Publications

There are significant challenges associated with manual assessment of hand MRI images in patients with inflammatory arthritis. These challenges include the time-consuming nature of manual scoring, the potential for inter-rater variability, and the need for specialized expertise. To overcome these limitations, we developed an artificial intelligence approach based on a convolutional neural network (CNN) to automatically assess bone erosions, osteitis, and synovitis.
The CNN was trained on a large dataset of hand MRI scans from patients with rheumatoid arthritis and psoriatic arthritis. The images were annotated by experienced rheumatologists, providing the CNN with accurate ground truth data. The trained CNN was then evaluated using cross-validation and an independent validation dataset.
The results demonstrated that the CNN achieved high accuracy in predicting erosions, osteitis, and synovitis scores, comparable to human experts. This suggests that the CNN can effectively identify and quantify these pathological changes in hand MRI images. Additionally, the CNN allowed for rapid scoring of MRI images, potentially reducing the time and cost associated with manual assessment.The CNN's performance was consistent across different patient cohorts and MRI scanners, indicating its generalizability. This suggests that the system could be applied in various clinical settings without requiring extensive retraining.
Overall, this study demonstrates the potential of artificial intelligence to significantly improve the assessment of hand MRI images in patients with inflammatory arthritis. By automating the scoring process, the CNN could streamline clinical workflows, reduce costs, and enhance the accuracy and consistency of diagnostic assessments.

The study investigated the role of fibroblast activation, a key process in synovitis and enthesitis, in the early stages of psoriatic arthritis (PsA) using 68Ga-FAPI-04 positron emission tomography/computed tomography (PET/CT) imaging, a novel whole-body imaging technique that allows the visualisation of cellular activity based on the used of radioactive tracers. 36 patients with psoriasis and arthralgia were included in the study and underwent whole-body PET/CT imaging with 68Ga-FAPI-04. We found that fibroblast activation was frequently observed in joints and entheses of patients with psoriasis and arthralgia. The intensity of fibroblast activation correlated with the number of tender joints and entheses, but not with ultrasound-detected inflammatory changes. Importantly, patients with significant fibroblast activation had a higher risk of developing PsA over time.
These findings suggest that fibroblast activation may serve as an early marker for the development of PsA, allowing for earlier intervention and potentially preventing disease progression. Targeting fibroblast activation could be a promising therapeutic strategy for PsA. Further research is needed to confirm these findings in larger studies and to explore the clinical significance of fibroblast activation in greater detail.

This study aimed to investigate the molecular changes occurring in the entheses (tendon attachment regions) of patients with psoriasis and psoriatic arthritis (PsA) using multispectral optoacoustic tomography (MSOT). MSOT is a novel non-invasive imaging technique that can measure various molecules within tissues, including hemoglobin, oxygen saturation, collagen, and lipids.
We compared the MSOT measurements of the ehtneses in patients with PsA, psoriasis, and healthy controls. We found that patients with psoriasis and PsA had higher levels of oxygenated hemoglobin and oxygen saturation, and lower levels of collagen in their entheses compared to healthy controls. These changes were more pronounced in patients with PsA. Additionally, the study found that tender entheses had lower collagen levels and higher lipid levels compared to nontender entheses. Erosions and enthesophytes on ultrasound were associated with specific changes in oxygen saturation and lipid signals.
These findings suggest that patients with psoriasis and PsA exhibit similar metabolic changes in their entheses, which may be exacerbated in the presence of inflammation. This supports the notion of a psoriatic disease spectrum characterized by common immunometabolic tissue changes. Furthermore, this suggests that MSOT may be a valuable tool for detecting early signs of enthesitis in patients with psoriasis, allowing for earlier intervention and potentially preventing the progression to PsA.

Objective

We know that genetic information can help doctors diagnose rheumatological diseases early. A tool called G-PROB has been developed to assist with diagnosis, but it has not been tested in real world settings yet. Our goal was to see if G-PROB could help diagnose patients with early inflammatory arthritis by using data from the Norfolk Arthritis Register (NOAR).

Methods

We collected genetic information from the blood of NOAR patients and compared it to the diagnosis made by their rheumatologist. G-PROB used this genetic data to calculate probabilities (from 0% to 100%) for six common rheumatic diseases: rheumatoid arthritis, lupus, psoriatic arthritis, spondyloarthropathy, gout, and other rheumatological diseases. We compared these probabilities to the doctor’s diagnosis for each patient.

