Elsevier

The Lancet Neurology

Volume 11, Issue 11, November 2012, Pages 999-1005
The Lancet Neurology

Rapid Review
Subgrouping of patients with neuropathic pain according to pain-related sensory abnormalities: a first step to a stratified treatment approach

https://doi.org/10.1016/S1474-4422(12)70189-8Get rights and content

Summary

Background

Patients with neuropathic pain present with various pain-related sensory abnormalities. These sensory features form different patterns or mosaics—the sensory profile—in individual patients. One hypothesis for the development of sensory profiles is that distinct pathophysiological mechanisms of pain generation produce specific sensory abnormalities. Several controlled trials of promising new drugs have produced negative results, but these findings could have been a result of heterogeneity in the patient population. Subgrouping patients on the basis of individual sensory profiles could reduce this heterogeneity and improve trial design.

Recent developments

A statistical categorisation of patients with neuropathic pain showed that subgroups of patients with distinct sensory profiles who perceive their pain differently do exist across a range of neuropathic disorders, although some distinct disorder-specific profiles were also detected. Results of the first clinical trials to use the subgroup approach at baseline could show a superior effect of the study drugs in specific subgroups, rather than in the entire cohort of patients.

Where next?

A new classification of neuropathic pain should take into account subgroups of patients with different sensory profiles. Sensory phenotyping has the potential to improve clinical trial design by enriching the study population with potential treatment responders, and might lead to a stratified treatment approach and ultimately to personalised treatment.

Introduction

Several controlled trials of drugs for neuropathic pain have produced negative results despite encouraging results from preclinical and early clinical studies.1, 2 But do such negative outcomes really show that these drugs are not efficacious, or were positive outcomes obscured by the heterogeneity of the study groups? Study drugs might be more efficacious in some subgroups than in the whole cohort. One option to reduce heterogeneity in a study cohort is to classify patients into more homogeneous subgroups before administering the study drug—a strategy that is referred to as stratified or personalised therapy. To undertake this approach, characteristics that make an individual or a subgroup of patients potentially more responsive to a specific treatment need to be identified.

Results from animal laboratory studies suggest that neuropathic pain is generated by several operating mechanisms working in concert.3 The specific pattern of mechanisms in an individual might cause a different treatment response. Thus, subgrouping of patients based on individual pathophysiological mechanisms could be a successful way to tailor treatment to the individual. Such mechanisms, however, cannot be readily examined in patients, although one hypothesis states that a surrogate marker (ie, the individual's sensory phenotype) might be indicative of the underlying pathophysiology of afferent processing. Since one specific symptom can be based on various completely disparate pathophysiological mechanisms, a specific sensory pattern, comprising spontaneous perceptions as well as negative and evoked sensory phenomena, might offer the best clue to the underlying mechanisms.

The aim of this Rapid Review is to summarise the methods that can be used to subgroup patients with neuropathic pain according to pain-related sensory profiles (table) and to critically discuss the value of the different assessment methods used to identify these subgroups. We will address the following questions: What are the most appropriate assessment methods for identification of sensory phenotypes in clinical trials? Do subgroups of patients with distinct sensory phenotypes exist? Which pain mechanisms might be linked to specific phenotypes? Is there evidence to show that a drug performs better in a subgroup of patients than in the entire cohort?

Section snippets

What are the best methods for identification of sensory phenotypes in clinical trials?

Pain-related sensory abnormalities can be assessed in several ways. Self-assessment methods (patient-reported outcomes) consist of a list of specifically designed questions addressing the quality and intensity of sensory symptoms perceived by patients. The same symptoms can be captured by the investigator in an interview format. Furthermore, sensory signs can be assessed by clinical examination of the affected skin area using various sensory stimuli.

Do subgroups of patients with distinct sensory phenotypes exist?

So far, PainDETECT, NPSI, StEP, and QST have been used in several neuropathic pain studies to subgroup patients according to their pattern of sensory abnormalities. These studies used a hierarchical cluster analysis or factor analysis in combination with multiple correspondence analysis for segmentation. This approach identifies patterns or dimensions of sensory symptoms that occur most frequently, without using a-priori hypotheses or other predefined assumptions.

PainDETECT identified five

Which pain mechanisms might be linked to specific phenotypes?

The proposed concept hypothesises that pain-related sensory deviations and (even more so) the individual sensory mosaic might be indicative of the underlying pathophysiological mechanisms of pain generation. Assuming that this is true, several statistical segmentation strategies have been used to discern subgroups of patients displaying distinctive sensory profiles, which have allowed some speculation about mechanisms and potential drug responses. In a large group of patients with different

Is there evidence that a drug performs better in a subgroup of patients than in the entire cohort?

So far, no large studies have used prospective phenotyping as an inclusion criterion. However, several encouraging trials have been published.26, 32, 33, 34, 35, 36, 37, 38, 39 In patients with traumatic nerve injury and postherpetic neuralgia given intracutaneous botulinum toxin, a good outcome correlated with the preservation of cutaneous innervations, documented by thermal thresholds at baseline.26 By contrast, response to systemic opioids correlated with a higher heat pain threshold in

Summary and future perspectives

Negative results of clinical trials of drugs for neuropathic pain could be the result of flawed study designs rather than the lack of efficacy of the study drugs. Thus, improvements in classification schemes and study designs are very important, otherwise drugs might not reach the patient despite being highly efficacious in defined subgroups of patients. The future challenge is to identify appropriate subgroups. Since no definite biomarkers have been identified to define individual pain

Search strategy and selection criteria

We searched PubMed for all studies of human beings, published in English, with the combined search terms “neuropathic pain”, “subgrouping”, “subgroups”, “questionnaire”, “QST”, “sensory phenotype”, “sensory profiles”, “mechanism-based”, “personalized”, “individualized”, “treatment”, and “therapy” from Jan 1, 2010, to July 16, 2012. We also identified relevant papers through searches of our files. We generated the final reference list on the basis of relevance to the scope of this Rapid Review.

References (41)

Cited by (219)

  • Angiotensin and pain

    2023, Angiotensin: From the Kidney to Coronavirus
View all citing articles on Scopus
View full text