Can Blood Test Determine Type of Multiple Sclerosis?

Research Summary: Kynurenine pathway metabolomics predicts and provides mechanistic insight into multiple sclerosis progression

A recent Australian research breakthrough in multiple sclerosis (MS) has drawn a lot of attention from the media. In these articles, the discovery has been described as something that will allow for people with MS to have a simple blood test to allow for diagnosis of their specific type of MS.  Additionally, it has been reported that it will provide much quicker feedback on whether medications are effective in individuals.

Considering the interest in this finding, we have gone through and analysed the journal article directly.  Below, we have broken this down into ‘Why’ and ‘What’, as well as a section on further questions that we think need to be addressed.

Why is this study important?

There were two main aims of the study.  Firstly, it is known that around 50% of people who are diagnosed with relapsing-remitting MS (RRMS) will transition to secondary progressive MS (SPMS) within 10-15 years.  As it is still unknown why some people progress and others don’t, this study aimed to find possible triggers for this switch.  As well as this, debate still exists as to the relevance of these different subtypes of MS, so improving our understanding of what is happening in these stages is important to determine how clinically important these defined phases are.

Secondly, it’s also thought that a shift happens in the disease mechanism between RRMS and progressive forms of MS.  Through analysing metabolic pathways that may be involved in these processes, the researchers hoped to potentially identify a stage-specific biomarker.

A biomarker is a molecule that can be used to identify a disease or stage of a disease (e.g. something that distinguishes RRMS from progressive MS).  You can think of this as a little bit like world landmarks.  For example, if you see the Eiffel Tower, you know you are in Paris.  If you see the Opera House, you know you are in Sydney.  By identifying this ‘landmark’ in the blood of specific types of MS, it is hoped that it would lead to being able to provide people with a better prognosis, as well as improve treatment decisions and monitoring.

What did they find?

The study focussed primarily on a metabolic pathway known as the kynurenine pathway.  This process involves the breakdown of an amino acid called tryptophan to NAD+ (which is important for energy production).  This process is of interest in MS because it is known to be influenced by inflammation and molecules made along the pathway are known to have either neuroprotective or neurotoxic effects.

Image: taken from the published article (linked at the bottom of the page).

The figure above outlines the results found as part of this research.  The main part to focus your attention on is the fact that the levels of different molecules in the pathway are altered in MS compared to healthy controls (HC) and that differences can also be observed between the different types of MS (RR, SP and PP).  The authors have also highlighted which steps in the process are either being increased or decreased and have provided some indicators as to how these changes could lead to symptoms in MS (e.g. mood abnormalities, neurodegeneration and energy depletion).

Considering the differences that they saw in those results, it was then of interest to see whether or not these molecules could be used as a biomarker for MS-type prediction.  The results from this part of the study can be seen in the figure below.

Image: taken from the published article (linked at the bottom of the page).

Whilst this may look a little confusing, we will focus on a few key aspects.  These are:

Part B:  this highlights the process of developing the ‘decision tree’ for making these predictions.  At the top, you can see the pie chart with the breakdown of participants – healthy controls (green), RRMS (yellow), SPMS (orange) and PPMS (red).  The purpose is to use the levels of the different molecules to end up with a ‘pie’ of each colour.

Let me explain this in a different way.   As an example, assume you have a bowl of bananas, grapes, apples and oranges and you want to find a way of getting someone to separate them into each fruit, so you give the following instructions:

Part C:  the prediction model that was generated was then applied to the original cohort of participants to see how accurately it could determine the group that each belonged to.  As you can see from the results, the researchers achieved a reasonably high level of sensitivity and specificity for all groups, although the SPMS group only had a 75% correct prediction rate.

Part D:  the prediction model was then used to perform the same analysis on a different, smaller cohort of samples that included healthy controls, as well as people with RRMS and SPMS.  As you can see, the sensitivity and specificity were decreased in this group, though this will be partly influenced by the quite small numbers of each group.

What questions remain?

This section is devoted to my thoughts and analysis of the paper and so is purely my opinion.  Firstly, I think the research and outcomes are interesting.  The pathway involved and the results seen provide some really interesting links to observed symptoms in people with MS, such as the decreased serotonin levels linking to mood issues and the decreased NAD⁺ leading to energy depletion.

I think the main question for me is how specific is this to MS and the different disease stages.  This pathway has been implicated in a number of different diseases, such as Alzheimer’s disease, cancer, schizophrenia, ALS, depression and Huntington’s disease.  This is not necessarily surprising when coupled with understanding the aspects that can alter this pathway, such as inflammation.

In this study, an important factor stands out to me in terms of the information provided about the participants.  On average, the people with progressive forms of MS had been living with the disease for approximately 10 years longer than those with RRMS.  There are some unavoidable reasons why this is the case, in particular for the SPMS group that have had to go through that ‘transition’ period.  However, this fact does lead me to one big question:

“Are the differences seen related to the type/stage of MS (i.e. RRMS, SPMS or PPMS) or more a result of disease duration?”

I believe this question is important when it comes to discussing these findings as a test for prognosis.  To be useful in this space, in particular to differentiate between RRMS and PPMS, these differences would need to be apparent at very early stages and not many years after diagnosis.  As it stands, it could be a good way of monitoring people with RRMS to try and check for when they transition to SPMS, but again this would depend on whether the levels of these biomarkers change around this timepoint or if it happens later in the disease process.  The researchers may already know the answer to this question, however, it is not something that I could see addressed in the journal article.

The other question relates to how this is applied on an individual basis.  It can be seen that when you use average numbers, differences can be observed.  However, quite a bit of variation exists between participants within each group.  As an example, imagine you are someone with RRMS who has a higher than expected level of the tested biomarker…do you get told you have SPMS?  If so, this has clear implications for decisions regarding your therapy.  This will be one of the complications in terms of taking this from a research study to a clinically applicable tool.

Interestingly, it should be noted that this aspect is probably not the key focus of the journal article.  Instead, the authors discuss the fact that trying to target this pathway and ‘rebalance’ the levels of these molecules could be a useful treatment option to slow neurodegeneration in people with MS.  There are already some therapies that work in this manner, most notably teriflunomide (Aubagio) and laquinimod (still in clinical trials).

In conclusion, it is great to see such exciting MS research coming out of Australia.  It will be really interesting to see how this develops over the next couple of years, as both a tool for prognosis and to help better understand the disease processes in MS and how they can be better treated.  We will be contacting the researchers involved in the hope of conducting an interview or Q&A to pose some of these questions to them.

Those interested in reading the published article in full, can do so here.