Results

We tested G-PROB on 1,047 patients and found that it worked well. The higher the G-probability for a disease, the more likely it matched the doctor’s diagnosis. G-PROB was particularly good at ruling out diseases. For example, if the probability of a disease was less than 5%, it was very unlikely to be the correct diagnosis. In about half the cases, G-PROB’s top result matched the doctor’s diagnosis.

Conclusion

We used G-PROB on a large group of real patients and found that it successfully turned complex genetic data into easy-to-understand probabilities. This tool may be especially useful in helping doctors rule out unlikely diagnoses in rheumatology clinics.

Understanding the Bigger Picture of Auto-inflammatory Skin Diseases

Auto-inflammatory skin diseases, such as atopic dermatitis, plaque-type psoriasis, and hidradenitis suppurativa, cause a lot of symptomatic and emotional burdens on patients and lead to increased healthcare costs. While the main symptoms are visible on the skin, these diseases involve underlying inflammation that affects the whole body, not just the skin. To better understand these diseases, there is a need for deeper investigation beyond the visible manifestations.
This study aimed to explore the differences and similarities in the blood of patients with these conditions compared to healthy individuals.
Using advanced techniques to analyze small molecules (metabolome) and fats (lipidome) in the blood, researchers performed lipidomic and metabolomic profiling using liquid chromatography-mass spectrometry (LC-MS) on plasma blood samples from affected patients and healthy individuals.
The analysis revealed significant changes related to the body’s natural defenses against oxidative stress. This shift was observed across all three diseases and highlighted the potential disruption in cysteine metabolism, a process crucial for the body's anti-oxidant defenses. Additionally, the study identified specific lipids that play important roles in inflammation, with different patterns seen in each disease.
These findings provide new insights into the broader, systemic impact of these auto-inflammatory skin diseases. By understanding these underlying changes, researchers hope to develop more personalized treatments that can better manage these conditions and improve patients' quality of life. Ultimately, this research could lead to more effective and targeted therapies, improving outcomes for those suffering from these challenging skin conditions.

Finding markers within blood that signal the development of psoriatic arthritis in those with psoriasis

Psoriatic Arthritis (PsA) is a chronic disease that causes painful swelling in joints and tendons, most often in the hands and feet, and that severely impacts the quality of life. A delayed diagnosis can lead to irreversible damage of joints and bones. The disease develops in 20-30% of people with psoriasis (PsO). Currently, there are no tests available to identify PsA development early, but new markers in the blood (genetic markers, lipids, proteins, or metabolites) could help.
In the BioCOM Study at the St. Vincent’s University Hospital (Dublin, Ireland), 90 individuals with PsO, with PsO and early joint pain symptoms but without a clear diagnosis of PsA, and those diagnosed with PsA were recruited (30 per group). Blood samples were collected and analysed for markers that could differentiate PsO from PsA.
The blood analysis showed that there are a few distinguishing proteins, lipids and metabolites that could be used to identify PsA. The next steps are to analyse this data with more advanced computational models, and to identify these markers in further larger study populations.
Once these findings are evaluated and confirmed, the blood markers could be developed into a diagnostic test for PsA. Such a test could help PsO patients who may be developing PsA to receive treatment sooner, and thereby reduce the chance of joint and bone damage.

