Skip to main content
ACR Open Rheumatology logoLink to ACR Open Rheumatology
. 2026 Sep 6;8(9):e90094. doi: 10.1002/acr2.90094

FADD Protein for the Early Diagnosis of Rheumatoid Arthritis

Géraldine Falgarone 1,2,3, Sara Mouasni 4, Sylvie Mistou 4, Valérie Devauchelle‐Pensec 5,6, Jacques‐Eric Gottenberg 7, Gilles Chiocchia †, Léa Tourneur 1,4,✉
PMCID: PMC13547749  PMID: 42702898

Abstract

Objective

FADD (Fas‐Associated Death Domain) protein in patients’ serum and synovial fluid could be a biomarker of inflammation in rheumatic diseases. We tested whether the baseline serum FADD level would be a predictive factor for meeting the American College of Rheumatology (ACR)/EULAR 2010 criteria for rheumatoid arthritis (RA) in order to facilitate the early diagnosis of RA.

Methods

We measured FADD by enzyme‐linked immunosorbent assay in the sera from 659 patients with early arthritis from the Etude et Suivi des Polyarthrites Indifférenciées Récentes (ESPOIR) cohort (at least two swollen joints for more than six weeks and less than six months, no previous exposure to glucocorticoids or disease‐modifying antirheumatic drugs). Patients were eligible for inclusion in the cohort if they had a definitive or probable clinical diagnosis of RA or a diagnosis of undifferentiated arthritis (UA) with a potential for progressing to RA, according to the 2010 ACR/EULAR criteria for RA.

Results

At inclusion, serum FADD protein was detected in 501 of 659 (76%) patients (range 0–1,503 ng/mL) and levels were independent of patient age and sex. Statistically significant higher levels of FADD were detected in the serum from patients with diagnosed RA at inclusion or who were diagnosed with RA thereafter, than from patients with UA. Receiver operating characteristic curve analysis indicated that a FADD >8 ng/mL had the best sensitivity (50.95%) and specificity (85.71%) for the diagnosis of RA. Moreover, FADD levels were positively associated with anti–citrullinated protein antibody (ACPA), rheumatoid factors, proinflammatory interleukin‐1β (IL‐1β), IL‐6 and IL‐2, and APRIL, a marker of B lymphocyte activation.

Conclusion

Baseline serum FADD level measurement appeared as a helpful biomarker for the early diagnosis of RA, particularly in patients who were ACPA negative.

INTRODUCTION

Rheumatoid arthritis (RA) is an autoimmune inflammatory disorder in which joints, bone, and cartilage are destroyed. 1 , 2 Early diagnosis of RA remains a challenge for clinicians in order to achieve the goal of remission. The windows of opportunity lead to improvement in therapeutic efficacy, 3 but although the earlier the treatment the better, the fear of overtreatment can also deter practitioners from starting treatments too early with the perspective of long‐lasting treatments. Indeed, there is still a need to reinforce the conviction in RA early diagnosis, whereas American College of Rheumatology (ACR)/EULAR criteria are still relying on a spectrum of clinical, biologic, and radiographic features. 4 When typical autoantibodies like anti–citrullinated protein antibodies (ACPAs) are positive, 5 this reinforces the diagnosis that is established if typical erosive synovitis is diagnosed on x‐rays. On the contrary, treatment of early arthritis is cautious and could be insufficient when criteria are not fulfilled.

The FADD protein is a common adapter molecule for the death receptors of the tumor necrosis factor receptor (TNF‐R) superfamily, including TNF‐R1 and Fas receptor. Although FADD is central in controlling several death signaling pathways, the role of FADD is not restricted to the transduction of death signals. 6 In particular, FADD can regulate cancerous and inflammatory processes. 7 Indeed, FADD can bind interleukin‐1 receptor 1 (IL‐1R1)/Toll‐like receptor (TLR) adapter protein myeloid differentiation factor 88 (MyD88). 8 , 9 This binding impairs MyD88/IL‐1 receptor–associated kinase (IRAK) interaction, which normally results in IRAK activation, then initiates the activation of a kinase cascade, ultimately leading to NF‐κB translocation to the nucleus and transcription of target genes. Because NF‐κB–targeted genes include proinflammatory cytokines like IL‐6, IL‐8, IL‐1β, and TNFα, FADD can be considered as an anti‐inflammatory protein. Confirming this hypothesis, increased NF‐κB activation and NF‐κB–dependent gene product (including IL‐6) expression were observed in FADD‐deficient mouse embryo fibroblasts in response to lipopolysaccharide or IL‐1β stimulation. 10

In RA, synovial tissues of patients are infiltrated by numerous macrophages involved in the disease pathogenesis. 11 , 12 Heat shock protein 60, extra domain A fibronectin, and hyaluronic acid, which are TLR4 ligands, are expressed in the RA joint. Proinflammatory cytokines including IL‐1β, IL‐6, and TNFα are also detected and are implicated in the joint destruction. 13 , 14 In the collagen‐induced arthritis mouse model, the schema proposed is that Fas/Fas ligand interaction among macrophages that express both molecules and are present in high number in the synovial tissues promotes chronic inflammation by sequestering FADD away from MyD88. 9 Because the antiapoptotic FLIP molecule is highly expressed in RA synovial macrophages, 15 Fas ligation does not induce cell death but instead exacerbates NF‐κB activation through the TLR4 and IL‐1R1 pathway. Thus, FADD sequestration away from MyD88 or FADD loss in the context of IL‐1R/TLR signaling would lead to an exacerbation of inflammatory processes.

