Stratifying patients with peripheral neuropathic pain based on sensory profiles: algorithm and sample size...

Area of Science:

• Neurology
• Pain Medicine
• Quantitative Sensory Testing

Background:

• Peripheral neuropathic pain (PNP) patients exhibit distinct sensory phenotypes.
• Previous cluster analysis identified three primary sensory profiles in PNP.

Purpose of the Study:

• To introduce a novel algorithm for classifying individual PNP patients into established sensory phenotypes.
• To evaluate the algorithm's performance in differentiating PNP patients from healthy controls.
• To determine phenotype frequencies across different PNP etiologies and propose sample sizes for stratified studies.

Main Methods:

• Development and application of a nondeterministic algorithm for patient-phenotype allocation.
• Validation of the algorithm using quantitative sensory testing (QST) data.
• Frequency analysis of sensory phenotypes in patients with painful diabetic polyneuropathy, peripheral nerve injury, and postherpetic neuralgia.

Main Results:

• The algorithm successfully distinguished neuropathic pain patients from healthy subjects with 78% sensitivity and 94% specificity.
• Phenotype frequencies varied significantly across etiologies: sensory loss was most common in diabetic polyneuropathy (83%).
• Recommended screening population sizes for phenotype-stratified studies were proposed based on phenotype prevalence and study design.

Conclusions:

• The developed algorithm provides a reliable method for classifying PNP patients into sensory phenotypes.
• Understanding phenotype distribution is crucial for designing effective, phenotype-stratified clinical trials in neuropathic pain.
• The findings support the potential for personalized treatment approaches based on distinct sensory profiles in neuropathic pain.