Beyond Skin, Gut, and Brain: Understanding Disease Connections by Why and How Diseases Happen, Not Just Where

Traditionally, diseases are classified based on the organs or tissues they most noticeably affect. For example, we distinguish between "skin" and "gut" diseases. However, research has revealed similarities between diseases that seem to have nothing in common. Many people today are affected by more than one disease. Understanding the connections and processes between these diseases, can help to develop new therapies or decide the best existing treatment for each individual patient. For example, understanding the link between diabetes and heart disease can help doctors identify patients at high risk and develop targeted treatments.

In our recently published study, we examined various diseases, in total about 500, to identify their similarities and differences, without focusing on the traditional way of grouping them by organ systems. We analysed six different types of data related to these diseases, which we gathered from various sources and public databases. One of the data types included information about which genes are associated with different diseases - for instance, whether certain genes are turned on, off, or remain neutral in a specific disease. We used this information to determine similar molecular patterns among the 500 diseases. We did this separately for each of the six data types and then combined the results.

In this combined map of diseases, we discovered several interesting connections, some of which were already known. For example, we found that psoriasis is closely related to Crohn's disease, ulcerative colitis, rheumatoid arthritis, and other autoimmune and inflammatory diseases, but it is less similar to other skin conditions. Based on this map, we grouped diseases into categories. Cancer and non-cancer diseases were clearly separated. Among non-cancers, we found a large group of inflammatory diseases with some links to neurodegenerative diseases such as Alzheimer's and Parkinson's.

While some of the observed disease relationships were known from epidemiological studies, we were surprised at how well our data-informed map replicated these findings. This approach, applied to individual patient data, could offer new insights and aid in precise medicine, guiding treatment decisions and drug development in the future.

Key findings

  • A map of 500 disease relationships based on their underlying mechanisms and processes challenges traditional views of diseases, since diseases only partially fit into “organ groups”.
  • Known relationships between diseases such as psoriasis and inflammatory bowel diseases could be replicated by this data-informed approach that does not rely on prior assumptions.
  • Inflammatory and metabolic diseases showed a link to neurodegenerative diseases, with type 2 diabetes and Alzheimer's displaying an interestingly high similarity.

Related publication:
A Multi-Dimensional Approach to Map Disease Relationships Challenges Classical Disease Views
Lena Möbus, Angela Serra, Michele Fratello, Alisa Pavel, Antonio Federico, Dario Greco Advanced Science (2024); 2401754
doi: 10.1002/advs.202401754