Because the first role described for FADD was to interact with death receptors localized at the plasma membrane, it was initially thought that FADD protein was mainly located in the cell cytoplasm. However, it was later demonstrated that human FADD (hFADD) protein can shuttle from the cytoplasm to the nucleus and reciprocally. 16 , 17 , 18 Thereafter, we described the existence of a third unexpected localization of FADD in microvesicles, accompanied with a new trafficking pathway of the protein by secretion. 19 , 20 , 21 In mice, we demonstrated that FADD protein could be released in the extracellular compartment following shedding of plasma membrane‐derived microvesicles, a process under the control of A2B and A3 adenosine receptors (ARs). 19 Interestingly, adenosine is present at high concentration in the synovial fluid from patients with RA (10–100 μM instead of <1 μM), and ARs are implicated in the regulation of inflammatory processes. 22 In particular, adenosine acting at A2AAR and/or A3AR mediates the anti‐inflammatory effects of methotrexate, a main drug used for the treatment of inflammatory arthritis. 23 Furthermore, A3AR agonists exert significant antirheumatic effects in different autoimmune arthritis murine models. 24 For that reason, we hypothesized that FADD protein expression in the joint of a patient with RA could be regulated through FADD release in synovial fluid following microvesicle shedding. 25 Confirming this hypothesis, soluble FADD was detected at high level in synovial fluid from patients with RA and could constitute a new biomarker of inflammation in rheumatic diseases. 21 Moreover, we detected a high level of FADD in the sera from patients of the Etude et Suivi des Polyarthrites Indifférenciées Récentes (ESPOIR) cohort, 21 which is a French national multicentric cohort of patients with early inflammatory arthritis of less than six months. 26 , 27 Moreover, serum FADD level was associated with the structural severity and the rapid radiographic progression in this cohort. 21 To go further with the use of FADD as a biomarker of inflammation, we wanted to determine if serum FADD level could be a baseline predictor of fulfilling 2010 ACR/EULAR criteria for patients with RA in the ESPOIR cohort. 4 Because we have previously shown that soluble FADD is a new marker of joint inflammation in gout and RA, 21 our goal was to assess for its role in early diagnosis in RA.

PATIENTS AND METHODS

Human samples

Serum samples were taken from 659 patients with early arthritis from the ESPOIR cohort, 26 , 27 a French national observational multicentric cohort of patients with arthritis of more than six weeks and less than six months (at least two swollen joints) and not undergoing any treatment with synthetic or biologic disease‐modifying antirheumatic drugs at inclusion (cohort registered under number NCT03666091 in ClinicalTrials.gov). Patients were included in the cohort between December 2002 and March 2005 and were observed for 10 years. Patients were eligible for inclusion in the cohort if they had a definitive or probable clinical diagnosis of RA or a diagnosis of undifferentiated arthritis (UA) with potential for progressing to RA. At baseline, 518 patients met the 2010 ACR criteria for RA and 141 were classified as having UA. 4 Serum samples were collected at enrollment and immediately stored at −80°C. One biologic resources center (Sarah Tubiana, Paris‐Bichat) was in charge of centralizing and managing the biologic data collection of the cohort. The Montpellier Ethics Committee approved the study in July 2002, and all patients provided informed consent.

FADD‐specific quantitative sandwich enzyme‐linked immunosorbent assay (ELISA)

Serum samples of 72 patients were simultaneously thawed daily, and FADD levels were assessed that day. To detect soluble FADD, phosphate buffered saline–Tween 20 (PBS‐T) was added to the biologic samples to obtain a final concentration of 0.05% Tween 20 that were then heated for five minutes at 100°C, which induced microvesicles rupture. Flat‐bottom microtiter plates (Costar, 3590) were coated overnight at 4°C with anti‐hFADD antibody (5 μg/mL in PBS, clone 125–140, Calbiochem‐Novabiochem Corp). After four washes with PBS‐T 0.05%, free protein‐binding sites were blocked by adding PBS‐2% bovine serum albumin (BSA) for two hours at room temperature (RT). Then, 50 μL of human sera was added either pure or half‐diluted in PBS‐T 0.05% containing 2% BSA for two hours at RT. After four more washes, anti‐hFADD antibody (1 μg/mL in PBS‐T 0.05% containing 2% BSA, clone A66‐2; BD Pharmingen, Becton Dickinson) was added for two hours at RT for detection. After extensive washing of the plates, peroxidase‐conjugated anti‐mouse IgG1 antibody (1:8 000 in PBS‐T 0.05% containing 2% BSA; Caltag) was added for one hour at RT for revelation. The colorimetric reaction was revealed by adding 3,3′,5,5′‐tetramethyl‐benzidine (Sigma‐Aldrich) for one hour at RT and protected from light. Reaction was stopped with 2N sulfuric acid, and the plates were read at 450 nm with a Titertek multiscan spectrophotometer (Dynatec MR5000). Recombinant hFADD protein (Sigma‐Aldrich) was used to perform standard range.

Statistical tests

Statistical analyses were performed using nonparametric tests because the data did not fulfill the criteria of normal distribution and equal variance. Quantitative data analyses using the Mann–Whitney test were performed with RStudio. P values <0.05 were considered to indicate statistical significance. In figures, the following P values are given: *P < 0.05, **P < 0.01, and ***P < 0.001.

