Characteristic central nervous system fingerprints and related molecular signatures in difficult-to-treat rheumatoid arthritis.

Characteristic central nervous system fingerprints and related molecular signatures in difficult-to-treat rheumatoid arthritis

Background

Despite guideline-recommended treatment, approximately 30% of rheumatoid arthritis (RA) patients remain symptomatic and are classified as difficult-to-treat (D2T), representing an unmet clinical need. D2T RA is a heterogeneous state potentially involving differences across several biological and psychological levels, but the exact underlying mechanisms remain unclear. To capture this complexity, our multimodal cross-sectional study aimed to characterize the contributing factors of the D2T state with a multimodal approach.

Methods

31 D2T RA (defined by EULAR, 2021), 18 non-D2T RA patients, and 32 healthy controls (HCs) were included. All participants underwent a clinical assessment, psychological analysis (in-depth interviews, the Rorschach test, and questionnaires), two resting-state functional MRI (fMRI) scans with standardized heat pain stimulation in between, and peripheral blood transcriptomic analysis.

Results

The activity and connectivity of brain regions, especially the somatosensory cortex and the posterior cingulate cortex, one of the main hubs of the default mode network (DMN), showed significant alterations in D2T RA. Furthermore, acute pain stimulation influenced connectivity patterns of D2T patients differently compared to non-D2T patients and HCs. These results are in line with the psychological profiles of D2T patients characterized by cognitive, coping, and emotional dysregulations as well as altered pain processing. Transcriptomic analysis identified mediators, pathways, and signaling mechanisms such as neuregulin 1 (NRG1) and type I interferon signaling associated with neuroinflammation and synaptic plasticity.

Conclusion

Our findings identify central nervous system (CNS) alterations as a meaningful contributor to the D2T RA phenotype, helping to explain persistent symptom burden beyond inflammation. These characteristic neuroimmune signatures support multidisciplinary, personalized interventions that target CNS processes alongside inflammation control to improve outcomes.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12974-026-03992-z.

Background

Rheumatoid arthritis (RA) is a systemic, autoimmune disease affecting up to 1% of the population [1]. Targeted anti-inflammatory and immune-modulating therapies have provided a breakthrough in the treatment of RA; however, symptoms persist in up to ~ 27.5% of patients, even with state-of-the-art anti-inflammatory therapy [2]. This „difficult-to-treat” (D2T) RA state, as defined by the European Alliance of Associations for Rheumatology (EULAR) [3–5], is characterized by symptom persistence despite guideline-recommended treatment, which represents a substantial unmet medical need. Though D2T RA refers to a heterogeneous population [6] with factors such as treatment non-compliance, comorbidities, fibromyalgia, and health-care related issues contributing, the underlying mechanisms are not fully understood. Central nervous system (CNS) factors, however, may be particularly relevant in relation to sustained pain, and could therefore play an important role in the background of symptom persistence [2–4, 7, 8].

Complex bidirectional crosstalk between chronic joint inflammation and the brain, referred as „joint-brain axis” is considered to contribute to symptoms in RA [9]. Autoantibodies and pro-inflammatory cytokines produced by peripheral immune cells and activated microglia in the spinal dorsal horn and other pain-related brain regions may lead to sustained neuroinflammation and consequent central sensitization in arthritis [10–13]. CNS changes associated with psychosocial distress, maladaptive coping mechanisms, cognitive malfunctions, or mood disorders are likely to contribute to D2T development, but these processes have not been thoroughly investigated [14–16]. The Rorschach inkblot test [17] is a performance-based psychological assessment in which responses to ambiguous visual stimuli are evaluated to characterize aspects of personality functioning, affective processing and cognitive organization, which, although debated, has recently regained scientific interest [18]. Neuroimaging studies indicate that its complex and ambiguous stimuli activate multiple brain networks, including those involved in vision, attention, emotion, and social cognition, supporting its value for exploring implicit cognitive-emotional processes beyond self-report tools [18, 19].

Functional magnetic resonance imaging (fMRI) revealed changes in brain activation patterns in RA related to pain and inflammation [20–24] as well as in response to treatment [25, 26]. Furthermore, it has also been demonstrated that CNS alterations are reflected in the peripheral blood [27–35]. Unbiased transcriptomics of peripheral blood mononuclear cells (PBMC) proved to be a valuable tool for determining neuroinflammatory pathways involved in CNS diseases such as migraine [27].

To explore the nature of the D2T state in RA, we performed an integrative, multimodal characterization of patients by determining (I) activity and connectivity changes with fMRI baseline and after acute painful stimulation, (II) bulk RNA sequencing (RNA-seq) analysis of PBMCs, and (III) psychodiagnostic profiling, including questionnaires, the Rorschach projective test, and in-depth interviews. Furthermore, potential molecular mechanisms of therapy resistance and links between persistent inflammation and pain were investigated to identify novel prognostic biomarkers and drug targets for D2T RA.

Methods

Participant selection

RA patients (both male and female) were identified from databases at the University of Pécs and the Hospital of the Hospitaller Brothers, Budapest. After pre-screening, eligible patients were contacted, gave written consent, and underwent clinical, psychological, blood, and fMRI assessments within 3 weeks. The participants had 3 separate visits during this time period: one visit for the clinical examination and blood collection, one visit for the fMRI investigation, and one separate visit for the psychological evaluation. Patients were classified as D2T RA (n = 31) or non-D2T RA (n = 18) based on EULAR D2T RA definition [4]. D2T patients were further divided into subgroups: high inflammation–high pain (n = 10), high inflammation–low pain (n = 11), and low inflammation–high pain (n = 10). High inflammation was defined as C-reactive protein (CRP) > 5 mg/L and/or erythrocyte sedimentation rate (ESR) > 25 mm/hr; high pain as visual analogue scale (VAS) > 50 mm (past 4 weeks). The non-D2T group had ≥ 2 prior biological/targeted synthetic disease-modifying antirheumatic drugs (b/tsDMARDs) and similar disease duration, but did not fulfill D2T criteria based on their overall clinical status and well-being. Healthy controls (HCs, n = 32) were age- and sex-matched. The study complied with the Declaration of Helsinki and was approved by the Hungarian National Ethics Committee (ETT-TUKEB; BMEÜ/3788–1/2022/EKU). Demographic characteristics including age, sex, and disease duration were comparable between sites, and all assessments (fMRI, transcriptomics, psychology) were performed using identical protocols regardless of recruitment site, which limits protocol-driven confounding. (Fig. 1, Suppl. Mat.1).

Procedures and data analysis

The study consisted of four major parts for all participants: (1) a clinical examination, (2) fMRI examinations, (3) a psychological evaluation and (4) PBMC transcriptomics. (Fig. 2a, Suppl. Mat.1).

Briefly, all participants underwent a comprehensive clinical evaluation, including a detailed medical interview, physical examination, laboratory test to assess inflammatory parameters and joint ultrasound when indicated. Pain characteristics were further investigated using the PainDetect Questionnaire [36] and 2016 ACR fibromyalgia criteria [37]. The detailed psychological analysis consisted of an in-depth interview, the Rorschach test, and self-reporting questionnaires, the results of which were individually analyzed and incorporated into the qualitative analysis.

Subjects were scanned using a 3 T MRI scanner (MAGNETOM PrismaFit, Siemens-Healthineers, Erlangen, Germany) using a 64-ch head-neck coil. The MRI study protocol included the following sequences: 3D T1 magnetization-prepared rapid acquisition with gradient echo (MPRAGE), 2D FID-EPI resting-state functional MRI (rs-fMRI) with field mapping to reduce image distortions due to B0 inhomogeneities.

Rs-fMRI evaluation was carried out by using CONN (21.a) [38, 39] and SPM (12.7771) [40]. For the neuroimaging analyses, sex, age and recruitment site were originally included as covariates in the general linear models. First-level analyses included local correlation maps (LCOR) calculation estimated as the weighted average of all short-range connections between a voxel and a 25 mm FWHM Gaussian neighborhood area; intrinsic connectivity (ICC) representing a measure of node centrality at each voxel, characterized by the strength of connectivity between a given voxel and the rest of the brain; interhemispheric correlation maps (IHC) calculation characterizing the strength of homotopic functional connectivity between the two hemispheres; fALFF analysis representing a relative measure of blood-oxygen-level dependent (BOLD) signal power within the frequency band of interest (0.008–0.09 Hz) compared to that over the entire frequency spectrum; ROI-to-ROI connectivity analysis characterizing the functional connectivity between each pair of regions, and seed-based connectivity maps estimation characterizing the spatial pattern of functional connectivity with a seed area. ROIs were defined by the modified AAL atlas distributed with CONN and the MIST122 atlas [41]. Between scans, standardized thermal pain stimulation was applied to both forearms. Thermal pulses were generated by the combined fMRI model of Pathway Pain Evaluation System (Medoc Advanced Medical Systems, Ramat Yishai, Israel) with a CHEPS (Contact Heat Evoked Potential Stimulator) thermode (Suppl. Mat.1).

Peripheral blood was collected once during the study period for the bulk RNA-seq measurements in EDTA containing tube (Vacuette, Greiner Bio-One, Kremsmünster, Austria). PBMCs were isolated by density gradient centrifugation using Ficoll Histopaque-1077 (Sigma-Aldrich, Merck KGaA, Darmstadt, Germany). The purification of total RNA was carried out as previously described [27]. RNA libraries were generated using the QuantSeq 3' mRNA-Seq Library Prep Kit FWD for Illumina (Lexogen, Vienna, Austria). Illumina sequencing was performed on the NovaSeq 6000 instrument (Illumina, San Diego, CA, USA) with 1 × 75 run configuration (Suppl. Mat.1).

Statistics

For both the clinical and psychological analyses, comparisons between two groups were conducted using a two-sample t-test and a non-parametric Mann–Whitney test, while comparisons involving more than two groups employed ANOVA or the Kruskal–Wallis test. Cohen’s d was used to quantify effect size, with values > 0.5 considered medium to support the interpretation of the magnitude and clinical relevance of group differences (Suppl. Mat.1). As a sensitivity analysis, we assessed the potential influence of sex on the psychological findings by testing interaction effects between group and sex where model estimation was feasible (Suppl. Mat.2).

Transcriptomic data were further analysed with bioinformatic tools. Functional analysis (gene annotations) was prepared using the Gene Ontology (GO) and Reactome databases. While multiple testing correction reduces the likelihood of false positives, it may also mask biologically meaningful signals; therefore, in our transcriptomic analysis, results with fold change ≥ 1.2 and raw p < 0.05 were considered for interpretation to capture potentially relevant findings (Suppl. Mat.1). For transcriptomic sensitivity analyses, statistical power was estimated using RNASeqPower based on the biological variability of the voom-normalised expression data (Suppl.Mat.3), and differential expression analyses were repeated with sex included as a covariate in the limma design matrix (Suppl.Mat.4). Ingenuity Pathway Analysis (IPA) software (QIAGEN, Venlo, Netherlands), version 111,725,566, was used to run a core analysis on differentially expressed genes to identify implicated canonical pathways, regulatory networks, and upstream regulators.