Psoriasis is a chronic skin condition that can also impact nails, joints, and areas where muscles and tendons attach to bones (entheses). Nail psoriasis is particularly concerning because it is often linked to more severe disease and reduced quality of life, and it can lead to psoriatic arthritis (a type of inflammatory arthritis). Recognizing and treating nail psoriasis effectively is crucial, but diagnosing and measuring the severity of nail symptoms can be challenging due to their varied manifestations, such as pitting, discoloration, and nail detachment.
To help healthcare providers assess nail psoriasis more systematically, a scoring system called the Nail Psoriasis Severity Index (NAPSI) was developed. This method quantifies changes in the nails, but it has limitations, such as requiring a physical examination during patient visits. To address these issues, a modified version called the modified Nail Psoriasis Severity Index (mNAPSI) allows for assessment using photographs, making it easier to evaluate nails remotely.
Recent advancements in artificial intelligence (AI)-based technology, particularly deep learning and neural networks, offer new avenues for automating the analysis of nail conditions based on simple hand photographs. This study aimed to develop a system that uses these AI technologies to classify nail psoriasis severity using photos of patients’ hands.
The research involved taking 1,154 photos of patients’ nails, which were then graded for severity by three independent experts. A neural network was trained using these images to predict the mNAPSI score based on the visual aspects of the nails. The system achieved strong performance, with a high degree of correlation between the neural network predictions and the expert assessments.
In conclusion, this study successfully created an AI-based method for assessing nail psoriasis severity automatically, which could potentially be used in both clinical settings and by patients at home. This advancement may facilitate more regular monitoring of the condition and improve management strategies while reducing the need for in-clinic visits. Future work aims to refine this system further and make it accessible to patients for easier self-evaluation.

Importance of Proper Sample Handling in Lipidomics and Metabolomics Research

The emerging fields of lipidomics and metabolomics, which study fats (lipids) and small molecules (metabolites) in the body, hold great promise for discovering new diagnostic biomarkers that can help to detect diseases early. However, to ensure the accuracy of these discoveries, it is crucial how biological samples are handled before analysis. Some of the molecules in biological samples, e.g., blood, can change or degrade if the samples are not handled properly. This can lead to inaccurate results.
This study investigated how the storage temperature and duration of plasma samples from blood affect the concentrations of various metabolites and lipids. Researchers collected blood samples from nine healthy volunteers and analyzed them using advanced liquid chromatography-mass spectrometry (LC-MS) techniques to measure a wide range of molecules.
The study found that while many molecules remained stable under less strict sample-handling conditions, others were more sensitive and required careful handling to maintain their integrity. Based on these findings, researchers developed four data-driven recommendations for handling blood samples. These guidelines vary in strictness, balancing the need to preserve the maximum number of analytes with practical considerations for routine clinical use.
These protocols also allow scientists to evaluate how vulnerable different potential biomarkers are to changes that might occur outside the body (ex vivo).

Understanding How Scientists Use Mass Spectrometry to Discover Disease Biomarkers

This review aims to provide a visual and easy-to-understand overview of the entire process involved in conducting an LC-MS-based clinical research project for biomarker discovery. By outlining the key steps from planning to data analysis and validation, it helps to guide researchers through the complex journey of finding new biomarkers that could lead to better disease understanding and more personalized treatment options.

Mass spectrometry (MS) is a powerful technology used by scientists to study small molecules in the body. This technique is crucial for discovering biomarkers (substances that can indicate the presence of a disease) helping us understand various diseases better and moving us closer to personalized medicine, where treatments are tailored to each individual. This summary outlines the key steps involved in using MS, particularly liquid chromatography-mass spectrometry (LC-MS), for clinical research.

Planning and Collaboration:
Before starting the research, scientists must carefully plan the study. This involves setting clear objectives and designing the study accordingly. They work with experts from different fields, including clinicians and data scientists, to ensure the study is well-rounded and thorough. Selecting the right subjects and deciding how to conduct the trials are essential steps that depend on the study’s goals and specific disease characteristics.

Handling Samples and Collecting Data:
Proper handling of biological samples, like blood or tissue, is critical because it affects the quality of the data collected later. The samples are then analyzed using LC-MS. This method can look for specific molecules (targeted), a broad range of molecules (semi-targeted), or any molecules present (non-targeted), depending on the study's focus. The choice of method impacts the detail and type of data collected.

Processing and Analyzing Data:
After data collection, scientists process the data to clean it up and make it ready for analysis. They use both traditional statistical methods and advanced techniques like machine learning to understand complex data patterns. Other tools, such as pathway analysis, help scientists see how these molecules interact in the body and how they might relate to the disease.

Validating Results and Ensuring Quality:
Before using these biomarkers in clinical settings, scientists must validate their findings to ensure that they are accurate and reliable. Throughout the research process, strict quality control measures are applied to make sure the data is trustworthy and that the findings can be confidently used to diagnose or predict disease.

