ABSTRACT
Objectives
Evidence indicates that hematological indices, such as neutrophil‐to‐lymphocyte ratio (NLR), systemic immune‐inflammation index (SII), and neutrophil‐to‐hemoglobin and lymphocyte (NHL), are valuable for assessing disease activity in rheumatoid arthritis (RA). This study investigated their potential as biomarkers for monitoring treatment response and disease improvement in RA patients receiving biological and targeted synthetic disease‐modifying antirheumatic drugs (b/tsDMARDs).
Methods
A retrospective analysis was conducted on 308 active RA patients receiving IL‐6 receptor inhibitors, TNF‐alpha inhibitors, cytotoxic T lymphocyte‐associated antigen‐4‐Ig (abatacept), or Janus kinase inhibitors. Hematological indices were collected at baseline and 6 months post‐treatment.
Results
Significant reductions in NLR, SII, and NHL were observed after initiation of b/tsDMARD therapy. These changes correlated with improvements in DAS28‐ESR and discriminated good/moderate responders from non‐responders. Each index predicted a > 1.2‐point reduction in DAS28‐ESR (area under the curve values ranging from 0.709 for NHL to 0.732 for SII). Adding any of these hematological indices to ESR increased sensitivity to 62.0%–63.4% versus 53.4% for ESR alone. Among patients with active disease but normal ESR/CRP, these indices (sensitivity 54.8%–66.1%) better reflected disease burden and treatment response than traditional markers (sensitivity 29.0%), which changed minimally despite DAS28‐ESR improvement.
Conclusion
NLR, SII, and NHL are potential biomarkers for assessing treatment response in RA patients receiving b/tsDMARDs. Used alongside ESR, these indices improve evaluation of disease improvement. Notably, in clinically active patients with normal ESR/CRP levels, these indices captured treatment‐related change that traditional acute‐phase reactants did not, supporting their use as alternative or complementary response indicators in this setting.
Keywords: CTLA4‐Ig, hematological indices, IL‐6 receptor inhibitor, JAK inhibitor, NHL, NLR, PLR, rheumatoid arthritis, SII, TNF‐alpha inhibitor
Key Points
Hematological indices show greater reduction in RA patients with good or moderate treatment responses, reflecting treatment efficacy.
Combining hematological indices with ESR provides higher accuracy in identifying treatment responses in RA.
In active RA with normal ESR/CRP, hematological indices predict treatment response more accurately than ESR/CRP.
1. Introduction
Rheumatoid Arthritis (RA) is an autoimmune disease causing chronic inflammatory arthritis. Biomarkers such as the erythrocyte sedimentation rate (ESR), C‐reactive protein (CRP), and the 28‐joint disease activity score (DAS28) are commonly utilized to assess disease severity and therapeutic response, with DAS28 serving as the primary endpoint in trials. Although ESR and CRP, both of which are components of the DAS28 scoring system, are widely used clinical markers of disease activity, they have certain limitations. Research has indicated that the sensitivity of ESR and CRP in detecting synovial inflammation is only 71.3% and 48.4%, respectively [1]. A study analyzing data from the Consortium of Rheumatology Researchers of North America (CORRONA) database revealed that, among RA patients with ongoing activity, approximately 58% exhibited normal levels of both ESR and CRP, 26% demonstrated elevated levels of either ESR or CRP, and only 16% showed elevated levels of both markers [2]. Therefore, identifying alternative markers for disease activity assessment is critical for advancing RA management.
Hematological indices derived from complete blood counts (CBCs), such as neutrophil‐to‐lymphocyte ratio (NLR), platelet‐to‐lymphocyte ratio (PLR), systemic immune‐inflammation index (SII), and neutrophil‐to‐hemoglobin and lymphocyte (NHL) score, have shown potential in assessing disease activity and treatment response in RA. These indices are significantly elevated in RA compared to healthy controls [3, 4, 5, 6, 7, 8, 9], and are positively correlated with disease activity [3, 5, 7, 8, 9, 10]. Furthermore, NLR and PLR have shown a positive correlation with ultrasound‐detected synovitis [10]. In RA patients with normal ESR and CRP levels, PLR, but not NLR, positively correlates with ultrasound‐detected synovitis and bone erosions, suggesting that PLR may serve as a valuable hematological marker for assessing ultrasound‐detected disease activity in RA patients with normal ESR and CRP [11].
Disease‐modifying antirheumatic drugs (DMARDs) are a class of medications used to treat RA, including conventional synthetic DMARDs (csDMARDs), targeted synthetic DMARDs (tsDMARDs), and biological DMARDs (bDMARDs). Among these, targeted synthetic agents inhibit specific intracellular signaling molecules, with Janus kinase inhibitors (JAKi) being the currently approved tsDMARDs for RA. Biologic agents are categorized by their mechanisms of action and targets, including TNF‐alpha (tumor necrosis factor‐alpha) inhibitors (TNFi), IL‐6 (Interleukin‐6) receptor inhibitors (IL‐6Ri), CD20‐depleting agents (anti‐CD20), and cytotoxic T lymphocyte‐associated antigen‐4‐Ig (CTLA4‐Ig; abatacept), a CD80/CD86 co‐stimulation inhibitor. Selection of biological and targeted synthetic DMARDs (b/tsDMARDs) is typically guided by recommendations from European League Against Rheumatism (EULAR) and the American College of Rheumatology (ACR) and individualized according to patient‐specific factors such as route of administration (oral vs. injectable), dosing frequency, tolerability, and pregnancy plans.
Hematological indices, including NLR, PLR, SII, and NHL, have been observed to decrease following treatment with b/tsDMARDs, such as IL‐6Ri, TNFi, and JAKi. These changes correlate with improvements in disease activity scores, suggesting their potential utility in reflecting disease improvement [8, 12, 13, 14]. However, the clinical utility of these indices in monitoring treatment response, especially in patients with normal ESR/CRP, remains incompletely understood. We therefore aimed to determine whether these indices complement DAS28, ESR, and CRP in evaluating treatment response to b/tsDMARDs, and whether, in clinically active RA with normal ESR/CRP, these indices can serve as alternative markers of improvement.
2. Patients and Methods
2.1. Study Design, Patients, and Treatments
The study was designed as a single‐center, retrospective observational cohort study. RA patients aged 18 years or older, who were receiving their first treatment with TNFi (etanercept, adalimumab, golimumab, certolizumab), IL‐6Ri (tocilizumab), CTLA4‐Ig (abatacept), or JAKi (tofacitinib, baricitinib, upadacitinib), were enrolled and followed for 6 months at Taipei Medical University‐Shuang Ho Hospital, Ministry of Health and Welfare, between January 1, 2020, and June 30, 2024. All enrolled patients met the ACR classification criteria for RA. Exclusion criteria included pregnancy, malignancy, active or chronic infections, hematologic disorders, organ transplantation, or recent blood transfusions. Patients who had missing any follow‐up data related to hematological indices or disease activity parameters were also excluded. Patients were allowed to receive corticosteroids during the study. In our cohort, the majority (67%) of patients were on low‐dose glucocorticoid (prednisolone ≤ 5 mg/day) at baseline. The study protocol was reviewed and approved by the Ethics Committee of the Taipei Medical University Joint Institutional Review Board (TMU‐JIRB N202502055).
