Urine cytokines as biomarkers for diagnosing interstitial cystitis/bladder pain syndrome and mapping its clinical characteristics
Abstract
The objective of the present study was to investigate the diagnostic values of urine cytokines in patients with interstitial cystitis/bladder pain syndrome (IC/BPS) and to identify their correlations with clinical characteristics. Urine samples were collected from 127 patients with IC/BPS [European Society for the Study of Interstitial Cystitis (ESSIC) types 1 and 2] and 28 controls. Commercially available multiplex immunoassays (MILLIPLEX map kits) were used to analyze 31 targeted cytokines. Cytokine levels between patients with IC/BPS and controls were analyzed using ANOVA. Receiver-operating characteristic curves of each cytokine to distinguish IC/BPS from controls were generated for calculation of the area under the curve. Patients with IC/BPS had urine cytokine profiles that differed from those of controls. Between patients with ESSIC type 1 and 2 IC/BPS, urine cytokine profiles were also different. Among cytokines with high diagnostic values (i.e., area under the curve > 0.7) with respect to distinguish patients with ESSIC type 2 IC/BPS from controls, regulated upon activation, normal T cell expressed and presumably secreted (RANTES), macrophage inflammatory protein (MIP)-1β, and IL-8 were of higher sensitivity, whereas macrophage chemoattractant protein (MCP)-1, chemokine (C-X-C motif) ligand 10 (CXCL10), and eotaxin-1 were of higher specificity. In multivariate logistic regression models controlling for age, sex, body mass index, and diabetes mellitus, the urine cytokines with high diagnostic values (MCP-1, RANTES, CXCL10, IL-7, and eotaxin-1) remained statistically significant in differentiating IC/BPS and controls. MCP-1, CXCL10, eotaxin-1, and RANTES were positively correlated with glomerulation grade and negatively correlated with maximal bladder capacity. In conclusion, patients with IC/BPS had urine cytokine profiles that clearly differed from those of controls. Urine cytokines might be useful as biomarkers for diagnosing IC/BPS and mapping its clinical characteristics.
INTRODUCTION
Interstitial cystitis/bladder pain syndrome (IC/BPS) is a chronic inflammatory disorder of the urinary bladder characterized by bladder pain associated with urinary frequency, urgency, nocturia, and sterile urine (16). The clinical characteristics of IC/BPS are heterogeneous, and the core symptom of IC/BPS is bladder pain; various phenotypes are associated with distinct outcomes (26). In addition to pain, IC/BPS has a variety of clinical phenotypes, including urinary, psychosocial, organ-specific, infection, neurological/systemic, and tenderness (UPOINT system) (22). IC/BPS is a disease affecting patients’ quality of life profoundly due to disabling symptoms (16). However, the etiology of IC/BPS remains unclear and is considered multifactorial, including defective/damaged bladder urothelium, activation of C-fibers, neurogenic inflammation with activation of mast cells, autoimmunity, and occult infection (16, 32).
The diagnostic criteria of IC/BPS are diverse (20). The National Institute of Diabetes and Digestive and Kidney Diseases uses strict criteria for IC/BPS (15), which limit their application outside of clinical trials. Many national and international societies have developed separate clinical guidelines. Because of many unknown factors in the pathophysiology and diverse clinical diagnostic criteria of IC/BPS, the development of relevant biomarkers is slow in patients with IC/BPS.
During the inflammatory process, human detrusor smooth muscle cells produce cytokines and chemokines, which are measured in the urine (6). Notably, urinary cytokine and chemokine levels are elevated in patients with IC/BPS (29). Gene expression analysis of urine sediment has been reported to discriminate patients with IC/BPS from controls (5). The upregulation of genes mainly associated with inflammation has been reported in patients with ulcerative IC/BPS but not in patients with nonulcerative IC/BPS or controls. Analysis of these proteins in urine is a noninvasive approach to assess the inflammatory status inside the bladder and has potential for the development of biomarkers in IC/BPS.
Urothelial dysfunction and suburothelial inflammation are important pathological findings in IC/BPS (24); neurotrophins, cytokines, and chemokines may play important roles in the pathogenesis of IC/BPS. Although many candidate biomarkers for IC/BPS have been suggested, the diagnostic values have not been clearly analyzed. Therefore, this study investigated the diagnostic values of urine cytokines in IC/BPS and assessed their correlations with clinical characteristics.
MATERIALS AND METHODS
Patients.
From February 2014 to December 2018, we prospectively enrolled 127 consecutive patients with clinical IC/BPS in the Department of Urology of a single medical center (Hualien Tzu Chi Hospital, Taiwan). All enrolled patients were Taiwanese (Asian). The diagnostic criteria for IC/BPS, based on the proposed guidelines of the European Society for the Study of Interstitial Cystitis (ESSIC), constituted “chronic pelvic pain, pressure, or discomfort perceived to be related to the urinary bladder accompanied by at least one other urinary symptom, such as persistent urge to void or urinary frequency, for more than 6 months” and the exclusion of potentially similar diseases (30). The workup included medical history, physical examinations, urinalysis, urine culture, prostate specific antigen (PSA) in male patients more than 40 yr old, postvoid residual urine volume, cystoscopy, and bladder biopsy. Enrolled patients underwent cystoscopy with hydrodistention under general anesthesia and were classified as ESSIC type 1 or 2 (without and with glomerulations, respectively). Patients with Hunner’s lesion (ESSIC type 3) were excluded from this study. Other exclusion criteria for analysis included active urinary tract infection, neurogenic voiding dysfunction (including cerebrovascular accident, spinal cord injury, multiple sclerosis, and Parkinson’s disease), a history of urinary tract malignancy or tuberculosis, a history of bladder surgery/or traumatic injury, a history of urethral or prostate surgery, a history of pelvic radiation, a history of nephrotic or nephritic syndrome, urolithiasis, and/or impaired renal function (serum creatinine > 2.0 mg/dL).