LC-MS-based biomarker studies can be broken down into five key phases, which often overlap:

• Planning and Design: Setting up the goals of the study, choosing the right participants, and deciding on the study's design.
• Sample Collection and Handling: Properly collecting and handling biological samples (like blood or tissue) to keep them in good condition.
• Data Collection (Using LC-MS): Using LC-MS to analyze the samples and collect detailed data about the small molecules they contain.
• Data Processing and Analysis: Cleaning and analyzing the data using various methods, including computer-based tools, to make sense of it.
• Validation and Interpretation: Confirming the findings to ensure they are accurate and making sense of the results to identify potential biomarkers.

This study explores how to improve the assessment of inflammation in joints and areas where muscles and tendons attach to bones (entheses) in people with inflammatory arthritis, such as rheumatoid arthritis (RA) and psoriatic arthritis (PsA). Accurate assessment is crucial for diagnosing these conditions and measuring disease activity. Traditional methods, including clinical examinations and ultrasound (US), provide valuable information but often focus on structural changes and not the deeper biological processes happening at the cellular level.
This study introduces a new imaging technique called Multispectral Optoacoustic Tomography (MSOT), which combines ultrasound and advanced imaging to visualize different chemical components in the body. MSOT can evaluate changes in tissue levels of important substances such as hemoglobin (which carries oxygen) and collagen (a protein important for tissue structure). This non-invasive technique allows researchers to better understand the metabolic processes involved in inflammation.
We conducted a study with 87 participants, including people with RA, PsA, and healthy individuals, to compare conventional US with MSOT imaging. The participants underwent both imaging methods to assess joints and entheseal sites for signs of inflammation such as swelling and tenderness. The results showed that MSOT could accurately measure levels of hemoglobin and collagen, highlighting metabolic differences between inflamed and non-inflamed areas.
The findings indicated that inflamed joints had increased blood flow (higher hemoglobin content) and signs of tissue breakdown (lower collagen levels), aligning with what is known about inflammatory processes. Additionally, the study found differences between the metabolic responses of joints and entheses, suggesting they may require different approaches for diagnosis and treatment.
In summary, this research emphasizes the potential of new imaging technologies to provide deeper insights into inflammatory arthritis, moving beyond just structural assessments to understand the underlying metabolic processes involved in these conditions.

The study aimed to investigate the effects of low-to-moderate physical activity on enthesitis in patients with psoriasis and psoriatic arthritis (PsA). Participants underwent a single badminton training session and were assessed for pain, inflammation, and disease activity before and after the session. We found that low-to-moderate physical activity did not lead to a significant increase in pain, tenderness, or inflammation in the entheses. Mild increases in blood flow to the entheses were observed in a few patients, but these changes were not associated with pain or other symptoms. One patient with active PsA experienced a temporary flare-up after the training session, but this was likely due to other factors rather than the physical activity itself. The findings suggest that low-to-moderate physical activity is generally well-tolerated in patients with psoriasis and PsA. While some individuals may experience mild and temporary changes in entheseal inflammation, these changes do not appear to be clinically significant. This study provides evidence to support the inclusion of physical activity as part of a comprehensive management plan for these conditions. However, further research is needed to define optimal exercise thresholds and to investigate the long-term effects of physical activity on enthesitis in PsA.

Improving outcomes in Psoriatic Arthritis

Psoriatic Arthritis (PsA) is a form of arthritis which is found in approximately 30% of people who have the skin condition, Psoriasis. Frequently debilitating and progressive, achieving a good outcome for a person with PsA is made difficult by late diagnosis, disease clinical features and in many cases, failure to adequately control features of inflammation. Research studies from individual centres have certainly contributed to our understanding of why people develop PsA but to adequately address the major areas of unmet need, multi-centre, collaborative research programmes are now required. HIPPOCRATES is a 5-year, Innovative Medicines Initiative (IMI) programme which includes 17 European academic centres experienced in PsA research, 5 pharmaceutical industry partners, 3 small-/medium-sized industry partners and 2 patient representative organisations. In this review, the ambitious programme of work to be undertaken by HIPPOCRATES is outlined and common approaches and challenges are identified. The participation of patient research partners in all stages of the work of HIPPOCRATES is highlighted. It is expected that, when completed, the results will ultimately allow for changes in the approaches to diagnosing, managing and treating PsA allowing for improvements in short-term and long-term outcomes.