2.2. Collection of Clinical Information
Demographic information, the 28‐joint disease activity score using ESR (DAS28‐ESR), and laboratory data, were retrieved from electronic medical records. Disease activity was assessed using the DAS28‐ESR, with disease activity categories defined as follows: low disease activity (LDA, DAS28‐ESR ≤ 3.2), moderate disease activity (MDA, 3.2 < DAS28‐ESR ≤ 5.1), and high disease activity (HDA, DAS28‐ESR > 5.1). Laboratory parameters included ESR, CRP, and CBCs with differentials. The CBC data collected encompassed white blood cell (WBC) count, platelet count, hemoglobin levels, and the percentages of neutrophils and lymphocytes, both prior to and 6 months post‐treatment. Absolute counts for neutrophils and lymphocytes were calculated by multiplying the WBC count by the respective percentage of each cell type. Hematological indices, including NLR, PLR, SII, and NHL score, were calculated as follows: NLR = neutrophil/lymphocyte ratio, PLR = platelet/lymphocyte ratio, SII = (neutrophil × platelet)/lymphocyte ratio, and NHL score = neutrophil/(lymphocyte × hemoglobin) ratio.
2.3. Categorization of Treatment Response
Treatment response to b/tsDMARDs was categorized based on the EULAR response criteria [15, 16] (Table S1). Patients were classified according to their individual change in DAS28‐ESR over 6 months (ΔDAS28‐ESR) and the level of DAS28‐ESR achieved at 6 months (LDA, MDA, or HDA as defined above). A “good response” was defined as a significant change (ΔDAS28‐ESR > 1.2 from baseline), with achievement of LDA (DAS28‐ESR ≤ 3.2). Patients with a minimal DAS28‐ESR change (ΔDAS28‐ESR ≤ 0.6 from baseline), or those DAS28‐ESR > 5.1 (HDA) with 0.6 < ΔDAS28‐ESR ≤ 1.2 from baseline, were classified as having “no response.” Any other condition not covered by these two criteria was classified as a “moderate response.”
2.4. Endpoints
2.4.1. Primary Endpoints
To investigate the correlation between changes in hematological indices (NLR, PLR, SII, and NHL score) and changes in disease activity indices (DAS28‐ESR, ESR, and CRP) at 6 months post‐treatment with b/tsDMARDs.
To evaluate the predictive value of hematological indices for treatment response and improvement in disease activity.
2.4.2. Secondary Endpoints
To observe the changes in hematological indices and disease activity indices after 6 months of b/tsDMARDs treatment.
2.5. Statistical Analysis
All statistical analyses were performed using SPSS version 30.0 (IBM SPSS Inc., Chicago, IL, USA), and graphs were generated with GraphPad Prism software version 10.3.1 (GraphPad Software, San Diego, CA, USA). The Kolmogorov–Smirnov test was applied to assess the normality of data distribution. Continuous variables were compared using the Mann–Whitney U test between two groups and the Kruskal‐Wallis H test for comparisons among four treatment groups. To identify specific group differences, a post hoc Dunn‐Bonferroni test was conducted. Changes in hematological indices and disease activity parameters from baseline to 6 months of treatment were analyzed using the Wilcoxon signed‐rank test. Effect sizes were estimated using Cohen's d. Categorical variables were compared using Pearson's chi‐squared test. Correlations between changes in hematological indices and changes in disease activity parameters during treatments were assessed using Spearman's rank correlation coefficient. Receiver operating characteristic (ROC) curve analysis, along with estimation of the area under the curve (AUC) and 95% confidence intervals (CIs), was used to determine the predictive value of changes in hematological indices for a reduction in DAS28 > 1.2. Comparisons of differences between the partial AUC (pAUC) under ROC curves were performed using the bootstrap method with 2000 bootstrap iterations in R software version 4.4.2 (R Foundation for Statistical Computing, Vienna, Austria). Multiple linear regression was used to estimate beta coefficients, thereby quantifying the relative contribution of each predictor to DAS28‐ESR. Variable selection for the final model was conducted using a backward elimination approach. Quantitative variables are presented as medians with interquartile ranges (IQR), while categorical variables are expressed as percentages. A p‐value of less than 0.05 was considered statistically significant, and “ns” denotes no statistical significance.
3. Results
3.1. Basic Characteristics of the Study Subjects
A total of 308 RA patients were enrolled in the study. Among them, 53 patients received treatment with IL‐6Ri, 118 with TNFi, 57 with CTLA4‐Ig, and 80 with JAKi. All patients were in the active stage of disease, with a DAS28‐ESR greater than 3.2 (96% had HDA, and 4% had MDA). The Kolmogorov–Smirnov test showed non‐normal distributions for disease‐activity measures and hematological indices. Table 1 summarizes demographic and laboratory data as well as the distribution of patients across the four treatment groups. No significant differences were observed in gender distribution, laboratory measurements, or hematological indices among groups. However, post hoc Dunn‐Bonferroni tests revealed significant differences in age between the CTLA4‐Ig‐ and TNFi‐treated groups (p = 0.001), as well as between the CTLA4‐Ig‐ and JAKi‐treated groups (p = 0.006). Additionally, the JAKi group had lower DAS28‐ESR than the CTLA4‐Ig group (p = 0.007), and CRP levels were significantly lower in the JAKi group compared to the IL‐6Ri group (p = 0.028).
TABLE 1.
The baseline characteristics and laboratory measurements of RA patients receiving b/tsDMARDs.
