Biomarkers of Pain: Quantitative Sensory Testing, Conditioned Pain Modulation, Punch Skin Biopsy

Introduction

Identification of potential biomarkers for chronic pain is essential for correct diagnosis, prediction, and evaluation of treatment response. A biomarker is defined as a “characteristic objectively measured and evaluated as an indicator of normal biologic processes, pathogenic processes, or pharmacologic responses to a therapeutic intervention.” The establishment of a new biomarker should follow a standardized process and requires the definition of the exact objective, i.e. what should be measured. The United States Food and Drug Administration (FDA) has defined seven types of biomarkers that can be used to assess different categories: susceptibility/risk, diagnostic, monitoring, prognostic, predictive, pharmacodynamic/response, safety ( Table 21.1 ).

TABLE 21.1

Categories Of Biomarkers According to the Biomarkers, EndpointS and other Tools (BEST) Glossary of the United States Food and Drug Administration (FDA)

Category	Definition
Susceptibility/risk	potential for developing a disease in an individual without clinically apparent disease
Diagnostic	detect/confirm the presence of a disease or condition of interest or identify individuals with a subtype of the disease
Monitoring	assess the status of a disease or medical condition or for evidence of exposure to (or effect of) a medical product or an environmental agent
Prognostic	identify the likelihood of a clinical event, disease recurrence, or progression in patients who have the disease of interest
Predictive	identify individuals who are more likely than similar individuals without the biomarker to experience a favorable or unfavorable effect from exposure to a medical product or an environmental agent
Pharmacodynamic/response	show that a biologic response has occurred in an individual who has been exposed to a medical product or an environmental agent
Safety	measured before or after an exposure to a medical product or an environmental agent to indicate the likelihood, presence, or extent of toxicity as an adverse effect

The correct diagnosis is a prerequisite prior to starting a suitable pain therapy. It is, for example, important to distinguish between chronic neuropathic and chronic nociceptive pain, as the treatment of these entities would be approached differently. While nociceptive pain, i.e. pain that arises from damage to non-neuronal tissue with activation of nociceptors, is likely to respond to non-steroidal anti-inflammatory drugs such as naproxen or ibuprofen, or in more severe cases opioids, treatment of neuropathic pain, i.e. pain as a direct consequence of a lesion or disease of the somatosensory system, is more challenging and requires the use of different co-analgesics, e.g. anti-convulsants, anti-depressants.

Many patients with chronic pain do not achieve adequate pain relief with available drugs. With regard to chronic neuropathic pain, the number needed to treat for a 50% pain reduction for recommended first-line agents, i.e. tricyclic anti-depressants, selective serotonin-norepinephrine reuptake inhibitors, and calcium alpha two delta agonist anti-convulsants, ranges from 3.5 to 7.7. In addition, many clinical trials involving newer agents have failed to show any treatment effect. Both inadequate pain relief with already available drugs and the failure of clinical trials involving newer agents might be explained by the heterogeneous clinical picture and the numerous underlying pathophysiologic pain mechanisms. A priori stratification of patients using a valid predictive biomarker could help identify patients who would benefit the most from a specific drug. In addition, to modify current treatment and/or initiate further measures, it is important to forecast disease progression.

To date, no valid biomarker has been identified for use in the assessment and treatment of chronic pain. However, the Initiative on Methods, Measurement, and Pain Assessment in Clinical Trials (IMMPACT) has recently considered three promising biomarkers for use in the development of analgesic treatments, i.e. sensory testing, skin punch biopsy, and brain imaging.

In this chapter, we will focus on sensory testing, including “static” quantitative sensory testing (QST) and “dynamic” conditioned pain modulation (CPM), as well as skin punch biopsy and their role as susceptibility, diagnostic, prognostic, predictive, and pharmacodynamic biomarkers. Although sensory testing does not completely correspond to the definition of a biomarker as an objective measurement because of its psychophysical nature, it is discussed here as there are promising hints for its use as a biomarker, and at least for QST, there exists a standardized protocol with reference values.