Data availability statement

The data underlying this article will be shared upon reasonable request to the corresponding author. The study protocol, consent form, case report forms, available data list, all scientific projects, newsletters, and some other documents are on the cohort ESPOIR website: http://www.lacohorteespoir.fr. The first patient was included one November 13, 2002, and the estimated completion date is June 1, 2025. The anonymized raw data are stored by the ESPOIR coordination center and are nonpublicly available.

RESULTS

Baseline characteristics

Overall, 659 of the 813 patients enrolled in the ESPOIR cohort 26 , 27 were analyzed for FADD measurement. From the 659 patients included at the baseline visit of the ESPOIR cohort, 517 were diagnosed as having RA according to the 2010 ACR/EULAR criteria for RA, 4 whereas 142 were diagnosed as having UA. Finally, 550 and 577 patients fulfilled the 2010 ACR/EULAR classification criteria for RA after two‐ and seven‐year follow‐up of the ESPOIR cohort, respectively. The baseline characteristics of the cohort are summarized in Table 1. Serum FADD was detected in 501 of 659 patients (76%) and ranged from 0 to 1,503 ng/mL. FADD levels in sera were independent of patient age and sex (Figure 1).

Table 1.

Comparison of baseline biologic characteristics of patients (n = 659) with RA and UA*

Characteristics RA (n = 517) UA (n = 142) Univariate analysis P value RA vs UA
Female, % 76 72 0.4
Age, y 50 (40–57) 50 (37–58) 0.2
One‐hour ESR, mm 24 (12–39) 18 (9–43) 0.014
CRP, mg/L 9 (3–25) 7 (0–18) 0.02
Positivity for anti‐CCP, % 50 1.4 <2.2 × 10−16
Anti‐CCP, U/mL 84 (0–570) 0 (0–0) <2.2 × 10−16
IgM‐RF positive, % 55 6.4 <2.2 × 10−16
IgM‐RF, IU/mL 17 (5–79) 4 (4–6) <2.2 × 10−16
IgA‐RF positive, % 50 7.1 <2.2 × 10−16
IgA‐RF, IU/mL 14 (6–38) 6 (5–7) <2.2 × 10−16
Initial erosion, % 55 37 6.6 × 10−5
DAS28 score 5.4 (4.6–6.2) 4.2 (3.5–4.8) <2.2 × 10−16
Detectable FADD, % 77 73 0.4
FADD, ng/mL 8.4 (0.8–20.1) 5.6 (0–13.9) 0.015
IL‐1β, pg/mL 1.6 (1.4–2.8) 1.5 (1.1–2) 0.2
IL‐1Ra, pg/mL 1,052 (759–1,504) 964 (742–1,413) 0.2
IL‐2, pg/mL 2.7 (1.3–4.7) 2.7 (1.3–3.3) 0.2
IL‐4, pg/mL 2.9 (2.2–4.8) 2.3 (2.3–2.3) 0.3
IL‐6, pg/mL 9 (3.7–20.5) 8 (3.3–18.2) 0.2
IL‐10, pg/mL 1.7 (1.3–3.2) 3.9 (2.5–4.2) 0.6
IL‐17, pg/mL 1.7 (1.2–3.8) 1.6 (1.1–3.4) 0.3
TNFα, pg/mL 2.3 (1.6–3.5) 2.3 (1.7–3.3) 0.3
IFNγ, pg/mL 1.7 (1.2–2.6) 1.4 (1.2–2.2) 0.4
MCP‐1, pg/mL 188 (136–248) 192 (139–241) 0.6
APRIL, pg/mL 2,410 (1,879–3,034) 2,356 (1,884–2,985) 0.3
BAFF, pg/mL 679 (502–918) 679 (478–841) 0.3
β2‐Microglobulin, mg/L 2 (1.7–2.4) 1.9 (1.6–2.1) 3.1 × 10−5
IgG, g/L 13.6 (11.7–16.3) 12.8 (10.9–15) 0.003
IgA, g/L 2.8 (2–3.6) 2.4 (1.8–3.2) 0.004
IgM, g/L 1.5 (1.1–2) 1.3 (0.9–1.8) 0.001
κ FLCs, mg/L 14.6 (11.1–19.2) 11.4 (9.3–15) 7.0 × 10−9
λ FLCs, mg/L 17.5 (13.3–23.2) 13.5 (11–18) 3.7 × 10−8
*

Results are expressed as median ± interquartile range. χ2 test was used to compare percentage data; otherwise, Mann–Whitney test was used. CCP, cyclic citrullinated peptide; CRP, C‐reactive protein; DAS28, Disease Activity Score using 28 joint counts; ESR, erythrocyte sedimentation rate; FLC, free light chain; IFNγ, interferon‐γ; IL‐1β, interleukin‐1β; IL‐1Ra, IL‐1 receptor antagonist; MCP‐1, monocyte chemoattractant protein 1; RA, rheumatoid arthritis; RF, rheumatoid factor; TNFα, tumor necrosis factor α; UA, undifferentiated arthritis.

Figure 1.

Figure 1

Baseline serum FADD level is independent of patient (A) age and (B) sex. FADD protein was measured by enzyme‐linked immunosorbent assay in the sera of patients from the Etude et Suivi des Polyarthrites Indifférenciées Récentes (ESPOIR) cohort at inclusion (baseline serum FADD). 26 , 27 All the patients who completed the seventh‐year visit of the ESPOIR cohort were analyzed (n = 591). The ESPOIR consortium cohort consisted of patients with untreated recently diagnosed arthritis. 2010 American College of Rheumatology/EULAR classification criteria for RA were applied. 4 Each symbol represents one patient. In (A), the bar represents mean ± SEM. Mann–Whitney test was performed. F, female; M, male; NS, not significant.