Results

Clinical and demographic characteristics

A total of 81 individuals were included: 32 healthy controls (HCs) and 49 RA patients, who were divided into D2T (n = 31) and non-D2T (also defined as low inflammation, low pain (LI-LP)) (n = 18) subgroups based on the EULAR’s 2020 D2T RA definition [4] (Fig. 1, Suppl.Mat.1). The D2T group was further categorized based on the level of inflammation and pain into high inflammation, high pain (HI-HP), high inflammation, low pain (HI-LP), low inflammation, high pain (LI-HP) subgroups. Groups were broadly similar in age and sex distribution. However, comorbidities such as hypertension, fibromyalgia, and neuropathic pain were more frequent in the RA population, particularly among D2T patients. Disease activity markers (DAS28-CRP/ESR), inflammatory parameters (CRP, ESR), as well as symptom- and pain-related outcomes (VAS, PGA, PGH, PTGA, MS) showed several differences both among the D2T subgroups and in comparison to non-D2T RA and HCs. Detailed demographic and clinical characteristics are summarized in Fig. 2 and Suppl.Mat.5,6.

Altered resting-state functional activity and connectivity patterns in the brain of D2T-RA patients

RA has been linked to altered brain connectivity, reflecting the impact of the disease on neural function [20–26]. To explore this further, we used a baseline rs- fMRI to investigate how brain connectivity alterations in RA may relate to the D2T state. Intrinsic connectivity measures how strongly a given voxel is functionally connected to the rest of the brain, allowing the identification of regions with altered neural activity. This revealed the importance of the posterior cingulate cortex (PCC), a central node of the default mode network (DMN), showing significantly reduced connectivity in all RA patients compared to HCs. (Fig. 3a, Suppl.Mat.7) When analyzing further, focusing on the differences of the D2T subgroup, ANOVA identified a significant cluster in the PCC with lower intrinsic connectivity in D2T RA patients compared to HCs (Fig. 3.e–f, Suppl.Mat.7–8).

Interhemispheric connectivity, reflecting the connections between corresponding areas across the two hemispheres, was also significantly reduced in all RA patients between many bilateral areas but most significantly in the sensory cortices, PCC, insula, and temporal regions (Fig. 3b, Suppl.Mat.7). Based on these findings, a seed-based connectivity analysis was conducted, using the left postcentral gyrus (PCG) as a seed to further explore the connections of the somatosensory areas, first in all RA patients compared to HCs. Functional connectivity strength was significantly increased in RA between the seed region and the right thalamus and left angular gyrus (AG). However, consistent with interhemispheric connectivity, connectivity strength was reduced between the seed and the contralateral somatosensory cortex and additionally between the seed and bilateral temporal poles and parts of the bilateral hippocampal formation (Fig. 3c, Suppl.Mat.7). To further explore whether these connectivity alterations are distinct in the D2T population, we conducted a seed-based connectivity analysis on the subgroups also using the PCG as seed (Fig. 3k, Suppl.Mat.7). This showed significantly altered connectivity between the PCG and several areas, including the medial prefrontal cortex (MPFC), superior frontal gyrus (SFG), and medial temporal gyrus (MTG), identifying the somatosensory cortex as the center of this altered functional network during baseline. Pairwise comparisons (Fig. 3l, Suppl.Mat.8) showed that high inflammation subgroups (HI-LP, HI-HP) differed significantly from low inflammation subgroups (non-D2T RA, LI-HP) in functional connectivity strength between the somatosensory cortex and the previously defined areas (MPFC, SFG, and MTG) suggesting that inflammation, rather than pain, may be an important factor associated with these changes in D2T RA.

Region of interest (ROI)-to-ROI analysis further confirmed this altered connectivity pattern in not only all RA patients (Fig. 3d, Suppl.Mat.7), but also when comparing RA subgroups with each other. In the four patient subgroups, significantly different connectivity strength was observed between the bilateral PCG and several temporal, parietal, and frontal regions (Fig. 3i, Suppl.Mat.7). Subsequent pairwise analysis (Fig. 3j, Suppl.Mat.8), explained the previous ANOVA results; the non-D2T RA subgroup (LI-LP) showed reduced connectivity strength as compared to the HI-LP subgroup of D2T RA population.

In addition to connectivity analyses, we further investigated regional brain activity using fractional amplitude of low-frequency fluctuations (fALFF) analysis. Given our focus on the D2T condition, fALFF analysis was conducted specifically in D2T and non-D2T RA subgroups to examine altered neural activity patterns beyond connectivity changes. Lower fALFF was identified in the non-D2T RA group compared to HCs in the following areas: lateral occipital cortex (LOC), lingual gyrus, caudate, middle frontal gyrus (MFG), MPFC, PCG, and frontal pole (FP). Compared to non-D2T RA, D2T RA patients exhibited significantly reduced fALFF in the frontal operculum, MFG, SFG, LOC, occipital pole (OP), and PCG (Fig. 3g,h, Suppl.Mat.7).

These results demonstrate that D2T RA is associated with characteristic connectivity alteration, particularly in brain regions involved in sensory processing, pain modulation, and cognitive control. Detailed statistical results are provided in Suppl.Mat.7–8.

Distinct perception and processing of an acute pain stimulus of the brain of D2T RA patients and their subgroups

Pain processing involves a dynamic interaction of different brain areas related to sensory, affective, and cognitive networks, which are often altered in chronic inflammatory and pain conditions [20–26, 42–50]. Our baseline resting-state analysis revealed alterations affecting these networks. Therefore, we next investigated how acute painful stimuli may influence these connectivity patterns.

Acute pain stimulation significantly enhanced the intrinsic connectivity of the PCC in contrast to baseline, along with MPFC, left AG, and left intracalcarine cortex in all RA patients compared to HCs (Fig. 4a, Suppl.Mat.9). Further analysis focusing on the D2T population revealed significantly different intrinsic connectivity in the main nodes of the DMN (MPFC, PCC, lateral parietal areas) between D2T RA, non-D2T RA, and HCs (Fig. 4e, Suppl.Mat.9), primarily driven by the D2T RA subgroup, as shown in post-hoc pairwise comparisons (Fig. 4f, Suppl.Mat.9).

Compared to baseline, interhemispheric connectivity was also significantly enhanced in all RA patients, primarily between the bilateral sensory areas, PCC, and MTG after acute pain stimulation (Fig. 4b, Suppl.Mat.9). To further explore the role of sensory areas, seed-based connectivity analysis was conducted using the left PCG as the seed, consistent with the baseline approach. This analysis showed altered resting-state connectivity within the bilateral somatosensory areas after painful stimulation driven by a marked reduction in HCs as compared to a slight increase in all RA patients (Fig. 4c, Suppl.Mat.9). Notably, the left sensorimotor cortex showed increased local correlation driven by the D2T RA subgroup, suggesting altered local functional integration in pain-related sensory areas potentially associated with the D2T state (Fig. 4g,h, Suppl.Mat.9).

To assess broader network-level changes related to RA, multivariate pattern analysis was conducted, which identified the left MTG and right AG as key areas showing altered connections in response to acute pain in all RA patients. These areas were used as seeds in a subsequent ROI-to-ROI analysis, including the nodes of the main resting-state networks (DMN, salience (SN), dorsal attention (DAN), sensorimotor (SMN) and frontoparietal (aka central executive, CEN) networks). This comparison (Fig. 4d, Suppl.Mat.9) reported increased resting-state connectivity in RA between the right AG and the DMN, along with weaker connectivity between the left MTG and DAN and SN.

To investigate how inflammation-driven connectivity changes observed at baseline evolved following pain stimulation, further subgroup analysis was conducted. Intrinsic connectivity was increased in the HI subgroups (HI-HP, HI-LP) in the left insula/frontal opercular area (Fig. 4i, Suppl.Mat.9), and seed-based connectivity analysis using this region as a seed revealed higher functional connectivity strength with the bilateral caudate after a painful event, compared to those with low inflammation (Fig. 4j, Suppl.Mat.9). Additionally, using the right orbitofrontal cortex (OFC) as a seed showed increased functional connectivity strength in patients with high inflammation (HI-HP, HI-LP) after pain, particularly with the PCC, MPFC/paracingulate cortex, SFG, and MTG. These findings were further supported by ROI-to-ROI connectivity analysis, suggesting a potential role of inflammation in modulating pain-related network connectivity (Fig. 4k,l, Suppl.Mat.9). Detailed statistical results are provided in Suppl.Mat.9.

Major psychological alterations revealed in the background of the D2T disease state

Our psychological assessment combined individual clinical interviews, standardized questionnaires, and the Rorschach projective test to examine mental health profiles across RA subgroups. Based on the individual psychological profiling among the 31 D2T RA patients, only two were free of clinically relevant psychological conditions. The most frequent diagnoses included histrionic personality disorder (16%), anxiety disorders (10%), and other specified personality disorders (10%), with isolated cases of adjustment disorder, cyclothymic traits, and schizofreniform features. D2T RA patients frequently showed signs of emotional instability, emotional suppression, and interpersonal challenges. About one-third (32%) demonstrated cognitive difficulties, including rigid thinking and less flexible coping strategies. A tendency toward heightened illness-focused attention appeared to be associated with lower self-esteem and a negative outlook. Future orientation was frequently pessimistic or uncertain (78%), and clinically relevant depressive symptoms were reported by 42%, often linked to illness burden (Fig. 5a).

A direct comparison of D2T to non-D2T RA revealed many overlapping psychological traits but also identified some differences. In the questionnaire-based comparison, the D2T group seemed to exhibit a greater tendency to magnify pain experiences and focus on somatic complaints (PCS-Enlarge). D2T RA patients also showed higher sensitivity to pain (PSQ-H, mild and overall score), suggesting a lower pain threshold and heightened sensitivity to painful stimuli. D2T patients tended to prefer emotion-focused strategies that provide quick relief when facing difficulties over problem-focused coping (WoC stress reduction) (Fig. 5b). Depressive symptoms could be observed for 13 D2T RA patients (42%), 4 non-D2T RA patients (22%), and 2 HCs (6%).

Rorschach comparisons further characterized the differences between D2T and non-D2T RA. D2T RA patients were characterized by emotional indistinction, less mature emotion regulation, and maladaptive coping strategies. In addition to internal emotional instability, lability, and increased suffering, a predominantly outward-directed, externalizing communication pattern was observed, in which extensive verbalization and dramatized expression of distress played a central role. D2T group showed a significantly altered personality organization compared to non-D2T RA, which is characterized by increased reliance on external help, externalization of responsibility, and psychological adjustment to the disease (Suppl.Mat.12).

Overall, these data together point toward a pattern of greater psychosocial burden, reduced emotional variability, and coping strategies that are rigid or maladaptive in D2T RA. This can properly be described by our newly proposed term “morbid adaptation”, characterized by a pathological attempt to adapt accompanied by hopelessness, motivational decline, and the lack of positive outlook (Fig. 5a). These findings shed light on the complex psychological challenges that patients with D2T RA experience, indicating that their treatment with pharmacotherapeutics only might be especially difficult.

RA characteristic psychological changes are associated with a specific brain connectivity network pattern

Evaluating different parameters of the Rorschach test allows us to identify a characteristic psychological profile of a patient with features that usually are not revealed in self-report questionnaires. We utilized the Rorschach test, applying Mérei's notation, which includes various statistically derived scales and indices [51]. Our group previously developed a formula called the Rorschach Probability Index (RPI) that predicts the probability of the subject suffering from chronic pain solely on Rorschach test indices [52], indicating that chronic pain patients exhibit measurable psychological profiles (Suppl.Mat.13). The RPI is generally low among HCs, and as the RPI increases, more characteristic features of chronic pain patients are present. We tested the RPI in our cohort and found an accuracy rate of 82% in successfully identifying RA patients based on Rorschach results only. This means that RA patients also exhibit a characteristic personality profile, showing several similarities with the pathological psychological changes of a chronic pain patient; however, it does not fully overlap. RPI alone also determined a higher degree of psychological involvement in D2T patients. In the case of RPI, we measured 82.7% in the D2T group and 63.6% in the non-D2T group (Fig. 5c). The significant difference with an appropriate effect size indicates that D2T patients are more likely to suffer from mental-emotional and cognitive difficulties than non-D2T patients.