Quantitative Sensory Testing

Sensory testing of chronic pain patients has become more relevant recently, as the heterogeneous clinical presentation of pain patients has become increasingly recognized.

Concerning neuropathic pain, patients with the same pain etiology can present with various symptoms and signs, for example, the simultaneous presence of mechanical allodynia and thermal hypesthesia in the same patient with postherpetic neuralgia. In contrast, different pain etiologies can share similar somatosensory abnormalities. The different somatosensory abnormalities are thought to reflect the underlying pathophysiologic mechanisms in chronic pain patients. For example, mechanical hyperalgesia, i.e. enhanced pain to punctate stimuli and dynamic allodynia, are thought to reflect central sensitization mechanisms, whereas increased sensitivity to heat can be related to primary hyperalgesia, i.e. peripheral sensitization of nociceptive afferents.

Up to now, treatment of chronic pain has been based on the underlying etiology. However, current recognition of the heterogeneous clinical presentations of chronic pain has led to the concept of a mechanism-based therapy with stratification and treatment of patients according to the underlying pathophysiologic mechanisms. An exact classification of patients according to their somatosensory profile is necessary to realize this concept and might lead to improved treatment response.

Diagnostic

QST allows for the quantification of the somatosensory function of the peripheral and central nervous system, including non-nociceptive (Aβ-) and nociceptive fibers (Aδ- and C-fibers) by applying calibrated mechanical (e.g. graded von Frey hairs) and thermal stimuli to the skin, whose intensity or painfulness are evaluated by the patient.

Many different QST procedures, mainly thermal testing procedures, have been used in the past to assess sensory function, especially in patients with neuropathic pain. One advantage of QST is the ability to assess the entire somatosensory system, including both large and small nerve fibers, in contrast to conventional electrophysiology (electromyogram/nerve conduction studies), which primarily assess large nerve fibers. Therefore QST has been used for the early diagnosis of diabetic polyneuropathy showing abnormal vibration and thermal detection thresholds. Compared to clinical examination or nerve conduction studies, the sensitivity of QST ranges from 27%–98% for cold or 58%–84% for vibration in diabetic neuropathies. Pure small fiber neuropathies are characterized by an isolated lesion of thin myelinated Aδ-fibers or unmyelinated C-fibers. Affected patients complain about burning or electric shock-like pain, tingling, pins and needles as well as impaired thermal perception. While nerve conduction studies are per definition normal in these entities, QST shows abnormal thermal thresholds indicative of small nerve fiber damage. Therefore QST has been recommended as a diagnostic tool to screen for cold and warm deficits in patients with suspected neuropathy with small fiber involvement, e.g. early diagnosis of diabetic neuropathy or diagnosis of small fiber neuropathy.

In addition, altered sensory function was shown to be present in different neuropathic pain etiologies such as postherpetic neuralgia (PHN) or central pain. Compared to reference data of 162 healthy controls, almost half of patients with PHN are characterized by dynamic mechanical allodynia. In addition, patients frequently exhibit abnormal loss to thermal or mechanical detection parameters. QST abnormalities were also found in pain conditions other than neuropathic pain, such as fibromyalgia, chronic low back pain, osteoarthritis, and migraine. A meta-analysis of 41 studies using QST for phenotyping patients with osteoarthritis showed that decreased pressure pain thresholds at both affected and unaffected sites could differentiate between patients and healthy controls. Many studies used sensory testing in fibromyalgia patients. Affected patients were shown to be characterized by altered thermal thresholds for heat and cold pain. ^, In addition, increased sensitivity to pressure pain is a common sign in fibromyalgia patients.

The correct interpretation of sensory abnormalities in pain patients requires a standardized testing protocol and comparison to reference data of healthy controls. In 2006, the German Research Network on Neuropathic Pain (DFNS) established a standardized QST protocol to analyze the exact somatosensory phenotype of neuropathic pain patients. This protocol consists of seven tests to assess 13 thermal and mechanical parameters, i.e. cold and warm detection thresholds (CDT, WDT), thermal sensory limen (TSL), paradoxical heat sensations (PHS), cold and heat pain thresholds (CPT, HPT), mechanical detection threshold (MDT; von Frey filament 0.25-512 mN), mechanical pain threshold (MPT, pinprick stimulators, 8-512 mN), mechanical pain sensitivity and dynamic mechanical allodynia (MPS and DMA; pinprick stimulators, standardized brush, cotton wisp, cotton wool tip), wind-up ratio (WUR), vibration detection threshold (VDT; Rydel-Seiffer graded tuning fork, 64 Hz, 8/8 scale), pressure pain sensitivity (PPT; pressure gauge device) ( Fig. 21.1 ).

