Inside the Science: Biomarkers

Timely diagnosis in ALS is critical to ensure early access to therapies and the best standard of care. Delayed diagnosis can also mean missed opportunities to participate in clinical trials. 

Currently, we don’t have reliable diagnostic biomarkers yet, and as a result, ALS diagnosis is usually done by eliminating other possible diseases. However, research is advancing. Studies of asymptomatic individuals at high genetic risk (i.e. carrying a known ALS-causing gene variant), finding and examining bio banked tissues and blood collected before onset of disease, and work with laboratory cell and animal models of ALS are all providing greater insights into how disease starts and possible ways to identify it soon after symptoms begin, or hopefully someday, even earlier.  

Further insights into disease progression can lead to more personalized and informed care planning. It can also lead to more balanced clinical trial groups and a more accurate interpretation of results. For example, if a clinical trial does not consider prognostic biomarkers at the beginning, it can unintentionally create uneven groups when randomizing participants to treatment or placebo. This imbalance can lead to misleading results. If one group mostly includes participants whose disease progresses slowly and the other group has mostly fast progressors, it might appear that the treatment helped or harmed them. In reality, the difference could simply be due to how the disease naturally varies from person to person. This is a particular problem in small clinical trials that is very common in ALS. Without biomarker data, it’s difficult to accurately interpret trial outcomes or determine whether a therapy truly had an effect. 

Neurofilament light chain (NfL), a protein released when nerve cells are damaged, is being increasingly investigated for its reliability as an ALS prognostic biomarker, with value in predicting disease progression. Research shows that NfL levels are linked to how quickly ALS progresses, with higher levels usually meaning faster progression, and lower levels suggesting a slower course. Right now, this works well when looking at averages across groups of people in clinical trials. But we still need to learn more about how useful NfL might be for understanding or treating ALS in an individual person. Other things like age, body weight, and how the samples are collected can also affect NfL levels, so figuring all of this out will be key to making the most of this biomarker in both research and care. 

Other types of neurofilaments, primarily a phosphorylated form of neurofilament heavy chain (pNfH), are also being studied for their potential as biomarkers, although NfL remains the most predominant. Despite our knowledge that blood neurofilament levels increase before an ALS diagnosis can be made, they can’t be used on their own as a diagnostic biomarker because this increase also occurs in other neurodegenerative diseases and is not specific to ALS. However, in people at a high genetic risk, clinicians are currently assessing the possibility of tracking NfL levels as a measure of predicting when ALS symptoms will develop. A real-world example is the ATLAS clinical trial, which is studying Qalsody (tofersen) in people who carry a SOD1 genetic variant but have not yet developed ALS symptoms. In the initial phase, participants are closely monitored for changes in NfL. If NfL levels rise beyond a certain threshold, suggesting possible early disease activity, participants are randomized to receive either Qalsody or a placebo, to explore if early intervention can delay or prevent symptom onset. The results of this trial will provide greater insights into the use of NfL, and help clinicians guide early intervention.  

Everyone experiences ALS differently, as it varies significantly from person to person in terms of symptom onset, progression, and possibly underlying biology. In clinical trials, a predictive biomarker could help identify people with similar biological profiles, to see if the therapy being tested would specifically help that group. For example, drugs targeting gut-neuroinflammation might show a greater benefit for people who show higher levels of inflammatory proteins in the gut and fit that profile.  

Currently, we don’t understand the disease well enough to have a set of biomarkers that can effectively stratify participants in clinical trials. However, research is ongoing, and the field is working towards greater consideration of using predictive biomarkers in trials. An example is the PRELUDE trial, which is investigating lithium carbonate as a potential therapeutic for a subset of people with ALS. Several previous trials were negative when tested in a wide group of individuals with ALS, but further analysis suggested people with a specific genetic change in the UNC13A gene might respond better to the therapy. In this case, the UNC13A genetic change acts as a potential predictive biomarker for lithium carbonate efficacy. Other, larger efforts, like the HEALEY ALS MyMatch program are attempting to involve predictive biomarkers as a key focus of a clinical trial platform. 

Note: sometimes these can be called pharmacodynamic or response biomarkers. There are subtle nuances to each, but the basic premise of seeing how the drug changes biological mechanisms, applies to all. 

In ALS clinical trials, these biomarkers are important to verify if the drug is reaching the brain and achieving its intended effect. This can give us several insights: 

• The drug or its delivery method didn’t work: If a drug is targeting neuroinflammation specifically, and after the trial no participants show a decrease in inflammatory proteins in the brain and spinal cord, this means that the drug’s mechanism was not effective, or it couldn’t effectively cross the blood-brain barrier (BBB)*. Since it can be very difficult to measure changes in the brain and/or spinal cord of living individuals, often biofluids like blood or cerebrospinal fluid (CSF) are collected and analyzed to see if the drug worked the way it was supposed to. While looking for these changes don’t always answer the BBB question, they can help determine if the drug did something. 

• The pathway isn’t right for ALS: Sometimes a treatment affects the biological pathway it’s meant to target, like reducing inflammation, but still doesn’t slow down the disease. In that case, it may mean the pathway itself isn’t the right one to focus on for ALS and not the way to a cure. This kind of result is very useful for researchers because it helps avoid repeating studies that target the same ineffective pathway, saving time and resources. Therefore, with target engagement biomarkers, researchers get more insights into the right things to target in the brain, even if a drug doesn’t help symptoms or progression. This helps guide future research toward more promising directions. 

These biomarkers depend on the type of drug being tested, and through which mechanism it hopes to slow down disease progression. For some, there are emerging biomarkers available, while others are harder to assess. Pharmaceutical companies and investigators need to thoughtfully include them in every trial so we can learn from the results. However, as noted above, there should be care when interpreting changes in these target engagement biomarkers. A positive change in target engagement biomarkers should not be confused with a demonstration of treatment efficacy.  

* The blood-brain barrier (BBB) is a protective layer that controls what can pass from the bloodstream into the brain, allowing in essential nutrients while blocking out harmful substances like toxins and bacteria. While this barrier is crucial for keeping the brain safe, it also makes it difficult for most drugs to reach the brain, which is a challenge for treating ALS. 

Currently, these are still being developed and validated for ALS. The closest we have to a promising monitoring biomarker is also NfL, which is increasingly being used as a biomarker in trials to see if the drug has reduced damage in the neurons. In ALS, we know that NfL levels in blood are increased, so the concept of seeing those levels decrease after treatment suggests that less motor neurons are dying. In fact, the approval of Qalsody for SOD1-ALS was notably based on its ability to lower NfL levels, establishing it as a valid surrogate endpoint. The field is still working to understand how effective NfL levels in blood can be as an indicator of neurons being rescued by a treatment, given that there is still more to learn about whether disease can be slowed without decreasing NfL. Some researchers remain hesitant to discard a treatment if it doesn’t significantly alter NfL levels. 

One group that is “all in” on NfL as a tool to measure treatment efficacy is the team behind the EXPERTS-ALS ‘pre-trial’ platform in the UK. In this unique study, participants are only assigned to an active treatment (no placebo) and their NfL levels are measured before and after to see if they were lowered in blood below a certain threshold. Any treatment that is successful will be fast-tracked to a large double-blind, placebo-controlled clinical trial that measures if it will truly affect disease progression.