Clinical characterization of interstitial cystitis/bladder pain syndrome in women based on the presence or absence of Hunner lesions and glomerulations

Risk Classification for Interstitial Cystitis/Bladder Pain Syndrome Using Machine Learning Based Predictions

OBJECTIVE

To improve diagnosis of interstitial cystitis (IC)/bladder pain syndrome(IC) we hereby developed an improved IC risk classification using machine learning algorithms.

METHODS

A national crowdsourcing resulted in 1264 urine samples consisting of 536 IC (513 female, 21 male, 2 unspecified), and 728 age-matched controls (318 female, 402 male, 8 unspecified) with corresponding patient-reported outcome (PRO) pain and symptom scores. In addition, 296 urine samples were collected at three academic centers: 78 IC (71 female, 7 male) and 218 controls (148 female, 68 male, 2 unspecified). Urinary cytokine biomarker levels were determined using Luminex assay. A machine learning predictive classification model, termed the Interstitial Cystitis Personalized Inflammation Symptom (IC-PIS) Score, that utilizes PRO and cytokine levels, was generated and compared to a challenger model.

RESULTS

The top-performing model using biomarker measurements and PROs (area under the curve [AUC]=0.87) was a support vector classifier, which scored better at predicting IC than PROs alone (AUC=0.83). While biomarkers alone (AUC=0.58) did not exhibit strong predictive performance, their combination with PROs produced an improved predictive effect.

CONCLUSION

IC-PIS represents a novel classification model designed to enhance the diagnostic accuracy of IC/bladder pain syndrome by integrating PROs and urine biomarkers. The innovative approach to sample collection logistics, coupled with one of the largest crowdsourced biomarker development studies utilizing ambient shipping methods across the US, underscores the robustness and scalability of our findings.

Progress and challenges in intravesical drug delivery

INTRODUCTION

Intravesical drug delivery (IDD) has gained recognition as a viable approach for treating bladder-related diseases over the years. However, it comes with its set of challenges, including voiding difficulties and limitations in mucosal and epithelial penetration. These challenges lead to drug dilution and clearance, resulting in poor efficacy. Various strategies for drug delivery have been devised to overcome these issues, all aimed at optimizing drug delivery. Nevertheless, there has been minimal translation to clinical settings.

AREAS COVERED

This review provides a detailed description of IDD, including its history, advantages, and challenges. It also explores the physical barriers encountered in IDD, such as voiding, mucosal penetration, and epithelial penetration, and discusses current strategies for overcoming these challenges. Additionally, it offers a comprehensive roadmap for advancing IDD into clinical trials.

EXPERT OPINION

Physical bladder barriers and limitations of conventional treatments result in unsatisfactory efficacy against bladder diseases. Nevertheless, substantial recent efforts in this field have led to significant progress in overcoming these challenges and have raised important attributes for an optimal IDD system. However, there is still a lack of well-defined steps in the workflow to optimize the IDD system for clinical settings, and further research is required to establish more comprehensive in vitro and in vivo models to expedite clinical translation.

Cystoscopic characteristic findings of interstitial cystitis and clinical implications

Interstitial cystitis/bladder pain syndrome (IC/BPS) is a chronic inflammatory bladder disease of unknown etiology, characterized by bladder pain and frequency urgency symptoms. Based on the cystoscopic findings after hydrodistention under anesthesia, the phenotype of IC/BPS includes no glamerulation, characteristic glomerulation, and with Hunner's lesion. IC is specifically defined if there are characteristic Hunner's lesion appeared in cystoscopy or after hydrodistention. If there are glomerulations without Hunner's lesion, BPS should be considered. The definition of Hunner's lesion and glomerulations differs based on different definition and observations. Currently, there has been no clear description and grading of the glomerulations and Hunner's lesion. Because the classification of IC/BPS has an impact on the treatment strategy and associated with therapeutic outcome, it is unmet to have a clear definition and consensus on the characteristic cystoscopic findings of IC/BPS. This article reviews the literature and presents the figures of Hunner's lesions and description of different mucosal lesions after cystoscopic hydrodistention.

