Sage Journals: Discover world-class research

Abstract

Cell-free DNA (cfDNA) has emerged as a potential biomarker for assessing disease activity and prognosis in rheumatoid arthritis (RA). However, the association between cfDNA levels and the established RA markers of inflammation and disease severity remains unclear. The current study aimed to detect plasma levels of cfDNA in patients with RA and to investigate their association with RA activity indicators (erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), disease activity score-28 (DAS28)), prognostic markers (rheumatoid factor (RF), anticitrullinated protein antibodies (ACPA)), and the musculoskeletal ultrasonographic (US7) scores. This controlled cross-sectional study included 108 RA patients and 108 healthy controls. Plasma levels of cfDNA were quantified by real-time PCR using ALU repeats. Levels of ESR, CRP, RF, and ACPA were measured using routine laboratory assays. Synovial inflammation and joint damage evaluation was performed using the US7 scoring system. Plasma levels of cfDNA were higher in RA patients than controls (P < 0.001) and significantly increased with higher DAS28 scores among all RA activity groups. Also, cfDNA levels were significantly positively correlated with ESR, CRP, RF, and ACPA levels (P-values <0.001). Regarding US7, cfDNA was significantly positively correlated with synovitis and erosion scores (P-values <0.05) but did not correlate significantly with tenosynovitis scores (P-values >0.05). In addition, plasma cfDNA was significantly higher in seropositive RA patients than in seronegative patients (P = 0.007). The odds ratio for cfDNA as a risk factor for erosions was 2.254. This study revealed that cfDNA levels are elevated in RA patients and positively associated with disease activity indicators and prognosis markers. Further research is warranted to validate these findings in larger cohorts and explore the clinical implications of cfDNA measurement in RA management.

Keywords

activity cell-free DNA DAS28 prognosis rheumatoid arthritis ultrasound

Introduction

The progressive destruction and erosion of joints is the hallmark of rheumatoid arthritis (RA), a chronic complex autoimmune inflammatory disease.¹ The timing of the diagnosis, early therapy initiation, disease activity, hereditary variables, and autoantibodies influence how quickly the disease progresses.² An essential part of diagnosing RA is done through serological testing, specifically the identification of rheumatoid factor (RF) and anti-citrullinated protein antibodies (ACPA).³ Musculoskeletal ultrasonography (US), in addition to clinical and serological evaluations, is becoming a crucial diagnostic and disease-progression tracking tool.² However, although the established RA diagnostic methods contribute to the early identification of RA, additional molecular markers are needed, especially in seronegative patients, to assess prognosis and responsiveness to treatment and help in the stratification of patients.

Circulating cell-free deoxyribonucleic acid (cfDNA) is now emerging as a useful tool for non-invasive diagnostic methods related to a wide range of clinical conditions, including autoimmune diseases. It is defined as the extracellular fragmented DNA that can be derived from healthy and diseased cells.⁴ Its release is mainly associated with cell death mechanisms, besides its active secretion. Moreover, the balance between cfDNA release and clearance mechanisms determines its levels in blood; when clearance is insufficient in high cell turnover, pathological conditions, and chronic inflammation, cfDNA accumulates.⁵ Previous studies reported that plasma cfDNA levels increase in RA and correlate with symptom severity and that the cfDNA concentration in synovial fluid is several times higher than in circulation, indicating the importance of local inflammation in the cfDNA release.^6,7 However, the dynamics of cfDNA in RA are still to be fully elucidated.

The current study aimed to detect plasma levels of cfDNA in seropositive and seronegative patients with RA and to investigate cfDNA level association with the established RA activity and prognostic markers, and the musculoskeletal US scoring system.

Methodology

Sample size calculation

Using the STATA 10 program for sample size calculation and based on the results of Duvvuri et al.,⁸ with a sensitivity of cfDNA in the detection of RA of 84% and postulated sensitivity of 90%, with an alpha error of 5% and power of study of 80%, the required sample size was 108 RA patients.

Ethical considerations

The study was carried out in accordance with the Declaration of Helsinki’s standards for research. Ain Shams University, Faculty of Medicine, Research Ethics Committee gave the study its ethical approval (#FMASU R 139/2024). Before being included in the study, each subject gave written, informed consent. All information was maintained confidential and was solely utilized for research.

Study subjects

The current controlled cross-sectional study was conducted on 108 adults (age ≥ 18 years) RA patients and 108 age- and sex-matched healthy controls in the period from. The RA patients were recruited from the Rheumatology Clinic, Ain Shams University, Cairo, Egypt; all fulfilled the ACR/EULAR 2010 classification criteria for RA.⁹ They were classified according to the disease activity score-28 (DAS28)¹⁰ into four groups: inactive (score <2.6), low activity (score 2.6–3.1), moderate activity (score 3.2–5), and high activity (score >5.1), with 27 patients in each group. All the selected RA patients were on conventional disease-modifying antirheumatic drug (DMARD) treatment (including Hydroxychloroquine, Leuflonamide, Methotrexate, and Sulfasalazine) either as monotherapy or combined. We excluded RA patients who were on biologic or target synthetic DMARDs. Patients with concomitant autoimmune diseases, acute or chronic infections and inflammation, co-existing malignancies, and pregnancy were also excluded from the study.