| IL‐6Ri (n = 53) | TNFi (n = 118) | CTLA4‐Ig (n = 57) | JAKi (n = 80) | p | Reference | |
|---|---|---|---|---|---|---|
| Characteristics and laboratory measurements | ||||||
| Type of drugs | Tocilizumab, n = 53 |
Adalimumab, n = 23 Certolizumab, n = 21 Etanercept, n = 30 Golimumab, n = 44 |
Abatacept, n = 57 |
Tofacitinib, n = 31 Baricitinib, n = 19 Upadacitinib, n = 30 |
— | — |
| Age (years), Median (Q1, Q3) | 55.0 (46.5, 65.0) | 52.0 (40.8, 63.0) | 64.0 (51.5, 70.0) | 53.0 (45.0, 62.0) | 0.001 | — |
| Gender | 0.097 | — | ||||
| Female (n, %) | 40 (75%) | 84 (71%) | 42 (74%) | 69 (86%) | ||
| Male (n, %) | 13 (25%) | 34 (29%) | 15 (26%) | 11 (14%) | ||
| DAS28‐ESR, Median (Q1, Q3) | 6.53 (6.16, 7.02) | 6.51 (5.99, 6.85) | 6.72 (6.33, 7.04) | 6.23 (5.82, 6.73) | 0.007 | — |
| ESR (mm/h), Median (Q1, Q3) | 29 (15, 48) | 25 (13, 44) | 28 (18, 56) | 25 (16, 46) | 0.348 | 0–20 |
| CRP (mg/dL), Median (Q1, Q3) | 1.18 (0.38, 3.09) | 0.69 (0.31, 2.02) | 0.90 (0.31, 2.46) | 0.52 (0.18, 1.70) | 0.034 | 0–0.5 |
| WBC count (103/μL), Median (Q1, Q3) | 7.29 (5.75, 9.55) | 7.30 (5.79, 8.72) | 6.00 (5.08, 8.31) | 6.85 (5.84, 8.21) | 0.089 | 4.8–10.8 |
| Neutrophil (%), Median (Q1, Q3) | 67.2 (58.4, 74.6) | 65.1 (58.5, 70.5) | 66.0 (57.5, 74.5) | 64.4 (58.0, 73.1) | 0.562 | 40–74 |
| Lymphocyte (%), Median (Q1, Q3) | 23.4 (16.1, 30.8) | 25.3 (19.4, 32.4) | 23.6 (16.0, 28.7) | 25.9 (18.5, 30.3) | 0.353 | 19–48 |
| Platelet (103/μL), Median (Q1, Q3) | 262 (226, 318) | 274 (236, 329) | 241 (204, 317) | 284 (238, 324) | 0.041 | 130–400 |
| Hemoglobin (g/dL), Median (Q1, Q3) | 12.3 (10.8, 13.5) | 12.3 (10.9, 13.4) | 11.7 (10.9, 12.6) | 12.4 (11.5, 13.2) | 0.237 | 14–18 |
| NLR, Median (Q1, Q3) | 2.85 (1.90, 4.41) | 2.63 (1.82, 3.62) | 2.84 (1.98, 4.57) | 2.49 (1.88, 3.85) | 0.497 | — |
| PLR, Median (Q1, Q3) | 1.69 (1.11, 2.40) | 1.55 (1.21, 1.94) | 1.67 (1.21, 2.70) | 1.67 (1.23, 2.21) | 0.431 | — |
| SII, Median (Q1, Q3) | 683 (479, 1354) | 741 (458, 1075) | 624 (437, 1272) | 717 (463, 1074) | 0.945 | — |
| NHL score, Median (Q1, Q3) | 0.23 (0.14, 0.40) | 0.22 (0.14, 0.31) | 0.23 (0.16, 0.43) | 0.21 (0.14, 0.31) | 0.442 | — |
Note: Except for the categorical variable of gender, which was analyzed using Pearson's chi‐squared test, all other p‐values were calculated using the Kruskal‐Wallis H test. Data are presented as median with interquartile ranges (Q1, Q3).
Abbreviations: b/tsDMARDs, biological and targeted synthetic disease‐modifying antirheumatic drugs; CRP, C‐reactive protein; CTLA4‐Ig, cytotoxic T lymphocyte‐associated antigen‐4‐Ig; DAS28, disease activity score for 28 joints; ESR, erythrocyte sedimentation rate; IL‐6Ri, IL‐6 receptor inhibitors; JAKi, Janus kinase inhibitors; NHL, neutrophil‐to‐hemoglobin and lymphocyte score; NLR, neutrophil‐to‐lymphocyte ratio; PLR, platelet‐to‐lymphocyte ratio; RA, rheumatoid arthritis; SII, systemic immune‐inflammation index; TNFi, TNF‐alpha inhibitors; WBC, white blood cell.
3.2. Changes in Hematological Indices Positively Correlate With Disease Activity Improvement
Six months after initiating b/tsDMARDs, ESR and CRP significantly decreased, accompanied by reductions in DAS28‐ESR (Figure S1). Hematological indices including NLR, PLR, SII, and NHL also declined. Detailed changes across different treatment groups are presented in Table S2. Spearman's rank correlation analyses (Table S3) revealed significant correlations between hematological indices and disease activity indicators, including DAS28‐ESR, ESR, and CRP. Specifically, changes in NLR (ΔNLR), PLR (ΔPLR), SII (ΔSII), and NHL (ΔNHL) positively correlated with changes in DAS28‐ESR (correlation coefficient ranging from 0.327 for ΔPLR to 0.435 for ΔSII, p < 0.01), suggesting that changes in these hematological indices reflect treatment response and clinical improvement in patients receiving b/tsDMARDs.
3.3. Hematological Indices Can Reflect the Treatment Response and Disease Improvement in RA Patients Receiving b/tsDMARDs
At baseline, NLR, PLR, SII, and NHL score showed no significant differences by EULAR response criteria. However, after 6 months of treatment, patients with a good or moderate response exhibited a greater reduction in hematological indices compared to those with no response (p < 0.01 to < 0.001) (Figure 1). Significant decreases in these indices after 6 months of treatment were observed in the good or moderate response group (p < 0.01 to < 0.001), indicating a potential association between these hematological indices and treatment efficacy. These findings suggest that while these hematological indices cannot serve as prognostic markers at baseline for predicting the response to b/tsDMARDs, they are reflective of the therapeutic response to these treatments.
FIGURE 1.
Hematological indices in each treatment response groups (Good response: n = 39, Moderate response: n = 242, and No response: n = 27). (A) NLR, (B) PLR, (C) SII, (D) NHL score. No significant differences in any of the hematological indices were observed among the treatment response groups at baseline. Data are presented as error bars representing the median, lower quartile (Q1), and upper quartile (Q3). Δ, change in indices; NHL, neutrophil‐to‐hemoglobin and lymphocyte score; NLR, neutrophil‐to‐lymphocyte ratio; PLR, platelet‐to‐lymphocyte ratio; SII, systemic immune‐inflammation index. *p < 0.05, **p < 0.01, ***p < 0.001, ns denotes no statistical significance.
3.4. Hematological Indices Show Comparable Predictive Ability to ESR in Assessing Disease Improvement
An ROC analysis was conducted to assess whether changes in these hematological indices predict a > 1.2 reduction in DAS28‐ESR. The area under the ROC curve (AUC) demonstrated moderate to good accuracy for the changes in these indices in distinguishing a DAS28 reduction > 1.2 in RA patients receiving b/tsDMARDs, with AUC (95% CI) values ranging from 0.707 (0.602–0.813) for ΔPLR to 0.732 (0.643–0.820) for ΔSII (Table 2). The cutoff values were 0.42 for ΔNLR, 0.20 for ΔPLR, 144.6 for ΔSII, and 0.05 for ΔNHL score. Combining hematological indices with ESR improved prediction, with AUCs (95% CI) up to 0.772 (0.680–0.864) for adding ΔNLR. To minimize the false positive rate (FPR), we constrained the FPR to around ≤ 0.2. The results demonstrated that the sensitivity of combining hematological indices with ESR was slightly elevated (62.0%–63.4% for combining hematological indices versus 53.4% for ESR alone). Furthermore, we computed the partial AUC (pAUC) with an FPR ≤ 0.2 and performed a bootstrap method to compare the pAUCs. The results indicated that combining ESR with NLR (p = 0.004) or NHL score (p = 0.018) significantly increased the pAUC compared to ESR alone. In conclusion, these results suggest that hematological indices demonstrate comparable predictive ability to ESR in assessing treatment response. Notably, combining ESR with hematological indices such as NLR and NHL score enhances predictive accuracy.
TABLE 2.
