Persistent, in particular neuropathic pain affects millions of people worldwide. However, the response rate of patients to existing analgesic drugs is less than 50%. There are several possibilities to increase this response rate, such as optimization of the pharmacokinetic and pharmacodynamic properties of analgesics. Another promising approach is to use prognostic biomarkers in patients to determine the optimal pharmacological therapy for each individual. Here, we discuss recent efforts to identify plasma and CSF biomarkers, as well as genetic biomarkers and sensory testing, and how these readouts could be exploited for the prediction of a suitable pharmacological treatment. Collectively, the information on single biomarkers may be stored in knowledge bases and processed by machine-learning and related artificial intelligence techniques, resulting in the optimal pharmacological treatment for individual pain patients. We highlight the potential for biomarker-based individualized pain therapies and discuss biomarker reliability and their utility in clinical practice, as well as limitations of this approach.
Keywords: CSF; Pain; QST; SNP; biomarker; neuropathic pain; plasma biomarker.
Copyright © 2019. Published by Elsevier Inc.