbiota in the MR analysis. Table S2. Full results of primary MR analysis between bacterial taxa and all-grade irAEs. Table S3. Full results of primary MR analysis between bacterial taxa and high-grade irAEs. Table S4. Results of ΜR PRESS Global test. Table S5. Instrumental variables for irAEs in the reverse MR analysis. Table S6. Results of reverse MR analysis between all-grade irAEs and the identified significant gut bacterial taxa. Table S7. Results of reverse MR analysis between high-grade irAEs and the identified significant gut bacterial taxa. Table S8. Results of heterogeneity and horizontal pleiotropy analysis in reverse MR analysis. Table S9. PhenoScanner results of instrumental variables used for significant bacterial taxa in the MR analysis. Table S10. Description of data sets used for weight-associated anthropometric and lifestyle factors in univariable MR analysis. Table S11. Multivariate MR estimates between gut microbiota (adjusted for BMI) and irAEs.

Rights and permissions

Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Liu, B., Liu, Z., Jiang, T. et al. Univariable and multivariable Mendelian randomization study identified the key role of gut microbiota in immunotherapeutic toxicity. Eur J Med Res 29, 161 (2024). https://doi.org/10.1186/s40001-024-01741-7

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1186/s40001-024-01741-7

Keywords

European Journal of Medical Research

ISSN: 2047-783X

Contact us