While the psychological assessment using the Rorschach test has been explored in many pathological conditions, there is growing interest in integrating somatic modalities, such as fMRI, with projective tests [18]. Therefore, we investigated how the relationship between RPI and functional brain connectivity changes across the whole study population at baseline and following acute pain. Multivariate pattern analysis identified the left SFG, which was used as a seed in the seed-based connectivity analysis, along with RPI as an independent variable of interest. This resulted in significant associations between the RPI and the functional connectivity between the seed and several brain regions (bilateral PCG/supramarginal gyrus (SMG), PCC, MPFC, AG, MTG, FP, TP, OFC, etc.). These data, together with a subsequent ROI-to-ROI analysis, showed both positive and inverse associations, e.g., higher RPI scores were associated with lower functional connectivity between the left SFG and bilateral PCG, but increased connectivity was noted with the PCC (Fig. 5d,e, Suppl.Mat.10).

After acute pain stimulus, we observed significant changes in this network (Fig. 5.f,g, Suppl.Mat.10). Notably, the previously reduced SFG-PCG connectivity in individuals with high RPI during the baseline (Fig. 5d, Suppl.Mat.10) became significantly stronger after acute pain (Fig. 5f, Suppl.Mat.10). Altogether, this identified network highlights the complex relationship between pain processing and psychological features in RA, further emphasizing the involvement of the DMN and somatosensory cortex in RA-related pathomechanisms.

Peripheral transcriptomic alterations in D2T RA reveal nervous-system-related functional annotations and associations with brain connectivity

Recent findings suggest that pathological processes in the brain may be mirrored in peripheral tissues [27–35]. To characterize peripheral transcriptional alterations associated with RA and the D2T disease state, bulk RNA-seq analysis was performed on PBMC samples of 28 D2T RA, 16 non-D2T RA and 29 HCs. When comparing RA to HCs, 269 differentially expressed (DE) mRNAs (174 upregulated and 95 downregulated in RA) were identified (Fig. 6.a,c). Functional enrichment analysis was performed using Gene Ontology (GO) and Reactome databases, which revealed the involvement of the DE genes in immune processes, oxygen transport, cell signaling, and communication (Suppl.Mat.14–15).

Next, we wanted to examine whether the transcriptomic profile of D2T patients could distinguish them from the non-D2T group. From the 87 DE mRNAs, 17 were upregulated, and 70 were downregulated (Fig. 6.b,d, Suppl.Mat.16). To identify transcriptomic alterations consistently associated with the D2T group, we compared the DE mRNAs identified in the D2T RA vs. non-D2T RA (Suppl.Mat.16) analysis with those identified in the D2T RA vs. HCs analysis (Suppl.Mat.17). The common DE transcripts include 16 downregulated mRNAs, such as EBF1, SERPINF1, or TNFRSF21, and three upregulated transcripts including NRG1, KIF3C, and CDK5 (Fig. 6.e–g). These transcripts and the related pathways potentially represent candidate peripheral molecular features associated with D2T RA; however, they require validation in independent cohorts.

Functional enrichment analysis identified GO terms categorized under “Biological Processes” such as chemical synaptic transmission, generation of neurons, memory, cognition, or behavior (Fig. 6.h–k). Additionally, enriched GO Molecular Function and Cellular Component terms included transmembrane transporter binding and neuronal cell body or synapse while Reactome analysis identified terms including nuclear signaling by ERBB4 (Suppl.Mat.18).

Given the occurrence of functional annotations related to neuronal processes in the peripheral transcriptomic results, we next examined whether selected mRNAs identified in the D2T-related transcriptomic analyses were associated with functional connectivity measures. Peripheral NRG1 expression was associated with alterations in baseline resting-state functional connectivity across all participants. Multivariate pattern analysis identified the right MFG and inferior temporal gyri (ITG) as seed regions with high functional connectivity heterogeneity, explained by NRG1 expression (Fig. 6l, Suppl.Mat.11). Seed-based connectivity analysis showed that higher NRG1 expression was linked to increased MFG connectivity with bilateral PCG, insular and opercular cortices, a cluster encompassing both the ACC and PCC, and supplementary motor cortex, with an inverse association in the left superior LOC. Similarly, higher NRG1 expression enhanced functional connectivity between ITG and frontal, parietal and temporal regions while inversely associating with medial prefrontal and posterior midline areas. These analyses identified an association between peripheral NRG1 expression and functional connectivity patterns involving brain regions relevant to pain processing, sensorimotor integration and cognitive-affective functions [21–23, 26, 45, 47, 49, 50, 53].

Peripheral transcriptomic profiles differ across pain and inflammation phenotypes in D2T RA

Due to the heterogeneity of the D2T population, transcriptomic analysis was conducted among RA subgroups with distinct pain and inflammatory parameters to identify potential subgroup-specific molecular signatures. Comparing the D2T RA HI-HP and HI-LP subgroups revealed 208 DE mRNAs mainly associated with disease-specific, inflammatory, and pain-related processes. Altogether 129 mRNAs were upregulated (e.g. GPSM1, CR2, HLA-DRB1 or IGF1) while 79 downregulated (e.g. HLA-DQ2, SIGLEC1 or IFIT1,2,3) (Fig. 7a). Functional analysis based on GO and Reactome databases revealed that transcripts downregulated in HI-HP were significantly enriched for interferon (IFN)-related pathways compared to HI-LP (Fig. 7c, Suppl.Mat.19–20). Consistently, Ingenuity Pathway Analysis (IPA) identified Interferon alpha/beta signaling as the most significantly enriched pathway in the HI-HP-HI-LP comparison and predicted its inhibition in HI-HP compared with HI-LP (Fig. 7.d,e).

In the LI-HP vs. non-D2T RA comparison 35 upregulated and 38 downregulated DE mRNAs were identified. Functional enrichment analysis identified terms related to both immune function and nervous system-related processes including the glutamate neurotransmitter release cycle, complement receptor-mediated signalling, or leukocyte migration. (Fig. 7.b, Suppl.Mat.21–22).

Discussion

To our knowledge, this is the first study providing a comprehensive, multimodal analysis of D2T RA integrating fMRI, clinical and psychological assessments, along with transcriptomic profiling. While RA primarily starts with synovial inflammation, it extends beyond the joints, impacting multiple organ systems, including the nervous system. These systemic manifestations contribute to the overall disease state, influencing symptom severity and treatment response [1, 6, 54]. Our findings support the presence of CNS-related alterations in RA, particularly in D2T patients, characterized by altered brain connectivity, changes in central pain processing, and cognitive-emotional disturbances, accompanied by peripheral molecular signatures. These alterations may contribute to symptom persistence and the complex clinical phenotype of the D2T disease state (Fig. 8).

Our results support and extend prior research demonstrating altered brain functioning in RA [20–26] and other chronic pain conditions [23, 42, 43, 46], as well as psychological disturbances [55–58]. We observed functional alterations in several brain areas, primarily the PCC, PCG, and SFG, alongside major brain networks (DMN, SMN, CEN, and SN). Previous studies in RA suggested that such alterations may be associated with a complex interplay between peripheral inflammation and central mechanisms, potentially influencing symptom severity, including pain and psychological dysfunction, as well as treatment response [20–26].

The PCC, as the central hub of the DMN, is active during rest and involved in cognitive processes, such as autobiographical memory retrieval and future planning, and regulates the shift from internal to external attention focus [53]. PCC disruptions in all RA patients, but especially observed in D2T RA, may contribute to enhanced self-referential thoughts and difficulties in switching the focus to task-related thoughts, which were also observed in the D2T population. Decreased PCC functional connectivity may also contribute to cognitive impairments in major depressive disorder (MDD) [56–58]. MDD has a higher prevalence among RA patients, which we can also observe in our cohort, especially in the D2T population and brain areas involved in MDD partially overlap with those affected in RA or in other chronic pain conditions, which may partially explain the higher prevalence of MDD and shared psychological characteristics [56, 58, 59].

We also observed significant functional connectivity disruptions involving the primary somatosensory cortex (S1) an important part of the “pain-matrix” located in the PCG [45, 47, 50]. Altered S1 function was reported in RA [21–23, 26, 49], and similar reductions in connectivity were linked to higher clinical pain severity in fibromyalgia (FM) and complex regional pain syndrome (CRPS) [43, 44, 48], suggesting that our findings of the reduced connectivity strength in RA may also be related to sustained chronic pain. Importantly, the prevalence of FM among RA patients is higher than in the normal population, which also could be observed in our cohort. FM often contributes to the D2T condition further complicating the interpretation of pain-related connectivity alterations [6]. However, this reduced functional connectivity between the bilateral somatosensory cortices changes dramatically after short-term painful stimulation. In contrast to the HC group, where the functional connectivity between the bilateral PCG is significantly reduced shortly after the painful event, the RA group exhibits a slight increase in connectivity strength. This difference alone may contribute to the emergence and the impact of central sensitization observed in RA [60]. In contrast, the enhanced hyperconnectivity with the thalamus, also observed in MDD, may be relevant to somatosensory dysregulation characteristic of both diseases [61].

Our observed fALFF reduction affecting many brain regions, particularly the LOC, LG, caudate, MFG, MPFC, precentral gyrus, and FP, potentially influences somatosensory processing [62, 63], cognitive functions [64, 65], attention [65, 66], learning [65, 67], memory [63, 65, 67] and motor control [68]. D2T patients exhibited a more pronounced reduction, potentially associated with a greater symptom severity and more complex alterations in pain modulation, behavior, and cognitive functions, aligning with their psychological profile.

Alterations of acute cerebral pain processing were previously demonstrated in RA and FM [21–23, 42, 43], and we found that D2T patients share some of these characteristic alterations after painful stimulation. Normally, an acute pain stimulus activates the SN with the involvement of the CEN in the regulatory process, in parallel with the deactivation of the DMN [69]. However, the enhanced connectivity strength in the AG and PCC in D2T patients suggests sustained DMN activity, which may contribute to enhanced pain-related self-referential processing and could be relevant to the observed pain catastrophizing [70]. In a study, painful stimulation of disease-affected joints in RA led to significantly reduced pain-related brain activation compared to HCs [22]. A similar pattern of reduced functional connectivity affecting the DMN and SMN could also be observed in the non-D2T group, potentially suggesting adaptation and compensatory mechanisms in pain perception and modulation. On the contrary, D2T patients exhibited a higher activation compared to HCs and non-D2T patients, indicating distinct alterations in how D2T patients may regulate and perceive pain. Inflammatory activity also seems to affect pain perception, as D2T subgroups with higher inflammation showed altered functional connectivity patterns in the insula/frontal opercular area and in the right OFC, which, being related to emotional processing, behavioral flexibility, and learning [71], might be relevant to pain modulatory mechanisms.