Baseline serum FADD levels as an early biomarker of RA

Statistically significant higher levels of FADD were detected in the serum from patients diagnosed with RA at inclusion than from patients with UA (mean ± SEM 36.5 ± 6.2 ng/mL vs 11.9 ± 1.7 ng/mL, respectively; Figure 2A; Table 1). Three patients classified as having UA with 46, 68, and 116 ng/mL of serum FADD, respectively, were not analyzed at years 2 and 7 because patients discontinued the study. Nevertheless, baseline serum FADD levels were also higher in patients who were negative for RA classification criteria at inclusion and who were diagnosed with RA thereafter (mean ± SEM 35 ± 5.8 ng/mL [n = 550 patients with RA] vs 10.7 ± 1.7 ng/mL [n = 67 patients with UA] at year 2, and 34 ± 5.6 ng/mL [n = 577 patients with RA] vs 5.9 ± 2.8 ng/mL [n = 14 patients with UA] at year 7; Figure 2A). Strikingly, only two patients with serum FADD >8 ng/mL—the threshold we determined using receiver operating characteristic curve to be distinguishing between RA and UA (sensitivity 50.95%, specificity 85.71%; Figure 2B)—were still diagnosed as having UA after seven‐year follow‐up (Figure 2A). Although the follow‐up of the ESPOIR cohort was at 10 years, 28 we stopped our analysis at 7 years because only 14 patients remained diagnosed as having UA at this time. Among patients fulfilling the 2010 ACR/EULAR classification criteria for RA4 after seven‐year follow‐up, 316 of 577 patients (54.8%) were negative for ACPA at inclusion. Baseline serum FADD levels were significantly higher in patients with ACPA‐positive RA than those with ACPA‐negative RA (mean ± SEM 49.8 ± 10.3 ng/mL vs 20.9 ± 5.5 ng/mL, respectively; Figure 2C). However, baseline serum FADD levels were considerably higher in patients with ACPA‐negative RA than in patients with UA (mean ± SEM ACPA‐negative RA 20.9 ± 5.5 ng/mL vs UA 5.9 ± 2.8 ng/mL; Figure 2A and C), suggesting that use of serum FADD level would have helped the early diagnosis of RA, particularly in patients who were ACPA negative. By the same way, baseline serum FADD levels of patients with no or low rheumatoid factor (RF) and/or ACPA were over those in patients with UA, whereas it was very elevated in patients with high RF or ACPA (mean ± SEM RF neg/ACPA neg 18.2 ± 5.8 vs RF low/ACPA low 18.3 ± 4.1 vs RF high/ACPA high 50.7 ± 10.2 ng/mL; Figure 2D). Moreover, high FADD baseline levels were significantly associated with both IgM‐ and IgA‐RF in patients with RA (Figure 2E). All these results demonstrated that measurement of baseline serum FADD levels could improve the early diagnosis of RA, particularly in patients with no ACPA or RF.

Figure 2.

Figure 2

Baseline serum FADD level can be used as an early diagnosis marker of rheumatoid arthritis (RA). (A) FADD protein was measured in the sera of patients from the Etude et Suivi des Polyarthrites Indifférenciées Récentes (ESPOIR) cohort at inclusion (baseline serum FADD). 26 , 27 Then, patients were diagnosed according to the 2010 American College of Rheumatology/EULAR criteria for RA as having RA or undifferentiated arthritis (UA), either at the time of inclusion or two or seven years later. 4 Some patients discontinued the study. *P < 0.05; ***P < 0.001. (B) Determination of a baseline serum FADD level threshold allowing to distinguish RA from UA. Receiver operating characteristic curve was performed using data from the patients of the ESPOIR cohort at seven years’ postinclusion. (C–E) Baseline FADD serum levels of patients from the ESPOIR cohort at inclusion were associated with (C) the absence (neg) and presence (pos) of anti–citrullinated protein antibodies (ACPA) in patients’ sera at inclusion, (D) the serology (ACPA and rheumatoid factor [RF]) or (E) IgM‐ or IgA‐ RF of patients at inclusion. *P < 0.05; **P < 0.01. Only patients diagnosed with RA at seven years’ postinclusion were analyzed. Each symbol represents one patient. Number of patients is indicated in brackets. Bar represents mean ± SEM. Mann–Whitney test was performed. CI, confidence interval; neg, negative; pos, positive.