As controls, we also included 28 women with genuine stress urinary incontinence without other significant lower urinary tract symptoms (defined as International Prostate Symptom Score < 6); all controls were prepared to undergo the suburethral sling procedure and had no other storage or voiding dysfunction in video-urodynamic studies.
Clinical investigation.
In patients with IC/BPS, assessment of clinical symptoms included O'Leary-Saint symptom score, interstitial cystitis symptom index, interstitial cystitis problem index, and visual analog scale pain score.
Urine biomarker investigation.
For biomarker investigation, urine samples were collected from all enrolled patients with IC/BPS and controls before the surgical procedures. Urine was self-voided when patients or controls reported a full bladder sensation. Urinalysis was performed to confirm an infection-free status before urine samples were stored. In total, 30 mL of urine were placed on ice immediately and transferred to the laboratory for preparation. Samples were centrifuged at 1,800 rpm for 10 min at 4°C. Supernatants were separated into aliquots in 1.5-mL tubes (1 mL/tube) and stored at −80°C. Before further analyses were performed, the frozen urine samples were centrifuged at 12,000 rpm for 20 min at 4°C, and the supernatants were used for subsequent measurements.
Inflammation-related cytokines and chemokines, as well as neurotrophins in urine, were assayed using commercially available microspheres with the Milliplex Human cytokine/chemokine magnetic bead-based panel kit (Millipore, Darmstadt, Germany). Thirty-one targeted analytes were used for the multiplex kit, including inflammatory cytokines and chemokines: catalog number HCYTMAG-60K-PX30 [epidermal growth factor, eotaxin-1, granulocyte-colony stimulating factor, granulocyte-macrophage colony-stimulating factor, interferon (IFN)-α2, IFN-γ, IL-10, IL-12p40, IL-12p70, IL-13, IL-15, IL-17A, IL-1 receptor antagonist, 1L-1α, IL-1β, IL-2, IL-3, IL-4, IL-5, IL-6, IL-7, IL-8, chemokine (C-X-C motif) ligand 10 (CXCL10), monocyte chemoattractant protein-1 (MCP-1), macrophage inflammatory protein (MIP)-1β, MIP-1α, regulated upon activation, normal T cell expressed and presumably secreted (RANTES), TNF-α, TNF-β, and VEGF] and catalog number HADK2MAG-61K (nerve growth factor).
The quantification of targeted analytes in urine samples was performed in accordance with the manufacturer’s instructions. Briefly, 25 μL assay buffer, 25 μL urine sample, and 25 μL beads were sequentially added to 96-well plates (panel kits), and the plates were incubated overnight in the dark at 4°C. The contents of the wells were removed, and the plates were washed twice with 200 μL wash buffer. Twenty-five microliters of detection antibody were added to each well, and the plates were incubated in the dark on a shaker plate for 1 h at room temperature. Twenty-five microliters of streptavidin-phycoerythrin solution were added into each well (to form a capture sandwich immunoassay); incubation was then performed in the dark for 30 min at room temperature. The well contents were again removed, and the plates were washed twice with 200 μL wash buffer. Finally, 150 μL of sheath fluid were added, and the plates were evaluated on the MAGPIX instrument with xPONENT software. Median fluorescence intensities of all cytokine/chemokine targets were analyzed to calculate the corresponding cytokine/chemokine concentrations in the urine samples.
This study was approved by the Institutional Review Board and Ethics Committee of Buddhist Tzu Chi General Hospital (IRB105-31-A). All patients with IC/BPS and controls were informed of the rationale and procedures of this study, and written informed consent was obtained from each participant.
Statistical analysis.
Continuous variables are presented as means ± SD, and categorical data are represented as numbers and percentages. For each targeted cytokine, the values of cytokines outside the range between means ± 3 SD in either IC/BPS or control groups were defined as outliers, which were excluded from further analysis. Among patients with IC/BPS, clinical data were compared between ESSIC types 1 and 2 using ANOVA. Differences in urine cytokines and chemokines between patients with IC/BPS and controls were also analyzed using ANOVA. The cytokines with the mean values in study groups below the minimum detectable concentrations as per the assay manufacturer were excluded for further analysis.
Receiver-operating characteristic (ROC) curves were generated for the ability of each cytokine to distinguish patients with ESSIC type 2 IC/BPS from controls, and the areas under the ROC curves (AUC) were calculated. Among patients with ESSIC type 2 IC/BPS, linear regression analysis with Pearson correlation was carried out to determine the relationship between clinical characteristics and urine cytokine levels. Multivariate logistic regression models for controlling the confounding factors were fit for each analyte, and odds ratios were calculated. All calculations were performed using SPSS Statistics for Windows (version 20.0, IBM, Armonk, NY). Differences were considered statistically significant if P values were <0.05.
RESULTS
Eligible patients with IC/BPS included 114 women and 13 men with a mean age of 54.6 ± 12.5 yr (range: 21–78 yr), which was similar to that of controls (mean age: 58.6 ± 9.9 yr, range: 39–75 yr; Table 1). Based on the findings of glomerulations in cystoscopic hydrodistention, 37 (29.1%) and 90 (70.9%) patients with IC/BPS were classified as ESSIC types 1 and 2, respectively. Both types of patients with IC/BPS had similar age and sex distributions, visual analog scale pain scores, and clinical symptom scores (interstitial cystitis symptom index, interstitial cystitis problem index, and O'Leary-Saint symptom score); however, patients with ESSIC type 2 IC/BPS had significantly lower body mass index and maximal bladder capacity under anesthesia (MBC).