A database with reference values for different body regions (face, dorsum of the foot, dorsum of the hand, trunk) was implemented ^, that allows direct comparison of patients with age- and sex-matched healthy controls. For this purpose, Z-transformation of every single parameter was performed (except for DMA, PHS), resulting in a QST profile where all parameters are presented as standard normal distribution independent of the different units. Z-values of zero represent a value corresponding to the mean of healthy controls. Z-values below zero indicate a loss of function, i.e. lower sensitivity compared to controls (hypoalgesia, hypoesthesia). Z-values above zero indicate a gain of function, i.e. higher sensitivity (hyperalgesia, hyperesthesia). A z-score of 0±1.96 represents the range that can be expected to include 95% of healthy control subject data (95% confidence interval). Z-values outside this 95% confidence interval are considered as absolute abnormalities (<–1.96 abnormal loss, >1.96 abnormal gain). Investigation of a large cohort of patients with different neuropathic pain symptoms revealed the presence of different combinations of loss and gain across all investigated pain etiologies. QST cluster analysis in patients with peripheral neuropathic pain of different etiology revealed the presence of three subgroups of patients that were characterized by different sensory profiles ( Fig. 21.1 ). The first cluster, “sensory loss,” was characterized by a loss of small and large fiber function in combination with PHS; the second cluster, “thermal hyperalgesia,” showed preserved sensory functions in combination with heat and cold hyperalgesia and mild dynamic mechanical allodynia; and the third cluster, “mechanical hyperalgesia,” was characterized by a loss of small fiber function in combination with pinprick hyperalgesia and dynamic mechanical allodynia. The three clusters were distributed across all included neuropathic pain etiologies. However, with different frequencies, e.g. the sensory loss cluster was the most frequent cluster in patients with polyneuropathy (about 51.8%), whereas patients with PHN were especially characterized by mechanical hyperalgesia, i.e. cluster three (46.6%). Interestingly, similar clusters were identified in human surrogate pain models, which supports the concept of a mechanism-based stratification approach.

Figure 21.1, Subgroup analysis of neuropathic pain. (Adapted from Baron et al. 19 ). Sensory profiles of the three clusters presented as mean z-scores ± 95% confidence interval for the test data set (n = 902). Note that z-transformation eliminates differences because of test site, sex, and age. Positive z-scores indicate positive sensory signs (hyperalgesia), whereas negative z-values indicate negative sensory signs (hypoesthesia and hypoalgesia). Dashed lines : 95% confidence interval for healthy subjects (−1.96 < z < +1.96). Note that if the mean of a cluster is within the shaded area , this does not imply that it does not differ from a healthy cohort. Values are significantly different from those of healthy subjects if their 95% marked interval does not cross the zero line. Insets show numeric pain ratings for dynamic mechanical allodynia (DMA) on a logarithmic scale (0-100) and frequency of paradoxical heat sensation (PHS) (0-3). Blue symbols: cluster one “sensory loss” (42%). Red symbols: cluster two “thermal hyperalgesia” (33%). Yellow symbols: cluster three “mechanical hyperalgesia” (24%). CDT , cold detection threshold; CPT , cold pain threshold; HPT , heat pain threshold; MDT , mechanical detection threshold; MPS , mechanical pain sensitivity; MPT , mechanical pain threshold; NRS , Numerical Rating Scale; PPT , pressure pain threshold; QST , quantitative sensory testing; TSL , thermal sensory limen; VDT , vibration detection threshold; WDT , warm detection threshold; WUR , wind-up ratio.