Construction and evaluation of a column chart model and a random forest model for predicting the prognosis of hydrodistention surgery in BPS/IC patients based on preoperative CD117, P2X3R, NGF, and TrkA levels

OBJECTIVE

This study seeks to investigate independent risk factors affecting the prognoses of patients with bladder pain syndrome/interstitial cystitis (BPS/IC) following hydrodistention surgery and to develop a column chart model and a random forest model to help predict clinical outcomes.

METHOD

A retrospective analysis was conducted on the clinical data of 1006 BPS/IC patients who visited the urology department of the Fifth Hospital of Shanxi Medical University (Shanxi Provincial People's Hospital) between June 2012 and June 2022. The patients were randomly divided into a model group (n = 704) and a validation group (n = 302). In the model group, logistic regression analysis was used to identify independent risk factors, which were used to construct a prognostic nomogram. The nomogram was evaluated by analyzing the area under the curve (AUC), calibration curve, and decision curve. These results were subsequently validated via consistency analysis (n = 302). And based on the random forest algorithm, we calculate the same data and construct a random forest model.

RESULT

Multivariate logistic regression analysis revealed that age and the expression of the biomarkers CD117, P2X3R, NGF, and TrkA were independent prognostic factors for patients with BPS/IC (P < 0.05). Using these five indicators, a nomogram was developed to predict the risk factors for BPS/IC (scores ranged from 0 to 400). Based on the indicators, the nomogram demonstrated good prognostic performance (AUC = 0.982 and 95% confidence interva is 0.960-0.100). The correction curve indicated a high level of differentiation in the model, and the decision curve suggested positive clinical benefits. The random forest model has high accuracy and good calibration in predicting the prognosis of patients with interstitial cystitis after hydrodistention surgery.

CONCLUSION

Age, CD117, P2X3R, NGF, and TrkA are independent prognostic factors for bladder pain syndrome/interstitial cystitis. The column chart model and random forest model constructed based on these indicators have good predictive performance for patient prognosis.

Genome-wide association study identifies risk loci within the major histocompatibility complex region for Hunner-type interstitial cystitis

Hunner-type interstitial cystitis (HIC) is a rare, chronic inflammatory disease of the urinary bladder with unknown etiology and genetic background. Here, we conduct a genome-wide association study of 144 patients with HIC and 41,516 controls of Japanese ancestry. The genetic variant, rs1794275, in the major histocompatibility complex (MHC) region (chromosome 6p21.3) is associated with HIC risk (odds ratio [OR] = 2.32; p = 3.4 × 10^-9). The association is confirmed in a replication set of 26 cases and 1,026 controls (p = 0.014). Fine mapping demonstrates the contribution to the disease risk of a completely linked haplotype of three human leukocyte antigen HLA-DQβ1 amino acid positions, 71, 74, and 75 (OR = 1.94; p = 5 × 10^-8) and of HLA-DPβ1 amino acid position 178, which tags HLA-DPB1∗04:02 (OR = 2.35; p = 7.5 × 10^-8). The three HLA-DQβ1 amino acid positions are located together at the peptide binding groove, suggesting their functional importance in antigen presentation. Our study reveals genetic contributions to HIC risk that may be associated with class II MHC molecule antigen presentation.

Deep Learning Models for Cystoscopic Recognition of Hunner Lesion in Interstitial Cystitis

Background

Accurate cystoscopic recognition of Hunner lesions (HLs) is indispensable for better treatment prognosis in managing patients with Hunner-type interstitial cystitis (HIC), but frequently challenging due to its varying appearance.

Objective

To develop a deep learning (DL) system for cystoscopic recognition of a HL using artificial intelligence (AI).

Design setting and participants

A total of 626 cystoscopic images collected from January 8, 2019 to December 24, 2020, consisting of 360 images of HLs from 41 patients with HIC and 266 images of flat reddish mucosal lesions resembling HLs from 41 control patients including those with bladder cancer and other chronic cystitis, were used to create a dataset with an 8:2 ratio of training images and test images for transfer learning and external validation, respectively. AI-based five DL models were constructed, using a pretrained convolutional neural network model that was retrained to output 1 for a HL and 0 for control. A five-fold cross-validation method was applied for internal validation.