Baseline serum FADD level as an inflammatory biomarker in RA

In patients with RA who fulfilled the 2010 ACR/EULAR classification criteria (n = 577),4 baseline serum FADD and C‐reactive protein (CRP) levels were not correlated (CRP < 6 mg/mL [n = 207], mean ± SEM FADD 32.6 ± 10 ng/mL; CRP ≥ 6 mg/L [n = 370], mean ± SEM FADD 34.7 ± 6.6 ng/mL; not significant). However, consistent with its previously described role as a marker of inflammation, 7 , 21 serum FADD levels were significantly associated with elevated levels of proinflammatory IL‐1β (mean ± SEM FADD IL‐1β < 1 pg/mL, 33.2 ± 6.1 ng/mL; FADD IL‐1β ≥ 1 pg/mL, 71.7 ± 35.6 ng/mL; Figure 3A), IL‐6 (mean ± SEM FADD IL‐6 < 1.5 pg/mL, 16.7 ± 5.1 ng/mL; FADD IL‐6 ≥ 1.5 pg/mL, 43.6 ± 8.5 ng/mL; Figure 3B), and even IL‐2 (mean ± SEM FADD IL‐2 < 1 pg/mL, 28.5 ± 6 ng/mL; FADD IL‐2 ≥ 1 pg/mL, 61.5 ± 18.7 ng/mL; Figure 3C). All other inflammatory cytokines tested were not associated with FADD level, including IL‐17, TNFα, and interferon‐γ (Figure 3D–F). Regarding IL‐10 and IL‐4 anti‐inflammatory cytokines, their levels were not significantly correlated with serum FADD levels (Figure 3G and H). Furthermore, there was no difference in serum levels of IL‐1Ra, MCP‐1, and BAFF between patients with RA with and without serum FADD detection (Figure 4A–C). By contrast, APRIL, a marker of B lymphocyte activation, was significantly higher in patients who were FADD positive than those who were FADD negative (mean± SEM 2,572 ± 44 pg/mL vs 2,297 ± 69 pg/mL, respectively; Figure 4D).

Figure 3.

Figure 3

Baseline serum FADD levels mirror the inflammatory state of patients with rheumatoid arthritis (RA). Baseline FADD levels were compared with baseline levels of pro‐ and anti‐inflammatory cytokines in the serum of patients from the Etude et Suivi des Polyarthrites Indifférenciées Récentes (ESPOIR) cohort 26 , 27 who fulfill the 2010 American College of Rheumatology/EULAR classification criteria for RA at seven years’ postinclusion. 4 FADD serum levels were associated with proinflammatory (A) interleukin‐1β (IL‐1β; <1 pg/mL, n = 498; ≥1 pg/mL, n = 26), (B) IL‐6 (<1.5 pg/mL, n = 165; ≥1.5 pg/mL, n = 359), and (C) T lymphocyte–activating IL‐2 (<1 pg/mL, n = 419; ≥1 pg/mL, n = 105) in patients’ sera at inclusion. *P < 0.05; **P < 0.01; ***P < 0.001. FADD serum levels were associated neither with (D) IL‐17 (<1 pg/mL, n = 475; ≥1 pg/mL, n = 49), (E) tumor necrosis factor α (TNFα; <1 pg/mL, n = 86; ≥1 pg/mL, n = 437), nor (F) interferon‐γ (IFNγ; <1 pg/mL, n = 485; ≥1 pg/mL, n = 39) proinflammatory cytokines, nor with (G) IL‐10 (<1 pg/mL, n = 493; ≥1 pg/mL, n = 34) nor (H) IL‐4 (<1 pg/mL, n = 471; ≥1 pg/mL, n = 53) anti‐inflammatory cytokines in patients’ sera at inclusion. Each symbol represents one donor. Bar represents mean ± SEM. Mann–Whitney test was performed. NS, not significant.

Figure 4.

Figure 4

The presence of FADD in the serum of patients with rheumatoid arthritis from the Etude et Suivi des Polyarthrites Indifférenciées Récentes (ESPOIR) cohort is correlated with a high concentration of APRIL, a marker of B lymphocyte activation. Baseline cytokines levels were compared with baseline FADD levels in the serum of patients from the ESPOIR cohort 26 , 27 who fulfill the 2010 ACR/EULAR classification criteria for rheumatoid arthritis at seven years’ postinclusion. 4 (A) Interleukin‐1 receptor antagonist protein (IL‐1Ra), (B) monocyte chemoattractant protein 1 (MCP‐1), and (C) BAFF serum levels were not associated with the absence (FADD negative [neg], n = 123, n = 123, and n = 120, respectively) and presence (FADD positive [pos], n = 401, n = 394, and n = 388, respectively) of FADD protein in patients’ sera at inclusion. (D) Serum levels of APRIL were associated with the absence (FADD neg, n = 119) and presence (FADD pos, n = 386) of FADD protein in patients’ sera at inclusion. **P < 0.01. Each symbol represents one patient. Bar represents mean ± SEM. Mann–Whitney test was performed. NS, not significant.

DISCUSSION

In an attempt to help the early diagnosis of RA, we looked at if baseline serum FADD level can improve the diagnosis of RA. Overall, we observed that a high FADD level at baseline in early arthritis could be a good predictor of fulfilling criteria of RA, independently of age, sex, and level of downstream inflammatory cytokines. We determined a threshold of 8 ng/mL could help to distinguish between RA and UA at diagnosis.

However, our study has limitations. A weakness in establishing the threshold of FADD is the low number of patients still classified as having UA at seven years’ postinclusion (n = 14). Increasing the cohort of patients who do not fulfill the 2010 ACR/EULAR classification criteria of RA could certainly improve the determination of an optimum baseline serum FADD threshold. 4 Nevertheless, we showed that FADD measurement at diagnosis would be particularly useful for patients with negative or low ACPA and/or RF. Furthermore, the imbalance between the number of patients with RA with and without detectable serum cytokine levels at the time of inclusion—for most cytokines—weakens the statistical significance of the association between cytokine levels and FADD levels.