Research has demonstrated psychosocial disturbances in RA [72, 73], which we also observed. Our psychological assessment based on the interviews, questionnaires, and the Rorschach test revealed characteristic alterations, including mood instability, emotional dysregulation, cognitive difficulties, and impaired coping. While questionnaires are the most common tools for assessing psychological state in RA, they primarily capture current symptoms and are less informative about predisposing personality traits or coping mechanisms that might influence disease course [74]. Historical models of RA-specific personality patterns have been proposed [75], but systematic confirmation has been inconsistent, and it remains unclear whether such features could contribute to more persistent or refractory disease manifestations.

The Rorschach test may provide information beyond conscious defenses, revealing deeper facets of emotional functioning [18] not captured by questionnaires. Although there was a debate in the 1990s about standardized systems (e.g., Exner Comprehensive System [76] and the Mérei Hungarian system [51]), the recent introduction of the Rorschach Performance Assessment System, the R-PAS [77] has resulted in renewed interest and enabled the broader application of the test. In parallel, neuroimaging studies have also provided further support for its validity in psychological assessments, demonstrating that the test is associated with brain networks involved in vision, attention, emotion, and social cognition [18]. Beyond personality assessment, the Rorschach test has also been applied in somatic conditions. Studies in chronic inflammatory disease have shown that specific standardized variables and derived indices can predict subsequent disease activity and psychological distress [78], as well as quantify difficulties in identifying and describing emotions through the Rorschach Alexithymia Scale [79]. More recently, certain Rorschach variables have been linked to reduced perceived competence, lower self-esteem, and avoidant coping styles in autoimmune cohorts [80]. These findings highlight the value of projective methods as a complementary tool alongside other psychological assessments in clinical settings, supporting the idea that selected Rorschach variables can offer clinically relevant information on personality and psychosocial functioning that may assist in medical management.

Extending these observations, our study also examined the RPI, a composite measure of selected Rorschach parameters originally developed to estimate risk for chronic musculoskeletal pain in this clinical context [52]. Our study suggests that RA patients exhibit a characteristic personality profile that also shares common features with chronic pain patients, and that this profile is accompanied by functional connectivity pattern mainly involving the relations of SFG, PCC and PCG, which dynamically changes after painful stimulation. This altered connectivity may be associated with increased pain-related rumination that disrupts cognitive control over self-focused thoughts [81, 82], consistent with maladaptive psychological adaptation in D2T RA.

The transcriptomic analysis of peripheral blood-derived cellular components has been successfully used in neurological and neuropsychiatric conditions to identify CNS-related peripheral molecular signatures [27–35]. PBMCs represent an easily accessible immune compartment that is also transcriptionally active and cross-tissue analyses have also demonstrated considerable overlap between transcripts detectable in blood and brain tissues, suggesting that, in addition to providing information on peripheral inflammatory processes, blood-derived expression profiles may also inform about alterations related to CNS functions [34, 35]. This can be of particular importance in D2T RA, representing a more complex disease phenotype [54], with a potential contribution of the interplay of “bottom-up” inflammatory signaling acting on the brain and “top-down” mechanisms as a consequence of CNS alterations leading to symptom persistence [83]. Circulating cytokines and other inflammatory mediators may activate responsive endothelial cells in the blood–brain barrier, leading to increased permeability and immune cell trafficking [84]. In addition, the activation of afferent neural pathways and neuroendocrine mechanisms can also contribute to the consequential neuroimmune alterations [83, 85]. In contrast, CNS-derived factors can be detected in the peripheral circulation, including extracellular vesicles (EVs) carrying proteins and nucleic acids, providing a potential route by which molecular changes related to CNS processes may be accessible in blood [86]. Furthermore, emerging data in RA have reported elevated serum levels of CNS-associated proteins, including neurofilament light chain, glial fibrillary acidic protein, and tau, together with associations between these markers and peripheral inflammatory proteins, providing evidence of blood-detected neuroinflammatory burden [87]. Importantly, EVs can also be taken up by recipient cells at the periphery, facilitating a potential transfer of bioactive cargo, including proteins and nucleic acids, which may provide a potential mechanism by which CNS-related molecular signals might influence peripheral cellular responses [88].

In our cohort, bulk RNA-seq identified peripheral transcriptomic alterations associated with the D2T disease state and with subgroups characterized by different inflammatory and pain phenotypes. Functional enrichment analysis revealed several nervous-system-related annotations, and among altered transcripts, NRG1 was significantly upregulated compared with both non-D2T RA and HCs. Moreover, peripheral NRG1 expression was found to show an association across all participants with functional connectivity patterns involving regions implicated in pain processing, sensorimotor integration, and cognitive-affective function [21–23, 26, 45, 47, 49, 50, 53]. This makes NRG1 a biologically plausible candidate for further investigation, given the reported involvement of NRG1/ErbB signaling in neuronal regulation and inflammatory responses [89–92]. In addition, comparison of high inflammatory subgroups with different extents of pain revealed downregulation of IFN-related transcriptional pathways in patients with high pain. In addition to their established immunomodulatory functions, type I IFNs can exhibit both pro- and antinociceptive effects, depending on concentration and site of action, suggesting alterations in these patterns may be relevant underlying pain heterogeneity among patients with D2T RA [93–95].

However, the transcriptional profile of PBMCs cannot be regarded as a direct surrogate for brain-specific molecular alterations, especially in a disease characterized by systemic immune dysregulation; therefore, nervous system-related functional annotations identified in peripheral samples require cautious interpretation. This is particularly relevant for NRG1, a complex gene with multiple alternatively spliced isoforms and tissue- and cell-type-specific expression. Moreover, blood-based DNA methylation differences involving NRG1 have previously been reported in post-traumatic stress disorder [96], indicating that peripheral molecular findings involving NRG1 are more complex and may not be disease-specific. Accordingly, the association between peripheral NRG1 expression and functional connectivity identifies NRG1 as a candidate peripheral molecular correlate of the observed neuroimaging phenotype, rather than evidence of altered CNS NRG1/ErbB signaling. Therefore, transcriptomic findings should be interpreted as complementary to neuroimaging and psychological results and require validation in independent cohorts before disease-specific or mechanistic conclusions can be drawn.

While our study provides a detailed multimodal characterization of D2T patients, several limitations should be noted. Given the heterogeneity of this population, we aimed to integrate complementary clinical, psychological, neuroimaging and peripheral transcriptomic data to capture different aspects of the D2T disease state. The relatively small sample size reflects the high-content, resource-intensive design of this multimodal study, which required repeated visits, MRI acquisition, experimental pain stimulation, psychological assessment and biological sampling. MRI-related contraindications, claustrophobia and incomplete datasets further reduced the number of participants available for some analyses (Fig. 1), which is particularly relevant for subgroup comparisons. Although the main demographic characteristics were broadly comparable across groups and recruitment sites, the limited sample size precluded a detailed assessment of recruitment-site-related effects, and residual site-related confounding cannot be fully excluded. Similarly, although sex and age were considered where applicable (Suppl. Mat. 4, 23-27), the cohort was not sufficiently powered to examine independent sex- or age-related effects, or their interactions with clinical subgroup status. This is particularly relevant in autoimmune diseases such as RA, where demographic and biological factors may influence inflammatory responses, pain processing and peripheral immune signatures. The potential contribution of relevant comorbid conditions, including fibromyalgia and depressive symptoms, to the observed functional connectivity alterations also cannot be excluded. In addition, the transcriptomic findings require cautious interpretation. Bulk PBMC expression patterns may be influenced by peripheral immune activation and cellular composition, and cannot be regarded as direct surrogates of molecular alterations in the CNS. Moreover, pathways annotated as neuroplastic, synaptic or nervous-system-related in neuronal contexts may have distinct functions in circulating immune cells. As this was an observational study without an independent replication cohort, the transcriptomic and subgroup-level findings should be considered hypothesis-generating and require confirmation in larger, independent populations.

Nevertheless, the present data support the relevance of altered central pain processing, psychological burden and peripheral immune-related molecular signatures in the D2T RA disease state. These observations highlight the need for further studies investigating integrated management strategies that extend beyond immunosuppression alone, including non-pharmacological interventions such as cognitive-behavioral therapy, mindfulness-based stress reduction and physiotherapy in selected patients [54, 97].

Overall, our research supports the relevance of CNS-related alterations in RA, particularly among D2T patients, which is associated with altered functional brain connectivity patterns, and a distinct clinical and psychological profile. These findings provide a basis for future studies examining the underlying mechanisms of the D2T state, including further investigation the link between the peripheral immune processes, central sensitization and psychological burden. Furthermore, our results highlight the need and serve as a foundation for future research on pathomechanisms, biomarkers, novel therapeutic targets, as well as nonpharmacological management strategies.

Supplementary Information

References

1.Gravallese EM Firestein GS Rheumatoid arthritis — common origins, divergent mechanisms N Engl J Med 2023 388 6 529 542 36780677 10.1056/NEJMra2103726 Gravallese EM, Firestein GS. Rheumatoid arthritis — common origins, divergent mechanisms. N Engl J Med. 2023;388(6):529–42.10.1056/NEJMra2103726 36780677

2.Hofman ZLM Roodenrijs NMT Nikiphorou E Kent AL Nagy G Welsing PMJ Difficult-to-treat rheumatoid arthritis: what have we learned and what do we still need to learn? Rheumatology (Oxford) 2024 10.1093/rheumatology/keae544 PMC11701314 39383505 Hofman ZLM, Roodenrijs NMT, Nikiphorou E, Kent AL, Nagy G, Welsing PMJ, et al. Difficult-to-treat rheumatoid arthritis: what have we learned and what do we still need to learn? Rheumatology (Oxford). 2024. 10.1093/rheumatology/keae544 .PMC11701314 39383505

3.Buch MH Eyre S McGonagle D Persistent inflammatory and non-inflammatory mechanisms in refractory rheumatoid arthritis Nat Rev Rheumatol 2021 17 1 17 33 10.1038/s41584-020-00541-7 33293696 Buch MH, Eyre S, McGonagle D. Persistent inflammatory and non-inflammatory mechanisms in refractory rheumatoid arthritis. Nat Rev Rheumatol. 2021;17(1):17–33.10.1038/s41584-020-00541-7 33293696

4.Nagy G Roodenrijs NMT Welsing PM Kedves M Hamar A van der Goes MC EULAR definition of difficult-to-treat rheumatoid arthritis Ann Rheum Dis 2021 80 1 31 35 33004335 10.1136/annrheumdis-2020-217344 PMC7788062 Nagy G, Roodenrijs NMT, Welsing PM, Kedves M, Hamar A, van der Goes MC, et al. EULAR definition of difficult-to-treat rheumatoid arthritis. Ann Rheum Dis. 2021;80(1):31–5.10.1136/annrheumdis-2020-217344 PMC7788062 33004335

5.Paudel ML, Li R, Naik C, Shadick N, Weinblatt ME, Solomon DH. Prevalence and Characteristics of Adults with Difficult-to-Treat Rheumatoid Arthritis in a Large Patient Registry. Rheumatology (Oxford). 2025 4 .Mar 1;64(3):1102–10. 10.1093/rheumatology/keae318 .PMC11879286 38837701

6.Nagy G Gunkl-Tóth L Dorgó AM McInnes IB The concept of difficult-to-treat disease in rheumatology: where next? Lancet Rheumatol 2025 10.1016/S2665-9913(24)00340-0 39848270 Nagy G, Gunkl-Tóth L, Dorgó AM, McInnes IB. The concept of difficult-to-treat disease in rheumatology: where next? Lancet Rheumatol. 2025. 10.1016/S2665-9913(24)00340-0 .39848270

