Development of a Cloud Platform for Advanced Diagnostics Based on Clinical and Omics Data.

Study Summary This is an open-label, multicentre, observational cohort study involving both retrospective and prospective collection of clinical data and biological samples.

Rare diseases represent a major healthcare challenge. Approximately 80% of currently known rare diseases are genetic in origin. These conditions are often complex, affect multiple organ systems, and are associated with significant morbidity, mortality, and healthcare costs. Reaching a correct diagnosis can take several years.

Systemic autoinflammatory diseases (SAIDs) are a growing group of rare disorders caused by dysfunction of the innate immune system. This dysfunction leads to recurrent or persistent episodes of sterile systemic inflammation, driven by the excessive production of inflammatory molecules called cytokines, including IL-1, IL-18, IL-6, TNF-α, and type I interferons. SAIDs may be rare or ultra-rare, but collectively they affect more than two million people in Europe. Up to 47 distinct disease entities have been identified, with both specific and overlapping clinical features.

Most autoinflammatory diseases have a strong genetic component, often involving mutations in a single gene. For known monogenic SAIDs, accurate disease classification is essential for selecting effective treatment. However, in approximately 40-60% of suspected cases, genetic testing does not provide a definitive diagnosis. These cases are classified as undifferentiated SAIDs (uSAIDs). Diagnostic uncertainty may delay appropriate treatment and increase the risk of irreversible organ damage. Undifferentiated cases may have a polygenic, multifactorial, or complex origin, with environmental factors also contributing to disease development and prognosis. Other conditions of interest in this study include systemic juvenile idiopathic arthritis (sJIA) and fibromyalgia (FMG).

Evidence suggests that patients with undifferentiated and complex SAIDs may benefit from personalised, stratified approaches to diagnosis and treatment, similar to those already used for some monogenic diseases. Achieving this requires the development of multi-omics profiles and artificial intelligence tools suitable for clinical practice.

Omics data are obtained by analysing groups of biological molecules, such as ions, nucleic acids, proteins, and enzymes, in samples including blood serum, urine, cerebrospinal fluid, saliva, and tissues. Different approaches, such as genomics, proteomics, and lipidomics, provide complementary information about how biological systems function. Integrating these data can improve understanding of disease mechanisms, natural history, prognosis, and treatment response.

However, interpreting molecular findings and combining them with clinical information remains challenging, particularly for rare diseases or cases in which the underlying cause is unknown. Although electronic health records (EHRs) are widely used in hospitals, omics data are rarely integrated into these systems. Clinical information is often shared with specialists through verbal communication or separate tools, such as email, rather than through a unified platform. This fragmentation can reduce efficiency, increase the risk of incomplete or inaccurate information exchange, and limit the reuse of data for future investigations. Identifying patients with similar clinical and genetic characteristics also often depends on the experience and memory of individual healthcare professionals.

This study aims to address these limitations by improving the speed, efficiency, and reliability of collaboration between clinical teams and specialists or organisations providing genomic and other molecular testing services.

Study Objectives

The study has three main objectives:

Improve diagnosis and patient stratification. Investigate whether omics data, including genomics, proteomics, and lipidomics, can help distinguish and classify patients with SAIDs, particularly undifferentiated and complex SAIDs, sJIA, and fibromyalgia. The ultimate goal is to support more accurate diagnoses and help identify the most appropriate treatments and care strategies for each patient.

Develop analytical algorithms and predictive models. Create computational tools capable of identifying relevant biological information in clinical and omics data and generating models to support the interpretation of individual patient cases.

Integrate clinical and molecular information. Develop a comprehensive, user-friendly clinical and omics record that combines patients' clinical histories, laboratory findings, omics data, and analytical results. Its usefulness in supporting patient care will then be evaluated.

The study will use leftover blood samples collected during routine laboratory testing, DNA samples already extracted for diagnostic genetic investigations, and, where available, previously collected and stored DNA, RNA, and serum samples held in the biobanks and repositories of the Istituto Giannina Gaslini. Clinical information will include data routinely collected to document patients' medical histories and current clinical conditions.

Expected Outcomes and Clinical Applications The project will develop an innovative clinical and omics electronic health record capable of systematically managing both conventional clinical information, including patient characteristics and medical history, and results from molecular investigations.

The platform will support searches of external databases and provide advanced ways to access and interpret patient information. It will also incorporate clinical decision-support functions to provide useful information to healthcare professionals and assist them in making informed decisions.

In addition, the project will develop computational pipelines to process raw genomic data and establish primary and secondary analysis protocols for omics data. These procedures will generate standardised results suitable for integration into the patient's electronic health record.

Overall, the project aims to improve the integration and interpretation of clinical and molecular information, facilitate collaboration between healthcare professionals and laboratory specialists, and support more timely, accurate, and personalised diagnosis and treatment for patients with rare and complex inflammatory conditions.