Musculoskeletal ultrasound examination (MSUS)

Ultrasound examination was performed by an expert sonographer who was blinded to the clinical and laboratory findings in each patient by utilizing equipment (Logic P5-R4.0, General Electric, Milwaukee, Wisconsin, USA with multifrequency linear probe ranging from 7 to 13 MHz using EULAR guidelines for grayscale (GS) and power doppler (PD), an assessment conducted in longitudinal and transverse views of the wrist, metacarpophalangeal joint (MCP) II and III, phalangeal interproximal (PIP) II and III, and metatarsophalangeal joint (MTP) II and V joints of the clinically most affected hand and forefoot. The German 7-joint musculoskeletal ultrasonographic (US7) score is a semiquantitative US scoring system that combines soft tissue changes (synovitis and tenosynovitis) and erosive bone lesions in seven preselected joints.¹¹

Laboratory investigations

Four milliliters (mL) of venous blood were withdrawn from each participant under aseptic conditions and subdivided into three tubes: the first was a vacutainer tube containing 3.2% sodium citrate for ESR analysis by Westergren method,¹² the second was a gel vacutainer tube, in which the blood was allowed to clot and then centrifuged for 10 min at 4000 rpm and the resultant serum was used to measure RF and c-reactive protein (CRP) using the Cobas 6000, c501 autoanalyzer (Roche Diagnostics, Switzerland) and ACPA using the Cobas e411 autoanalyzer (Roche Diagnostics, Switzerland). The normal cut-off for CRP was <5 mg/L, for RF was <14 IU/mL, and for APCA was <17.0 U/mL.

For cfDNA analysis, eight millilitres (8 mL) of blood were withdrawn from each participant into two k2-2EDTA vacutainer tubes (4 mL each). Each k2-EDTA tube was subjected to two centrifugation steps within 2 h of collection. The first centrifugation was at 1600 × g for 10 min and the resultant plasma supernatant was subjected to a second high-speed centrifugation at 16,000 × g for 10 min.

PCR workflow

Extraction of cfDNA

We used 4 mL of the collected plasma to extract cfDNA by the QIAamp^® Circulating Nucleic Acid Kit (QIAGEN, Germany). The extraction steps were done as directed by the kit manual. The concentration of the extracted cfDNA was determined using Qubit™ 4 Fluorometer (Thermo Fisher Scientific Inc., USA). The purity of cfDNA was assessed using a Nanodrop One Spectrophotometer (Thermo Fisher Scientific Inc., USA) by calculating the ratio between the absorbance value measured at 260 nm and the absorbance value measured at 280 nm. Any ratio from 1.8 to 2 was considered appropriate.

Amplification of cfDNA by real-time PCR using ALU repeats

As DNA repeats make up more than 50% of the human genome and constitute a major part of cfDNA fragments, it was expected to be a good marker for cfDNA measurement.¹³ The Alu repeat 115 fragments were amplified using Maxima SYBR Green qPCR Master Mix (2X) (Thermo Scientific, USA) and the Alu 115 primer set (Invitrogen, USA) (forward primer: 5’ CCTGAGGTCAGGAGTTCGAG’3; reverse primer: 5’CCCGAGTAGCTGGGATTACA’3). TaqMan™ Control Genomic Human DNA (10 ng/µL) (Applied Biosystems, Thermofisher, USA) was utilized to create a calibration curve to assess quantities of Alu 115. The final reaction volume (20 μL) was configured as follows: 10 µL of master mix, 0.6 µL of each Alu 115 forward and reverse primers, 5 µL of the cDNA extract, and 3.8 µL of nuclease-free water.

The amplification of the PCR was performed in real-time using a 5 Plex Rotor-Gene Real-Time PCR Analyzer (Qiagen, Germany) according to the following protocol: initial activation (10 min at 95°C) then 35 cycles of denaturation (30 s at 95°C), annealing (30 s at 64°C), and extension (30 s at 72°C). After PCR amplifications, we performed a melting curve analysis on every PCR amplicon to confirm the specificity of the reaction products.

The concentration of Alu 115 fragments was determined using the created standard curve by ten-fold serial dilution of Taqman human genomic DNA (10 ng/uL). The standard concentrations were expressed in pg/uL. by multiplying by 1000 to (10,000, 1000, 100, 10, and 1) pg/uL.

Statistical methodology

Data analysis was performed using IBM SPSS Statistics (V. 26.0, IBM Corp., 2019). Categorical variables were expressed as numbers and percentages. The non-parametric variables were presented as medians and interquartile ranges (IQR). For Data comparison, the Chi-square test, the Mann–Whitney U test, and the Kruskal-Wallis test were applied. The receiver operating characteristic (ROC) curve analysis was used to assess the diagnostic performance of cfDNA. Logistic regression analysis was done to identify independent risk factors for erosions. A P-value of <0.05 is statistically significant.

Results

Levels of cfDNA in RA patients and healthy controls

A significant increase in the cfDNA concentration was found in RA patients compared to the age- and sex-matched healthy control (P < 0.001). The median (IQR) cfDNA level in RA patients was 37.80 (21.63–87.29) pg/µL, while in controls was 16.96 (9.98–25.94) pg/µL. Figure 1.

Figure 1.

Box plots comparing levels of cfDNA among RA patients and controls.