7.Roodenrijs NMT Hamar A Kedves M Nagy G van Laar JM van der Heijde D Pharmacological and non-pharmacological therapeutic strategies in difficult-to-treat rheumatoid arthritis: a systematic literature review informing the EULAR recommendations for the management of difficult-to-treat rheumatoid arthritis RMD Open 2021 10.1136/rmdopen-2020-001512 33419871 PMC7798678 Roodenrijs NMT, Hamar A, Kedves M, Nagy G, van Laar JM, van der Heijde D, et al. Pharmacological and non-pharmacological therapeutic strategies in difficult-to-treat rheumatoid arthritis: a systematic literature review informing the EULAR recommendations for the management of difficult-to-treat rheumatoid arthritis. RMD Open. 2021. 10.1136/rmdopen-2020-001512 .PMC7798678 33419871

8.Roodenrijs NMT de Hair MJH van der Goes MC Jacobs JWG Welsing PMJ van der Heijde D Characteristics of difficult-to-treat rheumatoid arthritis: results of an international survey Ann Rheum Dis 2018 77 12 1705 1709 30194273 10.1136/annrheumdis-2018-213687 Roodenrijs NMT, de Hair MJH, van der Goes MC, Jacobs JWG, Welsing PMJ, van der Heijde D, et al. Characteristics of difficult-to-treat rheumatoid arthritis: results of an international survey. Ann Rheum Dis. 2018;77(12):1705–9.10.1136/annrheumdis-2018-213687 30194273

9.Süß P Rothe T Hoffmann A Schlachetzki JCM Winkler J The Joint-Brain Axis: insights from rheumatoid arthritis on the crosstalk between chronic peripheral inflammation and the brain Front Immunol 2020 10.3389/fimmu.2020.612104 33362800 Süß P, Rothe T, Hoffmann A, Schlachetzki JCM, Winkler J. The Joint-Brain Axis: insights from rheumatoid arthritis on the crosstalk between chronic peripheral inflammation and the brain. Front Immunol. 2020. 10.3389/fimmu.2020.612104 .PMC7758283 33362800

10.Mifflin KA Kerr BJ Pain in autoimmune disorders J Neurosci Res 2017 95 6 1282 1294 27448322 10.1002/jnr.23844 Mifflin KA, Kerr BJ. Pain in autoimmune disorders. J Neurosci Res. 2017;95(6):1282–94.10.1002/jnr.23844 27448322

11.Taylor P Manger B Alvaro-Gracia J Johnstone R Gomez-Reino J Eberhardt E Patient perceptions concerning pain management in the treatment of rheumatoid arthritis J Int Med Res 2010 38 4 1213 1224 20925993 10.1177/147323001003800402 Taylor P, Manger B, Alvaro-Gracia J, Johnstone R, Gomez-Reino J, Eberhardt E, et al. Patient perceptions concerning pain management in the treatment of rheumatoid arthritis. J Int Med Res. 2010;38(4):1213–24.10.1177/147323001003800402 20925993

12.Bersellini Farinotti A Wigerblad G Nascimento D Bas DB Morado Urbina C Nandakumar KS Cartilage-binding antibodies induce pain through immune complex-mediated activation of neurons J Exp Med 2019 216 8 1904 1924 31196979 10.1084/jem.20181657 Bersellini Farinotti A, Wigerblad G, Nascimento D, Bas DB, Morado Urbina C, Nandakumar KS, et al. Cartilage-binding antibodies induce pain through immune complex-mediated activation of neurons. J Exp Med. 2019;216(8):1904–24.10.1084/jem.20181657 PMC6683987 31196979

13.Goebel A Andersson D Helyes Z Clark JD Dulake D Svensson C The autoimmune aetiology of unexplained chronic pain Autoimmun Rev 2022 21 3 103015 34902604 10.1016/j.autrev.2021.103015 PMC13390206 Goebel A, Andersson D, Helyes Z, Clark JD, Dulake D, Svensson C. The autoimmune aetiology of unexplained chronic pain. Autoimmun Rev. 2022;21(3):103015.10.1016/j.autrev.2021.103015 PMC13390206 34902604

14.Nerurkar L Siebert S McInnes IB Cavanagh J Rheumatoid arthritis and depression: an inflammatory perspective Lancet Psychiatry 2019 6 2 164 173 30366684 10.1016/S2215-0366(18)30255-4 Nerurkar L, Siebert S, McInnes IB, Cavanagh J. Rheumatoid arthritis and depression: an inflammatory perspective. Lancet Psychiatry. 2019;6(2):164–73.10.1016/S2215-0366(18)30255-4 30366684

15.Yalcin I Barrot M The anxiodepressive comorbidity in chronic pain Curr Opin Anaesthesiol 2014 27 5 520 527 25188218 10.1097/ACO.0000000000000116 Yalcin I, Barrot M. The anxiodepressive comorbidity in chronic pain. Curr Opin Anaesthesiol. 2014;27(5):520–7.10.1097/ACO.0000000000000116 25188218

16.Humo M Lu H Yalcin I The molecular neurobiology of chronic pain-induced depression Cell Tissue Res 2019 377 1 21 43 10.1007/s00441-019-03003-z 30778732 Humo M, Lu H, Yalcin I. The molecular neurobiology of chronic pain-induced depression. Cell Tissue Res. 2019;377(1):21–43.10.1007/s00441-019-03003-z 30778732

17.Rorschach H. Psychodiagnostics, a diagnostic test based on perception, including Rorschachs paper, The application of the form interpretation test (published posthumously by Dr. Emil Oberholzer): Berne, Switzerland : H. Huber; New York, N. Y. : Grune & Stratton inc., [1942] [©1942]; 1942.

18.Vecchio A Roberti E Pulido SG Orlandi M Provenzi L Mensi M Only an inkblot? A literature review of the neural correlates of the Rorschach inkblot test Neurosci Biobehav Rev 2023 152 105281 37311472 10.1016/j.neubiorev.2023.105281 Vecchio A, Roberti E, Pulido SG, Orlandi M, Provenzi L, Mensi M. Only an inkblot? A literature review of the neural correlates of the Rorschach inkblot test. Neurosci Biobehav Rev. 2023;152:105281.10.1016/j.neubiorev.2023.105281 37311472

19.Muzio E Inkblots and neurons: Correlating typical cognitive performance with brain structure and function [Editorial] Rorschachiana 2016 37 1 1 6 Muzio E. Inkblots and neurons: Correlating typical cognitive performance with brain structure and function [Editorial]. Rorschachiana. 2016;37(1):1–6.

20.Basu N Kaplan CM Ichesco E Larkin T Schrepf A Murray AD Functional and structural magnetic resonance imaging correlates of fatigue in patients with rheumatoid arthritis Rheumatology (Oxford) 2019 58 10 1822 1830 31330004 10.1093/rheumatology/kez132 Basu N, Kaplan CM, Ichesco E, Larkin T, Schrepf A, Murray AD, et al. Functional and structural magnetic resonance imaging correlates of fatigue in patients with rheumatoid arthritis. Rheumatology (Oxford). 2019;58(10):1822–30.10.1093/rheumatology/kez132 31330004

21.Flodin P Martinsen S Altawil R Waldheim E Lampa J Kosek E Intrinsic Brain Connectivity in Chronic Pain: A Resting-State fMRI Study in Patients with Rheumatoid Arthritis Front Hum Neurosci 2016 10 107 27014038 10.3389/fnhum.2016.00107 Flodin P, Martinsen S, Altawil R, Waldheim E, Lampa J, Kosek E, et al. Intrinsic Brain Connectivity in Chronic Pain: A Resting-State fMRI Study in Patients with Rheumatoid Arthritis. Front Hum Neurosci. 2016;10:107.10.3389/fnhum.2016.00107 PMC4791375 27014038

22.Sandström A Ellerbrock I Jensen KB Martinsen S Altawil R Hakeberg P Altered cerebral pain processing of noxious stimuli from inflamed joints in rheumatoid arthritis: An event-related fMRI study Brain Behav Immun 2019 81 272 279 31228612 10.1016/j.bbi.2019.06.024 Sandström A, Ellerbrock I, Jensen KB, Martinsen S, Altawil R, Hakeberg P, et al. Altered cerebral pain processing of noxious stimuli from inflamed joints in rheumatoid arthritis: An event-related fMRI study. Brain Behav Immun. 2019;81:272–9.10.1016/j.bbi.2019.06.024 31228612

23.Sandström A Ellerbrock I Löfgren M Altawil R Bileviciute-Ljungar I Lampa J Distinct aberrations in cerebral pain processing differentiating patients with fibromyalgia from patients with rheumatoid arthritis Pain 2022 163 3 538 547 34224497 10.1097/j.pain.0000000000002387 Sandström A, Ellerbrock I, Löfgren M, Altawil R, Bileviciute-Ljungar I, Lampa J, et al. Distinct aberrations in cerebral pain processing differentiating patients with fibromyalgia from patients with rheumatoid arthritis. Pain. 2022;163(3):538–47.10.1097/j.pain.0000000000002387 PMC8832547 34224497

24.Schrepf A Kaplan CM Ichesco E Larkin T Harte SE Harris RE A multi-modal MRI study of the central response to inflammation in rheumatoid arthritis Nat Commun 2018 9 1 2243 10.1038/s41467-018-04648-0 29884867 Schrepf A, Kaplan CM, Ichesco E, Larkin T, Harte SE, Harris RE, et al. A multi-modal MRI study of the central response to inflammation in rheumatoid arthritis. Nat Commun. 2018;9(1):2243.10.1038/s41467-018-04648-0 PMC5993749 29884867

25.Hess A Axmann R Rech J Finzel S Heindl C Kreitz S Blockade of TNF-α rapidly inhibits pain responses in the central nervous system Proc Natl Acad Sci 2011 108 9 3731 3736 21245297 10.1073/pnas.1011774108 Hess A, Axmann R, Rech J, Finzel S, Heindl C, Kreitz S, et al. Blockade of TNF-α rapidly inhibits pain responses in the central nervous system. Proc Natl Acad Sci. 2011;108(9):3731–6.10.1073/pnas.1011774108 PMC3048151 21245297

26.Rech J Hess A Finzel S Kreitz S Sergeeva M Englbrecht M Association of brain functional magnetic resonance activity with response to tumor necrosis factor inhibition in rheumatoid arthritis Arthritis Rheum 2013 65 2 325 333 23238986 10.1002/art.37761 Rech J, Hess A, Finzel S, Kreitz S, Sergeeva M, Englbrecht M, et al. Association of brain functional magnetic resonance activity with response to tumor necrosis factor inhibition in rheumatoid arthritis. Arthritis Rheum. 2013;65(2):325–33.10.1002/art.37761 23238986

27.Aczél T Körtési T Kun J Urbán P Bauer W Herczeg R Identification of disease- and headache-specific mediators and pathways in migraine using blood transcriptomic and metabolomic analysis J Headache Pain 2021 22 1 117 10.1186/s10194-021-01285-9 34615455 Aczél T, Körtési T, Kun J, Urbán P, Bauer W, Herczeg R, et al. Identification of disease- and headache-specific mediators and pathways in migraine using blood transcriptomic and metabolomic analysis. J Headache Pain. 2021;22(1):117.10.1186/s10194-021-01285-9 PMC8493693 34615455