ROC analysis of hematological changes for DAS28 reduction > 1.2 in RA patients receiving b/tsDMARDs (n = 308).
| Change of indices | AUC (95% CI) | Sensitivity | Specificity | Cut‐off value | Positive likelihood ratio | Negative likelihood ratio | p | pAUC (FPR ≤ 0.2) | p a |
|---|---|---|---|---|---|---|---|---|---|
| Traditional inflammatory markers | |||||||||
| △ESR | 0.737 (0.634–0.839) | 53.4% | 79.3% | 8.5 | 2.58 | 0.59 | < 0.001 | 0.060 | — |
| △CRP | 0.651 (0.528–0.774) | 27.2% | 82.8% | 0.29 | 1.22 | 0.88 | 0.016 | 0.021 | 0.038 |
| Hematological indices | |||||||||
| △NLR | 0.715 (0.628–0.802) | 59.5% | 79.3% | 0.42 | 2.88 | 0.51 | < 0.001 | 0.068 | 0.738 |
| △PLR | 0.707 (0.602–0.813) | 55.2% | 79.3% | 0.20 | 2.67 | 0.56 | < 0.001 | 0.048 | 0.692 |
| △SII | 0.732 (0.643–0.820) | 60.9% | 79.3% | 144.6 | 2.95 | 0.49 | < 0.001 | 0.073 | 0.554 |
| △NHL score | 0.709 (0.620–0.799) | 57.3% | 79.3% | 0.05 | 2.77 | 0.54 | < 0.001 | 0.065 | 0.816 |
| Traditional inflammatory markers + Hematological indices | |||||||||
| △ESR + △NLR |
0.772 (0.680–0.864) |
62.0% | 79.3% | — | 3.00 | 0.48 | < 0.001 | 0.075 | 0.004 |
| △ESR + △PLR | 0.756 (0.657–0.856) | 62.7% | 79.3% | — | 3.03 | 0.47 | < 0.001 | 0.063 | 0.761 |
| △ESR + △SII | 0.767 (0.672–0.861) | 63.4% | 79.3% | — | 3.07 | 0.46 | < 0.001 | 0.070 | 0.118 |
| △ESR + △NHL | 0.769 (0.675–0.862) | 63.1% | 79.3% | — | 3.05 | 0.47 | < 0.001 | 0.072 | 0.018 |
Abbreviations: AUC, area under the curve; b/tsDMARDs, biological and targeted synthetic disease‐modifying antirheumatic drugs; CI, confidence interval; CRP, C‐reactive protein; DAS28, disease activity score for 28 joints; ESR, erythrocyte sedimentation rate; FPR, false positive rate; NHL, neutrophil‐to‐hemoglobin and lymphocyte score; NLR, neutrophil‐to‐lymphocyte ratio; pAUC, partial area under the curve; PLR, platelet‐to‐lymphocyte ratio; RA, rheumatoid arthritis; ROC, receiver operating characteristic; SII, systemic immune‐inflammation index.
The p‐values for pAUC were determined between ESR and other indices.
3.5. Hematological Indices Can Reflect Disease Severity, Whereas Traditional Inflammatory Markers May Not Be Reliable in Active RA Patients With Normal ESR and CRP Levels
Over half of active RA patients exhibit normal ESR and CRP levels [2]. In this subset, traditional inflammatory markers may not accurately reflect the severity of the disease. We assessed whether hematological indices could serve as alternative indicators for disease improvement and treatment response in patients with active RA (DAS28‐ESR > 3.2) but normal ESR (≤ 20 mm/h) and CRP (≤ 0.5 mg/dL) at baseline. Seventy‐two of 308 (23%) met these criteria, encompassing IL‐6Ri (n = 9, 13%), TNFi (n = 34, 47%), CTLA4‐Ig (n = 10, 14%), and JAKi (n = 19, 26%); the proportions mirrored the overall treatment distribution. In Table 3, all hematological indices except PLR significantly decreased after b/tsDMARDs, while ESR/CRP showed no significant changes, despite a reduction in DAS28‐ESR. Cohen's d indicated greater responsiveness of hematological indices than traditional markers. Although the effect size for PLR was numerically larger than that for ESR or CRP, its reduction was not significant (borderline downward trend), implying substantial variability and warranting cautious interpretation of PLR's responsiveness. By contrast, NLR, SII, and NHL exhibited consistent, significant declines, suggesting that these indices capture aspects of disease activity or improvement not fully reflected by ESR/CRP in this context. Even though ESR and CRP remained normal and static, NLR, SII, and NHL significantly decreased in tandem with clinical improvement, supporting their potential utility as alternative or complementary markers of treatment response.
TABLE 3.
Levels of hematological indices and disease activity markers pre‐ and post‐b/tsDMARDs in RA patients with normal ESR and CRP.
| Baseline (n = 72) | After 6 months (n = 72) | p | Cohen's d (95% CI) | |
|---|---|---|---|---|
| Disease activity parameters | ||||
| DAS28‐ESR | 5.71 (5.21, 6.05) | 3.85 (3.27, 4.29) | < 0.001 | 2.017 (1.611–2.418) |
| ESR (mm/h) | 9 (4, 13) | 8 (3, 15) | 0.606 | −0.218 (−0.451 to 0.017) |
| CRP (mg/dL) | 0.12 (0.10, 0.31) | 0.10 (0.10, 0.24) | 0.615 | 0.037 (−0.195 to 0.268) |
| Hematological indices | ||||
| NLR | 2.06 (1.41, 2.88) | 1.66 (1.23, 2.14) | 0.003 | 0.300 (0.063–0.535) |
| PLR | 1.34 (0.97, 1.72) | 1.15 (0.97, 1.55) | 0.096 | 0.153 (−0.080 to 0.385) |
| SII | 474 (333, 718) | 390 (290, 490) | 0.012 | 0.218 (−0.017 to 0.451) |
| NHL score | 0.16 (0.11, 0.23) | 0.12 (0.09, 0.18) | 0.003 | 0.295 (0.058–0.530) |
Note: Data are presented as median with interquartile ranges in parenthesis (Q1, Q3).
Abbreviations: b/tsDMARDs, biological and targeted synthetic disease‐modifying antirheumatic drugs; CI, confidence interval; CRP, C‐reactive protein; DAS28, disease activity score for 28 joints; ESR, erythrocyte sedimentation rate; NHL, neutrophil‐to‐hemoglobin and lymphocyte score; NLR, neutrophil‐to‐lymphocyte ratio; PLR, platelet‐to‐lymphocyte ratio; RA, rheumatoid arthritis; SII, systemic immune‐inflammation index.
3.6. Hematological Indices Can Reflect the Therapeutic Efficacy in Active RA Patients With Normal ESR and CRP Levels Receiving b/tsDMARDs
The ROC curve analysis results, presented in Table 4 and illustrated in Figure S2, demonstrated that changes in hematological indices can also reflect disease improvement and treatment response in active RA patients with normal ESR and CRP. The AUC (95% CI) values ranged from 0.685 (0.535–0.836) for ΔPLR to 0.752 (0.617–0.886) for ΔSII in predicting ΔDAS28 > 1.2. To minimize the FPR, we constrained FPR ≤ 0.2. The sensitivity of hematological indices (ranging from 46.8% for PLR to 66.1% for NLR and NHL score) was significantly higher than that of ESR and CRP (both 29%). Additionally, the bootstrap method was applied to validate the differences in pAUCs, and hematological indices were found to have a significantly larger pAUC than the traditional inflammatory markers in predicting the treatment response (ESR vs. hematological indices = 0.019 vs. 0.089 for PLR to 0.103 for NHL). These findings suggest that, when the FPR is constrained to ≤ 0.2, hematological indices provide superior accuracy (Table 4). Similar results were also observed in active RA patients with normal ESR and either normal or elevated CRP, as detailed in Tables S4 and S5.
