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

Abstract

In a recent cluster analysis, it has been shown that patients with peripheral neuropathic pain can be grouped into 3 sensory phenotypes based on quantitative sensory testing profiles, which are mainly characterized by either sensory loss, intact sensory function and mild thermal hyperalgesia and/or allodynia, or loss of thermal detection and mild mechanical hyperalgesia and/or allodynia. Here, we present an algorithm for allocation of individual patients to these subgroups. The algorithm is nondeterministic-ie, a patient can be sorted to more than one phenotype-and can separate patients with neuropathic pain from healthy subjects (sensitivity: 78%, specificity: 94%). We evaluated the frequency of each phenotype in a population of patients with painful diabetic polyneuropathy (n = 151), painful peripheral nerve injury (n = 335), and postherpetic neuralgia (n = 97) and propose sample sizes of study populations that need to be screened to reach a subpopulation large enough to conduct a phenotype-stratified study. The most common phenotype in diabetic polyneuropathy was sensory loss (83%), followed by mechanical hyperalgesia (75%) and thermal hyperalgesia (34%, note that percentages are overlapping and not additive). In peripheral nerve injury, frequencies were 37%, 59%, and 50%, and in postherpetic neuralgia, frequencies were 31%, 63%, and 46%. For parallel study design, either the estimated effect size of the treatment needs to be high (>0.7) or only phenotypes that are frequent in the clinical entity under study can realistically be performed. For crossover design, populations under 200 patients screened are sufficient for all phenotypes and clinical entities with a minimum estimated treatment effect size of 0.5.

Figures

Figure 1.

Receiver Operating Characteristic analysis of the discriminatory power of the proposed algorithm to separate between patients with neuropathic pain and healthy subjects. Black line: full sensory testing, gray line: reduced protocol, using only warm detection threshold and mechanical pain sensitivity. The green dotted diagonal line indicates random classification (“coin flipping”). The area marked by dashed lines indicates the optimum ratio of sensitivity and specificity at 64% for probability for being healthy for full phenotyping (reduced phenotyping: 63%).

Figure 2.

Sensory phenotype probabilities and probability of being healthy for (A) n = 902 patients with neuropathic pain and (B) n = 188 healthy subjects. Gray line: probability for being healthy, blue line: sensory loss, red line: thermal hyperalgesia, yellow line: mechanical hyperalgesia. Subjects on the x-axis are sorted by their individual probability of being healthy. Dotted line: a phenotype with a probability over 64% should be considered relevant in the individual patient. Thirteen healthy subjects (7%) did not reach this criterion, 198 patients (22%) had profiles consistent with being normal.

Figure 3.

Sensory phenotype frequency and overlap between phenotypes for (A) diabetic polyneuropathy, (B) peripheral nerve injury, and (C) postherpetic neuralgia. Gray: healthy (H), blue: sensory loss (SL), red: thermal hyperalgesia (TH), yellow: mechanical hyperalgesia (MH). First bar (DET): deterministic algorithm (adds to 100%), 3 subsequent bars: probabilistic approach (a patient may be allocated to more than one phenotype, percentages are not additive). Bars are to scale, sizes of the circles in Venn diagrams and their overlaps are illustrative, not to scale.