Outcome measurements and statistical analysis

True- and false-positive rates were plotted as a receiver operating curve when the threshold changed from 0 to 1. Accuracy, sensitivity, and specificity were evaluated at a threshold of 0.5. Diagnostic performance of the models was compared with that of urologists as a reader study.

Results and limitations

The mean area under the curve of the models reached 0.919, with mean sensitivity of 81.9% and specificity of 85.2% in the test dataset. In the reader study, the mean accuracy, sensitivity, and specificity were, respectively, 83.0%, 80.4%, and 85.6% for the models, and 62.4%, 79.6%, and 45.2% for expert urologists. Limitations include the diagnostic nature of a HL as warranted assertibility.

Conclusions

We constructed the first DL system that recognizes HLs with accuracy exceeding that of humans. This AI-driven system assists physicians with proper cystoscopic recognition of a HL.

Patient summary

In this diagnostic study, we developed a deep learning system for cystoscopic recognition of Hunner lesions in patients with interstitial cystitis. The mean area under the curve of the constructed system reached 0.919 with mean sensitivity of 81.9% and specificity of 85.2%, demonstrating diagnostic accuracy exceeding that of human expert urologists in detecting Hunner lesions. This deep learning system assists physicians with proper diagnosis of a Hunner lesion.

Can We Use Urinary Cytokine/Chemokine Analysis in Discriminating Ulcer-Type Interstitial Cystitis/Bladder Pain Syndrome?

Purpose: Interstitial cystitis/bladder pain syndrome (IC/BPS) has ulcer (HIC) and non-ulcer subtypes. Differentiation of these two subtypes could only be based by cystoscopy. This study analyzed the urinary cytokines and chemokines among IC/BPS subtypes and controls for discriminating HIC from non-HIC and controls. Materials and Methods: A total of 309 consecutive patients with clinically diagnosed IC/BPS were enrolled. All patients received cystoscopic hydrodistention under anesthesia and urine samples were collected prior to the procedure. Enrolled patients were classified into subtypes based on the glomerulation grade, maximal bladder capacity (MBC), and presence of Hunner’s lesion. Inflammation-related cytokines and chemokines in urine samples, including interleukin-8 (IL-8), C-X-C motif chemokine ligand 10 (CXCL10), monocyte chemoattractant protein-1 (MCP-1), brain-derived neurotrophic factor (BDNF), eotaxin-1 (eotaxin), IL-6, macrophage inflammatory protein-1 beta (MIP-1β), regulated upon activation, normally T-expressed, and presumably secreted (RANTES), tumor necrosis factor-alpha (TNF-α), and prostaglandin E2 (PGE2) were assayed using commercially available microspheres with the Milliplex® Human Cytokine/Chemokine Magnetic Bead-based Panel kit. The clinical data and urine levels of analytes between IC/BPS patients and controls, and among HIC, non-HIC, and controls were analyzed. Results: Among the 10 proteins, MCP-1, eotaxin, MIP-1β, TNF-α, and PGE2 were significantly different between IC/BPS and control, while IL-8, CXCL10, BDNF, IL-6, and RANTES were significantly higher in HIC than non-HIC patients. The receiver operating characteristic curve was used to analyze each urine biomarker in the patients with IC/BPS and controls. Among the 10 urine biomarkers, MIP-1β and TNF-α had an area under curve of >0.70 to predict IC/BPS from controls, however, the predictive values of these urine biomarkers to predict HIC from non-HIC were low. Combined cut-off values of MIP-1β and TNF-α can only have a 50% sensitivity and 39.6% specificity in identifying HIC from non-HIC. Conclusion: The results of this study demonstrate that urine cytokines and chemokines may be useful to discriminate patients with HIC from controls. An elevation of urine levels of IL-8, CXCL 10, BDNF, IL-6, and RANTES in IC/BPS patients should prompt physicians to consider the diagnosis of HIC.

