The Casual Associations Between Brain Functional Networks and Fibromyalgia: A Large-Scale Genetic Correlation and Mendelian Randomization Study

Abstract

While the central mechanisms of fibromyalgia have gained attention, the causal effects between brain networks and fibromyalgia remain unclear. Two-sample Mendelian randomization and Linkage Disequilibrium Score Regression were performed to investigate the relationship between 191 rsfMRI traits and 8 fibromyalgia-related traits. A total of 4 rsfMRI traits were genetically correlated with trouble falling asleep, 11 with back pain for 3+ months, 16 with pain all over the body, 14 with insomnia, 5 with fibromyalgia, 4 with fibromyalgia, and 3 with malaise and fatigue. Pheno801 has significant causal effects on malaise and fatigue (OR = 1.0022, p = 0.01), fibromyalgia (finngen) (OR = 1.5055, p = 0.03), and insomnia (OR = 1.4063, p = 0.04). Pheno1696 significantly impacts fibromyalgia-related comorbidities (OR = 1.002, p = 0.02), trouble falling asleep (OR = 1.0285, p = 0.04), malaise and fatigue (OR = 1.0011, p = 0.04), and pain all over the body (OR = 0.9967, p = 0.04). Pheno103 has marked effects on fibromyalgia (finngen) (OR = 0.7477, p = 0.02), malaise and fatigue (OR = 0.9987, p = 0.03), and pain all over the body (OR = 1.0033, p = 0.03). Our findings suggest that targeting these networks could effectively prevent or alleviate fibromyalgia.

Keywords: brain functional networks; fibromyalgia; genetic correlations; mendelian randomization; resting-state functional MRI.

Figure 1

Workflow of two-sample mendelian randomization. Assumption 1: the IVs must be strongly associated with the exposures; Assumption 2: the IVs must be independent of the potential confounders of the association between the exposure and outcome; Assumption 3: the IVs should not be associated with the outcomes directly.

Figure 2

LDSC analysis of the genetic correlation between rsfMRI traits and fibromyalgia-related traits. Significant correlations were observed between 4 rsfMRI traits and “Trouble falling asleep”, 11 rsfMRI traits and “Back pain for 3+ months”, 16 rsfMRI traits and “Pain all over the body”, 14 rsfMRI traits and “Insomnia”, 5 rsfMRI traits and “Fibromyalgia (finngen)”, 4 rsfMRI traits and “Fibromyalgia (GCST)”, and 3 rsfMRI traits and “Malaise and fatigue”. p < 0.05 was considered significant.

Figure 3

Mendelian randomization results of rsfMRI traits and fibromyalgia-related traits. Left: the forest plot shows the significant causal relationships estimated using five MR methods (inverse variance weighted, weighted mode, weighted median, simple mode, and MR Egger). The red dots represent rsfMRI traits that have a significant effect on fibromyalgia-related traits, while the black dots represent those with no significant effect. p < 0.05 was considered significant. OR represents the effect size of a 1 standard deviation change in the mean rsfMRI phenotype on the risk of fibromyalgia, with the error bars indicating 95% confidence intervals. p-values are from the Mendelian randomization analyses, and all analyses were two-sided. Right: the pattern diagram illustrates the brain anatomical region corresponding to the rsfMRI phenotypes.

Figure 4

Interaction between rsfMRI traits and fibromyalgia-related traits. This figure highlights a total of ten causal effects involving three rsfMRI traits and six fibromyalgia-related traits. The red plus signs (+) and arrows signify risk factors, whereas the green minus signs (−) and arrows denote protective factors.