28.Gerring Z Rodriguez-Acevedo AJ Powell JE Griffiths LR Montgomery GW Nyholt DR Blood gene expression studies in migraine: Potential and caveats Cephalalgia 2016 36 7 669 678 26862113 10.1177/0333102416628463 Gerring Z, Rodriguez-Acevedo AJ, Powell JE, Griffiths LR, Montgomery GW, Nyholt DR. Blood gene expression studies in migraine: Potential and caveats. Cephalalgia. 2016;36(7):669–78.10.1177/0333102416628463 26862113

29.Rollins B Martin MV Morgan L Vawter MP Analysis of whole genome biomarker expression in blood and brain Am J Med Genet B Neuropsychiatr Genet 2010 153b 4 919 936 20127885 10.1002/ajmg.b.31062 Rollins B, Martin MV, Morgan L, Vawter MP. Analysis of whole genome biomarker expression in blood and brain. Am J Med Genet B Neuropsychiatr Genet. 2010;153b(4):919–36.10.1002/ajmg.b.31062 PMC3098564 20127885

30.Sullivan PF Fan C Perou CM Evaluating the comparability of gene expression in blood and brain Am J Med Genet B Neuropsychiatr Genet 2006 141b 3 261 268 16526044 10.1002/ajmg.b.30272 Sullivan PF, Fan C, Perou CM. Evaluating the comparability of gene expression in blood and brain. Am J Med Genet B Neuropsychiatr Genet. 2006;141b(3):261–8.10.1002/ajmg.b.30272 16526044

31.Achiron A Gurevich M Peripheral blood gene expression signature mirrors central nervous system disease: The model of multiple sclerosis Autoimmun Rev 2006 5 8 517 522 17027885 10.1016/j.autrev.2006.02.009 Achiron A, Gurevich M. Peripheral blood gene expression signature mirrors central nervous system disease: The model of multiple sclerosis. Autoimmun Rev. 2006;5(8):517–22.10.1016/j.autrev.2006.02.009 17027885

32.Tikhonova MA Shvaikovskaya AA Zhanaeva SY Moysak GI Akopyan AA Rzaev JA Concordance between the In Vivo Content of Neurospecific Proteins (BDNF, NSE, VILIP-1, S100B) in the Hippocampus and Blood in Patients with Epilepsy Int J Mol Sci 2024 25 1 502 10.3390/ijms25010502 PMC10779095 38203674 Tikhonova MA, Shvaikovskaya AA, Zhanaeva SY, Moysak GI, Akopyan AA, Rzaev JA, et al. Concordance between the In Vivo Content of Neurospecific Proteins (BDNF, NSE, VILIP-1, S100B) in the Hippocampus and Blood in Patients with Epilepsy. Int J Mol Sci. 2024;25(1):502.10.3390/ijms25010502 PMC10779095 38203674

33.Jansen R Penninx BWJH Madar V Xia K Milaneschi Y Hottenga JJ Gene expression in major depressive disorder Mol Psychiatry 2016 21 3 339 347 10.1038/mp.2015.57 26008736 Jansen R, Penninx BWJH, Madar V, Xia K, Milaneschi Y, Hottenga JJ, et al. Gene expression in major depressive disorder. Mol Psychiatry. 2016;21(3):339–47.10.1038/mp.2015.57 26008736

34.Melé M Ferreira PG Reverter F DeLuca DS Monlong J Sammeth M The human transcriptome across tissues and individuals Science 2015 348 6235 660 665 25954002 10.1126/science.aaa0355 Melé M, Ferreira PG, Reverter F, DeLuca DS, Monlong J, Sammeth M, et al. The human transcriptome across tissues and individuals. Science. 2015;348(6235):660–5.10.1126/science.aaa0355 PMC4547472 25954002

35.Gerring ZF Powell JE Montgomery GW Nyholt DR Genome-wide analysis of blood gene expression in migraine implicates immune-inflammatory pathways Cephalalgia 2018 38 2 292 303 28058943 10.1177/0333102416686769 Gerring ZF, Powell JE, Montgomery GW, Nyholt DR. Genome-wide analysis of blood gene expression in migraine implicates immune-inflammatory pathways. Cephalalgia. 2018;38(2):292–303.10.1177/0333102416686769 28058943

36.Freynhagen R Baron R Gockel U Tölle TR painDETECT: a new screening questionnaire to identify neuropathic components in patients with back pain Curr Med Res Opin 2006 22 10 1911 1920 17022849 10.1185/030079906X132488 Freynhagen R, Baron R, Gockel U, Tölle TR. painDETECT: a new screening questionnaire to identify neuropathic components in patients with back pain. Curr Med Res Opin. 2006;22(10):1911–20.10.1185/030079906X132488 17022849

37.Wolfe F Clauw DJ Fitzcharles MA Goldenberg DL Häuser W Katz RL 2016 Revisions to the 2010/2011 fibromyalgia diagnostic criteria Semin Arthritis Rheum 2016 46 3 319 329 27916278 10.1016/j.semarthrit.2016.08.012 Wolfe F, Clauw DJ, Fitzcharles MA, Goldenberg DL, Häuser W, Katz RL, et al. 2016 Revisions to the 2010/2011 fibromyalgia diagnostic criteria. Semin Arthritis Rheum. 2016;46(3):319–29.10.1016/j.semarthrit.2016.08.012 27916278

38.Nieto-Castanon A, Whitfield-Gabrieli S. CONN functional connectivity toolbox: RRID SCR_009550, release 222022.10.1089/brain.2012.0073 22642651

39.Whitfield-Gabrieli S Nieto-Castanon A Conn: a functional connectivity toolbox for correlated and anticorrelated brain networks Brain Connect 2012 2 3 125 141 22642651 10.1089/brain.2012.0073 Whitfield-Gabrieli S, Nieto-Castanon A. Conn: a functional connectivity toolbox for correlated and anticorrelated brain networks. Brain Connect. 2012;2(3):125–41.10.1089/brain.2012.0073 22642651

40.Penny W, Friston K, Ashburner J, Kiebel S, Nichols T. Statistical Parametric Mapping: The Analysis of Functional Brain Images 2007. 10.1016/B978-0-12-372560-8.X5000-1 https://www.sciencedirect.com/book/edited-volume/9780123725608/statistical-parametric-mapping.

41.Urchs S Armoza J Moreau C Benhajali Y St-Aubin J Orban P MIST: a multi-resolution parcellation of functional brain networks MNI Open Res 2019 1 3 Urchs S, Armoza J, Moreau C, Benhajali Y, St-Aubin J, Orban P, et al. MIST: a multi-resolution parcellation of functional brain networks. MNI Open Res. 2019;1:3.

42.Flodin P Martinsen S Löfgren M Bileviciute-Ljungar I Kosek E Fransson P Fibromyalgia is associated with decreased connectivity between pain- and sensorimotor brain areas Brain Connect 2014 4 8 587 594 24998297 10.1089/brain.2014.0274 Flodin P, Martinsen S, Löfgren M, Bileviciute-Ljungar I, Kosek E, Fransson P. Fibromyalgia is associated with decreased connectivity between pain- and sensorimotor brain areas. Brain Connect. 2014;4(8):587–94.10.1089/brain.2014.0274 PMC4202907 24998297

43.Kim J Loggia ML Cahalan CM Harris RE Beissner F Garcia RG The Somatosensory Link in Fibromyalgia: Functional Connectivity of the Primary Somatosensory Cortex Is Altered by Sustained Pain and Is Associated With Clinical/Autonomic Dysfunction Arthritis & Rheumatology 2015 67 5 1395 1405 10.1002/art.39043 PMC4414820 25622796 Kim J, Loggia ML, Cahalan CM, Harris RE, Beissner F, Garcia RG, et al. The Somatosensory Link in Fibromyalgia: Functional Connectivity of the Primary Somatosensory Cortex Is Altered by Sustained Pain and Is Associated With Clinical/Autonomic Dysfunction. Arthritis & Rheumatology. 2015;67(5):1395–405.10.1002/art.39043 PMC4414820 25622796

44.Kim J-H Choi S-H Jang JH Lee D-H Lee K-J Lee WJ Impaired insula functional connectivity associated with persistent pain perception in patients with complex regional pain syndrome PLoS ONE 2017 12 7 e0180479 28692702 10.1371/journal.pone.0180479 Kim J-H, Choi S-H, Jang JH, Lee D-H, Lee K-J, Lee WJ, et al. Impaired insula functional connectivity associated with persistent pain perception in patients with complex regional pain syndrome. PLoS ONE. 2017;12(7):e0180479.10.1371/journal.pone.0180479 PMC5503260 28692702

45.Kim W Kim SK Nabekura J Functional and structural plasticity in the primary somatosensory cortex associated with chronic pain J Neurochem 2017 141 4 499 506 28278355 10.1111/jnc.14012 Kim W, Kim SK, Nabekura J. Functional and structural plasticity in the primary somatosensory cortex associated with chronic pain. J Neurochem. 2017;141(4):499–506.10.1111/jnc.14012 28278355

46.Ng SK Urquhart DM Fitzgerald PB Cicuttini FM Kirkovski M Maller JJ Examining resting-state functional connectivity in key hubs of the default mode network in chronic low back pain Scand J Pain 2021 21 4 839 846 34378878 10.1515/sjpain-2020-0184 Ng SK, Urquhart DM, Fitzgerald PB, Cicuttini FM, Kirkovski M, Maller JJ, et al. Examining resting-state functional connectivity in key hubs of the default mode network in chronic low back pain. Scand J Pain. 2021;21(4):839–46.10.1515/sjpain-2020-0184 34378878

47.Legrain V Iannetti GD Plaghki L Mouraux A The pain matrix reloaded: A salience detection system for the body Prog Neurobiol 2011 93 1 111 124 21040755 10.1016/j.pneurobio.2010.10.005 Legrain V, Iannetti GD, Plaghki L, Mouraux A. The pain matrix reloaded: A salience detection system for the body. Prog Neurobiol. 2011;93(1):111–24.10.1016/j.pneurobio.2010.10.005 21040755

48.Pujol J Macià D Garcia-Fontanals A Blanco-Hinojo L López-Solà M Garcia-Blanco S The contribution of sensory system functional connectivity reduction to clinical pain in fibromyalgia Pain 2014 155 8 1492 1503 24792477 10.1016/j.pain.2014.04.028 Pujol J, Macià D, Garcia-Fontanals A, Blanco-Hinojo L, López-Solà M, Garcia-Blanco S, et al. The contribution of sensory system functional connectivity reduction to clinical pain in fibromyalgia. Pain. 2014;155(8):1492–503.10.1016/j.pain.2014.04.028 24792477

49.Zheng Y Hou Z Ma S Huang Z Peng J Huang S Altered dynamic functional network connectivity in rheumatoid arthritis associated with peripheral inflammation and neuropsychiatric disorders RMD Open 2024 10 1 e003684 38428977 10.1136/rmdopen-2023-003684 Zheng Y, Hou Z, Ma S, Huang Z, Peng J, Huang S, et al. Altered dynamic functional network connectivity in rheumatoid arthritis associated with peripheral inflammation and neuropsychiatric disorders. RMD Open. 2024;10(1):e003684.10.1136/rmdopen-2023-003684 PMC10910624 38428977

50.Auvray M Myin E Spence C The sensory-discriminative and affective-motivational aspects of pain Neurosci Biobehav Rev 2010 34 2 214 223 18718486 10.1016/j.neubiorev.2008.07.008 Auvray M, Myin E, Spence C. The sensory-discriminative and affective-motivational aspects of pain. Neurosci Biobehav Rev. 2010;34(2):214–23.10.1016/j.neubiorev.2008.07.008 18718486

51.Mérei F The Rorschach-test 2002 Medicina724 Mérei F. The Rorschach-test. Medicina; 2002. p. 724.

