An examination of the neuroimmune interactions of amitriptyline, spinal cord stimulation and opioid therapy in human neuropathic pain

Abstract

Background: The objective of this thesis was to explore mechanistic processes of amitriptyline, opioids and "Burst" spinal cord stimulation (SCS) for neuropathic pain in vivo. These three therapies are commonly employed for the management of neuropathic pain without a global agreed consensus with regard to mechanism of action. Whilst there is growing body of pre-clinical evidence supporting the concept that chronic neuropathic pain is associated with neuroimmune dysfunction and / or neuroinflammation, this has yet to be translated into the clinical domain. Epidural steroid injections have demonstrated an attenuation of neuroinflammation associated with a short-term reduction in pain. To demonstrate dynamic changes within the neuroimmune interface (neuronal-glia-immune cell communications) we compared the constituents of cerebrospinal fluid (CSF) before and after treatment with amitriptyline and Burst SCS. The mechanisms and side effects of opioids proposed for chronic pain are also related to neuroimmune interactions. Due to the fact many patients were taking opioids before enrolment, we compared the CSF of patients taking and not taking opioids prior to commencing the proposed interventional treatment. Not only would this provide data of confounding variables but give insight into opioid related phenomena and mechanisms. A more detailed characterisation of the effects of currently employed therapies may facilitate better stratification and phenotyping; as well as providing information which may enhance our understanding of the pathophysiology of neuropathic pain. Methods: This study involved taking cerebrospinal fluid samples before and after treatment and determining the cellular constituents of central peptides and immune cells. Patients were screened for neuropathic pain at outpatient clinics at St. James?s Hospital, Dublin and enrolled according to strict inclusion/exclusion criteria of the individual study. Lumbar radicular pain was the most common diagnosis for enrolled patients. Demographics and pains scores according to numerical rating scale (NRS) were recorded. After a baseline CSF sample the interventional treatment was commenced for 8 weeks. A second sample was subsequently taken after 8 weeks of treatment as well as outcome data. We compared the baseline and post intervention samples examining cellular data of T cells via flow cytometry, specific neuropeptides by ELISA and the proteome via Mass Spectrometry. In vitro work was also performed to examine the effect of amitriptyline on T cells directly. Results: In a cohort of patients with lumbar radicular pain who were deemed responders to amitriptyline, GO analysis identified immune system process as the most modulated proteins. This was associated with an attenuation of pro-inflammatory pathways, including the MAPK pathway identified by KEGG analysis. The chemokine eotaxin-1 was also significantly reduced in responders to amitriptyline. In-vitro work on T cells demonstrated an attenuation of the TH1/Th17 response and a modulation of other cytokine networks. This was largely in na?ve populations however postulating that peripheral and central actions of amitriptyline may differ. According to GO analysis proteins related to synapse assembly were the most modulated after Burst-SCS. This is consistent with neurophysiological data that suggests burst neuronal firing strengthens synaptic transmission and efficacy. Individual protein analysis also indicated that Burst SCS demonstrated potential supraspinal effects and modulation of immune effectors. GO analysis identified expression of proteins related to positive regulation of nervous system development and myeloid leucocyte activation in patients medicated with opioids. The largest decrease expression of proteins in the patients medicated with opioids related to neutrophil mediated immunity. Conclusion: Although effective, no treatment achieved full remission of symptoms in any patient. This would suggest that current therapies only target selective pathways involved in the pathophysiology of neuropathic pain. Although evidence of immuno-modulation was present with all therapies examined; this was not consistent. We have demonstrated that immune effectors or neuro-glial communications are likely contributory in the pathogenesis and attenuation of neuropathic pain. What is also evident was molecular evidence of neuroplasticity which involves many pathways and cells within the central nervous system. The chronicity of pain in humans is inherently complex but advances in technology with CSF analysis and neuroimaging in conjunction with sham/placebo-controlled trials will enable better phenotyping of patients and a better understanding of treatment mechanisms and the pathophysiology of neuropathic pain.