Among the RA patients, cfDNA concentrations did not differ significantly according to participants’ sex (P = 0.071). The median (IQR) cfDNA level in males was 77.73 (21.63–246.28) pg/µL, while in females was 36.96 (24.11–66.50) pg/µL.

Association of cfDNA with disease activity by DAS28

Plasma levels of cfDNA significantly increased with higher disease activity scores (P < 0.001) and differed significantly between all activity groups (inactive, low, moderate, and high). Therefore, cfDNA was superior on ESR, which did not differ significantly between low and moderate activity groups (P = 0.096), RF, which did not differ significantly between inactive and low activity groups (P = 0.807), and ACPA, which did not differ significantly between inactive and low activity groups (P = 0.271) and between low and moderate activity groups (P = 0.163). Tables 1 and 2.

Table 1.

Comparison of the different characteristics of RA patients according to disease activity by DAS28 score.

Parameter	DAS28 score								Chi-square
	Inactive		Low		Moderate		High		Chi-square
	n	%	n	%	n	%	n	%	X ²	P-value
Sex n, (%)
Male	7	25.93	2	7.41	4	14.81	10	37.04	8.121	0.044*
Female	20	74.07	25	92.59	23	85.19	17	62.96	8.121	0.044*
Erosion n, (%)
No erosions	20	74.07	11	40.74	9	33.33	3	11.11	22.991	<0.001*
Erosion	7	25.93	16	59.26	18	66.67	24	88.89	22.991	<0.001*
Serology n, (%) (RF + ACPA)
Seronegative	5	18.52	0	0.00	6	22.22	0	0.00	12.450	0.006*
Seropositive	22	81.48	27	100.00	21	77.78	27	100.00	12.450	0.006*
	DAS28 score								Kruskal-Wallis test
	Range (min-max)				Median	IQR			X ²	P-value
Age (years)
Inactive	22–57				47.00	42.00–53.00			6.992	0.072
Low	24–74				52.00	33.50–65.00
Moderate	22–77				34.00	27.50–55.50
High	20–67				46.00	33.50–50.00
Disease duration (years)
Inactive	3–18				10.00	8.00–16.00			5.184	0.159
Low	3–30				12.00	8.50–24.50
Moderate	4–27				10.00	5.50–17.00
High	4–25				9.00	6.50–12.00
ESR (mm/h)
Inactive	10–25				15.00	12.00–19.00			63.523	<0.001*
Low	10–75				34.00	20.00–41.50
Moderate	12–80				44.00	30.00–62.00
High	34–125				75.00	45.00–110.00
CRP (mg/L)
Inactive	1–48				3.00	2.00–4.00			77.973	<0.001*
Low	6–24				12.00	6.00–24.00
Moderate	6–65				26.00	12.00–48.00
High	24–132				75.00	48.00–96.00
RF (IU/mL)
Inactive	8–256				48.00	45.00–54.00			26.766	<0.001*
Low	20–287				45.00	28.00–90.00
Moderate	8–117				30.00	10.00–38.00
High	1.666–564				256.00	94.00–340.50
ACPA (U/mL)
Inactive	5–100				22.00	8.00–45.00			44.430	<0.001*
Low	10–111				26.00	12.00–33.00
Moderate	8–130				25.00	22.00–74.00
High	22–500				118.00	105.50–177.00
cfDNA (pg/µL)
Inactive	1.52–39.5				16.96	9.63–25.31			84.183	<0.001*
Low	13.3–50.58				24.11	17.76–36.48
Moderate	35.5–221.79				46.55	37.80–66.00
High	82.22–320.22				150.44	120.50–248.14
Synovitis GSUS scores
Inactive	1–3				2.00	1.00–2.00			29.374	<0.001*
Low	1–16				6.00	1.50–10.00
Moderate	1–10				5.00	2.00–6.00
High	1–16				8.00	3.00–15.00
Synovitis PDUS scores
Inactive	0–3				1.00	0.00–1.00			29.600	<0.001*
Low	0–11				2.00	1.00–4.00
Moderate	0–6				2.00	1.50–3.00
High	0–11				2.00	2.00–4.50
Erosion scores
Inactive	0–4				0.00	0.00–0.50			17.931	<0.001*
Low	0–4				1.00	0.00–2.00
Moderate	0–4				1.00	0.00–2.00
High	0–4				2.00	1.00–2.50
Tenosynovitis GSUS scores
Inactive	0–4				0.00	0.00–1.00			7.788	0.061
Low	0–7				0.00	0.00–1.00
Moderate	0–4				1.00	0.00–1.00
High	0–7				1.00	0.00–2.00
Tenosynovitis PDUS scores
Inactive	0–3				0.00	0.00–0.00			4.497	0.213
Low	0–5				0.00	0.00–0.00
Moderate	0–2				0.00	0.00–1.00
High	0–5				0.00	0.00–1.00

ACPA: anticitrullinated protein antibodies; cfDNA: cell-free DNA; CRP: c-reactive protein; DAS28: disease activity score-28; ESR: erythrocyte sedimentation rate; GSUS: grayscale ultrasound; PDUS: power Doppler ultrasound; RF: rheumatoid factor.

Significant P-value < 0.05.

Table 2.

Significance of differences in characteristics observed between RA activity groups based on DAS28 score.