| Patients With IC/BPS | ||||||
|---|---|---|---|---|---|---|
| ESSIC type 1 (n = 37, 29.1%) | ESSIC type 2 (n = 90, 70.9%) | Overall (n = 127, 100%) | Controls (n = 28) | P Value Between Patients With ESSIC Type 1 and Type 2 IC/BPS | P Value Between Overall Patients With IC/BPS and Controls | |
| Age | 57.0 ± 11.1 (28−78) | 53.6 ± 13.0 (21−76) | 54.6 ± 12.5 (21−78) | 58.6 ± 9.9 (39−75) | 0.174 | 0.119 |
| Sex | 35 women and 2 men | 79 women and 11 men | 114 women and 13 men | 28 women | 0.343 | 0.127 |
| Diabetes mellitus, n (%) | 7 (18.9) | 9 (10) | 16 (12.6) | 5 (17.9) | 0.237 | 0.541 |
| Body mass index | 25.5 ± 4.2 | 23.1 ± 4.1 | 23.8 ± 4.3 | 25.7 ± 4.0 | 0.003 | 0.039 |
| Visual analog scale pain score | 4.3 ± 2.7 | 4.2 ± 2.7 | 4.2 ± 2.7 | 0.893 | ||
| Interstitial cystitis symptom index | 9.8 ± 4.7 | 10.4 ± 4.4 | 10.2 ± 4.5 | 0.490 | ||
| Interstitial cystitis problem index | 10.4 ± 4.6 | 10.1 ± 3.7 | 10.2 ± 4.0 | 0.779 | ||
| O’Leary-Saint score | 20.1 ± 8.8 | 20.6 ± 7.5 | 20.5 ± 7.9 | 0.747 | ||
| Maximal bladder capacity under anesthesia, mL | 766.2 ± 194.4 | 688.9 ± 169.1 | 711.4 ± 179.5 | 0.027 | ||
Table 2 shows the targeted urine cytokine levels of patients with IC/BPS and controls. For each targeted cytokine, the numbers of outliers within IC/BPS and control groups ranged from 0 to 6 and from 0 to 1, respectively. Patients with IC/BPS had urine cytokine profiles distinct from those of controls; patients with ESSIC type 2 IC/BPS also had urine cytokine levels distinct from those of patients with ESSIC type 1 IC/BPS, which included MCP-1, IFN-γ, and IL-10.
| Patients With IC/BPS | ||||||
|---|---|---|---|---|---|---|
| Urine Cytokine | ESSIC type 1 (n = 37) | ESSIC type 2 (n = 90) | Overall (n = 127) | Control (n = 28) | P Value Between Patients With ESSIC Type 1 and Type 2 IC/BPS | P Value Between Overall Patients With IC/BPS and Controls |
| MCP-1 | 243.8 ± 214.4 (2)* | 366.3 ± 387.8 (2)* | 332.0 ± 351.6 (4)* | 142.3 ± 93.0 (1) | 0.027 | <0.001 |
| MIP-1α† | 1.39 ± 0.68 (2)* | 1.19 ± 0.66 (1)* | 1.24 ± 0.67 (3)* | 0.88 ± 0.59 (0) | 0.120 | 0.010 |
| MIP-1β | 3.03 ± 2.24 (1) | 3.51 ± 3.41 (2) | 3.37 ± 3.12 (3) | 2.44 ± 1.59 (0) | 0.439 | 0.128 |
| RANTES | 7.80 ± 6.63 (1) | 10.58 ± 8.40 (1)* | 9.79 ± 8.01 (2)* | 5.34 ± 4.56 (1) | 0.078 | <0.001 |
| Eotaxin-1 | 8.08 ± 7.89 (1) | 9.61 ± 8.12 (2)* | 9.17 ± 8.06 (3)* | 4.88 ± 3.21 (1) | 0.338 | <0.001 |
| G-CSF | 8.43 ± 9.09 (1) | 10.08 ± 15.73 (2) | 9.61 ± 14.14 (3) | 9.31 ± 9.95 (1) | 0.556 | 0.918 |
| GM-CSF† | 1.13 ± 0.44 (1) | 1.21 ± 0.43 (1) | 1.19 ± 0.44 (2) | 1.11 ± 0.30 (1) | 0.307 | 0.384 |
| VEGF† | 16.50 ± 6.64 (1)* | 13.51 ± 7.29 (1)* | 14.36 ± 7.21 (2)* | 10.61 ± 4.98 (0) | 0.035 | 0.002 |
| NGF | 0.36 ± 0.11 (0)* | 0.42 ± 0.32 (4)* | 0.40 ± 0.27 (4)* | 0.30 ± 0.06 (0) | 0.107 | <0.001 |
| EGF | 7.11 ± 4.36 (0) | 6.60 ± 4.77 (1) | 6.75 ± 4.64 (1) | 6.00 ± 5.20 (0) | 0.575 | 0.450 |
| CXCL10 | 28.81 ± 40.35 (1) | 63.16 ± 123.36 (1)* | 53.35 ± 107.39 (2) | 14.76 ± 18.63 (1) | 0.105 | 0.065 |
| IFN-α2 | 3.54 ± 1.70 (1) | 3.55 ± 2.05 (1) | 3.55 ± 1.95 (2) | 2.84 ± 1.40 (0) | 0.978 | 0.068 |
| IFN-γ | 1.34 ± 0.32 (1) | 1.17 ± 0.26 (1) | 1.22 ± 0.29 (2) | 1.22 ± 0.17 (0) | 0.006 | 0.921 |
| TNF-α | 0.78 ± 0.42 (1) | 0.79 ± 0.48 (2) | 0.79 ± 0.46 (3) | 0.74 ± 0.25 (1) | 0.864 | 0.617 |
| TNF-β† | 0.80 ± 0.16 (1) | 0.75 ± 0.14 (2) | 0.76 ± 0.15 (3) | 0.76 ± 0.10 (0) | 0.077 | 0.889 |
| IL-1α† | 1.48 ± 0.81 (0) | 1.60 ± 1.36 (1) | 1.56 ± 1.22 (1) | 1.40 ± 0.73 (1) | 0.617 | 0.514 |
| IL-1β† | 0.57 ± 0.24 (3) | 0.63 ± 0.54 (3) | 0.62 ± 0.47 (6) | 0.57 ± 0.27 (1) | 0.481 | 0.633 |
| IL-1RA | 384.1 ± 353.8 (0) | 566.6 ± 750.6 (2) | 513.4 ± 663.8 (2) | 343.1 ± 478.8 (1) | 0.160 | 0.208 |
| IL-2† | 0.69 ± 0.17 (0)* | 0.79 ± 0.18 (0) | 0.76 ± 0.18 (0)* | 0.85 ± 0.18 (0) | 0.002 | 0.016 |
| IL-3† | 0.46 ± 0.18 (0)* | 0.53 ± 0.18 (0) | 0.51 ± 0.18 (0)* | 0.61 ± 0.21 (0) | 0.037 | 0.011 |
| IL-4 | 12.18 ± 6.64 (0)* | 12.68 ± 9.38 (0)* | 12.53 ± 8.65 (0)* | 7.37 ± 4.76 (1) | 0.739 | <0.001 |
| IL-5† | 0.49 ± 0.12 (0) | 0.45 ± 0.16 (0) | 0.46 ± 0.15 (0) | 0.45 ± 0.06 (0) | 0.157 | 0.691 |
| IL-6 | 2.05 ± 2.10 (0) | 2.71 ± 5.59 (3) | 2.52 ± 4.84 (3) | 1.40 ± 1.36 (1) | 0.490 | 0.234 |