52.Császár-Nagy N Lovas É Takács S Nagy Z An analysis of the personal characteristics of patients with chronic pain syndrome with special regard to self-empowerment and the self-regulatory and intrapsychic control system Magyar Pszichológiai Szemle 2020 75 1 103 126 Császár-Nagy N, Lovas É, Takács S, Nagy Z. An analysis of the personal characteristics of patients with chronic pain syndrome with special regard to self-empowerment and the self-regulatory and intrapsychic control system. Magyar Pszichológiai Szemle. 2020;75(1):103–26.

53.Leech R Sharp DJ The role of the posterior cingulate cortex in cognition and disease Brain 2013 137 1 12 32 23869106 10.1093/brain/awt162 Leech R, Sharp DJ. The role of the posterior cingulate cortex in cognition and disease. Brain. 2013;137(1):12–32.10.1093/brain/awt162 PMC3891440 23869106

54.Gunkl-Tóth L McInnes IB Nagy G Bridging the gap: combining treat-to-target and difficult-to-treat strategies in the management of rheumatoid arthritis Nat Rev Rheumatol 2026 22 5 319 327 10.1038/s41584-026-01354-w 41772290 Gunkl-Tóth L, McInnes IB, Nagy G. Bridging the gap: combining treat-to-target and difficult-to-treat strategies in the management of rheumatoid arthritis. Nat Rev Rheumatol. 2026;22(5):319–27.10.1038/s41584-026-01354-w 41772290

55.Yan C-G Chen X Li L Castellanos FX Bai T-J Bo Q-J Reduced default mode network functional connectivity in patients with recurrent major depressive disorder Proc Natl Acad Sci 2019 116 18 9078 9083 30979801 10.1073/pnas.1900390116 Yan C-G, Chen X, Li L, Castellanos FX, Bai T-J, Bo Q-J, et al. Reduced default mode network functional connectivity in patients with recurrent major depressive disorder. Proc Natl Acad Sci. 2019;116(18):9078–83.10.1073/pnas.1900390116 PMC6500168 30979801

56.Yang R Gao C Wu X Yang J Li S Cheng H Decreased functional connectivity to posterior cingulate cortex in major depressive disorder Psychiatry Res Neuroimaging 2016 255 15 23 27500452 10.1016/j.pscychresns.2016.07.010 Yang R, Gao C, Wu X, Yang J, Li S, Cheng H. Decreased functional connectivity to posterior cingulate cortex in major depressive disorder. Psychiatry Res Neuroimaging. 2016;255:15–23.10.1016/j.pscychresns.2016.07.010 27500452

57.Parlar M Densmore M Hall GB Frewen PA Lanius RA McKinnon MC Relation between patterns of intrinsic network connectivity, cognitive functioning, and symptom presentation in trauma-exposed patients with major depressive disorder Brain and Behavior 2017 7 5 e00664 28523217 10.1002/brb3.664 Parlar M, Densmore M, Hall GB, Frewen PA, Lanius RA, McKinnon MC. Relation between patterns of intrinsic network connectivity, cognitive functioning, and symptom presentation in trauma-exposed patients with major depressive disorder. Brain and Behavior. 2017;7(5):e00664.10.1002/brb3.664 PMC5434180 28523217

58.Zhang L Qin K Pan N Xu H Gong Q Shared and distinct patterns of default mode network dysfunction in major depressive disorder and bipolar disorder: A comparative meta-analysis J Affect Disord 2025 368 23 32 39260575 10.1016/j.jad.2024.09.021 Zhang L, Qin K, Pan N, Xu H, Gong Q. Shared and distinct patterns of default mode network dysfunction in major depressive disorder and bipolar disorder: A comparative meta-analysis. J Affect Disord. 2025;368:23–32.10.1016/j.jad.2024.09.021 39260575

59.Matcham F Rayner L Steer S Hotopf M The prevalence of depression in rheumatoid arthritis: a systematic review and meta-analysis Rheumatology (Oxford) 2013 52 12 2136 2148 24003249 10.1093/rheumatology/ket169 Matcham F, Rayner L, Steer S, Hotopf M. The prevalence of depression in rheumatoid arthritis: a systematic review and meta-analysis. Rheumatology (Oxford). 2013;52(12):2136–48.10.1093/rheumatology/ket169 PMC3828510 24003249

60.Mesci N Mesci E Kandemir EU Kulcu DG Celik T Impact of central sensitization on clinical parameters in patients with rheumatoid arthritis North Clin Istanb 2024 11 2 140 146 38757102 10.14744/nci.2023.81231 Mesci N, Mesci E, Kandemir EU, Kulcu DG, Celik T. Impact of central sensitization on clinical parameters in patients with rheumatoid arthritis. North Clin Istanb. 2024;11(2):140–6.10.14744/nci.2023.81231 PMC11095328 38757102

61.Brown EC Clark DL Hassel S MacQueen G Ramasubbu R Thalamocortical connectivity in major depressive disorder J Affect Disord 2017 217 125 131 28407555 10.1016/j.jad.2017.04.004 Brown EC, Clark DL, Hassel S, MacQueen G, Ramasubbu R. Thalamocortical connectivity in major depressive disorder. J Affect Disord. 2017;217:125–31.10.1016/j.jad.2017.04.004 28407555

62.Grill-Spector K, Kourtzi Z, Kanwisher N. The lateral occipital complex and its role in object recognition. Vision Res. 2001;41(10-11):1409–22. 10.1016/s0042-6989(01)00073-6 .11322983

63.Palejwala AH Dadario NB Young IM O'Connor K Briggs RG Conner AK Anatomy and White Matter Connections of the Lingual Gyrus and Cuneus World Neurosurgery 2021 151 e426 e437 33894399 10.1016/j.wneu.2021.04.050 Palejwala AH, Dadario NB, Young IM, O’Connor K, Briggs RG, Conner AK, et al. Anatomy and White Matter Connections of the Lingual Gyrus and Cuneus. World Neurosurgery. 2021;151:e426–37.10.1016/j.wneu.2021.04.050 33894399

64.Badre D D'Esposito M Is the rostro-caudal axis of the frontal lobe hierarchical? Nat Rev Neurosci 2009 10 9 659 669 10.1038/nrn2667 19672274 Badre D, D’Esposito M. Is the rostro-caudal axis of the frontal lobe hierarchical? Nat Rev Neurosci. 2009;10(9):659–69.10.1038/nrn2667 PMC3258028 19672274

65.Jobson DD Hase Y Clarkson AN Kalaria RN The role of the medial prefrontal cortex in cognition, ageing and dementia Brain Commun 2021 10.1093/braincomms/fcab125 34222873 Jobson DD, Hase Y, Clarkson AN, Kalaria RN. The role of the medial prefrontal cortex in cognition, ageing and dementia. Brain Commun. 2021. 10.1093/braincomms/fcab125 .PMC8249104 34222873

66.Demeter E Hernandez-Garcia L Sarter M Lustig C Challenges to attention: A continuous arterial spin labeling (ASL) study of the effects of distraction on sustained attention Neuroimage 2011 54 2 1518 1529 20851189 10.1016/j.neuroimage.2010.09.026 PMC2997179 Demeter E, Hernandez-Garcia L, Sarter M, Lustig C. Challenges to attention: A continuous arterial spin labeling (ASL) study of the effects of distraction on sustained attention. Neuroimage. 2011;54(2):1518–29.10.1016/j.neuroimage.2010.09.026 PMC2997179 20851189

67.Grahn JA Parkinson JA Owen AM The role of the basal ganglia in learning and memory: Neuropsychological studies Behav Brain Res 2009 199 1 53 60 19059285 10.1016/j.bbr.2008.11.020 Grahn JA, Parkinson JA, Owen AM. The role of the basal ganglia in learning and memory: Neuropsychological studies. Behav Brain Res. 2009;199(1):53–60.10.1016/j.bbr.2008.11.020 19059285

68.Gordon EM Chauvin RJ Van AN Rajesh A Nielsen A Newbold DJ A somato-cognitive action network alternates with effector regions in motor cortex Nature 2023 617 7960 351 359 10.1038/s41586-023-05964-2 37076628 Gordon EM, Chauvin RJ, Van AN, Rajesh A, Nielsen A, Newbold DJ, et al. A somato-cognitive action network alternates with effector regions in motor cortex. Nature. 2023;617(7960):351–9.10.1038/s41586-023-05964-2 PMC10172144 37076628

69.De Ridder D Vanneste S Smith M Adhia D Pain and the triple network model Front Neurol 2022 10.3389/fneur.2022.757241 35321511 De Ridder D, Vanneste S, Smith M, Adhia D. Pain and the triple network model. Front Neurol. 2022. 10.3389/fneur.2022.757241 .PMC8934778 35321511

70.Lee J Protsenko E Lazaridou A Franceschelli O Ellingsen D-M Mawla I Encoding of Self-Referential Pain Catastrophizing in the Posterior Cingulate Cortex in Fibromyalgia Arthritis & Rheumatology 2018 70 8 1308 1318 10.1002/art.40507 PMC6105462 29579370 Lee J, Protsenko E, Lazaridou A, Franceschelli O, Ellingsen D-M, Mawla I, et al. Encoding of Self-Referential Pain Catastrophizing in the Posterior Cingulate Cortex in Fibromyalgia. Arthritis & Rheumatology. 2018;70(8):1308–18.10.1002/art.40507 PMC6105462 29579370

71.Rudebeck PH Rich EL Orbitofrontal cortex Curr Biol 2018 28 18 R1083 R1088 30253144 10.1016/j.cub.2018.07.018 Rudebeck PH, Rich EL. Orbitofrontal cortex. Curr Biol. 2018;28(18):R1083–8.10.1016/j.cub.2018.07.018 PMC9253859 30253144

72.Sturgeon JA Finan PH Zautra AJ Affective disturbance in rheumatoid arthritis: psychological and disease-related pathways Nat Rev Rheumatol 2016 12 9 532 542 10.1038/nrrheum.2016.112 27411910 Sturgeon JA, Finan PH, Zautra AJ. Affective disturbance in rheumatoid arthritis: psychological and disease-related pathways. Nat Rev Rheumatol. 2016;12(9):532–42.10.1038/nrrheum.2016.112 PMC5449457 27411910

73.Moudi S Heidari B Yousefghahari B Gholami R Gholinia H Babaei M The prevalence and correlation of depression and anxiety with disease activity in rheumatoid arthritis Reumatologia 2023 61 2 86 91 37223374 10.5114/reum/154905 Moudi S, Heidari B, Yousefghahari B, Gholami R, Gholinia H, Babaei M. The prevalence and correlation of depression and anxiety with disease activity in rheumatoid arthritis. Reumatologia. 2023;61(2):86–91.10.5114/reum/154905 PMC10201383 37223374

74.Keefe FJ Somers TJ Psychological approaches to understanding and treating arthritis pain Nat Rev Rheumatol 2010 6 4 210 216 10.1038/nrrheum.2010.22 20357790 Keefe FJ, Somers TJ. Psychological approaches to understanding and treating arthritis pain. Nat Rev Rheumatol. 2010;6(4):210–6.10.1038/nrrheum.2010.22 20357790

75.Anderson KO Bradley LA Young LD McDaniel LK Wise CM Rheumatoid arthritis: review of psychological factors related to etiology, effects, and treatment Psychol Bull 1985 98 2 358 387 3901066 Anderson KO, Bradley LA, Young LD, McDaniel LK, Wise CM. Rheumatoid arthritis: review of psychological factors related to etiology, effects, and treatment. Psychol Bull. 1985;98(2):358–87.3901066

76.Exner JE Exner J Andronikof A Fontan P The Rorschach: a comprehensive system - revised: administration and coding manual 2022 Rorschach WorkshopsExner JE, Exner J, Andronikof A, Fontan P. The Rorschach: a comprehensive system - revised: administration and coding manual. Rorschach Workshops; 2022.