Parameter	Mann-Whitney test
	Inactive & low	Inactive & moderate	Inactive & high	Low & moderate	Low & high	Moderate & high
ESR (mm/h)	<0.001*	<0.001*	<0.001*	0.096	<0.001*	<0.001*
CRP (mg/L)	<0.001*	<0.001*	<0.001*	<0.001*	<0.001*	<0.001*
RF (IU/mL)	0.807	0.006*	0.001*	0.001*	0.004*	<0.001*
ACPA (U/mL)	0.271	0.028*	<0.001*	0.163	<0.001*	<0.001*
cfDNA (pg/µL)	0.006*	<0.001*	<0.001*	<0.001*	<0.001*	<0.001*
Synovitis GSUS scores	0.001*	<0.001*	<0.001*	0.512	0.120	0.011*
Synovitis PDUS scores	<0.001*	<0.001*	<0.001*	0.895	0.302	0.267
Erosion scores	0.030*	0.008*	<0.001*	0.588	0.027*	0.088

ACPA: anticitrullinated protein antibodies, cfDNA: cell-free DNA, CRP: c-reactive protein, DAS28: disease activity score-28, ESR: erythrocyte sedimentation rate, GSUS: Grayscale ultrasound, PDUS: power Doppler ultrasound, RF: rheumatoid factor.

Significant P-value < 0.05.

Regarding the US7 score, tenosynovitis GSUS and PDUS scores did not differ significantly between the different activity groups (P = 0.061 & 0.213, respectively). Synovitis GSUS and synovitis PDUS scores did not differ significantly between the low and moderate activity groups (P = 0.512 & 0.895, respectively) and between the low and high activity groups (P = 0.120 & 0.302, respectively). In addition, synovitis PDUS scores did not differ significantly between high and moderate activity groups (P = 0.267). Erosion scores did not differ significantly between the low and moderate activity groups (P = 0.588) and between the high and moderate activity groups (P = 0.088). Tables 1 and 2.

Association of cfDNA with erosions

Higher cfDNA levels were significantly associated with erosions (P < 0.001), as were increased RF (P = 0.043), ACPA (P < 0.001), CRP (P < 0.001), and ESR (P < 0.001) levels. Furthermore, erosions were significantly linked to higher US7 (P-values <0.05) and DAS28 (P < 0.001) scores. On the other hand, erosions were not associated with patients’ age (P = 0.604), sex (P = 0.578), disease duration (P = 0.104), and serology status (RF + ACPA) (P = 0.089). Table 3.

Table 3.

Comparison of the different characteristics of RA patients according to the presence of erosions.

Parameter	Erosion				Chi-square
	No erosions		Erosion		Chi-square
	n	%	n	%	X ²	P-value
Sex n, (%)
Male	8	18.60	15	23.08	0.309	0.578
Female	35	81.40	50	76.92	0.309	0.578
Serology (RF + ACPA) n, (%)
Seronegative	7	16.28	4	6.15	2.900	0.089
Seropositive	36	83.72	61	93.85	2.900	0.089
	Erosion				Mann-Whitney test
	Range		Median	IQR	Z	P-value
Age (years)
No erosions	22–77		44.00	32.00–55.00	0.518	0.604
Erosion	20–74		48.00	35.00–55.00	0.518	0.604
Disease duration (years)
No erosions	3–26		9.00	8.00–16.00	1.627	0.104
Erosion	4–30		11.00	7.00–19.00	1.627	0.104
ESR (mm/h)
No erosions	10–110		20.00	13.50–35.00	4.296	<0.001*
Erosion	10–125		45.00	30.00–75.00	4.296	<0.001*
CRP (mg/L)
No erosions	1–75		6.00	3.00–24.00	5.101	<0.001*
Erosion	1–132		45.00	12.00–56.00	5.101	<0.001*
RF (IU/mL)
No erosions	8–256		45.00	28.00–51.00	2.023	0.043*
Erosion	1.666–564		81.00	23.00–256.00	2.023	0.043*
ACPA (U/mL)
No erosions	5–130		22.00	11.00–36.50	3.851	<0.001*
Erosion	6–500		74.00	23.00–112.00	3.851	<0.001*
DAS28 score
No erosions	1–8		2.60	1.00–4.20	4.412	<0.001*
Erosion	1–8.3		4.30	2.80–5.90	4.412	<0.001*
cfDNA (pg/µL)
No erosions	1.52–150.44		24.11	16.00–37.38	5.179	<0.001*
Erosion	8.39–320.22		55.50	36.79–140.21	5.179	<0.001*
Synovitis GSUS scores
No erosions	1–15		2.00	1.00–6.00	2.969	0.003*
Erosion	1–16		5.00	2.00–10.00	2.969	0.003*
Synovitis PDUS scores
No erosions	0–5		1.00	1.00–2.00	2.632	0.008*
Erosion	0–11		2.00	1.00–4.00	2.632	0.008*
Tenosynovitis GSUS scores
No erosions	0–2		0.00	0.00–1.00	2.884	0.004*
Erosion	0–7		1.00	0.00–1.00	2.884	0.004*
Tenosynovitis PDUS scores
No erosions	0–5		0.00	0.00–0.00	3.326	0.001*
Erosion	0–5		0.00	0.00–1.00	3.326	0.001*

ACPA: anticitrullinated protein antibodies; cfDNA: cell-free DNA; CRP: c-reactive protein; DAS28: disease activity score-28; ESR: erythrocyte sedimentation rate; GSUS: Grayscale ultrasound; PDUS: power Doppler ultrasound; RF: rheumatoid factor.