| IL-7 | 1.44 ± 0.53 (0) | 1.90 ± 2.14 (1) | 1.77 ± 1.84 (1) | 1.33 ± 0.65 (1) | 0.199 | 0.221 |
| IL-8 | 12.66 ± 14.03 (1) | 15.62 ± 22.10 (2) | 14.78 ± 20.12 (3) | 14.62 ± 24.13 (1) | 0.458 | 0.971 |
| IL-10 | 1.23 ± 0.38 (2)* | 0.99 ± 0.30 (1) | 1.06 ± 0.34 (3) | 1.03 ± 0.14 (1) | 0.002 | 0.527 |
| IL-12p40† | 1.02 ± 0.47 (0)* | 0.96 ± 0.38 (2)* | 0.98 ± 0.41 (2)* | 0.68 ± 0.31 (0) | 0.474 | <0.001 |
| IL-12p70 | 1.43 ± 0.52 (0) | 1.21 ± 0.90 (1) | 1.28 ± 0.81 (1) | 1.19 ± 0.21 (0) | 0.161 | 0.566 |
| IL-13† | 1.17 ± 0.31 (2) | 1.17 ± 0.33 (2) | 1.17 ± 0.32 (4) | 1.28 ± 0.35 (1) | 0.998 | 0.134 |
| IL-15 | 1.25 ± 0.45 (1) | 1.38 ± 0.69 (1) | 1.35 ± 0.63 (2) | 1.25 ± 0.36 (1) | 0.282 | 0.459 |
| IL-17A | 0.98 ± 0.35 (0) | 0.93 ± 0.78 (2) | 0.94 ± 0.68 (2) | 0.92 ± 0.21 (1) | 0.684 | 0.846 |
ROC curves were constructed showing the abilities of urine cytokines to distinguish patients with ESSIC type 2 IC/BPS from controls. Table 3 shows the diagnostic values of each urine cytokine, including AUC, cutoff value, sensitivity, specificity, positive predictive value, and negative predictive value. The cytokines that provided acceptable discrimination for use in the diagnostic test (i.e., AUC > 0.7) included MCP-1, RANTES, IL-7, CXCL10, MIP-1β, eotaxin-1, IL-1 receptor antagonist, and IL-8. Among the cytokines with high diagnostic values, RANTES, MIP-1β, and IL-8 had high sensitivity (>82%), while MCP-1, CXCL10, and eotaxin-1 had high specificity (>78%). Figure 1 shows the clustered display of urine cytokine levels useful for diagnosis of IC/BPS, based on respective cuffoff values. The color distribution of patients with ESSIC type 2 IC/BPS differed from that of controls. Figure 2 shows violin plots of significant cytokines with high diagnostic values of IC/BPS.
| Urine Cytokine | Area Under the Curve | Cutoff Value | Sensitivity, % | Specificity, % | Positive Predictive value, % | Negative Predictive Value, % |
|---|---|---|---|---|---|---|
| MCP-1 | 0.780 | 200.335 | 58.9 | 78.6 | 89.8 | 37.3 |
| RANTES | 0.767 | 3.445 | 82.2 | 53.6 | 85.1 | 48.4 |
| IL-7 | 0.756 | 1.285 | 71.1 | 67.9 | 87.7 | 42.2 |
| CXCL10 | 0.754 | 24.705 | 56.7 | 78.6 | 89.5 | 36.1 |
| MIP-1β | 0.751 | 1.57 | 85.6 | 50.0 | 84.6 | 51.9 |
| Eotaxin-1 | 0.720 | 7.055 | 52.2 | 85.7 | 92.2 | 35.8 |
| IL-1RA | 0.703 | 130.645 | 76.7 | 53.6 | 84.1 | 41.7 |
| IL-8 | 0.701 | 2.785 | 85.6 | 42.9 | 82.8 | 48.0 |
| IFN-α2 | 0.670 | 2.86 | 51.1 | 71.4 | 85.2 | 31.3 |
| IL-17A* | 0.661 | 0.765 | 44.4 | 85.7 | 90.9 | 32.4 |
| IL-6 | 0.660 | 1.185 | 46.7 | 82.1 | 89.4 | 32.4 |
| IL-10* | 0.637 | 0.865 | 38.9 | 92.9 | 94.6 | 32.1 |
| NGF | 0.615 | 0.295 | 62.2 | 60.7 | 83.6 | 33.3 |
| IL-4 | 0.609 | 17.47 | 37.8 | 96.4 | 97.1 | 32.5 |
| EGF | 0.589 | 3,708 | 58.9 | 53.6 | 80.3 | 28.8 |
| IFN-γ* | 0.583 | 1.09 | 46.7 | 78.6 | 87.5 | 31.4 |
| IL-15 | 0.575 | 1.61 | 18.9 | 92.9 | 89.5 | 26.3 |
| IL-12p70* | 0.557 | 0.935 | 25.6 | 96.4 | 95.8 | 28.7 |
| TNF-α | 0.527 | 0.765 | 41.1 | 71.4 | 82.2 | 27.4 |
| G-CSF | 0.515 | 3.245 | 75.6 | 42.9 | 81.0 | 35.3 |
Fig. 1.Clustered display of urine cytokines levels diagnosing interstitial cystitis/bladder pain syndrome (IC/BPS) by the respective cutoff values for European Society for the Study of Interstitial Cystitis (ESSIC) type 2 IC/BPS and controls. Each row represents one patient. From top to bottom, patients are listed by increasing age. Each column represents one cytokine target. From left to right, cytokines are listed by the decrease of the area under the curve. Red and green colors indicate that the diseased status was diagnosed by higher and lower than the cutoff value of the cytokine, respectively. MCP-1, macrophage chemoattractant protein-1; RANTES, regulated upon activation, normal T cell expressed and presumably secreted; IL, interleukin; IP-10, also known as chemokine (C-X-C motif) ligand 10 (CXCL10); MIP, macrophage inflammatory protein; IL-1RA, IL-1 receptor antagonist; VEGF, vascular endothelial growth factor; IFN, interferon; BDNF, brain-derived neurotrophic factor; GM-CSF, granulocyte-macrophage colony-stimulating factor; NGF, nerve growth factor; EGF, epidermal growth factor; TNF, tumor necrosis factor; G-CSF, granulocyte-colony stimulating factor.