TABLE 4.
ROC analysis of hematological changes for DAS28 reduction > 1.2 in RA patients with normal ESR and CRP (n = 72).
| Change of indices | AUC (95% CI) | Sensitivity | Specificity | Cut‐off value | Positive likelihood ratio | Negative likelihood ratio | p | pAUC (FPR ≤ 0.2) | p a |
|---|---|---|---|---|---|---|---|---|---|
| Traditional inflammatory markers | |||||||||
| △ESR | 0.617 (0.398–0.836) | 29.0% | 80.0% | 4.5 | 1.45 | 0.89 | 0.295 | 0.019 | — |
| △CRP | 0.556 (0.342–0.771) | 29.0% | 80.0% | 0.02 | 1.45 | 0.89 | 0.606 | 0.016 | 0.930 |
| Hematological indices | |||||||||
| △NLR | 0.726 (0.598–0.854) | 66.1% | 80.0% | 0.18 | 3.31 | 0.42 | 0.001 | 0.097 | 0.026 |
| △PLR | 0.685 (0.535–0.836) | 46.8% | 90.0% | 0.15 | 4.68 | 0.59 | 0.016 | 0.089 | 0.013 |
| △SII | 0.752 (0.617–0.886) | 54.8% | 80.0% | 52.5 | 2.74 | 0.56 | < 0.001 | 0.094 | 0.014 |
| △NHL score | 0.747 (0.626–0.868) | 66.1% | 80.0% | 0.01 | 3.31 | 0.42 | < 0.001 | 0.103 | 0.012 |
Abbreviations: AUC, area under the curve; b/tsDMARDs, biological and targeted synthetic disease‐modifying antirheumatic drugs; CI, confidence interval; CRP, C‐reactive protein; DAS28, disease activity score for 28 joints; ESR, erythrocyte sedimentation rate; FPR, false positive rate; NHL, neutrophil‐to‐hemoglobin and lymphocyte score; NLR, neutrophil‐to‐lymphocyte ratio; pAUC, partial area under the curve; PLR, platelet‐to‐lymphocyte ratio; RA, rheumatoid arthritis; ROC, receiver operating characteristic; SII, systemic immune‐inflammation index.
The p‐values for pAUC were determined between ESR and other indices.
3.7. Hematological Indices May Be More Reliable Than ESR/CRP in Assessing Disease Activity in IL‐6Ri‐ or JAKi‐Treated Patients
RA patients treated with IL‐6Ri or JAKi often exhibit substantial reduction in ESR and CRP, potentially underestimating actual disease activity. In many of these patients, CRP normalizes (≤ 0.5 mg/dL) while DAS28‐ESR remains elevated (> 3.2), indicating low CRP may not reflect adequate disease control. After adjusting for CRP and each hematological index in a multiple linear regression model, SII was significantly associated with DAS28‐ESR reduction (standardized beta coefficient = 0.590, p = 0.031), whereas CRP and other hematological indices did not show significant associations (Table 5). These findings suggest that hematological indices, particularly SII, may provide a more reliable assessment of disease severity and serve as surrogate markers in patients receiving IL‐6Ri or JAKi.
TABLE 5.
Multiple linear regression to determine the relative contribution of each predictor to DAS28‐ESR reduction in RA patients receiving IL‐6Ri or JAKi (n = 102).
| Predictors | Beta coefficients | 95% CI | Standardized beta | p |
|---|---|---|---|---|
| △CRP (mg/dL) | 0.016 | −0.070 to 0.102 | 0.038 | 0.709 |
| △NLR | −0.108 | −0.424 to 0.209 | −0.277 | 0.501 |
| △PLR | −0.052 | −0.336 to 0.232 | −0.056 | 0.716 |
| △SII | 0.001 | 0.000–0.001 | 0.590 | 0.031 |
| △NHL score | 0.460 | −3.505 to 4.425 | 0.105 | 0.818 |
Note: Multivariate linear regression model adjusted for CRP and individual hematological indices. Dependent variable: DAS28‐ESR. RA patients whose CRP levels returned to normal (≤ 0.5 mg/dL) while DAS28‐ESR remained elevated (> 3.2) after 6 months of treatment with IL‐6Ri or JAKi were analyzed.
Abbreviations: CI, confidence interval; CRP, C‐reactive protein; DAS28, disease activity score for 28 joints; ESR, erythrocyte sedimentation rate; IL‐6Ri, IL‐6 receptor inhibitors; JAKi, Janus kinase inhibitors; NHL, neutrophil‐to‐hemoglobin and lymphocyte score; NLR, neutrophil‐to‐lymphocyte ratio; PLR, platelet‐to‐lymphocyte ratio; RA, rheumatoid arthritis; SII, systemic immune‐inflammation index.
4. Discussion
Our study provides important insight into the role of hematological indices, including NLR, PLR, SII, and NHL, in assessing disease improvement and treatment response in RA patients receiving b/tsDMARDs. We demonstrated that changes in these indices significantly correlate with improvements in disease activity as assessed by DAS28‐ESR. They also complement DAS28, ESR, and CRP in monitoring efficacy, especially in patients with normal ESR and/or CRP, where traditional inflammatory markers may not adequately reflect disease severity.
While previous studies suggested that hematological indices such as NLR and PLR may not outperform ESR and CRP in distinguishing active (DAS28 ≥ 2.6) from inactive (DAS28 < 2.6) RA [17], our findings highlight a different and potentially more clinically relevant application. Specifically, we used these indices not merely for diagnosis but to monitor treatment response and disease improvement over time. By evaluating both the magnitude of DAS28 reduction and achievement of EULAR‐defined response criteria, we demonstrated that changes in NLR, SII, and NHL effectively reflect treatment efficacy, particularly in active patients with normal ESR and CRP. This extends applicability beyond diagnostic classification, positioning them as complementary tools to ESR, CRP, and DAS28 in disease monitoring. Moreover, incorporating these indices in subpopulations with normal ESR/CRP, our findings suggest broader real‐world clinical applicability, where discordance between clinical presentation and laboratory findings is common.
Although limited studies have reported that NLR, PLR, SII, and NHL decrease following treatment with IL‐6Ri, TNFi, and JAKi, correlating with clinical improvements [8, 12, 13, 14], our results are consistent with and extend these findings (Tables S2 and S3). Specifically, we observed similar trends in CTLA4‐Ig‐treated patients, as well as those with normal ESR and CRP.
