Andreas Goebel, PhD, is a reader in Pain Medicine, director of the Pain Research Institute, and an Honorary Consultant in Pain Medicine at the University of Liverpool, UK. After training in anesthesia and pain medicine in Germany and the UK, Goebel completed further training in post-trauma immunology at Harvard Medical School, Boston, US. His research focuses on the role of the adaptive immune system in chronic pain. Goebel recently took time to talk with Lincoln Tracy, a research fellow from Monash University, Australia, to discuss his path to post-trauma immunology, developing the concept of autoantibody pain for complex regional pain syndrome, and more. Read on for an edited transcript of their conversation.
What drew you to train in anesthesia and pain medicine?
I wanted to do internal medicine initially, or perhaps something to do with cancer. But when I was a student I read The House of God [by the American author and psychiatrist Stephen Joseph Bergman under the pen name of Samuel Shem] and realized that the realities he described, such as constantly finding himself in situations where there was nothing he could do for his patients, were exactly what was happening in our clinics. I vaguely remember him saying that a good specialty to avoid a lot of these issues was anesthesia. I thought about it, and it’s quite nice because you can instantly relieve pain, if you want. I also remember some of my relatives having chronic pain; that wasn’t what drove me to it, but I could see the impact it had on daily life.
Why did you then go on to train in post-trauma immunology?
After I graduated and started working in the pain clinic, the professor who led the unit often talked about the immune system having something to do with unexplained chronic pain. This was at the time when Linda Watkins published her studies on interleukin-1 [IL-1] and the vagus nerve, which was very exciting and inspired me to an extent. Not being able to find a clear explanation really fascinated me, and to pursue a research career I needed to find something to study. As a result, we decided that immunology might be a good field to go for.
What are the risk factors for complex regional pain syndrome (CRPS)?
We still don’t know a lot about the risk factors for CRPS, unfortunately. For the subset of patients who develop persistent CRPS, we know that having acute CRPS at a younger age and having a cold limb [with respect to temperature] are risk factors for the development of persistent CRPS. There is some very preliminary work suggesting that patients who go on to develop persistent CRPS are more anxious, but we don’t know if this is the chicken or the egg. It’s challenging to do proper prospective studies as around 85% of patients with CRPS get better. I see the 15%-20% of patients who don’t get better, and I think one of the main reasons why so little is known about the risk factors for persistent CRPS is because it’s a very rare disorder.
What therapeutic options are available for CRPS, and what are their limitations?
The most important option is exercise, which can improve function and may, to a degree, reduce pain. Exercise may also play a role in preventing CRPS. In a study led by Fiona Cowell and Sharon Gillespie, colleagues of mine here in Liverpool, they were able to prevent almost all cases of CRPS in patients with a distal radius fracture by using early exercise and simple measures such as removing casts where needed. We are trying to reproduce these results now in other centers. However, it’s not right to say that a patient develops CRPS because they didn’t move enough. There are other factors, many of which are still not very clear.
There’s no licensed drug for CRPS here in the UK. I sometimes offer drugs used in neuropathic pain patients, but very few patients display meaningful benefits over a longer period. We also offer a pain management program, which has psychological elements to limit how much the pain affects people. In the UK, the only recommended treatment for persistent CRPS is spinal cord stimulation. While it’s not a brilliant option, it can help to reduce the pain.
How did you develop the concept of autoantibody pain as a framework for understanding CRPS?
This originally stemmed from observations at the pain clinic at Würzburg University. Patients typically had all sorts of low-level increases in post-infectious antibodies. At some point I thought that maybe I needed to look at this, because I couldn’t see a pattern that explained these abnormal antibody profiles in a meaningful way. There were too many patients to claim that they had all experienced a variety of different infections.
After thinking about it for a long time, it became clear that if we thought there was something behind these observations, we needed to change our thinking. We had to look at whether there are pathogenic antibodies in CRPS that could cause the problem, rather than pursuing the idea of an underlying post-infectious syndrome.
Then in 2000, I read a paper from a neurology team which showed you can transfer antibodies produced in other diseases from men to mice. Suddenly, I felt super excited. I thought that this might allow us to find out about the pathogenicity we were seeing in our patients, and I thought we could transfer antibodies in order to cause pain as well.
How do the autoantibodies cause pain?
We aren’t completely sure how the autoantibodies cause pain in CRPS. One thing we know for sure is that there is no systemic inflammatory response, so they must belong to a special group of autoantibodies. But how do autoantibodies cause pain without causing major structural damage to nerves or systemic inflammatory issues?
One possibility is that the autoantibodies bind directly to the nerves and change their function, like an autoimmune channelopathy. We initially thought that this might be what happens in CRPS. But the other possibility is that the autoantibodies bind to cells that are close to the nerve, causing the cells to secrete something that excites the nerve. Again, this is an alteration of cell function rather than the unspecific inflammation typically associated with autoantibody binding. The binding could happen either peripherally, including in the dorsal root ganglia, or in the spinal cord.
In 2014 you presented a PRF webinar about the autoimmune causes of chronic pain. How has CRPS research evolved since then?
We probably have more evidence of what doesn’t work, I suppose. We probably know more about the brain abnormalities that occur in CRPS, and that they are probably not the underlying cause of the condition. When we first started publishing our work, everyone was talking about how CRPS is a primary brain disorder that only manifests after trauma. But the evidence that has come since then has largely shown this isn’t the case. Another thing that has changed since the webinar is that CRPS is now an orphan disease. This is important, as now industry has really become interested in the condition. It’s very encouraging.
You’ve previously written that drugs targeting the immune system may be effective in CRPS. What evidence is there to support this?