Fig. 2.Violin plots of significant cytokines with high diagnostic values of interstitial cystitis/bladder pain syndrome (IC/BPS). A: urine levels of macrophage chemoattractant protein-1 (MCP-1; A) were significantly increased in patients with both European Society for the Study of Interstitial Cystitis (ESSIC) type 1 IC/BPS [IC (1)] and ESSIC type 2 IC/BPS [IC (2)] versus controls. B, C, and E: urine levels of regulated upon activation, normal T cell expressed and presumably secreted (RANTES; B), chemokine (C-X-C motif) ligand 10 (CXCL10; C), and eotaxin-1 (E) were significantly increased in patients with ESSIC type 2 IC/BPS versus controls. D and F: urine levels of macrophage inflammatory protein (MIP-1β; D) and IL-8 (F) were not significantly increased in patients with ESSIC type 1 or type 2 IC/BPS versus controls. IC(T), total IC/BPS.
Controlling for age, sex, body mass index, and the disease of diabetes mellitus, multivariate logistic regression models revealed that numerous urine cytokines remained statistically significant in differentiating IC/BPS and controls, including those with high diagnostic values [MCP-1 (odds ratio: 2.121), CXCL10 (odds ratio: 1.262), eotaxin-1 (odds ratio: 1.146), RANTES (odds ratio: 1.140),and IL-7 (odds ratio: 1.098)], and another cytokine [nerve growth factor (odds ratio: 1.066)] (Table 4). Urine cytokines not only have the potentials to differentiate IC/BPS from controls but also to discriminate ESSIC type 2 IC/BPS from ESSIC type 1 IC/BPS.
| P Value | Odds Ratio | 95% Confidence Interval | Odds Ratio Units, pg/mL | |
|---|---|---|---|---|
| IC/BPS (total) vs. control | ||||
| MCP-1 | 0.002 | 2.121 | 1.037−1.224 | 100 |
| CXCL10 | 0.016 | 1.262 | 1.043−1.527 | 10 |
| Eotaxin-1 | 0.015 | 1.146 | 1.027−1.279 | 1 |
| RANTES | 0.016 | 1.140 | 1.025−1.268 | 1 |
| IL-7 | 0.041 | 1.098 | 1.004−1.202 | 0.1 |
| NGF | 0.024 | 1.066 | 1.008−1.127 | 0.01 |
| IC/BPS (ESSIC type 2) vs. control | ||||
| MCP-1 | 0.002 | 2.423 | 1.401−4.192 | 100 |
| CXCL10 | 0.010 | 1.303 | 1.066−1.593 | 10 |
| Eotaxin-1 | 0.010 | 1.169 | 1.038−1.316 | 1 |
| RANTES | 0.013 | 1.154 | 1.031−1.291 | 1 |
| IL-7 | 0.028 | 1.113 | 1.011−1.225 | 0.1 |
| NGF | 0.041 | 1.060 | 1.002−1.120 | 0.01 |
| IC/BPS (ESSIC type 2) vs. IC/BPS (ESSIC type 1) | ||||
| IFN-γ | 0.004 | 0.810 | 0.702−0.934 | 0.1 |
| IL-12p70 | 0.001 | 0.810 | 0.719−0.912 | 0.1 |
| IL-10 | 0.001 | 0.823 | 0.729−0.928 | 0.1 |
| IL-17A | 0.028 | 0.980 | 0.962−0.998 | 0.01 |
Table 5 shows the significant correlations between urine cytokine levels and the clinical characteristics of patients with ESSIC type 2 IC/BPS, based on linear regression analysis. Among the cytokines with high diagnostic values, MCP-1, eotaxin-1, CXCL10 (high specificity), and RANTES (high sensitivity) were positively correlated with glomerulation grade and negatively correlated with MBC.