As previously mentioned, the sensitivity of ESR and CRP for detecting synovial inflammation is suboptimal (71.3% and 48.4%, respectively) [1], and approximately 58% of RA patients with ongoing activity have normal levels of both [2]. Consequently, DAS28‐ESR/CRP may not accurately reflect disease severity. In active RA with normal ESR and CRP, despite severe synovitis, DAS28 may underestimate disease severity, as both ESR and CRP contribute significantly to its calculation. Although disease activity markers other than DAS28, such as the clinical disease activity index (CDAI) or simplified disease activity index (SDAI), were not incorporated into our study, our results suggest these hematological indices may better reflect disease activity than ESR and CRP in this subgroup, supporting their potential as alternative biomarkers for disease monitoring (Tables 3 and 4). Hematological indices derived from routine CBCs reflect inflammatory activity through pathways distinct from acute‐phase reactants and can track disease activity even when ESR/CRP are normal. For instance, platelet count acts as an acute‐phase responsive cellular marker, rising via IL‐6 driven, thrombopoietin‐mediated megakaryopoiesis; accordingly, PLR and SII may reflect inflammatory burden despite normal ESR/CRP. Likewise, NLR, SII, and NHL may rise in active disease due to stress leukocytosis, characterized by neutrophilia with relative lymphopenia, even in the absence of elevated ESR/CRP.
Cheng et al. reported that, among RA patients with normal ESR/CRP, higher PLR correlated with ultrasound‐detected synovitis and erosions [11]. In our clinically active ESR/CRP‐normal subgroup, however, PLR did not decrease significantly with treatment (Table 3). The study populations differ substantially: many in Cheng et al. were in remission or LDA, whereas 96% of ours had HDA. Inflammation in our patients was overt despite flat acute‐phase reactants; under such conditions, PLR may contribute limited information. Moreover, a proportion of our patients received IL‐6Ri or JAKi, therapies that can lower platelets and thereby blunt or confound PLR changes. Taken together, these contextual and pharmacologic factors may explain the discrepancy, suggesting PLR is more variable in clinically active RA with normal ESR/CRP.
Additionally, IL‐6Ri/JAKi often markedly reduce ESR/CRP, which may lead to disproportionately high remission rates when assessed by DAS28, compared to CDAI or SDAI. Since ESR or CRP is heavily weighted in DAS28, while CRP has a low weight in SDAI, DAS28 may underestimate synovitis with these therapies [18, 19, 20]. This may explain why a significant proportion of patients classified as in clinical remission by DAS28 still show ultrasound synovitis [21, 22]. Future studies incorporating joint ultrasound may provide a more accurate assessment of synovitis, as it provides superior sensitivity compared to laboratory measurements. Given therapy‐suppressed ESR/CRP, hematological indices, especially SII, may better reflect disease activity in IL‐6Ri/JAKi recipients.
IL‐6Ri and JAKi directly affect hematopoiesis. IL‐6 is crucial for megakaryopoiesis and maintaining circulating neutrophil counts via mobilization and survival signals [23, 24, 25, 26, 27]. Accordingly, IL‐6 blockade reduces platelet production and may cause neutropenia. JAK2 is essential for myelopoiesis; its inhibition (e.g., baricitinib; tofacitinib to a lesser extent) may lower neutrophil counts and hemoglobin [28]. JAK3 is predominantly expressed in lymphoid cells, and its inhibition (e.g., tofacitinib) can cause lymphopenia. Clinical trials document dose‐dependent decreases in lymphocyte and neutrophil counts with tofacitinib and upadacitinib and transient lymphocytosis with baricitinib followed by stabilization or decline to baseline [29, 30]. Given these pharmacologic effects, single biomarkers such as ESR, CRP, NLR, or PLR may not reliably indicate inflammation under IL‐6Ri/JAKi because the therapies themselves modulate the components of these measures. SII, integrating three parameters, seems more robust. Among patients receiving IL‐6Ri, for example, CRP may be suppressed to near zero and platelets reduced, while ongoing inflammation limits neutrophil decline and keeps lymphocytes relatively low due to stress response; SII remains elevated compared with well‐controlled disease. When inflammation is controlled, neutrophils decline and/or lymphocytes rise, and SII decreases, as observed in our cohort (Table S2). In the IL‐6Ri/JAKi subgroup, PLR alone was less informative because platelet suppression with IL‐6Ri lowers its numerator, whereas such platelet effects are not typical with JAKi. Likewise, NLR alone can be distorted by JAKi‐related lymphopenia. In contrast, SII captured overall hematologic balance and correlated with DAS28 improvements.
The three JAKi in our study have distinct JAK‐selectivity profiles, potentially differently affecting hematological indices. Tofacitinib primarily inhibits JAK1/JAK3, baricitinib inhibits JAK1/JAK2, and upadacitinib is relatively JAK1‐selective. In this study, however, we analyzed JAKi as a class because sample size was insufficient to support robust, drug‐specific subgroup comparisons. Notably, a head‐to‐head study reported no significant differences in changes in hematological indices, including NLR, PLR, SII, and NHL, between tofacitinib and baricitinib over 24 weeks, except platelet count [8], suggesting class effects may predominate. Future studies with larger cohorts should evaluate potential drug‐specific effects.
In our cohort, 162 of 308 patients (52.6%) received glucocorticoids at baseline (median prednisolone dose in the overall cohort, 3.05 mg/day). After 6 months, this proportion declined to 47 of 308 (15.3%), and median dose fell to 0 mg/day, consistent with the steroid‐sparing effect by b/tsDMARDs (Table S6). Prednisolone dose changes in ESR/CRP‐normal patients are shown in Table S7. A large proportion of patients received low‐dose prednisolone (≤ 5 mg/day) at baseline, especially in the ESR/CRP‐normal subgroup. Evidence indicates that oral prednisolone < 10 mg/day does not meaningfully change neutrophil or lymphocyte counts [31, 32, 33]. Nevertheless, glucocorticoids raise circulating neutrophils and lower lymphocytes, thereby elevating NLR; tapering after b/tsDMARD initiation likely contributed, in conjunction with clinical improvement, to the observed reductions in NLR and related indices. This pattern aligns with routine practice, where effective disease control enables gradual glucocorticoid tapering or withdrawal.
Our study has some limitations. First, the retrospective study design may involve potential confounding factors, such as comorbidities, previous medication use, or concomitant therapies, which may influence hematological indices and treatment outcomes. For instance, corticosteroids are known to increase neutrophil levels while decreasing lymphocyte counts, resulting in an elevated NLR. Additionally, all enrolled patients were from a single center and the relatively small sample sizes in individual treatment subgroups may limit the generalizability of our findings. Therefore, future prospective studies with larger cohorts are required to validate our results. Second, the reliance on DAS28‐ESR to define treatment response is susceptible to inherent biases in clinical scoring systems, such as differences in patients' pain tolerance and inter‐physician variability in assessing joint tenderness or swelling. Furthermore, we did not investigate the relationship of hematological indices with disease activity markers other than DAS28, such as CDAI or SDAI. Nonetheless, our findings support the potential clinical utility of hematological indices in monitoring treatment response and disease improvement, particularly in patients with normal ESR and CRP levels, where traditional inflammatory markers may be less informative.
5. Conclusion
In conclusion, hematological indices—particularly NLR, SII, and NHL—can aid evaluation of disease activity and treatment response in RA patients receiving b/tsDMARDs. These indices correlated with improvements in DAS28‐ESR and may complement traditional markers, especially in clinically active patients with normal ESR/CRP, where ESR/CRP underperform. Baseline values were not reliable predictors; however, on‐treatment changes in NLR, SII, and NHL tracked clinical improvement. In IL‐6Ri‐ or JAKi‐treated patients, SII was the only index consistently tracking response, whereas PLR and NLR were variably affected by drug‐related platelet and lymphocyte changes. SII may therefore be the preferred hematological index when IL‐6 or JAK pathways are targeted.
