Identifying and characterising a cellular blood biomarker in treatment-resistant depression

Depression has an enormous impact on personal well-being, families and society. Current treatments sometimes control symptoms but many patients do not respond well to existing antidepressant drugs. We know that the immune system, the body's defence system, can affect mood and brain function. Immune cells circulate in the blood. If there is infection or damage, immune cells can be recruited to tissues including the brain. They also release chemical messengers called pro-inflammatory cytokines. These cytokines can affect mood - for example, you might have noticed feeling tired and unhappy when you have the flu. In some patients with depression, the levels of these cytokines are higher than normal, even though there is no infection. And that's especially the case for people with 'treatment-resistant depression', who do not recover with standard anti-depressants. This makes us think that, for some people, inflammation might be contributing to their low mood, and that if we treat the inflammation, it might help them to recover.

At the moment, however, there are no tests that doctors can use to detect whether the immune system is contributing to depression in a particular patient. Using the blood tests we have at the moment, the differences between healthy and depressed people are too small to be used as a biomarker (a biological test) of inflammation. We aim to develop a better biomarker, which doctors could use to identify patients who have an 'inflammatory' type of depression that requires a different treatment. This project will be led by Dr Mary-Ellen Lynall, a psychiatrist, working with Professor Ed Bullmore, a psychiatrist and neuroscientist, and Dr Menna Clatworthy, an immunologist.

We want to find out which of the many different kinds of immune cells in the blood are the 'culprit cells' responsible for the changes in inflammatory proteins and gene expression that we can see in depression. We are particularly interested in a group of immune cells called monocytes, because stress is a potent cause of depressive behaviour in humans and other animals, and in stressed animals we know that abnormal monocytes travel from the blood to the brain where they contribute to depressive behaviour. We predict that the same is true in humans.

To test this theoretical prediction, we will use a technique called RNA sequencing, which tells us which genes are 'expressed' (switched on). Most studies of blood gene expression in depression take all the blood immune cells as a group, then look at which genes are switched on, on average. Pooling cells together like this means that gene signals from abnormal cells can be swamped by signals from the unaffected cells. We think that by zooming in on particular types of cells, and even individual cells, we can find a stronger signature of inflammation in treatment-resistant depression. We will start out looking at gene expression in the simplest way, taking all the immune cells together. Next, we will focus on particular groups of cells, including monocytes. Finally, we will use cutting edge technology called 'single cell RNA sequencing' to analyse gene expression within individual cells.

Once we have found a cell-based signature of inflammation in depression, we aim to develop a simple, practical way of testing for it, that normal hospitals could use. We will then put this new biomarker to the acid test by seeing whether we can use it to predict whether patients with depression will respond to a new anti-inflammatory drug. To do this my fellowship will be closely linked to a trial of a new drug for treatment-resistant depression that will take place at the same time.

This project will help us to understand how the immune system is altered in some people with depression. The final goal is to develop a blood test that can help doctors to find new anti-inflammatory treatments for depression, and to tailor the treatment they give to each individual patient.

Background: There is an increasing appreciation that interactions between the immune system and the brain are clinically important. Animals and humans exposed to inflammatory insults demonstrate sickness behaviours that are similar to clinical features of major depressive disorder (MDD). Some patients with MDD have increased circulating levels of pro-inflammatory markers. Treatment resistant depression (TRD) is particularly associated with immune abnormalities including increased circulating CRP and IL6, and increased whole-blood IL1B and TNF gene expression. However, effect sizes from current whole-blood assays are too small for their use as clinical biomarkers in depression, possibly because the disease-related signals are diluted or obscured by the heterogeneous mixture of cells in peripheral blood.

Objectives: We aim to develop a cell-subset specific biomarker of inflammation in TRD, with a larger effect-size than current whole-blood markers. On the basis of animal models and preliminary human data, we hypothesise that patients with TRD have increased numbers of inflammatory, glucocorticoid-resistant monocytes in peripheral blood. We thus expect to find this biomarker within a monocyte subset.

Methods: To develop this biomarker, I will conduct a series of experiments. Initially, I will use cytometry and RNA sequencing data to investigate the peripheral immune profile of TRD at progressively finer scales of cellular resolution, from PBMCs to sorted cell subsets to single cells. Subsequently I will optimise a more pragmatic qPCR or cytometry biomarker of TRD, and test whether this new biomarker can predict treatment response in an upcoming trial of a novel anti-inflammatory drug for TRD.

Opportunities: This proposal leverages newly available datasets and the latest cellular processing technologies to advance the development of biomarkers, and to provide mechanistic insight into the cellular changes underlying the inflammation associated with TRD.

Our research aims to develop a biomarker of inflammation in treatment-resistant depression (TRD). There are currently no biomarkers used to guide treatment for any psychiatric disorder. A mechanistically specific, peripherally accessible biomarker for an inflamed sub-group of patients with treatment-resistant depression would transform the prospects for therapeutic progress in this area.

Our work could therefore potentially benefit:
1) Patients with depression. The World Health Organization estimates that, globally, 350 million people suffer from depression, which is the second leading cause of disability worldwide. In the UK, in 2013-2014, work-related stress, depression and anxiety accounted for 39% of all work-related illnesses. Depression has a profound negative influence on individual quality of life and impacts family members and carers. Current treatments are ineffective in one third of patients. At present, such patients typically take multiple monoaminergic antidepressants, each for many weeks, before alternative treatments are offered, with an attendant long duration of distressing symptoms and impact on their personal, professional and family life. A blood biomarker for TRD would lead to more effective, person-centred care for patients with depression. A biomarker would allow earlier identification of patients unlikely to respond to monoaminergic antidepressants, and hence earlier consideration of alternative treatments. This would allow patients to have better control of their condition and improved quality of life. This would also impact their families and carers.
2) Patients participating in our study, and other patients with mood disorders, by increasing their understanding of the science underpinning depression.
3) Society, by enabling patients suffering from mood disorders to realise their full potential, both personally and professionally.
4) Health care providers and the economy. Depression has a substantial health-economic impact. The UK Department of Health estimated (2007) that the cost of depression to England alone - including health service use, informal care and lost productivity - amounted to £20-24 billion per year. The NHS, social care and welfare system could benefit from significant savings if a biomarker could be used to tailor treatments according to likely response, decreasing the overall burden of depressive disorders and reducing the high costs associated with TRD.
5) The pharmaceutical industry and other organizations conducting clinical trials. The development of therapies for depression is hampered by disease heterogeneity. By identifying blood-based biomarkers for the phenotypic stratification of depressive disorders, our results could inform the design of more targeted clinical trials for both psychological and biological treatments for depression. Providing validated immune biomarkers could stimulate industry interest in re-purposing anti-inflammatory drugs for depressive disorders in a precisely defined subgroup of patients.
6) Medical researchers investigating other conditions. Our methods for peripheral blood biomarker identification are methodologically innovative. Our technical work and results will inform research strategies in other diseases with an immune component, or where the tissue of primary interest is inaccessible to sampling (especially other brain-based disorders).
7) The general public. A clear immune biomarker which guides treatment for depression could help to reduce the considerable stigma still associated with psychiatric disease.

Timescale: We aim to develop a biomarker and test it to the point of first validation in an independent dataset by the end of the fellowship period. If this is successful, the biomarker could rapidly be translated for clinical use in the UK and internationally within a few years. We anticipate that the economic and societal benefits from this and related research will occur within our professional lifetimes.