| Urine Cytokine | Grade of Glomerulation | Maximal Bladder Capacity Under Anesthesia | Visual Analog Scale Pain Score | Interstitial Cystitis Symptom Index | Interstitial Cystitis Problem Index | O’Leary-Saint Score |
|---|---|---|---|---|---|---|
| MCP-1 | +0.215 | −0.253 | NS | NS | NS | NS |
| MIP-1β | NS | NS | NS | NS | NS | NS |
| RANTES | +0.247 | −0.258 | NS | NS | NS | NS |
| Eotaxin | +0.190 | −0.282 | NS | NS | NS | NS |
| G-CSF | NS | NS | NS | NS | NS | NS |
| NGF | NS | NS | NS | NS | −0.206 | −0.197 |
| EGF | NS | NS | NS | NS | NS | NS |
| CXCL10 | +0.258 | −0.249 | NS | NS | NS | NS |
| IFN-α2 | NS | NS | NS | NS | NS | NS |
| IFN-γ | −0.222 | NS | NS | NS | NS | NS |
| TNF-α | NS | NS | NS | NS | NS | NS |
| IL-1RA | NS | NS | NS | NS | NS | NS |
| IL-4 | NS | NS | −0.242 | −0.288 | −0.349 | −0.348 |
| IL-6 | NS | NS | NS | NS | NS | NS |
| IL-7 | NS | NS | NS | NS | NS | NS |
| IL-8 | NS | NS | NS | NS | −0.209 | −0.189 |
| IL-10 | −0.296 | NS | NS | −0.198 | −0.225 | −0.229 |
| IL-12p70 | −0.275 | NS | NS | NS | NS | NS |
| IL-15 | NS | NS | NS | NS | NS | NS |
| IL-17A | −0.263 | NS | NS | NS | NS | NS |
DISCUSSION
To the best of our knowledge, this clinical study has the largest number of patients with IC/BPS and investigates the greatest number of potential targets. This study assessed the diagnostic values of multiple urine cytokines in patients with ESSIC type 2 IC/BPS and determined which have the high sensitivity (RANTES, MIP-1β, and IL-8) and which have the high specificity (MCP-1, CXCL10, and eotaxin-1). Multivariate models for controlling confounding factors revealed that numerous urine cytokines remained statistically significant in differentiating IC/BPS and controls. Significant correlations between urine cytokine levels and clinical characteristics of patients with ESSIC type 2 IC/BPS (including symptom scores, pain severity in visual analog scale, glomerulation grade, and MBC) were demonstrated. This noninvasive approach with urine cytokine analysis provides important information regarding the pathological bladder condition in patients with IC/BPS. Low bladder capacity is an important marker for a bladder centric manifestation of IC/BPS (31), and the patients with lower bladder capacity have significant different molecular characteristics related to inflammatory and immune responses (10). Urine cytokines have important roles in the diagnosis and mapping of the clinical characteristics of IC/BPS, with the potential to serve as novel biomarkers.
MCP-1, also referred to as chemokine (C-C motif) ligand (CCL)2, is a member of the chemokine (C-C motif) family and is a potent chemotactic factor for monocytes. Many cell types, including endothelial cells, epithelial cells, fibroblasts, and smooth muscle cells, can produce MCP-1 (11). MCP-1 is reportedly involved in many inflammatory diseases (17, 25), including inflammatory bowel disease, allergic asthma, and rheumatoid arthritis, which are also associated with IC/BPS (2, 9).
CXCL10, also referred to as IFN-γ-induced protein 10, is a member of the chemokine (C-X-C motif) family and is a chemotactic factor for monocytes/macrophages and activated T cells (12). CXCL10 is secreted by monocytes, endothelial cells, and fibroblasts in response to IFN-γ and is involved in afferent sensitization (4). In patients with IC/BPS, increased serum CXCL10 levels have been reported (23). Additionally, expression levels of CXCL10 in bladder tissue have been shown to increase in mice with cyclophosphamide-induced cystitis and subsequently decrease after CXCL10 antibody treatment is provided (23). CXCL10 therefore appears to play an important role in the molecular mechanism of IC/BPS and may be a useful treatment target, as suggested by its upregulation in patients with IC/BPS.
RANTES, also known as CCL5, is a member of the chemokine (C-C motif) family that is a chemotactic factor for T cells, eosinophils, and basophils; it plays an active role in recruiting leukocytes into inflammatory sites (21). A central role has been identified for RANTES in the pathophysiology of neurogenic cystitis (8). In neurogenic cystitis, TNF signaling mediates RANTES expression in the urothelium, which directs mast cell chemotaxis resulting in mast cell microlocalization juxtaposed with the urothelium. Additionally, anti-RANTES antibodies can block mast cell trafficking and stabilize bladder barrier function. These indicate that RANTES might be an important target in the pathophysiology and treatment of IC/BPS.
MCP-1, CXCL10, and RANTES are known to be upregulated and involved in chemokine signaling in peripheral neuroinflammatory responses (4). They are suspected to play a central role in the maintenance of afferent hypersensitivity and neuropathic pain, similar to the pathophysiology of IC/BPS. In a small study of human clinical cases, the urine CXCL10 level was shown to be significantly higher in patients with ulcerative IC/BPS but not in patients with nonulcerative IC/BPS compared with controls (29). In addition, urine MCP-1 and RANTES levels tended to be higher in patients with nonulcerative IC/BPS than in controls; however, this difference was not statistically significant. In our study, these cytokines were increased in patients with ESSIC type 2 IC/BPS, which suggests the presence of neuropathic inflammation within the bladder. Positive correlations with glomerulation grade and negative correlations with MBC were also noted for these cytokines. More severe inflammation and afferent hypersensitivity within the urinary bladder might result in smaller bladder capacity and greater clinical severity of glomerulation. These cytokines can diagnose IC/BPS and can provide important information regarding the clinical characteristics of IC/BPS. In contrast to invasive procedures, including bladder biopsy and cystoscopic hydrodistention, the analysis of urine proteins is a feasible and noninvasive method for assessment of pathological conditions within the bladder.