77.Meyer G Eblin J An overview of the Rorschach performance assessment system (R-PAS) Psychol Inj Law 2012 10.1007/s12207-012-9130-y Meyer G, Eblin J. An overview of the Rorschach performance assessment system (R-PAS). Psychol Inj Law. 2012. 10.1007/s12207-012-9130-y .

78.Porcelli P Rorschach Comprehensive System Predictors of Biopsychological Distress in Patients with Chronic Disease Rorschachiana 2010 31 143 171 Porcelli P. Rorschach Comprehensive System Predictors of Biopsychological Distress in Patients with Chronic Disease. Rorschachiana. 2010;31:143–71.

79.Porcelli P Mihura JL Assessment of Alexithymia With the Rorschach Comprehensive System: The Rorschach Alexithymia Scale (RAS) J Pers Assess 2010 92 2 128 136 20155562 10.1080/00223890903508146 Porcelli P, Mihura JL. Assessment of Alexithymia With the Rorschach Comprehensive System: The Rorschach Alexithymia Scale (RAS). J Pers Assess. 2010;92(2):128–36.10.1080/00223890903508146 20155562

80.Giusy M A Study Of Rorschach’s Test In Patients With Autoimmune Diseases And Possible Relation With Opioid And Cannabinoid Systems American Journal of Medical and Clinical Research & Reviews 2024 03 01 9 Giusy M. A Study Of Rorschach’s Test In Patients With Autoimmune Diseases And Possible Relation With Opioid And Cannabinoid Systems. American Journal of Medical and Clinical Research & Reviews. 2024;03:01–9.

81.Nekovarova T Fajnerova I Horacek J Spaniel F Bridging disparate symptoms of schizophrenia: a triple network dysfunction theory Front Behav Neurosci 2014 8 171 24910597 10.3389/fnbeh.2014.00171 Nekovarova T, Fajnerova I, Horacek J, Spaniel F. Bridging disparate symptoms of schizophrenia: a triple network dysfunction theory. Front Behav Neurosci. 2014;8:171.10.3389/fnbeh.2014.00171 PMC4038855 24910597

82.Hemington KS Wu Q Kucyi A Inman RD Davis KD Abnormal cross-network functional connectivity in chronic pain and its association with clinical symptoms Brain Struct Funct 2016 221 8 4203 4219 10.1007/s00429-015-1161-1 26669874 Hemington KS, Wu Q, Kucyi A, Inman RD, Davis KD. Abnormal cross-network functional connectivity in chronic pain and its association with clinical symptoms. Brain Struct Funct. 2016;221(8):4203–19.10.1007/s00429-015-1161-1 26669874

83.Kelleher EM Meouchi R Irani A Beyond inflammation: why understanding the brain matters in inflammatory arthritis Arthritis Care Res 2026 78 1 3 14 10.1002/acr.25694 PMC12826092 41236146 Kelleher EM, Meouchi R, Irani A. Beyond inflammation: why understanding the brain matters in inflammatory arthritis. Arthritis Care Res. 2026;78(1):3–14.10.1002/acr.25694 PMC12826092 41236146

84.Galea I The blood-brain barrier in systemic infection and inflammation Cell Mol Immunol 2021 18 11 2489 2501 10.1038/s41423-021-00757-x 34594000 Galea I. The blood-brain barrier in systemic infection and inflammation. Cell Mol Immunol. 2021;18(11):2489–501.10.1038/s41423-021-00757-x PMC8481764 34594000

85.Wang D, Li T, Wang W, Ruan Y, Cai J, Yan X. Pain mechanism and management strategy of rheumatoid arthritis. Frontiers in Pain Research. 2025; 6: 2025.10.3389/fpain.2025.1693399 PMC12602523 41229441

86.Shi M Sheng L Stewart T Zabetian CP Zhang J New windows into the brain: Central nervous system-derived extracellular vesicles in blood Prog Neurobiol 2019 175 96 106 30685501 10.1016/j.pneurobio.2019.01.005 Shi M, Sheng L, Stewart T, Zabetian CP, Zhang J. New windows into the brain: Central nervous system-derived extracellular vesicles in blood. Prog Neurobiol. 2019;175:96–106.10.1016/j.pneurobio.2019.01.005 PMC6546433 30685501

87.J. Brock AM, F. Sunzini, K. Stefanov, M. Arnott, J. Cavanagh, N. Basu. Central Nervous System (CNS)-Derived proteins in rheumatoid arthritis: evidence for a blood-detectable neuroinflammatory burden. Ann Rheum Dis. 2026. [ conference abstract]. 10.1136/annrheumdis-2026-eular.A.512 .

88.Maas SLN Breakefield XO Weaver AM Extracellular Vesicles: Unique Intercellular Delivery Vehicles Trends Cell Biol 2017 27 3 172 188 27979573 10.1016/j.tcb.2016.11.003 PMC5318253 Maas SLN, Breakefield XO, Weaver AM. Extracellular Vesicles: Unique Intercellular Delivery Vehicles. Trends Cell Biol. 2017;27(3):172–88.10.1016/j.tcb.2016.11.003 PMC5318253 27979573

89.Ma Y Fan P Zhao R Zhang Y Wang X Cui W Neuregulin-1 regulates the conversion of M1/M2 microglia phenotype via ErbB4-dependent inhibition of the NF-κB pathway Mol Biol Rep 2022 49 5 3975 3986 10.1007/s11033-022-07249-9 35166983 Ma Y, Fan P, Zhao R, Zhang Y, Wang X, Cui W. Neuregulin-1 regulates the conversion of M1/M2 microglia phenotype via ErbB4-dependent inhibition of the NF-κB pathway. Mol Biol Rep. 2022;49(5):3975–86.10.1007/s11033-022-07249-9 35166983

90.Mencel M Nash M Jacobson C Neuregulin Upregulates Microglial α7 Nicotinic Acetylcholine Receptor Expression in Immortalized Cell Lines: Implications for Regulating Neuroinflammation PLoS ONE 2013 8 7 e70338 23936190 10.1371/journal.pone.0070338 Mencel M, Nash M, Jacobson C. Neuregulin Upregulates Microglial α7 Nicotinic Acetylcholine Receptor Expression in Immortalized Cell Lines: Implications for Regulating Neuroinflammation. PLoS ONE. 2013;8(7):e70338.10.1371/journal.pone.0070338 PMC3728089 23936190

91.Mouton-Liger F Dumurgier J Cognat E Hourregue C Zetterberg H Vanderstichele H CSF levels of the BACE1 substrate NRG1 correlate with cognition in Alzheimer’s disease Alzheimer's Research & Therapy 2020 12 1 88 10.1186/s13195-020-00655-w PMC7372801 32690068 Mouton-Liger F, Dumurgier J, Cognat E, Hourregue C, Zetterberg H, Vanderstichele H, et al. CSF levels of the BACE1 substrate NRG1 correlate with cognition in Alzheimer’s disease. Alzheimer’s Research & Therapy. 2020;12(1):88.10.1186/s13195-020-00655-w PMC7372801 32690068

92.Tian J Geng F Gao F Chen Y-H Liu J-H Wu J-L Down-Regulation of Neuregulin1/ErbB4 Signaling in the Hippocampus Is Critical for Learning and Memory Mol Neurobiol 2017 54 6 3976 3987 10.1007/s12035-016-9956-5 27295274 Tian J, Geng F, Gao F, Chen Y-H, Liu J-H, Wu J-L, et al. Down-Regulation of Neuregulin1/ErbB4 Signaling in the Hippocampus Is Critical for Learning and Memory. Mol Neurobiol. 2017;54(6):3976–87.10.1007/s12035-016-9956-5 27295274

93.Donnelly CR Jiang C Andriessen AS Wang K Wang Z Ding H STING controls nociception via type I interferon signalling in sensory neurons Nature 2021 591 7849 275 280 10.1038/s41586-020-03151-1 33442058 Donnelly CR, Jiang C, Andriessen AS, Wang K, Wang Z, Ding H, et al. STING controls nociception via type I interferon signalling in sensory neurons. Nature. 2021;591(7849):275–80.10.1038/s41586-020-03151-1 PMC7977781 33442058

94.Tan PH Ji J Yeh CC Ji RR Interferons in Pain and Infections: Emerging Roles in Neuro-Immune and Neuro-Glial Interactions Front Immunol 2021 12 783725 34804074 10.3389/fimmu.2021.783725 Tan PH, Ji J, Yeh CC, Ji RR. Interferons in Pain and Infections: Emerging Roles in Neuro-Immune and Neuro-Glial Interactions. Front Immunol. 2021;12:783725.10.3389/fimmu.2021.783725 PMC8602180 34804074

95.Su J Zhang MD Kupari J Kwak D Picton L Xu B Persistent interferon signaling causes sensory neuron plasticity and pain before and during arthritis Nat Neurosci 2026 29 5 1095 1108 10.1038/s41593-026-02234-y 41807847 Su J, Zhang MD, Kupari J, Kwak D, Picton L, Xu B, et al. Persistent interferon signaling causes sensory neuron plasticity and pain before and during arthritis. Nat Neurosci. 2026;29(5):1095–108.10.1038/s41593-026-02234-y PMC13156039 41807847

96.Uddin M Ratanatharathorn A Armstrong D Kuan PF Aiello AE Bromet EJ Epigenetic meta-analysis across three civilian cohorts identifies NRG1 and HGS as blood-based biomarkers for post-traumatic stress disorder Epigenomics 2018 10 12 1585 1601 30456986 10.2217/epi-2018-0049 Uddin M, Ratanatharathorn A, Armstrong D, Kuan PF, Aiello AE, Bromet EJ, et al. Epigenetic meta-analysis across three civilian cohorts identifies NRG1 and HGS as blood-based biomarkers for post-traumatic stress disorder. Epigenomics. 2018;10(12):1585–601.10.2217/epi-2018-0049 PMC6331697 30456986

97.Majnik J Császár-Nagy N Böcskei G Bender T Nagy G Non-pharmacological treatment in difficult-to-treat rheumatoid arthritis Front Med (Lausanne) 2022 9 991677 36106320 10.3389/fmed.2022.991677 Majnik J, Császár-Nagy N, Böcskei G, Bender T, Nagy G. Non-pharmacological treatment in difficult-to-treat rheumatoid arthritis. Front Med (Lausanne). 2022;9:991677.10.3389/fmed.2022.991677 PMC9465607 36106320