We have previously demonstrated that human monocytes can secrete FADD protein through microvesicle shedding following the NLRP3 inflammasome activation. 21 Moreover, this secretion can occur concomitantly to IL‐1β, although it was not the case for peripheral blood mononuclear cell (PBMC)‐derived monocytes from patients with RA. 21 Here, we showed that high FADD levels were associated with high IL‐1β in the sera from patients with RA at inclusion. However, the dose of IL‐1β is tricky (only 26 patients with RA had IL‐1β ≥1 pg/mL), and it seemed not to be helpful in RA diagnosis (Table 1). For that reason, baseline serum FADD level is a better marker of inflammation than IL‐1β. FADD level could indicate the level of inflammasome‐dependent inflammation that appeared to be linked to the future RA status. Furthermore, the high baseline serum FADD levels measured in patients with RA with high baseline serum IL‐6 levels are consistent with the previously described role of IL‐6 as a marker of synovial inflammation in this ESPOIR cohort. 29 In addition, both FADD and IL‐6 were described as markers of structural progression in ESPOIR cohort. 21 , 29 For that reason, besides its role as a biomarker, FADD protein could constitute a therapeutic target in RA. 30 Recently, it was demonstrated in a Freund's complete adjuvant‐induced rat model of RA that trimetazidine (TMZ), by targeting the epigenetic expression of microRNA128a, can decrease the secretion of FADD‐containing microvesicles into synovial fluid, thereby increasing FADD expression in paw tissue. 31 Consistent with intracellular FADD's previously described role as an anti‐inflammatory molecule, 9 , 10 TMZ thus attenuates the TLR4/MyD88‐mediated inflammatory signaling pathway, ultimately ameliorating RA disease. 31

It is not surprising that serum FADD level is associated with ACPA and/or RF production because they are key factors for RA diagnosis and pathogenesis. Indeed, both ACPA and RF are involved in the activation of macrophages and consequently the secretion of inflammatory cytokines, including IL‐1β, IL‐6, and TNFα. 32 , 33 Recently, it has been shown in a citrullinated collagen‐induced arthritis mice model that ACPA can trigger both the priming and activation of the NLRP3 inflammasome. 34 Because FADD is secreted by human RA primary PBMC‐derived monocytes in response to the NLRP3 inflammasome activation, 21 ACPA would be an indirect inducer of FADD secretion. In line with this hypothesis, baseline serum FADD levels were significantly higher in patients with ACPA‐positive RA than those with ACPA‐negative RA.

In conclusion, allowing RA diagnosis independently of conventional cytokines involved in the pathophysiology of RA, which are difficult to detect, could help in early diagnosis to choose therapeutic strategy. With this work, we propose the first proof of concept of a novel biomarker to help with early diagnosis in RA, particularly for patients for whom approved biomarkers such as ACPA and RF are negative. This paves the way for the design of a clinical study to evaluate the diagnostic performance of FADD in cases of early arthritis, taking into account the choice of treatment in order to avoid any disabling sequelae in the event of postponement.

AUTHOR CONTRIBUTIONS

All authors contributed to at least one of the following manuscript preparation roles: conceptualization AND/OR methodology, software, investigation, formal analysis, data curation, visualization, and validation AND drafting or reviewing/editing the final draft. As corresponding author, Dr Tourneur confirms that all authors have provided the final approval of the version to be published and takes responsibility for the affirmations regarding article submission (eg, not under consideration by another journal), the integrity of the data presented, and the statements regarding compliance with institutional review board/Declaration of Helsinki requirements.

ROLE OF THE STUDY SPONSOR

Merck Sharp and Dohme, Pfizer, AbbVie, and Roche‐Chugai had no role in the study design or in the collection, analysis, or interpretation of the data or the decision to submit the manuscript for publication. Publication of this article was not contingent upon approval by Merck Sharp and Dohme, Pfizer, AbbVie, or Roche‐Chugai.

Supporting information

Disclosure Form:

ACR2-8-e90094-s001.pdf (901.1KB, pdf)

ACKNOWLEDGMENTS

We thank N. Rincheval (Centre Hospitalier Universitaire de Montpellier and EA 2415), who did expert monitoring and data management, and all the investigators who recruited and observed the patients: F. Berenbaum, Paris‐Saint Antoine; M. C. Boissier, Paris‐Bobigny; A. Cantagrel, Toulouse; B. Combe, Montpellier; M. Dougados, Paris‐Cochin; P. Fardelone and P. Boumier, Amiens; B. Fautrel, Paris‐La Pitié; R. M. Flipo, Lille; P. Goupille, Tours; F. Liote, Paris‐Lariboisière; O. Vittecoq, Rouen; X. Mariette, Paris‐Bicêtre; O. Meyer and P. Dieude, Paris‐Bichat; A. Saraux, Brest; T. Schaeverbeke, Bordeaux; J. Sibilia, Strasbourg; and C. Lukas, Montpellier, for expert x‐ray reading.

Supported by the Fondation ARTHRITIS. Regarding the Etude et Suivi des Polyarthrites Indifférenciées Récentes (ESPOIR) cohort, an unrestricted grant from Merck Sharp and Dohme was allocated for the first five years. Two additional grants from INSERM were obtained to support part of the biologic database. The French Society of Rheumatology, Pfizer, AbbVie, and Roche‐Chugai also supported the ESPOIR cohort study.

1Unité de Recherche SynKoMIC, Université Sorbonne Paris Nord, Bobigny, France; 2Unité de Médecine Ambulatoire, Hôpital Avicenne, Assistance Publique ‐ Hôpitaux de Paris, Bobigny, France; 3Institut Fédératif de Recherche Synerco, Université Sorbonne Paris Nord, Bobigny, France; 4Université Paris Cité, INSERM U1016, CNRS 8104, Cochin Institute, Paris, France; 5Department of Rheumatology, Hôpital de la Cavale Blanche, Centre Hospitalier Universitaire de Brest, Brest, France; 6Université de Bretagne‐Occidentale, Brest, France; 7Rheumatology Department, Strasbourg University Hospitals, Strasbourg, France.