Using Artificial Intelligence to improve assessment of Psoriatic Arthritis

Psoriatic arthritis (PsA) is a long-lasting inflammatory disease that affects the joints and often requires careful monitoring to manage its symptoms. Doctors commonly use a tool called the Disease Activity Score 28 (DAS28 CRP) to evaluate how active the disease is, based on various symptoms. This study aimed to find out which specific symptoms in this score are most important for assessing the activity of PsA.
The researchers analyzed data from 80 patients with PsA, including both men and women, with ages averaging around 56 years. These patients showed a range of disease activity levels, from remission (where symptoms are minimal or absent) to moderate activity. Using advanced computer techniques, including machine learning, the researchers analyzed the different symptoms included in the DAS28 CRP score.
They discovered that tenderness in the right index finger’s metacarpophalangeal joint (the joint at the base of the finger) was the most important symptom for determining PsA activity. This single symptom alone allowed a computer-based model to predict whether a new patient was in remission with 67% accuracy.
Further analysis using artificial intelligence (AI) techniques (emergent self-organizing map, a type of artificial neural network) helped identify specific subgroups of patients who showed unique patterns of symptoms, such as tenderness or swelling in certain joints. By using AI, the study was able to pinpoint these differences, which could lead to more tailored treatment approaches.
These findings demonstrate the potential of AI in refining the DAS28 CRP scoring system for PsA. By focusing on the most relevant symptoms and recognizing unique patterns in certain patients, doctors can more accurately assess PsA and provide more effective treatments, especially for those whose symptoms do not match the common patterns. These findings are an important step towards precision medicine, which aims to customize healthcare to individual patients.

ALISTER - Helping Researchers Handle Blood Samples Better

Scientific studies focusing on fats (lipidomics) and small molecules (metabolomics) in our blood hold significant potential for advancing clinical diagnostics and biomarker discovery. However, these studies are highly sensitive to how blood samples are handled prior to the analysis (pre-analytical conditions). If the blood samples are not managed properly, the data obtained might be inaccurate and unreliable, leading to wasted time, money, and resources. There is a critical need for guidance on sample handling to ensure the accurate translation of omics technologies into routine clinical practice. The goal of this study was to support decision-making regarding the stability of blood samples across all phases of lipidomics- and metabolomics-focused research. Data from multiple pre-analytic studies were put together to form a comprehensive database. An interactive, flexible approach to evaluate this data was developed using an RShiny-based web application. This application is designed to be broadly applicable in clinical and bioanalytic research, offering a user-friendly interface for assessing sample stability.
The resulting online tool called ALISTER (Application for Lipid Stability Evaluation & Research). ALISTER helps scientists and healthcare professionals make informed decisions concerning the stability of blood samples in lipidomic and metabolomic studies. It provides guidance on the best ways to collect and store blood samples to ensure that the results of these studies are accurate and reliable.
Using data collected from various studies, ALISTER allows users to:

• Plan how to collect blood samples in the best possible way to maintain sample quality.

• Check the quality of blood samples used in past experiments.

• ALISTER is easy to use and accessible online, making it a practical tool for researchers and clinicians working on studies involving blood samples. By using ALISTER, researchers can ensure that their findings are based on high-quality samples, leading to more trustworthy results. This tool ultimately helps in the accurate development of new diagnostic tests and enhances our understanding of health and disease.
  ALISTER is available for use by anyone interested at https://itmp.shinyapps.io/alister/.

Particular T cells play a role in intestinal inflammation related to spondyloarthritis

• Spondyloarthritis (SpA) - including psoriatic arthritis (PsA) - patients, often show signs of gut inflammation, linked to more progressive joint pathology
• In this study we found a marked proinflammatory (“type17”) skewed T cell profile in gut biopsies from SpA patients in relationship to gut inflammation
• Remarkably, an enrichment of particular T cell subsets (“γδT cells”) is associated with higher SpA disease activity
• This study adds to our understanding of dysregulated immunity in combined gut-joint SpA disease
• It enforces the prominent role of these specific T cell subsets in driving joint pathology.
• This information may be helpful for optimizing targeted treatment strategies for SpA patients with or without extraarticular disease manifestations at barrier sites (gut, skin, eye)

Key findings/highlights/impact

• Marked “type 17” immunity in gut-joint SpA disease
• Enrichment of specific T cells linked to joint pathology
• Targeting “gut immune cells” in joint disease could have therapeutic potential