Significant P-value < 0.05.

Association of cfDNA with seropositivity

There was a significant difference in cfDNA levels according to the serology status (RF + ACPA) of RA patients (P = 0.007), being significantly higher in seropositive patients (median (IQR): 40.23 (24.11–90.50) pg/µL) than seronegative patients (median (IQR): 36.79 (8.39–37.30) pg/µL). Table 4.

Table 4.

Comparison of cfDNA levels according to the serology (RF+ACPA) status of RA patients.

Parameter	cfDNA (pg/µL)			Mann-Whitney test
	Range	Median	IQR	Z	P-value
Serology (RF + ACPA)
Seronegative	1.52–37.8	36.79	8.39–37.30	−2.677	0.007*
Seropositive	4.11–320.22	40.23	24.11–90.50	−2.677	0.007*

ACPA: anticitrullinated protein antibodies; cfDNA: cell-free DNA; RF: rheumatoid factor.

Significant P-value < 0.05.

Correlations of cfDNA with routine laboratory markers and radiological findings

Significant positive correlations were observed between plasma levels of cfDNA and ESR level (r = 0.508, P < 0.001), CRP level (r = 0.672, P < 0.001), RF titre (r = 0.515, P < 0.001), ACPA level (r = 0.753, P < 0.001), DAS28 scores (r = 0.750, P < 0.001), synovitis GSUS scores (r = 0.515, P < 0.001), synovitis PDUS scores (r = 0.296, P = 0.002), and erosion scores (r = 0.382, P < 0.001). No significant correlations were observed between plasma levels of cfDNA and tenosynovitis GSUS scores (r = 0.115, P = 0.238), tenosynovitis PDUS scores (r = 0.143, P = 0.141), patients’ age (r = −0.082, P = 0.401), and disease duration (r = −0.091, P = 0.348).

Diagnostic abilities of cfDNA

ROC curve analysis was performed to determine the diagnostic ability of cfDNA to discriminate between RA patients and controls; the best cut-off point was 36.79 pg/µL, and the area under the curve (AUC) was 0.801. Figure 2.

Figure 2.

ROC curve for discriminating ability of cfDNA between RA patients and healthy controls (cut-off: 36.79 pg/µL, area under curve: 0.801, diagnostic sensitivity: 52.78%, diagnostic specificity: 100.00%, positive predictive value: 100.00%, and negative predictive value: 34.6%).

The diagnostic ability of cfDNA to discriminate RA patients according to disease activity by DAS28 and the presence of erosions was also assessed. The best cut-off point of plasma cfDNA in discriminating between inactive and active (low, moderate, and high activity) RA patients was 34.73 pg/µL with an AUC of 0.898 (Figure 3). The best cut-off point of cfDNA in discriminating between RA patients with erosions and those without was 37.8 pg/µL with an AUC of 0.795. Figure 4.

Figure 3.

ROC curve for discriminating ability of cfDNA between RA patients with inactive disease and those with active disease (low, moderate, and high activity) (cut-off: 34.73 pg/µL, area under curve: 0.898, diagnostic sensitivity: 77.78%, diagnostic specificity: 88.89%, positive predictive value: 95.5%, and negative predictive value: 57.1%).

Figure 4.

ROC curve for discriminating ability of cfDNA between RA patients with erosions and those without (cut-off: 37.8 pg/µL, area under curve: 0.795, diagnostic sensitivity: 67.69%, diagnostic specificity: 81.40 %, positive predictive value: 84.6%, and negative predictive value: 62.5%).

Finally, the best cut-off point of cfDNA in discriminating between seronegative and seropositive RA patients was 37.8 pg/µL with an AUC of 0.747. Figure 5.

Figure 5.

ROC curve for discriminating ability of cfDNA between seronegative and seropositive RA patients (cut-off: 37.8 pg/µL, area under curve: 0.747, diagnostic sensitivity: 53.61%, diagnostic specificity: 100%, positive predictive value: 100%, and negative predictive value: 19.6%).

Risk factors for erosions

The logistic multivariate regression analysis was applied and revealed that RF and cfDNA were significant independent risk factors for erosions among RA patients. The odds ratio (OR) and 95% confidence interval for RF were 2.012 (1.003–3.021) (P = 0.012), while for cfDNA were 2.254 (1.258–4.042) (P = 0.006). Table 5.

Table 5.

Multivariate logistic regression analysis for independent risk factors for erosions among RA patients.

Parameter	Odd’s ratio	95% C.I. for odd’s ratio	P-value
ESR	0.984	0.947–1.024	0.428
CRP	1.039	0.994–1.086	0.087
RF	2.012	1.003–3.021	0.012*
ACPA	0.994	0.977–1.011	0.470
DAS28 Score	0.815	0.459–1.448	0.486
cfDNA	2.254	1.258–4.042	0.006*
Synovitis GSUS scores	0.989	0.807–1.211	0.912
Synovitis PDUS scores	1.102	0.733–1.655	0.640
Tenosynovitis GSUS scores	1.023	0.989–1.058	0.184
Tenosynovitis PDUS scores	1.493	0.864–2.580	0.151

Significant P-value < 0.05.