Eotaxin-1, also known as CCL11, is a member of the chemokine (C-C motif) family that acts as a selective chemoattractant for eosinophils (14). It has been implicated in many eosinophilic inflammatory diseases in the respiratory tract, gastrointestinal tract, and skin, such as asthma, allergic rhinitis, inflammatory bowel disease, and atopic dermatitis (1, 3). Eotaxin-1 has been reported as a potential urinary biomarker for IC/BPS (18) but is not well studied. In the present study, eotaxin-1 had a high specificity for diagnosis of ESSIC type 2 IC/BPS as well as a positive correlation with glomerulation grade and negative correlation with MBC. These findings suggested that eotaxin-1-associated inflammation plays an important role in the pathophysiology of IC/BPS.
MIP-1β, also known as CCL4, is a member of the chemokine (C-C motif) family and is a chemoattractant for natural killer cells, monocytes, and a variety of other immune cells (7). MIP-1β was reportedly elevated in urine samples from patients with overactive bladder and was presumed to be related to bladder inflammation (28). However, the specific roles of MIP-1β in the pathogenesis of overactive bladder and IC/BPS have been uncertain. In addition, IL-8 is a chemoattractant of neutrophils and T cells (27), which can regulate angiogenesis through direct enhancement of the survival and proliferation of endothelial cells (19). IL-8 is also essential for normal urothelial growth and survival (27). Higher expression of IL-8 in urine (13) but lower in bladder tissue (27) was noted in patients with IC/BPS. Additional evidence is needed to elucidate the specific expression patterns of IL-8 in patients with IC/BPS. In this study, both MIP-1β and IL-8 had high sensitivity for diagnosis of patients with ESSIC type 2 IC/BPS but did not show significant correlations with the clinical characteristics of IC/BPS.
There were several limitations in this study. First, intraindividual variation may have contributed to bias within the data. There are no clear data regarding changes in urine cytokine expression levels under different bladder conditions (i.e., a state of urgency or nonurgency) within the same individual. Second, systemic inflammatory diseases and comorbidities (e.g., diabetes, cardiovascular disease, metabolic syndrome, and aging) might affect the bladder and related urine conditions. Other proinflammatory conditions may confound the results. Third, most study patients and all controls were women, so the findings might not be generalizable to men. Finally, the expression levels of some proteins in urine exhibited extreme values, and such outliers were excluded from analysis in this study, although the number of outliers was small.
In conclusion, compared with controls, patients with IC/BPS had distinct urine cytokine profiles, which could reflect clinical bladder conditions such as glomerulation and MBC. The use of urine cytokines as biomarkers might facilitate the diagnosis and mapping of the clinical characteristics of IC/BPS.
GRANTS
This work was supported by Buddhist Tzu Chi Medical Foundation Grant TCMMP105-02-03.
DISCLOSURES
No conflicts of interest, financial or otherwise, are declared by the author(s).
AUTHOR CONTRIBUTIONS
H.-C.K. conceived and designed research; H.-C.H. and Y.-H.W. performed experiments; Y.-H.J., J.-F.J., and Y.-H.H. analyzed data; Y.-H.J. and J.-F.J. interpreted results of experiments; Y.-H.W. prepared figures; Y.-H.J. drafted manuscript; Y.-H.J., Y.-H.H., and H.-C.H. edited and revised manuscript; J.-F.J. and H.-C.K. approved final version of manuscript.
REFERENCES
- 1. . From airway inflammation to inflammatory bowel disease: eotaxin-1, a key regulator of intestinal inflammation. Clin Immunol 153: 199–208, 2014. doi:10.1016/j.clim.2014.04.012.
Crossref | PubMed | ISI | Google Scholar - 2. . Interstitial cystitis: unexplained associations with other chronic disease and pain syndromes. Urology 49, Suppl: 52–57, 1997. doi:10.1016/S0090-4295(99)80332-X.
Crossref | PubMed | ISI | Google Scholar - 3. . Eotaxins and CCR3 receptor in inflammatory and allergic skin diseases: therapeutical implications. Curr Drug Targets Inflamm Allergy 2: 81–94, 2003. doi:10.2174/1568010033344480.
Crossref | PubMed | Google Scholar - 4. . Delayed functional expression of neuronal chemokine receptors following focal nerve demyelination in the rat: a mechanism for the development of chronic sensitization of peripheral nociceptors. Mol Pain 3: 38, 2007. doi:10.1186/1744-8069-3-38.
Crossref | PubMed | ISI | Google Scholar - 5. . Gene expression analysis of urine sediment: evaluation for potential noninvasive markers of interstitial cystitis/bladder pain syndrome. J Urol 187: 725–732, 2012. doi:10.1016/j.juro.2011.09.142.
Crossref | PubMed | ISI | Google Scholar - 6. . Monocyte chemoattractant protein-1 production by human detrusor smooth muscle cells. J Urol 171: 462–466, 2004. doi:10.1097/01.ju.0000090192.36436.d5.
Crossref | PubMed | ISI | Google Scholar - 7. . B cells and professional APCs recruit regulatory T cells via CCL4. Nat Immunol 2: 1126–1132, 2001. doi:10.1038/ni735.
Crossref | PubMed | ISI | Google Scholar - 8. . RANTES mediates TNF-dependent lamina propria mast cell accumulation and barrier dysfunction in neurogenic cystitis. Am J Physiol Renal Physiol 292: F1372–F1379, 2007. doi:10.1152/ajprenal.00472.2006.
Link | ISI | Google Scholar - 9. . Bladder pain syndrome/interstitial cystitis is associated with asthma: A case-control study. Neurourol Urodyn 37: 1773–1778, 2018. doi:10.1002/nau.23520.
Crossref | PubMed | ISI | Google Scholar - 10. . Correlation of gene expression with bladder capacity in interstitial cystitis/bladder pain syndrome. J Urol 192: 1123–1129, 2014. doi:10.1016/j.juro.2014.05.047.
Crossref | PubMed | ISI | Google Scholar - 11. . Monocyte chemoattractant protein-1 (MCP-1): an overview. J Interferon Cytokine Res 29: 313–326, 2009. doi:10.1089/jir.2008.0027.