Improving Early Diagnosis and Treatment of Psoriatic Arthritis with Biomarkers

Diagnosing and treating immune-mediated inflammatory diseases (IMIDs) like psoriatic arthritis (PsA) can be challenging due to the overlap and variability in their symptoms. PsA is a condition that affects about 30% of people with psoriasis and often impacts joints and tendons. Diagnosing PsA usually requires the expertise of clinical specialists or becomes more evident when the disease has significantly progressed. Identifying specific biomarkers (molecules that can signal the presence of a disease) could greatly improve early diagnosis and treatment.
Currently, there are no specific biomarkers available to reliably diagnose PsA. However, research into various types of biomarkers offers hope. Scientists are exploring different ways to identify these biomarkers using advanced techniques such as:
• Genomic and Epigenomic Profiling: Studying changes in the DNA and how genes are expressed.
• Proteomics: Analyzing proteins present in blood and tissues samples.
• Metabolomics and Lipidomics: Examining small molecules (metabolites) and fats (lipids) in blood and tissue samples.
These methods could lead to the discovery of complex molecular patterns specific to PsA, which would help distinguish it from other similar diseases. Integrating these biomarkers with data from high-throughput histology (the study of tissues), imaging, and standardized clinical assessments could significantly enhance the ability to diagnose PsA early, predict its onset in people with psoriasis, differentiate it from other immune-mediated inflammatory diseases, and tailor treatment more effectively.
This review focuses on the innovative technologies being used in the HIPPOCRATES project, a European Union initiative, to identify biomarker profiles specific to PsA. By combining data from multiple scientific approaches (multi-omics), researchers aim to develop comprehensive molecular profiles that could revolutionize how PsA is diagnosed and managed, ultimately improving the outcomes for patients.

The study aimed to develop a minimally invasive biopsy procedure for obtaining entheseal tissue (i.e. tissue from tendon attachment regions) from patients with psoriatic arthritis (PsA) and establish a method to validate the retrieved tissue. A novel biopsy procedure was developed and tested on both cadavers and patients with PsA. The technique involved using ultrasound guidance to identify the lateral epicondyle enthesis and then performing a minimally invasive biopsy. The retrieved tissue was analyzed using tissue analysis techniques and second harmonic generation (SHG) microscopy, which is an advanced microscopy techniwue can differentiate entheseal tissue from other tissues based on its unique collagen structure. The study found that the developed biopsy procedure was safe, well-tolerated, and minimally invasive. SHG microscopy was successfully used to identify entheseal tissue within the biopsy samples. Additionally, the retrieved tissue was suitable for molecular analysis, allowing for the study of cellular and molecular changes in enthesitis. This study presents a valuable method for obtaining entheseal tissue from PsA patients. The ability to study entheseal tissue will contribute to a better understanding of the pathophysiology of enthesitis and inform the development of targeted therapies.

The study aimed to investigate whether deep learning (DL), which is an advanced artificial intelligence-based methods, could be used to differentiate between different forms of inflammatory arthritis using data from magnetic resonance imaging (MRI) scans of the hand of patients with arthritis. A neural network was trained on a large dataset of hand MRI scans from patients with seropositive rheumatoid arthritis, seronegative rheumatoid arthritis, psoriatic arthritis, and psoriasis. The network was designed to identify patterns in the MRI images that could distinguish between the different arthritis types. The neural network successfully identified patterns in the MRI images that were associated with each arthritis type. However, the accuracy was moderate, suggesting that larger datasets are needed to improve performance. The study also found that clinical data provided little additional benefit in distinguishing between the arthritis types, suggesting that the differences in inflammation distribution are primarily visible in the MRI images. The findings suggest that deep learning could be a valuable tool for classifying unclear cases of inflammatory arthritis, particularly in distinguishing psoriatic arthritis from seronegative rheumatoid arthritis. Additionally, the study found that only a subset of MRI sequences were necessary for accurate classification, potentially leading to shorter scan times in the future.
Overall, this study demonstrates the potential of deep learning for classifying inflammatory arthritis types using hand MRIs. However, further research is needed to improve accuracy and explore clinical applications.