Author Contributions
All authors contributed to the conceptualization and design of the study. J.‐P.J. drafted the manuscript, and J.‐P.J., Y.‐H.H., Y.‐S.C., and W.‐C.S. revised it. All authors reviewed and approved the final version for publication. J.‐P.J., Y.‐H.H., and W.‐C.S. had full access to all study data and took responsibility for the integrity of the data and the accuracy of the data analysis. J.‐P.J. and Y.‐H.H. were responsible for the acquisition, analysis, and interpretation of the data.
Funding
This work was supported by Taipei Medical University–Shuang Ho Hospital, Ministry of Health and Welfare (Grant No. 114HCP‐14).
Ethics Statement
The study protocol was reviewed and approved by the Ethics Committee of the Taipei Medical University Joint Institutional Review Board (TMU‐JIRB N202502055).
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Table S1: The classification of treatment response (good, moderate, or no response).
Table S2: The levels of hematological indices and disease activity markers before and after use of b/tsDMARDs.
Table S3: Correlation between changes in hematological indices and disease activity indices in RA patients treated with b/tsDMARDs (n = 308).
Table S4: Levels of hematological indices and disease activity markers pre‐ and post‐b/tsDMARDs in RA patients with normal ESR.
Table S5: ROC analysis of hematological changes for DAS28 reduction > 1.2 in RA patients with normal ESR (n = 108).
Table S6: Daily prednisolone dose at baseline and 6 months after b/tsDMARD initiation in overall cohort (n = 308).
Table S7: Daily prednisolone dose at baseline and 6 months after b/tsDMARD initiation in RA patients with normal ESR and CRP (n = 72).
Figure S1: Levels of hematological indices and disease activity markers pre‐ and 6 months post‐b/tsDMARDs treatments (n = 308).
Figure S2: ROC analysis of hematological changes for DAS28 reduction > 1.2 in RA patients with normal ESR and CRP (n = 72).
Acknowledgments
We acknowledge Pharm. D. Yen‐Ying Lee for providing valuable and insightful suggestions for this study. We also appreciate the statistical consultation services provided by Dr. Ying‐Hsuan Tai from the Department of Medical Research, Shuang Ho Hospital, Taipei Medical University.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Orr C. K., Najm A., Young F., et al., “The Utility and Limitations of CRP, ESR and DAS28‐CRP in Appraising Disease Activity in Rheumatoid Arthritis,” Frontiers in Medicine 5 (2018): 185. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Kay J., Morgacheva O., Messing S. P., et al., “Clinical Disease Activity and Acute Phase Reactant Levels Are Discordant Among Patients With Active Rheumatoid Arthritis: Acute Phase Reactant Levels Contribute Separately to Predicting Outcome at One Year,” Arthritis Research & Therapy 16, no. 1 (2014): R40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Choe J. Y., Lee C. U., and Kim S. K., “Association Between Novel Hematological Indices and Measures of Disease Activity in Patients With Rheumatoid Arthritis,” Medicina (Kaunas, Lithuania) 59, no. 1 (2023): 117. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Erre G. L., Buscetta G., Mangoni A. A., et al., “Diagnostic Accuracy of Different Blood Cells‐Derived Indexes in Rheumatoid Arthritis: A Cross‐Sectional Study,” Medicine (Baltimore) 99, no. 44 (2020): e22557. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Obaid J., Almjydy M. M. A., Garban M. A. Q., Al‐Hebari F. S. Q., and Al‐Washah N. A. H., “Neutrophil‐To‐Monocyte Ratio Is the Better New Inflammatory Marker Associated With Rheumatoid Arthritis Activity,” Health Science Reports 6, no. 8 (2023): e1478. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Peng Y. F., Cao L., Zeng Y. H., et al., “Platelet to Lymphocyte Ratio and Neutrophil to Lymphocyte Ratio in Patients With Rheumatoid Arthritis,” Open Medicine (Warsaw) 10, no. 1 (2015): 249–253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Xu Y., He H., Zang Y., et al., “Systemic Inflammation Response Index (SIRI) as a Novel Biomarker in Patients With Rheumatoid Arthritis: A Multi‐Center Retrospective Study,” Clinical Rheumatology 41, no. 7 (2022): 1989–2000. [DOI] [PubMed] [Google Scholar]
- 8. Choe J. Y. and Kim S. K., “Association Between Hematological Indicesand Disease Activity in Patients With Rheumatoid Arthritis Treated With Janus Kinase Inhibitors for 24 Weeks,” Medicina (Kaunas, Lithuania) 58, no. 3 (2022): 426. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Dervisevic A., Fajkic A., Jahic E., et al., “Systemic Immune‐Inflammation Index in Evaluation of Inflammation in Rheumatoid Arthritis Patients,” Medeniyet Medical Journal 39, no. 3 (2024): 183–191. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Targońska‐Stępniak B., Zwolak R., Piotrowski M., Grzechnik K., and Majdan M., “The Relationship Between Hematological Markers of Systemic Inflammation (Neutrophil‐To‐Lymphocyte, Platelet‐To‐Lymphocyte, Lymphocyte‐To‐Monocyte Ratios) and Ultrasound Disease Activity Parameters in Patients With Rheumatoid Arthritis,” Journal of Clinical Medicine 9, no. 9 (2020): 2760. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Cheng J. H., Cai W. X., Xiang X. H., et al., “Platelet‐To‐Lymphocyte Ratios as a Haematological Marker of Synovitis in Rheumatoid Arthritis With Normal Acute Phase Reactant Level,” Annals of Medicine 56, no. 1 (2024): 2346546. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Pan Y. J., Su K. Y., Shen C. L., and Wu Y. F., “Correlation of Hematological Indices and Acute‐Phase Reactants in Rheumatoid Arthritis Patients on Disease‐Modifying Antirheumatic Drugs: A Retrospective Cohort Analysis,” Journal of Clinical Medicine 12, no. 24 (2023): 7611. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Yetişir A., Sariyildiz A., Türk İ., and Coskun Benlidayi I., “Evaluation of Inflammatory Biomarkers and the Ratio of Hemoglobin‐Red Cell Distribution Width in Patients With Rheumatoid Arthritis Treated With Tumor Necrosis Factor‐Alpha Inhibitors,” Clinical Rheumatology 43, no. 6 (2024): 1815–1821. [DOI] [PubMed] [Google Scholar]
- 14. Zhou L., Xiao D. M., Qin W., et al., “The Clinical Value of Hematological Markers in Rheumatoid Arthritis Patients Treated With Tocilizumab,” Journal of Clinical Laboratory Analysis 33, no. 5 (2019): e22862. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Fransen J. and van Riel P. L., “The Disease Activity Score and the EULAR Response Criteria,” Clinical and Experimental Rheumatology 23, no. 5 Suppl 39 (2005): S93–S99. [PubMed] [Google Scholar]