Author disclosures are available at https://onlinelibrary.wiley.com/doi/10.1002/acr2.90094.

References

  • 1. Boissier MC, Semerano L, Challal S, et al. Rheumatoid arthritis: from autoimmunity to synovitis and joint destruction. J Autoimmun 2012;39(3):222–228. doi: 10.1016/j.jaut.2012.05.021 [DOI] [PubMed] [Google Scholar]
  • 2. Smolen JS, Aletaha D, McInnes IB. Rheumatoid arthritis. Lancet 2016;388(10055):2023–2038. doi: 10.1016/S0140-6736(16)30173-8 [DOI] [PubMed] [Google Scholar]
  • 3. Burgers LE, Raza K, van der Helm‐van Mil AH. Window of opportunity in rheumatoid arthritis ‐ definitions and supporting evidence: from old to new perspectives. RMD Open 2019;5(1):e000870. doi: 10.1136/rmdopen-2018-000870 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Aletaha D, Neogi T, Silman AJ, et al. 2010 Rheumatoid arthritis classification criteria: an American College of Rheumatology/European League Against Rheumatism collaborative initiative. Arthritis Rheum 2010;62(9):2569–2581. doi: 10.1002/art.27584 [DOI] [PubMed] [Google Scholar]
  • 5. De Rycke L, Peene I, Hoffman IEA, et al. Rheumatoid factor and anticitrullinated protein antibodies in rheumatoid arthritis: diagnostic value, associations with radiological progression rate, and extra‐articular manifestations. Ann Rheum Dis 2004;63(12):1587–1593. doi: 10.1136/ard.2003.017574 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Tourneur L, Chiocchia G. FADD: a regulator of life and death. Trends Immunol 2010;31(7):260–269. doi: 10.1016/j.it.2010.05.005 [DOI] [PubMed] [Google Scholar]
  • 7. Mouasni S, Tourneur L. FADD at the crossroads between cancer and inflammation. Trends Immunol 2018;39(12):1036–1053. doi: 10.1016/j.it.2018.10.005 [DOI] [PubMed] [Google Scholar]
  • 8. Aliprantis AO, Yang RB, Weiss DS, et al. The apoptotic signaling pathway activated by Toll‐like receptor‐2. EMBO J 2000;19(13):3325–3336. doi: 10.1093/emboj/19.13.3325 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9. Ma Y, Liu H, Tu‐Rapp H, et al. Fas ligation on macrophages enhances IL‐1R1‐Toll‐like receptor 4 signaling and promotes chronic inflammation. Nat Immunol 2004;5(4):380–387. doi: 10.1038/ni1054 [DOI] [PubMed] [Google Scholar]
  • 10. Bannerman DD, Tupper JC, Kelly JD, et al. The Fas‐associated death domain protein suppresses activation of NF‐κ B by LPS and IL‐1 β. J Clin Invest 2002;109(3):419–425. doi: 10.1172/JCI0214774 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Mulherin D, Fitzgerald O, Bresnihan B. Synovial tissue macrophage populations and articular damage in rheumatoid arthritis. Arthritis Rheum 1996;39(1):115–124. doi: 10.1002/art.1780390116 [DOI] [PubMed] [Google Scholar]
  • 12. Cuda CM, Pope RM, Perlman H. The inflammatory role of phagocyte apoptotic pathways in rheumatic diseases. Nat Rev Rheumatol 2016;12(9):543–558. doi: 10.1038/nrrheum.2016.132 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Tak PP, Smeets TJ, Daha MR, et al. Analysis of the synovial cell infiltrate in early rheumatoid synovial tissue in relation to local disease activity. Arthritis Rheum 1997;40(2):217–225. doi: 10.1002/art.1780400206 [DOI] [PubMed] [Google Scholar]
  • 14. Feldmann M. The cytokine network in rheumatoid arthritis: definition of TNF alpha as a therapeutic target. J R Coll Physicians Lond 1996;30(6):560–570. doi: 10.1016/S0035-8819(25)00312-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Perlman H, Pagliari LJ, Liu H, et al. Rheumatoid arthritis synovial macrophages express the Fas‐associated death domain‐like interleukin‐1beta‐converting enzyme‐inhibitory protein and are refractory to Fas‐mediated apoptosis. Arthritis Rheum 2001;44(1):21–30. doi: 10.1002/1529-0131(200101)44:1<21:AID-ANR4>3.0.CO;2-8 [DOI] [PubMed] [Google Scholar]
  • 16. Gómez‐Angelats M, Cidlowski JA. Molecular evidence for the nuclear localization of FADD. Cell Death Differ 2003;10(7):791–797. doi: 10.1038/sj.cdd.4401237 [DOI] [PubMed] [Google Scholar]
  • 17. Screaton RA, Kiessling S, Sansom OJ, et al. Fas‐associated death domain protein interacts with methyl‐CpG binding domain protein 4: a potential link between genome surveillance and apoptosis. Proc Natl Acad Sci U S A 2003;100(9):5211–5216. doi: 10.1073/pnas.0431215100 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18. Sheikh MS, Huang Y. The FADD is going nuclear. Cell Cycle 2003;2(4):346–347. doi: 10.4161/cc.2.4.421 [DOI] [PubMed] [Google Scholar]
  • 19. Tourneur L, Mistou S, Schmitt A, et al. Adenosine receptors control a new pathway of Fas‐associated death domain protein expression regulation by secretion. J Biol Chem 2008;283(26):17929–17938. doi: 10.1074/jbc.M802263200 [DOI] [PubMed] [Google Scholar]