According to the updated grading system for neuropathic pain proposed by the International Association for the Study of Pain (IASP) Special Interest Group on Neuropathic Pain (NeuPSIG), a confirmation of sensory signs is necessary to raise the certainty of neuropathic pain presence from “possible” to “probable,” which can be sometimes only obtained using QST. However, only a few studies have examined the sensory profile in painful versus painless neuropathy so far. As their results showed similar thermonociceptive impairment in both cases, QST seems limited in its ability to differentiate between neuropathy without pain and neuropathic pain.

Prognostic

There is evidence that QST has a prognostic value. For example, thermal deficits as indicative of small nerve fiber loss were associated with painfulness of diabetic neuropathy, while mechanical deficits were shown to predict the development of foot ulcerations. The prognostic value of QST was also demonstrated in a group of patients with colorectal cancer who were examined by QST before receiving oxaliplatin-based chemotherapy and were then followed up for 26 weeks. Pretreatment touch sensation deficits were found to predict severity and painfulness of oxaliplatin-induced neuropathy.

Susceptibility/Predictive

In addition, QST seems to be a promising susceptibility and predictive biomarker for chronic pain (for definition, see Table 21.2 ). A recent prospective study aimed to identify potential risk factors for PHN via characterization of the sensory profile and physical and psychosocial functioning during acute herpes zoster. The sensory profile at the affected site and a distal contralateral control site (four dermatomes above or rostral) of 74 patients with acute herpes zoster was compared to 20 healthy controls. Patients with PHN at the six-month follow-up point, defined as persisting pain in the affected area (22.4%), were then compared to those without PHN. Besides older age, higher pain intensity, reduced quality of life, and physical functioning during acute herpes zoster, the development of PHN seems to be predicted by sensory alterations. The authors concluded that mechanical hyposthesia and/or hyperalgesia/mechanical allodynia combined with preserved thermal detection, especially at body sites distal to the acute herpes zoster area, may be potential risk factors for PHN.

TABLE 21.2

Quantitative Sensory Testing Parameters and Suspected Location of the Underlying Pathophysiologic Mechanisms

QST Parameter	QST Testing Device	Suspected location of an underlying pathophysiologic mechanism		QST Finding
QST Parameter	QST Testing Device	Peripheral	Central	QST Finding
Thermal
Detection threshold for innocuous thermal stimuli (cold, warm)	Thermotest	Aδ (cold) C (warm)	Spinothalamic	Increased thermal detection threshold, i.e. cold/warm hypoesthesia Decreased thermal detection threshold, i.e. cold/warm hyperesthesia
Detection of the threshold for noxious thermal stimuli (cold, heat)	Thermotest	(Aδ) C	Spinothalamic	Increased thermal pain threshold, i.e. cold/heat hypoalgesia/analgesia Decreased thermal pain threshold, i.e. cold/heat hyperalgesia
Paradoxical heat sensations	Thermotest		Central disinhibition	Heat sensation as a response to a cold stimulus
Mechanical
Detection threshold for innocuous mechanical stimuli (static light touch)	Calibrated von Frey hairs	Aβ	Lemniscal	Increased mechanical detection threshold, i.e. mechanical hypoesthesia Decreased mechanical detection threshold, i.e. mechanical hyperesthesia
Detection threshold for innocuous vibration	Graduated tuning fork	Aβ	Lemniscal	Increased vibration detection threshold
Detection threshold for noxious mechanical stimuli (pinprick)	Calibrated needles, e.g. pinprick stimulators	Aδ	Spinothalamic	Increased mechanical pain threshold, i.e. pinprick hypoalgesia/analgesia Decreased mechanical pain threshold, i.e. pinprick hyperalgesia
Pain rating suprathreshold stimulation	Calibrated needles, e.g. pinprick stimulators	Aδ	Spinothalamic	Decreased pain/absent pain, i.e. pinprick hypoalgesia/analgesia Increased pain, i.e. hyperalgesia
Dynamic mechanical allodynia	Cotton wool, Q-tip, brush	Aβ	Lemniscal	Allodynia, i.e. pain because of a normally non-painful stimulus
Pain induced by repetitive stimulation (wind-up ratio)	Calibrated needles, e.g. pinprick stimulator		Spinothalamic	Temporal summation
Blunt pressure	Algometer	Aδ, C	Spinothalamic	Increased pressure pain threshold, i.e. pressure hypoalgesia Decreased pressure pain threshold, i.e. pressure hyperalgesia