The study aimed to develop a neural network model to classify different types of arthritis based on the shape of articular bone using high-resolution peripheral quantitative computer tomography (HR-pQCT) scans, a novel imaging technology based on high-resolution X-ray imaging. The model was trained and validated on a large dataset of rheumatoid Arthritis (RA), Psoriatic Arthritis (PsA) and healthy controls. The results showed that the neural network could accurately differentiate between these conditions and even classify undifferentiated arthritis patients into RA, PsA, or healthy controls. The study also highlighted the importance of articular entheseal regions in PsA and suggested their potential for diagnosis using ultrasound. Overall, the research demonstrates the potential of neural networks in improving arthritis classification and diagnosis.

The HIPPOCRATES consortium is finding new ways to diagnose psoriatic arthritis and hopes to improve the outcomes for those with the disease

Psoriatic arthritis affects 1-2% of the population, and develops in joints and tendons, mostly in patients who are already diagnosed with psoriasis. The main symptoms of psoriatic arthritis are pain and reduced mobility in the joints, and tiredness. As there are currently no tests available to diagnose psoriatic arthritis, there are often delays in diagnosis and therefore also delays in finding effective treatment.
HIPPOCRATES is a European research project (https://hippocrates-imi.eu/), funded by the Innovative Medicines Initiative, that aims to find specific blood markers to diagnose psoriatic arthritis, identify which individuals with psoriasis are at risk of developing psoriatic arthritis, and find optimal treatment options for people living with psoriatic arthritis.
As part of the HIPPOCRATES project, the HIPPOCRATES Prospective Observational Study (HPOS - https://www.hpos.study/) is recruiting 25,000 individuals with psoriasis in Europe, to study the early risk factors for the development of psoriatic arthritis. Multiple industry partners and researcher associations, such as the Group for Research and Assessment of Psoriasis and Psoriatic Arthritis (GRAPPA), are involved in the project and have been sharing data and samples from their clinical trials with psoriasis and psoriatic arthritis patients. The HIPPOCRATES project will combine these samples to be able to analyze bigger populations of psoriasis and psoriatic arthritis patients and use advanced mathematical models to find markers for risk factors and diagnosis.
If new tests are developed that could identify the risk of developing psoriatic arthritis in psoriasis patients the progression to psoriatic arthritis could possibly be delayed or prevented.
The HIPPOCRATES project involves many psoriasis patients in the planning, execution, analysis and sharing of the work that is being performed. This ensures that the patient’s perspective and opinion remain a central focus of the project.

Psoriatic arthritis (PsA) is a chronic inflammatory condition that affects the joints and often accompanies psoriasis, a skin disease. Recent advances in treatment, particularly the introduction of biological medications, have significantly improved the management of PsA. These treatments, known as biological disease-modifying antirheumatic drugs (bDMARDs), target specific proteins in the immune system to reduce inflammation. However, while many patients respond well to these therapies, some do not improve significantly and continue to experience high levels of disease activity.
In this study, we explored a novel approach for treating patients with difficult-to-manage PsA by alternating therapies that target two different proteins in the immune system: interleukin-23 (IL-23) and interleukin-17 (IL-17). These proteins play a role in the inflammatory processes associated with PsA. The study involved three patients who did not sufficiently respond to other standard treatments, including tumork necrosis factor alpha (TNFα) inhibitors, IL-23, and IL-17 inhibitors.
The treatment regimen used in this study involved giving patients IL-23 and IL-17 inhibitors in a rotating cycle, with patients receiving injections of IL-23 (guselkumab) followed by IL-17 (secukinumab) over a series of months. This approach was based on the idea that these cytokines can work well together without significantly raising the risk of infections, which can be a concern with some therapies.
After starting this alternating treatment regimen, all three patients showed significant improvement in their PsA symptoms across various affected areas including joints, skin, and entheses (the areas where tendons attach to bones). They experienced a marked reduction in disease activity, achieving a state of minimal disease activity—a significant goal in managing PsA—within a period of six months. Notably, during this time, none of the patients experienced infections or serious side effects. The findings from this small study suggest that alternating between IL-23 and IL-17 inhibitors may provide a new and effective treatment strategy for patients with PsA who do not respond adequately to conventional therapies. In summary, this research highlights a promising potential treatment strategy for challenging cases of psoriatic arthritis that could help improve patient outcomes while minimizing risks.