Blog

Optimization of rheumatoid arthritis treatments plan through clinical and transcriptomic data exploration and iTwin’s artificial intelligence

With an incidence on the general population of about 1%, rheumatoid arthritis (RA) is the most common autoimmune disease. In Europe there are almost 3 million patients with RA, about 400 thousand only in Italy. Around 75% are women of any age, with a peak in the manifestation of symptoms for the 30 – 50 years old group.

Autoimmune diseases are characterized by a malfunction of the immune system which fails to recognize healthy tissues as “self” and mistakenly attacks the body.

The RA disease determines a chronic inflammation of the synovial membrane contained inside the joints, which grows up to cause the destruction of cartilage and the erosion of bone tissue. This leads to a drop in patient’s quality of life, due to severe pain and motor stiffness; in the advanced stage of the disease, the patient may develop joint deformities and ankylosis.

Rheumatoid arthritis reduces life expectancy by 10 to 15 years, also for a higher risk of developing heart problems caused by a more rapid and severe atherosclerosis. RA is a systemic disease and can involve a large number of tissues such as blood vessels, serosa, muscles, lungs, kidneys, heart, central and peripheral nervous systems, visual and haematopoietic systems.

Today we don’t have a definitive cure but doctors can manage RA with a large range of drugs: DMARDs (disease-modifying antirheumatic drugs), JAK inhibitors, NSAIDs (non-steroidal anti-inflammatory drugs) and glucocorticoids. DMARDs can be conventional synthetic (cs-DMARD) and biological (b-DMARD); b-DMARDs are new generation complex protein drugs that block inflammation by acting as inhibitors of cytokines, in particular TNF- α and IL-1.

Guidelines from EULAR1 (European League Against Rheumatism) recommend to initially treat the patient with cs-DMARDs and, in case of no response, switching to b-DMARDs or JAK inhibitors; the latter inhibits the activity of Janus kinases by blocking the cytokine signaling pathway. Among the cs-DMARDs we find well-known drugs such as Methotrexate, Leflunomide and Sulfasalzina, which are usually administered in combination with NSAIDs and glucocorticoids.

The variety of these drugs is quite wide and patients, due to their basal variability, tend to respond differently to a same therapy. In particular way 50% of patients strating a new cs-DMARD must stop it in the following 12/18 months due to ineffectiveness or adverse events1. In fact we should not forget that drugs used for RA can lead to serious side effects that manifest themselves differently in treated patients.

Even if a drug works on average for different patients, it does not mean that it will have the same effect for a specific individue: 45% of patients treated with methotrexate are unable to solve RA problems2; on the other hand, 30 – 40% of them fail to respond to biologics drugs2.

Today we know that each patient has a unique clinical and transciptomic profile that – if properly analyzed – can help physicians to select more effective treatments. Enabling precision medicine in RA is very important as it can improve patients’ quality of life: only 1 out of 4 patients is able to reach a low level of disease activity1. Identifying as quick as possible the best treatment for the patient is therefore essential to avoid disease progression and serious symptoms worsening : 1 out of 3 RA patients can’t work full time after 5 years of disease onset2.

Clinical studies conducted on blood and synovial tissue samples have shown that through the sequencing of RNA (RNA-seq or single cell RNA-seq) it is possible to categorize RA patients on the basis of their genetic expression and optimize treatments management.

In a recent paper of 20203, it was observed that the composition of MerTK positive (“firefighters”) and MerTK negative (“pyromaniacs”) macrophages – due to their antagonistic action on inflammation – within the RA patients synovial tissue is an important indicator of disease remission. If the ratio of MerTK positive to MerTK negative macrophages is less than 2.5 times, the risk of relapse increases by 16 times if the patient stop the treatment. The genetic characterization of these two classes of macrophages, obtained through single cell RNA-seq, could so help physicians to scale up therapies in a more effective way avoiding flare-ups of the disease.

In another paper of 20204, a group of researchers highlighted that through RNA-seq of blood samples it is possible to divide RA patients into two subgroups based on their probability of responding positively or negatively to anti-TNF treatments. Anti-TNFs are the most sold drugs in the world but around 70% of RA patients who are administered with them fail to respond to the therapy5 and can be interested by a dangerous progression of the disease. With an accuracy level close to 90%, it would therefore be possible to suggest to anti-TNFs immune RA patients alternative working treatments.

At iCareX, we are working to make RNA sequencing and treatment personalization accessible to all RA patients. We collaborate with affiliated clinics and partner laboratories to collect synovial tissue samples from patients, extract the RNA and sequence it.

Our teams work to provide physicians with personalized medicine solutions based on transcriptomic data and state of the art artificial intelligence. Through the “Exploring” module of our iTwin platform, rheumatologists can explore the digitized clinical, radiological, ultrasound, biological and transcriptomic data of their patients to get a wide overview of their health profile; they also have also the possibility to view the anonymized data of the subjects being treated in our network of doctors, in order to identify those most similar to their patients and develop more effective treatment strategies.

Using powerful AI-based and data-driven scenario analyzes, rheumatologists can use iTwin’s “Personalizing” module to evaluate in advance impacts of different treatments on RA biomarkers and possible side effects of administered drugs. This is essential to give the patient a therapy that satisfies the trade-off between slowing the inflammation and limiting the side effects.

We also allow the doctor to control the effects of the treatment through the information collected by our iTwin Monitoring mobile application, where the patient can provide hig-frequency feedbacks on his “disease activity score” and communicate how he feels and what side effects he has encountered by completing qualitative and quantitative gamified medical questionnaries.

With the Twins we want to help rheumatology experts take more concrete steps towards optimizing RA treatmens, improving patients’ quality of life and increasing their satisfaction.

NOTES

[1] “EULAR recommendations for the management of rheumatoid arthritis with synthetic and biological disease-modifying antirheumatic drugs: 2019 update”

[2] “Maximising Therapeutic Utility for Rheumatoid Arthritis using genetic and genomic tissue responses to stratify medicines”- Queen Mary, University of London, Department William Harvey Research Institute

[3] “Distinct synovial tissue macrophage subsets regulate inflammation and remission in rheumatoid arthritis” – Stefano Aliverini et al. (2020)

[4] “Clinical Validation of a Blood-Based Predictive Test for Stratification of Response to Tumor Necrosis Factor Inhibitor Therapies in Rheumatoid Arthritis Patients” – Theodore Mellors et al. (2020)

[5] “Efficacy and safety of adalimumab as monotherapy in patients with rheumatoid arthritis for whom previous disease modifying antirheumatic drug treatment has failed”- Van de Putte LB, Atkins C, Malaise M, et al. (2004)

Share on facebook
Facebook
Share on twitter
Twitter
Share on linkedin
LinkedIn
Share on whatsapp
WhatsApp
Share on email
Email