A priori stratification of patients according to their sensory profile could help to predict treatment response in clinical trials and clinical practice. Consequently, the Committee for Medicinal Products for Human Use (CHMP) of the European Medicines Agency (EMA) has acknowledged the sensory subgrouping of patients by use of standardized QST, an adequate stratification tool for determination of specific sensory phenotypes of patients in clinical trials on neuropathic pain. Although no analgesic drug has been developed based on the prediction by sensory testing, several studies have reported promising results. In a Phase 2a proof of concept study with retrospective phenotyping of patients with painful diabetic neuropathy, a novel selective TRPA1 antagonist (GRC 17536) was shown to be effective, especially in patients with preserved sensory function (irritable nociceptor). A retrospective analysis of a randomized controlled trial investigating the effect of pregabalin vs. placebo in patients with painful human immunodeficiency virus (HIV) neuropathy showed significant pain reduction only in a subgroup of patients with marked pinprick hyperalgesia (corresponding to QST cluster three). In contrast, there was no difference in the entire patient cohort. In a first placebo-controlled trial with a priori phenotype-stratification, oxcarbazepine was more effective for relief of neuropathic pain in patients with the “irritable nociceptor” phenotype defined as hypersensitivity and preserved small nerve fiber function (corresponding to QST cluster two) compared to the “non-irritable phenotype” (NNT for 50% pain relief: 3.9 vs. 13). However, the same authors failed to show a phenotype-dependent treatment effect of 5% lidocaine patch in patients with peripheral neuropathic pain because of PHN or peripheral nerve injury using the same patient stratification (irritable vs. non-irritable phenotype). An additional prospective crossover study investigated the effect of pregabalin versus placebo in patients with painful chemotherapy induced neuropathy. At baseline, all study participants received sensory testing on the dorsum of the feet, assuming that patients with increased mechanical pain sensitivity would more likely respond to treatment with pregabalin. Since there was no correlation of the MPT with the analgesic response, the authors concluded that the MPT was not a useful treatment predictor in this specific case.

To examine the value of predictive sensory testing, future trials of drug development will need to include a prospective stratification of patients according to their somatosensory profile.

Besides the above-mentioned advantages, several limitations should be kept in mind when performing and interpreting QST ( Table 21.3 ). One major limitation is that QST only assesses stimulus-evoked negative and positive sensory phenomena, whereas spontaneous pain is not detectable. Since the whole somatosensory pathways are assessed, QST cannot determine the exact localization of the lesion, i.e. peripheral versus central. Finally, as a psychophysical method, QST depends on the patient's cooperation and may be influenced by psychological factors that may result in false-negative findings. Therefore it is even more important to use a standardized testing protocol with standardized stimuli and instructions and reference values of healthy controls. Because of high commitments of time and costs and the need for a good training to ensure standardization, the DFNS QST protocol is only used by certain centers. To overcome these limitations, easy-to-use bedside protocols have recently been presented that enables increased applicability of sensory testing in clinical practice and clinical trials of drug development. ^,

TABLE 21.3

Advantages and Limitation of QST

Advantages	Limitations
Non-invasive	Psychophysical method
Examination of large and small fibers	No information about the localization of the lesion (central versus peripheral)
Assessment of gain and loss of function	No information about the extent of the lesion (predefined testing area)
Characterization of sensory phenotypes	Only stimulus-evoked phenomena, not spontaneous pain
Standardized German Research Network on Neuropathic Pain (DFNS) protocol with reference values	Standardized DFNS protocol needs training, is time-consuming, and cost intensive
Subgrouping of patients

You're Reading a Preview

Become a Clinical Tree membership for Full access and enjoy Unlimited articles

Become membership

If you are a member. Log in here