  • 20. Cimino Y, Costes A, Damotte D, et al. FADD protein release mirrors the development and aggressiveness of human non‐small cell lung cancer. Br J Cancer 2012;106(12):1989–1996. doi: 10.1038/bjc.2012.196 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Mouasni S, Gonzalez V, Schmitt A, et al. The classical NLRP3 inflammasome controls FADD unconventional secretion through microvesicle shedding. Cell Death Dis 2019;10(3):190. doi: 10.1038/s41419-019-1412-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22. Ohta A, Sitkovsky M. Role of G‐protein‐coupled adenosine receptors in downregulation of inflammation and protection from tissue damage. Nature 2001;414(6866):916–920. doi: 10.1038/414916a [DOI] [PubMed] [Google Scholar]
  • 23. Montesinos MC, Desai A, Delano D, et al. Adenosine A2A or A3 receptors are required for inhibition of inflammation by methotrexate and its analog MX‐68. Arthritis Rheum 2003;48(1):240–247. doi: 10.1002/art.10712 [DOI] [PubMed] [Google Scholar]
  • 24. Baharav E, Bar‐Yehuda S, Madi L, et al. Antiinflammatory effect of A3 adenosine receptor agonists in murine autoimmune arthritis models. J Rheumatol 2005;32(3):469–476. [PubMed] [Google Scholar]
  • 25. Vilmont V, Tourneur L, Chiocchia G. Fas‐associated death domain protein and adenosine partnership: fad in RA. Rheumatology (Oxford) 2012;51(6):964–975. doi: 10.1093/rheumatology/ker402 [DOI] [PubMed] [Google Scholar]
  • 26. Combe B, Benessiano J, Berenbaum F, et al. The ESPOIR cohort: a ten‐year follow‐up of early arthritis in France: methodology and baseline characteristics of the 813 included patients. Joint Bone Spine 2007;74(5):440–445. doi: 10.1016/j.jbspin.2007.06.001 [DOI] [PubMed] [Google Scholar]
  • 27. Combe B, Rincheval N. Early lessons from the recent‐onset rheumatoid arthritis cohort ESPOIR. Joint Bone Spine 2015;82(1):13–17. doi: 10.1016/j.jbspin.2014.07.003 [DOI] [PubMed] [Google Scholar]
  • 28. Combe B, Rincheval N, Berenbaum F, et al. Current favourable 10‐year outcome of patients with early rheumatoid arthritis: data from the ESPOIR cohort. Rheumatology (Oxford) 2021;60(11):5073–5079. doi: 10.1093/rheumatology/keab398 [DOI] [PubMed] [Google Scholar]
  • 29. Baillet A, Gossec L, Paternotte S, et al. Evaluation of serum interleukin‐6 level as a surrogate marker of synovial inflammation and as a factor of structural progression in early rheumatoid arthritis: results from a French national multicenter cohort. Arthritis Care Res (Hoboken) 2015;67(7):905–912. doi: 10.1002/acr.22513 [DOI] [PubMed] [Google Scholar]
  • 30. Kobayashi T, Okamoto K, Kobata T, et al. Novel gene therapy for rheumatoid arthritis by FADD gene transfer: induction of apoptosis of rheumatoid synoviocytes but not chondrocytes. Gene Ther 2000;7(6):527–533. doi: 10.1038/sj.gt.3301127 [DOI] [PubMed] [Google Scholar]
  • 31. Omran E, Alzahrani AR, Ezzat SF, et al. Deciphering the therapeutic potential of trimetazidine in rheumatoid arthritis via targeting mi‐RNA128a, TLR4 signaling pathway, and adenosine‐induced FADD‐microvesicular shedding: in vivo and in silico study. Front Pharmacol 2024;15:1406939. doi: 10.3389/fphar.2024.1406939 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Sokolove J, Zhao X, Chandra PE, et al. Immune complexes containing citrullinated fibrinogen costimulate macrophages via Toll‐like receptor 4 and Fcγ receptor. Arthritis Rheum 2011;63(1):53–62. doi: 10.1002/art.30081 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Laurent L, Anquetil F, Clavel C, et al. IgM rheumatoid factor amplifies the inflammatory response of macrophages induced by the rheumatoid arthritis‐specific immune complexes containing anticitrullinated protein antibodies. Ann Rheum Dis 2015;74(7):1425–1431. doi: 10.1136/annrheumdis-2013-204543 [DOI] [PubMed] [Google Scholar]
  • 34. Cai L, Zhang K, Gao J, et al. ACPA‐induced ATP release and K+ efflux trigger NLRP3 inflammasome activation in rheumatoid arthritis. Cell Commun Signal 2025;23(1):302. doi: 10.1186/s12964-025-02331-8 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Disclosure Form:

ACR2-8-e90094-s001.pdf (901.1KB, pdf)

Data Availability Statement

The data underlying this article will be shared upon reasonable request to the corresponding author. The study protocol, consent form, case report forms, available data list, all scientific projects, newsletters, and some other documents are on the cohort ESPOIR website: http://www.lacohorteespoir.fr. The first patient was included one November 13, 2002, and the estimated completion date is June 1, 2025. The anonymized raw data are stored by the ESPOIR coordination center and are nonpublicly available.


Articles from ACR Open Rheumatology are provided here courtesy of Wiley

RESOURCES