Finding immune drugs that are effective in persistent CRPS hasn’t proven to be successful thus far. We’ve tried intermediate-dose steroids, and we didn’t have much luck. People have tried strong tumor necrosis factor alpha [TNF-α] antagonists, which didn’t work either. We tried low-dose immunoglobulins, which were not effective, but theoretically they might be at a higher dose. We also found that mycophenolate, an immune-suppressing drug, may be effective but I suspect that the number needed to treat for long-term treatment is high.
We, and others, have found that plasma exchange may work. In 2016 the American Society for Apheresis listed CRPS as a new plasma exchange treatment indication as a result of these clinical trials. I am skeptical about the general pragmatic value of this treatment because the repeated small trauma needed to insert needles can unfortunately cause problems in CRPS. I also suspect that antibody levels need to be held down for relatively long periods of time to allow dissociation from their targets and healing processes to occur. Unfortunately, this is not always possible using this technology.
Together with Zsuzsanna Helyes, you recently published a study where you transferred human autoantibodies from CRPS patients into mice, which produced CRPS in the animals. Can you tell me more about this work?
This study was really a continuation of the research that we first published in 2014. For this study we wanted to produce a mouse model of CRPS that lasted longer than a day or two, which is what we have previously gotten from immunoglobulin G [IgG] injections over five days. Then, if the model worked, we wanted to try immune system-targeting drugs to explore the mechanisms by which the antibodies cause hypersensitivity, swelling, and so on.
And both worked well, which was a surprise. The animals remained hypersensitive in their injured paw through 14 days, which was the maximum we looked at, even though the peripheral, injury-related inflammation had already subsided by day seven post-injury. Interestingly, the injected CRPS-IgG did not enhance peripheral inflammation. Zsuzsanna really insisted on looking at the central nervous system, and she was right: We found quite a strong glial cell activation in the dorsal horn spinal cord, on the same side as the paw injury, which went all the way up to the brain. This activation occurred in a two-hit model: animals that had been injected with IgG from patients with CRPS before having their paw injured via a surgical incision. However, we still don’t know exactly where these antibodies bind.
The other thing we found is that you could prevent this activation with a drug that targeted the immune system. But, interestingly, you needed to use the IL-1 antagonist; using steroids was only temporarily effective. Data from our animal models suggest that anakinra, an IL-1 receptor antagonist, can reverse the hypersensitivity resembling CRPS. We’re now trying anakinra in an open-label study in humans.
However, it’s important to also acknowledge a paper we published in PAIN with David Andersson as co-senior author. This paper was important to us, as it was done in an independent lab that confirmed what Zsuzsanna and I had previously found – that the behavioral changes induced by IgG after injury are real. We also found that if you take IgG from patients with higher pain sensitivity and transfer it to mice, then the mice display stronger hypersensitivities than if you transfer IgG from patients with lower pain intensities. And we found that you can also transfer cold hypersensitivity, a common problem for patients with CRPS. To us, it’s amazing; it means you can reproduce clinical phenotypes in animals and confirms that the antibodies must be pathogenic.
What does the future hold for this line of research?
David Andersson and I have also looked at antibodies in fibromyalgia. We found that if you transfer IgG from people with fibromyalgia to mice, they become profoundly hypersensitive without needing to injure them. This suggests that there is a different mechanism in CRPS; the CRPS antibodies only work when an injury occurs. It looks like we will be able to apply some of the lessons learned from CRPS and fibromyalgia to other chronic primary pain conditions we see in the clinic.
And the lessons don’t stop there. Together with Camilla Svensson, we have been able to identify abnormal binding patterns of fibromyalgia antibodies to DRG cells. I had tried to find similar binding patters for the CRPS antibodies in the past and failed. It might be harder to identify the binding patterns in CRPS because the affected area is smaller compared to fibromyalgia. But now, given the success in the fibromyalgia work, we are looking at CRPS again in more detail.
Is there anything you would like to add?
I would like to acknowledge the many people who have supported me; my wife for support, discussion of these concepts over the years, and developing together with me the initial idea that we need to look for antibody pathogenicity.
Lincoln Tracy is a research fellow and freelance writer based in Melbourne, Australia. You can follow him on Twitter @lincolntracy.
Additional Reading
Autoantibodies produce pain in Complex Regional Pain Syndrome by sensitizing nociceptors.
Cuhadar U, Gentry C, Vastani N, Sensi S, Bevan S, Goebel A, Andersson D
Pain. 2019 Dec; 160(12):2855-2865.
Passive transfer of fibromyalgia pain from patients to mice.
Goebel A, Gentry C, Cuhadar U, Krock E, Vastani N, Sensi S, Sandor K, Jurczak A, Baharpoor A, Brieskorn L, Morado Urbina C, Sandstrom A, Tour J, Kadetoff D, Kosek E, Bevan S, Svensson CI, Andersson DA
bioRxiv, 2019 Jul 24. doi: 10.1101/713495
Helyes Z, Tékus V, Szentes N, Pohóczky K, Botz B, Kiss T, Kemény Á, Környei Z, Tóth K, Lénárt N, Ábrahám H, Pinteaux E, Francis S, Sensi S, Dénes Ádám, Goebel A
Proc Natl Acad Sci U S A. 2019 Jun 25; 116(26):13067-13076.
Cowell F, Gillespie S, Cheung G, Brown D.
J Hand Ther. 2018 Apr-Jun;31(2):201-205. PMID: 29706197
Tékus V, Hajna Z, Borbély E, Markovics A, Bagoly T, Szolcsányi J, Thompson V, Kemény A, Helyes Z, Goebel A
Pain. 2014 Feb; 155(2):299-308. Epub 2013 Oct 18.