Discussion

Due to its association with inflammation and tissue damage, cfDNA has gained attention as a potential biomarker of RA activity and prognosis. The current study revealed that plasma levels of cfDNA were significantly higher in RA patients than controls. Levels of cfDNA significantly increased with higher DAS28 scores and differed significantly between all RA activity groups (inactive, low, moderate, and high). In addition, cfDNA levels were significantly positively correlated with RA prognosis markers (RF, ACPA), and activity indicators (CRP, ESR, DAS28). Furthermore, cfDNA in the current study was revealed to be an independent risk factor for erosions.

To our knowledge, this study takes the lead in investigating the correlation of plasma cfDNA levels with the US7 scoring system items and the serology status (RF + ACPA) in RA patients. It revealed that cfDNA was positively correlated with synovitis GSUS and PDUS scores and erosion scores but did not correlate significantly with tenosynovitis GSUS and PDUS scores, indicating a relationship between cfDNA release and synovial inflammation and joint damage. Also, plasma cfDNA levels were significantly higher in seropositive RA patients than in seronegative ones.

Our findings go along with several studies that have investigated cfDNA levels in RA activity.^6,14–16 A study by Abdelal et al.,¹⁶ reported a significant correlation between plasma cfDNA levels and DAS28 scores in RA patients, with higher cfDNA levels observed in patients with active disease compared to those in remission.¹⁶ Similarly, Lauková et al.,¹⁷ reported a positive association between cfDNA and ESR, CRP, and RA activity by DAS28. They also reported higher levels of cfDNA in the active group than in the remission group.¹⁷ Also, in a study by Eldosoky et al.,¹⁸ they had three groups of participants (active RA patients, RA patients in remission, and healthy controls) and found that cfDNA levels reached the highest in active RA patients and the lowest in controls with significant differences between the three groups included.¹⁸ They reported positive correlations of plasma cfDNA levels with DAS28 and ESR.¹⁸ Another study by Dong et al.,¹⁹ found that cfDNA levels were associated with increased synovial inflammation and joint damage, suggesting a potential pathological role for cfDNA in RA.¹⁹

Previously, cfDNA was reported to be released by active secretion from RA synovial cells in vitro.²⁰ This finding supports the association between GSUS and cfDNA found in the current study. Hashimoto et al.,²⁰ suggested that cfDNA in RA synovium may originate from synovial cells or inflammatory cells such as neutrophils.²⁰

Furthermore, the release of cfDNA in RA is a multifactorial process that is related to the pathophysiology of the disease and involves various mechanisms, including apoptosis and necrosis; synovial inflammation and tissue damage in RA lead to the release of cellular debris, including fragmented DNA, into circulation.^8,21 Moreover, neutrophil extracellular traps (NETs), which are web-like structures composed of chromatin and antimicrobial proteins released by activated neutrophils, contribute to the release of cfDNA and propagate inflammation in the synovium.^22,23 In addition, dysregulated clearance mechanisms may further exacerbate cfDNA accumulation in RA patients, perpetuating immune-mediated tissue damage and joint inflammation.^5,8 Also, oxidative stress, which is the imbalance between reactive oxygen species (ROS) production and antioxidant defense mechanisms, is a hallmark of RA. ROS-induced DNA damage may lead to the release of cfDNA into circulation.²⁴ Furthermore, dysregulated immune responses in RA, including aberrant cytokine production and immune cell activation, contribute to tissue damage and cfDNA release. Pro-inflammatory cytokines such as tumor necrosis factor-alpha (TNF-α) and interleukin-6 (IL-6) are implicated in the pathogenesis of RA and may promote cfDNA release.²⁵ Moreover, Alvarado-Vásquez et al.,²⁶ reported that cfDNA may cause endothelial dysfunction and other adverse consequences on the vascular system by activating toll-like receptor (TLR)-9 which induces NF-κB signaling pathway. A finding that supports the involvement of cfDNA in the pathogenesis of RA.²⁶

In contrast to our results, Hajizadeh et al.,²⁷ reported no significant association between cfDNA and CRP, erosivity, or extra-articular manifestations in RA patients.²⁷ Also, Dunaeva et al.,¹ measured cfDNA in serum samples and observed a decrease in cfDNA levels in patients with established RA in comparison with healthy controls.¹

In the current study, at a cut-off of 36.79 pg/µL, cfDNA discriminated RA patients from healthy controls with an AUC of 0.801. While at a cut-off of 34.73 pg/µL, cfDNA discriminated active RA patients from those with inactive disease with an AUC of 0.898. Dunaeva et al.,¹ reported, in the ROC curve analysis, that the AUC for cfDNA to discriminate RA patients from healthy controls was 0.914.¹ Eldosoky et al.,¹⁸ also reported that cfDNA at a cut-off of 14.41 ng/µL (=14,400 pg/µL) had an AUC of 0.931, 86% diagnostic sensitivity, and 84% diagnostic specificity for the discrimination of active RA patients from those in remission.¹⁸ The variation in cfDNA cut-off values across studies contributes to discrepancies, indicating that the elevation of cfDNA in different populations is not standardized. Although cfDNA levels increase, there is no well-established cut-off that can be universally applied for the diagnosis and prognosis of RA. Differences in DNase activity among different sample types and different disease statuses can contribute to the discrepancy in cfDNA levels.^1,28