- 16. Wells G., Becker J. C., Teng J., et al., “Validation of the 28‐Joint Disease Activity Score (DAS28) and European League Against Rheumatism Response Criteria Based on C‐Reactive Protein Against Disease Progression in Patients With Rheumatoid Arthritis, and Comparison With the DAS28 Based on Erythrocyte Sedimentation Rate,” Annals of the Rheumatic Diseases 68, no. 6 (2009): 954–960. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Lijuan W., Yuting Z., Chaoyang L., and Ju Y., “Neutrophil‐Lymphocyte, Platelet‐Lymphocyte and Lymphocyte‐Monocyte Ratios May Not Be Useful Markers to Assess Disease Activity in Rheumatoid Arthritis: A STROBE‐Compliant Article,” Medicine (Baltimore) 100, no. 45 (2021): e27631. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Smolen J. S. and Aletaha D., “Interleukin‐6 Receptor Inhibition With Tocilizumab and Attainment of Disease Remission in Rheumatoid Arthritis: The Role of Acute‐Phase Reactants,” Arthritis and Rheumatism 63, no. 1 (2011): 43–52. [DOI] [PubMed] [Google Scholar]
- 19. Asai S., Takahashi N., Kobayakawa T., et al., “Comparison of the Effects of Baricitinib and Tocilizumab on Disease Activity in Patients With Rheumatoid Arthritis: A Propensity Score Matching Analysis,” Clinical Rheumatology 40, no. 8 (2021): 3143–3151. [DOI] [PubMed] [Google Scholar]
- 20. Frade‐Sosa B., Ponce A., Ruiz‐Esquide V., et al., “High Sensitivity C Reactive Protein in Patients With Rheumatoid Arthritis Treated With Antibodies Against IL‐6 or Jak Inhibitors: A Clinical and Ultrasonographic Study,” Diagnostics (Basel) 12, no. 1 (2022): 182. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Scirè C. A., Montecucco C., Codullo V., Epis O., Todoerti M., and Caporali R., “Ultrasonographic Evaluation of Joint Involvement in Early Rheumatoid Arthritis in Clinical Remission: Power Doppler Signal Predicts Short‐Term Relapse,” Rheumatology (Oxford, England) 48, no. 9 (2009): 1092–1097. [DOI] [PubMed] [Google Scholar]
- 22. Brown A. K., Quinn M. A., Karim Z., et al., “Presence of Significant Synovitis in Rheumatoid Arthritis Patients With Disease‐Modifying Antirheumatic Drug‐Induced Clinical Remission: Evidence From an Imaging Study May Explain Structural Progression,” Arthritis and Rheumatism 54, no. 12 (2006): 3761–3773. [DOI] [PubMed] [Google Scholar]
- 23. Gordon M. S., Nemunaitis J., Hoffman R., et al., “A Phase I Trial of Recombinant Human Interleukin‐6 in Patients With Myelodysplastic Syndromes and Thrombocytopenia,” Blood 85, no. 11 (1995): 3066–3076. [PubMed] [Google Scholar]
- 24. Ishibashi T., Kimura H., Shikama Y., et al., “Interleukin‐6 Is a Potent Thrombopoietic Factor In Vivo in Mice,” Blood 74, no. 4 (1989): 1241–1244. [PubMed] [Google Scholar]
- 25. Van Zaanen H. C., Lokhorst H. M., Aarden L. A., Rensink H. J., Warnaar S. O., and Van Oers M. H., “Blocking Interleukin‐6 Activity With Chimeric Anti‐IL6 Monoclonal Antibodies in Multiple Myeloma: Effects on Soluble IL6 Receptor and Soluble gp130,” Leukemia & Lymphoma 31, no. 5–6 (1998): 551–558. [DOI] [PubMed] [Google Scholar]
- 26. Rodríguez Mdel C., Bernad A., and Aracil M., “Interleukin‐6 Deficiency Affects Bone Marrow Stromal Precursors, Resulting in Defective Hematopoietic Support,” Blood 103, no. 9 (2004): 3349–3354. [DOI] [PubMed] [Google Scholar]
- 27. Jones M. R., Quinton L. J., Simms B. T., Lupa M. M., Kogan M. S., and Mizgerd J. P., “Roles of Interleukin‐6 in Activation of STAT Proteins and Recruitment of Neutrophils During Escherichia coli Pneumonia,” Journal of Infectious Diseases 193, no. 3 (2006): 360–369. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Bonelli M., Kerschbaumer A., Kastrati K., et al., “Selectivity, Efficacy and Safety of JAKinibs: New Evidence for a Still Evolving Story,” Annals of the Rheumatic Diseases 83, no. 2 (2024): 139–160. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Taylor P. C., Keystone E. C., van der Heijde D., et al., “Baricitinib Versus Placebo or Adalimumab in Rheumatoid Arthritis,” New England Journal of Medicine 376, no. 7 (2017): 652–662. [DOI] [PubMed] [Google Scholar]
- 30. Genovese M. C., Kremer J., Zamani O., et al., “Baricitinib in Patients With Refractory Rheumatoid Arthritis,” New England Journal of Medicine 374, no. 13 (2016): 1243–1252. [DOI] [PubMed] [Google Scholar]
- 31. Dale D. C., Fauci A. S., Guerry D. I., and Wolff S. M., “Comparison of Agents Producing a Neutrophilic Leukocytosis in Man. Hydrocortisone, Prednisone, Endotoxin, and Etiocholanolone,” Journal of Clinical Investigation 56, no. 4 (1975): 808–813. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Kauh E., Mixson L., Malice M. P., et al., “Prednisone Affects Inflammation, Glucose Tolerance, and Bone Turnover Within Hours of Treatment in Healthy Individuals,” European Journal of Endocrinology 166, no. 3 (2012): 459–467. [DOI] [PubMed] [Google Scholar]
- 33. Fleishaker D. L., Mukherjee A., Whaley F. S., Daniel S., and Zeiher B. G., “Safety and Pharmacodynamic Dose Response of Short‐Term Prednisone in Healthy Adult Subjects: A Dose Ranging, Randomized, Placebo‐Controlled, Crossover Study,” BMC Musculoskeletal Disorders 17 (2016): 293. [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
Table S1: The classification of treatment response (good, moderate, or no response).
Table S2: The levels of hematological indices and disease activity markers before and after use of b/tsDMARDs.
Table S3: Correlation between changes in hematological indices and disease activity indices in RA patients treated with b/tsDMARDs (n = 308).
Table S4: Levels of hematological indices and disease activity markers pre‐ and post‐b/tsDMARDs in RA patients with normal ESR.
Table S5: ROC analysis of hematological changes for DAS28 reduction > 1.2 in RA patients with normal ESR (n = 108).
Table S6: Daily prednisolone dose at baseline and 6 months after b/tsDMARD initiation in overall cohort (n = 308).
Table S7: Daily prednisolone dose at baseline and 6 months after b/tsDMARD initiation in RA patients with normal ESR and CRP (n = 72).
Figure S1: Levels of hematological indices and disease activity markers pre‐ and 6 months post‐b/tsDMARDs treatments (n = 308).
Figure S2: ROC analysis of hematological changes for DAS28 reduction > 1.2 in RA patients with normal ESR and CRP (n = 72).
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.