Gulf War Illness: How can disease modelling aid the search for a cure?

Biological systems are hugely complex, with a staggering number of pathways interacting every millisecond. The downstream effects of just one biological change are enormous. So, when a new disease emerges, researchers have a near-impossible task working upstream from symptoms to identify the underlying cause and biological mechanisms causing disease. Not only does the appearance of a common symptom – such as a fever, for example – lead to potential misdiagnoses, but it can also lead to clinicians choosing ineffective therapies that treat symptoms, rather than the true cause of a disease.

This problem is compounded when researchers are presented with a complex set of symptoms, with no easily identifiable infectious agent as the cause. Often with rare or emerging diseases, clinicians will treat the most obvious symptoms without understanding the underlying cause of the disease or having a full picture of the condition itself. Complex problems require complex solutions. And with an unprecedented volume of data at our fingertips today, combined with new opportunities provided by computer modelling, the answers could be easier to find than ever before.

Disease modelling Gulf War Illness

In the mid-to-late 1990s, an unexplained illness began to emerge in army bases across the US. Symptoms and their severity ranged across individuals, and included fatigue, musculoskeletal pain, skin rashes, headaches, and gastrointestinal issues. The only thing the 170,000 patients had in common was that they had served during the 1990-91 Persian Gulf War. The condition, which became known as Gulf War Illness (GWI), has been seen in troops from all sides of the conflict and has been associated with chemical exposure.

Thirty years on, GWI has proven itself to be a challenging clinical condition; chronic and multisymptomatic by nature, the disease affects around a third of all veterans who served during the Gulf War. A complex range of factors are thought to contribute to the development of the illness. These include exposure to various chemicals associated with the war, from chemical warfare agents to pyridostigmine bromide – a preventative drug given to soldiers at risk of exposure to toxic chemicals. GWI has also been associated with pesticide exposure, breathing in smoke from oil well fires, exposure to uranium, and psychological factors, such as post-traumatic stress disorder.

With such a wide range of symptoms and contributing factors, understanding the mechanism of GWI and thus finding a treatment has proven a significant challenge. Our academic partner Gordon Broderick, Director of the Center for Clinical Systems Biology, Rochester General Hospital, began to study this using a novel approach.

In an effort to identify treatments for GWI when there is no clear cause, researchers have been working on a conventional disease pathway for the condition. They began by extracting prior knowledge and existing data from over 20,000 research papers, with the hope of using this to identify therapies, or combinations of therapies, that can alleviate symptoms.

After developing a conventional disease pathway model for GWI, researchers then used Elsevier’s data and technology to create a model to act as a basis for in silico modelling of the disease. By reducing the disease to its key components, a computer-readable network of disease markers was characterised using genetic, biomarker, and clinical data; the model was also populated with ‘disease vs control’ data. Once the model was created, accuracy was enhanced by drawing on existing information available in other biomedical and literature databases.

Tinker, tailor, soldier, therapy

By introducing various drugs into the model, and adjusting their dosages, the model was able to demonstrate which drugs work best together, map potential side effects and drug interactions, and predict the efficacy of therapies. By constantly adjusting doses and drug combinations within the model, therapies that propel the model back into a healthy state can be more reliably predicted before moving on to in vivo studies.

Carrying out primary research in silico in this way not only limits risk, but it saves time and money and allows research teams to test more ideas. Computer modelling provides more and new treatment options. It helps to reduce unnecessary and overlapping experiments and aids researchers in understanding which approaches are likely to fail, so they can be avoided. Not only does this make drug discovery more efficient, but it is also less resource intensive. Treatments for GWI identified using in silico modelling have now been fast-tracked for testing in Phase I clinical trials as a direct result of the successful outcomes seen in mouse models.

The future of in silico

Manually reviewing the huge amount of data on symptoms of disease indications, and biomarkers associated with a disease as complex as Gulf War Illness, is a multifaceted challenge. Mathematical modelling can help researchers overcome this barrier whilst also reducing the need and costs for computing power. Researchers can use machine understanding (semantic) approaches to summarise and extract critical information.

Using in silico modelling before moving on to in vivo studies can fast-track research, helping find better treatments for unmet medical needs. The research technique also has potential applications across other areas. By developing conventional disease pathways, and using in silico modelling to identify potential therapies for other complex or rare diseases, there is a real promise of new breakthroughs on the horizon.

About the author

Dr. Chris Cheadle is the Director of Research for Biology Products, Elsevier Life Sciences. He trained in Molecular Biology and earned a doctorate in Genetics from George Washington University, followed by a postdoctoral fellowship at the National Institute of Aging in the Microarray Core. He served as Director of the Lowe Family Genomics Core at the Johns Hopkins University from 2005-15. In his role at Elsevier, Chris has been responsible for helping to formulate the vision of the New Biology Strategy (NBS) expected to guide development for new products and make more effective use of present day assets in the future.