Differences observed between studies could also be attributed to the variability in disease activity and severity among RA patients, treatment regimens (e.g. disease-modifying antirheumatic drugs, biologics), patient demographics (e.g. age, sex, ethnicity) and characteristics (e.g. comorbidities, disease duration, genetic factors), preanalytical variables including sample collection and processing methods that can introduce bias and affect cfDNA measurements, the sensitivity and specificity of analytical assays used to quantify cfDNA levels (e.g. quantitative PCR, next-generation sequencing), the type of cfDNA studied (nuclear or mitochondrial), and the matrix in which cfDNA was detected (serum, plasma, synovial fluid).^1,29

The major limitations of our study were its single-center design with a relatively small sample size and that it did not consider the different treatment regimens and duration of treatment of the included participants. Moreover, since nuclear origin has been suggested as the primary source of circulating cfDNA,⁶ we analyzed cfDNA exclusively using ALU-PCR. It is recommended to assess total cfDNA concentrations, including cfDNA from minor sources like mitochondrial cfDNA to give a more comprehensive view of total cfDNA in RA patients as well as to evaluate DNase activity. Despite these limitations, our findings contribute to the growing body of evidence supporting the utility of cfDNA as a promising biomarker for assessing RA activity and prognosis, as evidenced by its association with DAS28, US7 scores, CRP, ESR, RF, and ACPA.

Conclusion

The study demonstrates that elevated plasma cfDNA levels correlate with RA activity, prognosis markers, and the presence of erosions, reflecting ongoing synovial inflammation, tissue damage, and immune dysregulation. These findings highlight plasma cfDNA as a potential reliable biomarker for disease monitoring, stratifying disease severity and guiding treatment approaches. Its role as an independent risk factor for erosions suggests a prognostic value, making cfDNA a critical tool for early intervention. Further research could solidify its application in clinical protocols and explore therapeutic implications.

Footnotes

Acknowledgements

None

Authors contribution

• SIT was responsible for the study’s conceptualization, methodology, investigation, data curation, formal analysis, project administration, writing of the original draft, and review and editing of the manuscript.

• SFS was responsible for the study’s methodology, investigation, data curation, formal analysis, project administration, writing of the original draft, and review and editing of the manuscript.

• SSH was responsible for the study’s investigation, data curation, formal analysis, project administration, and review and editing of the manuscript.

• EME was responsible for the study’s investigation, data curation, formal analysis, project administration, and review and editing of the manuscript.

• SS was responsible for the study’s investigation, data curation, formal analysis, project administration, and review and editing of the manuscript.

• DHHE was responsible for the study’s investigation, data curation, formal analysis, project administration, and review and editing of the manuscript.

• BO was responsible for the study’s investigation, data curation, formal analysis, project administration, and review and editing of the manuscript.

• MAA was responsible for the study’s investigation, data curation, formal analysis, project administration, and review and editing of the manuscript.

• RAG was responsible for the study’s conceptualization, methodology, investigation, data curation, formal analysis, project administration, writing of the original draft, and review and editing of the manuscript.

Availability of data and materials

The datasets used and/or analyzed during the current study available from the corresponding author on reasonable request.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Ethical approval

Consent for publication

Not applicable

Informed consent

Written informed consent was obtained from all subjects before the study.

Trial registration

Not applicable.

ORCID iDs

Sara I Taha

Dalia Hussein Helmy Elsheikh

References

Dunaeva

Buddingh’

Toes

, et al (2015) Decreased serum cell-free DNA levels in rheumatoid arthritis. Autoimmunity Highlights 6(1–2): 23–30.

Backhaus

Ohrndorf

Kellner

, et al (2013) The US7 score is sensitive to change in a large cohort of patients with rheumatoid arthritis over 12 months of therapy. Annals of the Rheumatic Diseases 72(7): 1163–1169.

Allard-Chamard

Boire

(2019) Serologic diagnosis of rheumatoid arthritis. Clinics in Laboratory Medicine 39(4): 525–537.

Ranucci

(2019) Cell-free DNA: Applications in different diseases. Methods in Molecular Biology 1909: 3–12.

Mondelo-Macía

Castro-Santos

Castillo-García

, et al (2021) Circulating free DNA and its emerging role in autoimmune diseases. Journal of Personalized Medicine 11(2): 151.

Hashimoto

Yoshida

Hashiramoto

, et al (2021) Cell-free DNA in rheumatoid arthritis. International Journal of Molecular Sciences 22(16): 8941.

Zhu

Kong

Che

, et al (2023) Bioinspired nanogels as cell-free DNA trapping and scavenging organelles for rheumatoid arthritis treatment. Proceedings of the National Academy of Sciences of the USA 120(33): e2303385120.

Duvvuri

Lood

(2019) Cell-free DNA as a biomarker in autoimmune rheumatic diseases. Frontiers in Immunology 10: 502.

Aletaha

Neogi

Silman

, et al (2010) 2010 rheumatoid arthritis classification criteria: An American College of Rheumatology/European League against rheumatism collaborative initiative. Arthritis & Rheumatology 62(9): 2569–2581.

10.