Crossref | PubMed | ISI | Google Scholar - 12. . IFN-gamma-inducible protein 10 (IP-10; CXCL10)-deficient mice reveal a role for IP-10 in effector T cell generation and trafficking. J Immunol 168: 3195–3204, 2002. doi:10.4049/jimmunol.168.7.3195.
Crossref | PubMed | ISI | Google Scholar - 13. . Do the National Institute of Diabetes and Digestive and Kidney Diseases cystoscopic criteria associate with other clinical and objective features of interstitial cystitis? J Urol 173: 93–97, 2005. doi:10.1097/01.ju.0000146466.71311.ab.
Crossref | PubMed | ISI | Google Scholar - 14. . Human eotaxin is a specific chemoattractant for eosinophil cells and provides a new mechanism to explain tissue eosinophilia. Nat Med 2: 449–456, 1996. doi:10.1038/nm0496-449.
Crossref | PubMed | ISI | Google Scholar - 15. . Summary of the National Institute of Arthritis, Diabetes, Digestive and Kidney Diseases Workshop on Interstitial Cystitis, National Institutes of Health, Bethesda, Maryland, August 28-29, 1987. J Urol 140: 203–206, 1988. doi:10.1016/S0022-5347(17)41529-1.
Crossref | PubMed | ISI | Google Scholar - 16. . Clinical guidelines for interstitial cystitis and hypersensitive bladder syndrome. Int J Urol 16: 597–615, 2009. doi:10.1111/j.1442-2042.2009.02326.x.
Crossref | PubMed | ISI | Google Scholar - 17. . Interleukin (IL)-4 and IL-13 up-regulate monocyte chemoattractant protein-1 expression in human bronchial epithelial cells: involvement of p38 mitogen-activated protein kinase, extracellular signal-regulated kinase 1/2 and Janus kinase-2 but not c-Jun NH2-terminal kinase 1/2 signalling pathways. Clin Exp Immunol 145: 162–172, 2006. doi:10.1111/j.1365-2249.2006.03085.x.
Crossref | PubMed | ISI | Google Scholar - 18. . Potential urine and serum biomarkers for patients with bladder pain syndrome/interstitial cystitis. Int J Urol 21, Suppl 1: 34–41, 2014. doi:10.1111/iju.12311.
Crossref | PubMed | ISI | Google Scholar - 19. . IL-8 directly enhanced endothelial cell survival, proliferation, and matrix metalloproteinases production and regulated angiogenesis. J Immunol 170: 3369–3376, 2003. doi:10.4049/jimmunol.170.6.3369.
Crossref | PubMed | ISI | Google Scholar - 20. . Guideline of guidelines: bladder pain syndrome. BJU Int 122: 729–743, 2018. doi:10.1111/bju.14399.
Crossref | PubMed | ISI | Google Scholar - 21. . Targeting CCL5 in inflammation. Expert Opin Ther Targets 17: 1439–1460, 2013. doi:10.1517/14728222.2013.837886.
Crossref | PubMed | ISI | Google Scholar - 22. . Clinical phenotyping of women with interstitial cystitis/painful bladder syndrome: a key to classification and potentially improved management. J Urol 182: 155–160, 2009. doi:10.1016/j.juro.2009.02.122.
Crossref | PubMed | ISI | Google Scholar - 23. . CXCL10 blockade protects mice from cyclophosphamide-induced cystitis. J Immune Based Ther Vaccines 6: 6, 2008. doi:10.1186/1476-8518-6-6.
Crossref | PubMed | Google Scholar - 24. . Higher levels of cell apoptosis and abnormal E-cadherin expression in the urothelium are associated with inflammation in patients with interstitial cystitis/painful bladder syndrome. BJU Int 108, 2b: E136–E141, 2011. doi:10.1111/j.1464-410X.2010.09911.x.
Crossref | PubMed | ISI | Google Scholar - 25. . Monocyte chemoattractant protein-1 (MCP-1) inhibits the intestinal-like differentiation of monocytes. Clin Exp Immunol 145: 190–199, 2006. doi:10.1111/j.1365-2249.2006.03113.x.
Crossref | PubMed | ISI | Google Scholar - 26. . Mapping of pain phenotypes in female patients with bladder pain syndrome/interstitial cystitis and controls. Eur Urol 62: 1188–1194, 2012. doi:10.1016/j.eururo.2012.05.023.
Crossref | PubMed | ISI | Google Scholar - 27. . Interleukin-8 is essential for normal urothelial cell survival. Am J Physiol Renal Physiol 297: F816–F821, 2009. doi:10.1152/ajprenal.90733.2008.
Link | ISI | Google Scholar - 28. . Urine cytokines suggest an inflammatory response in the overactive bladder: a pilot study. Int Urol Nephrol 42: 629–635, 2010. doi:10.1007/s11255-009-9647-5.
Crossref | PubMed | ISI | Google Scholar - 29. . Urinary chemokines as noninvasive predictors of ulcerative interstitial cystitis. J Urol 187: 2243–2248, 2012. doi:10.1016/j.juro.2012.01.034.
Crossref | PubMed | ISI | Google Scholar - 30. . Diagnostic criteria, classification, and nomenclature for painful bladder syndrome/interstitial cystitis: an ESSIC proposal. Eur Urol 53: 60–67, 2008. doi:10.1016/j.eururo.2007.09.019.
Crossref | PubMed | ISI | Google Scholar - 31. . Bladder capacity is a biomarker for a bladder centric versus systemic manifestation in interstitial cystitis/bladder pain syndrome. J Urol 198: 369–375, 2017. doi:10.1016/j.juro.2017.02.022.
Crossref | PubMed | ISI | Google Scholar - 32. . Bladder afferent hyperexcitability in bladder pain syndrome/interstitial cystitis. Int J Urol 21, Suppl 1: 18–25, 2014. doi:10.1111/iju.12308.
Crossref | PubMed | ISI | Google Scholar