Supervised machine learning algorithm identified KRT20, BATF and TP63 as biologically relevant biomarkers for bladder biopsy specimens from interstitial cystitis/bladder pain syndrome patients

OBJECTIVES

This study was carried out to identify biomarkers that distinguish Hunner-type interstitial cystitis from non-Hunner-type interstitial cystitis patients.

METHODS

Total ribonucleic acid was purified from 212 punch biopsy specimens of 89 individuals who were diagnosed as interstitial cystitis/bladder pain syndrome. To examine the expression profile of patients' bladder specimens, 68 urothelial master transcription factors and nine known markers (E-cadherin, cytokeratins, uroplakins and sonic hedgehog) were selected. To classify the biopsy samples, principal component analysis was carried out. A decision tree algorithm was adopted to identify critical determinants, in which 102 and 116 bladder specimens were used for learning and validation, respectively.

RESULTS

Principal component analysis segregated tissues from Hunner-type and non-Hunner-type interstitial cystitis specimens in principal component axes 2 and 4. Principal components 2 and 4 contained urothelial stem/progenitor transcription factors and cytokeratins, respectively. A decision tree identified KRT20, BATF and TP63 to classify non-Hunner-type and Hunner-type interstitial cystitis specimens. KRT20 was lower in tissues from Hunner-type compared with non-Hunner-type interstitial cystitis specimens (P < 0.001). TP63 was lower in Hunner's lesions compared with adjacent mucosa from Hunner-type interstitial cystitis patients (P < 0.001). Blinded validation using additional biopsy specimens verified that the decision tree showed fairly precise concordance with cystoscopic diagnosis.

CONCLUSION

KRT20, BATF and TP63 were identified as biologically relevant biomarkers to classify tissues from interstitial cystitis/bladder pain syndrome specimens. The biologically explainable determinants could contribute to defining the elusive interstitial cystitis/bladder pain syndrome pathogenesis.

[Interstitial cystitis: the latest findings on its aetiopathogenesis]

New findings provide progress in the understanding of the complicated aetiopathogenesis of interstitial cystitis/bladder pain syndrome (IC/BPS), whose causalities have only been deciphered in fragments so far. An increasingly complex network of pathomechanisms is emerging, in which the frequently mentioned mast cells and urothelial changes seem to be only a fragment of the pathological changes. The latest findings regarding a possible genetic and epigenetic predisposition are based on pedigree analyses, detection of single nucleotide polymorphisms and significant changes in differentially expressed genes. Multiple alterations can be detected at the molecular level. Platelet-activating factor, VEGF, corticotropin-releasing hormone and the inflammasome are important players in understanding the disease, but the pathomechanism underlying the "activation" of IC remains unclear. New starting points could be the detection of viruses (Epstein-Barr virus, BK polyomaviruses) or bacterial inflammation by pathogens that cannot be detected in standard cultures.

Relationship between the frequency of electrocautery of Hunner lesions and changes in bladder capacity in patients with Hunner type interstitial cystitis

Electrocautery is a promising treatment option for patients with Hunner type interstitial cystitis (HIC), but frequently requires multiple sessions due to recurrence of the lesions. In the present study, we assessed the relationship between the frequency of electrocautery of Hunner lesions and changes in maximum bladder capacity (MBC) at hydrodistension in a large cohort of 118 HIC patients. Three mixed-effect linear regression analyses were conducted for MBC against (1) the number of sessions; (2) the number of sessions and the time between each session and the first session; and (3) other relevant clinical parameters in addition to the Model (2). The mean number of sessions was 2.8 times. MBC decreased approximately 50 mL for each additional electrocautery session, but this loss was offset by 10 mL for each year the subsequent session was postponed. MBC of < 400 mL at the first session was a significant risk factor for MBC loss with further sessions. No other clinical parameters were associated with MBC over time. This study demonstrates a significant relationship between the frequency of electrocautery of Hunner lesions and MBC changes in HIC patients. Low MBC at the first session is a poor prognostic marker for MBC loss over multiple sessions.