Taha

Samaan

Ibrahim

, et al (2022) Can complete blood count picture tell us more about the activity of rheumatological diseases? Clinical Medicine Insights: Arthritis and Musculoskeletal Disorders 15: 11795441221089182.

11.

Backhaus

Ohrndorf

Kellner

, et al (2009) Evaluation of a novel 7-joint ultrasound score in daily rheumatologic practice: A pilot project. Arthritis & Rheumatology 61(9): 1194–1201.

12.

Taha

Samaan

Ibrahim

, et al (2021) Post-COVID-19 arthritis: Is it hyperinflammation or autoimmunity? European Cytokine Network 32(4): 83–88.

13.

Gezer

Bronkhorst

Holdenrieder

(2022) The utility of repetitive cell-free DNA in cancer liquid biopsies. Diagnostics (Basel) 12(6): 1363.

14.

Leon

Ehrlich

Shapiro

, et al (1977) Free DNA in the serum of rheumatoid arthritis patients. Journal of Rheumatology 4(2): 139–143.

15.

Leon

Revach

Ehrlich

, et al (1981) DNA in synovial fluid and the circulation of patients with arthritis. Arthritis & Rheumatology 24(9): 1142–1150.

16.

Abdelal

Zakaria

Sharaf

, et al (2016) Levels of plasma cell-free DNA and its correlation with disease activity in rheumatoid arthritis and systemic lupus erythematosus patients. The Egyptian Rheumatologist 38(4): 295–300.

17.

Lauková

Konečná

Vlková

, et al (2018) Anti-cytokine therapy and plasma DNA in patients with rheumatoid arthritis. Rheumatology International 38(8): 1449–1454.

18.

Eldosoky

Hammad

Kotb

, et al (2018) Cell free DNA and CD38 expression on T helper cells as biomarkers of disease activity in rheumatoid arthritis. American Journal of Biochemistry 8: 60–68.

19.

Dong

Liu

Sun

, et al (2020) Identification of specific joint-inflammatogenic cell-free DNA molecules from synovial fluids of patients with rheumatoid arthritis. Frontiers in Immunology 11: 662.

20.

Hashimoto

Yoshida

Yokoyama

, et al (2023). Tocilizumab suppresses NF-kappa B activation via toll-like receptor 9 signaling by reducing cell-free DNA in rheumatoid arthritis. Clinical and Experimental Immunology 213(2): 209–220.

21.

Smolen

Aletaha

Barton

, et al (2018) Rheumatoid arthritis. Nature Reviews Disease Primers 4: 18001.

22.

Song

Pan

, et al (2021) Neutrophil extracellular traps tied to rheumatoid arthritis: Points to ponder. Frontiers in Immunology 11: 578129.

23.

Khandpur

Carmona-Rivera

Vivekanandan-Giri

, et al (2013) NETs are a source of citrullinated autoantigens and stimulate inflammatory responses in rheumatoid arthritis. Science Translational Medicine 5(178): 178ra40.

24.

da Fonseca

LJS

Nunes-Souza

Goulart

MOF

, et al (2019) Oxidative stress in rheumatoid arthritis: What the future might hold regarding novel biomarkers and add-on therapies. Oxidative Medicine and Cellular Longevity 2019: 7536805.

25.

Lubberts

van den Berg

(2003) Cytokines in the pathogenesis of rheumatoid arthritis and collagen-induced arthritis. Advances in Experimental Medicine and Biology 520: 194–202.

26.

Alvarado-Vásquez

(2015) Circulating cell-free mitochondrial DNA as the probable inducer of early endothelial dysfunction in the prediabetic patient. Experimental Gerontology 69: 70–78.

27.

Hajizadeh

DeGroot

TeKoppele

, et al (2003) Extracellular mitochondrial DNA and oxidatively damaged DNA in synovial fluid of patients with rheumatoid arthritis. Arthritis Research and Therapy 5(5): R234-40.

28.

Shakhmaeva

Saifullina

Sattarova

, et al (2011) Electrochemical sensor for blood deoxyribonucleases: Design and application to the diagnosis of autoimmune thyroiditis. Analytical and Bioanalytical Chemistry 401(8): 2591–2597.

29.

Ghorab

Abd-Elmotalib Mansy

Saleh

, et al (2024) DNA integrity index as a molecular marker for prediction of tumor burden in breast cancer. Ain Shams Medical Journal 75(1): 103–116.

Circulating cell-free DNA as a potential biomarker for prediction of disease activity and prognosis among Egyptian rheumatoid arthritis patients

Abstract

Keywords

Introduction

Methodology

Sample size calculation

Ethical considerations

Study subjects

Musculoskeletal ultrasound examination (MSUS)

Laboratory investigations

PCR workflow

Extraction of cfDNA

Amplification of cfDNA by real-time PCR using ALU repeats

Statistical methodology

Results

Levels of cfDNA in RA patients and healthy controls

Association of cfDNA with disease activity by DAS28

Association of cfDNA with erosions

Association of cfDNA with seropositivity

Correlations of cfDNA with routine laboratory markers and radiological findings

Diagnostic abilities of cfDNA

Risk factors for erosions

Discussion

Conclusion

Footnotes

Acknowledgements

Authors contribution

Availability of data and materials

Declaration of conflicting interests

Funding

Ethical approval

Consent for publication

Informed consent

Trial registration

ORCID iDs

References