Biomarkers in the Degenerative Human Intervertebral Disc Tissue and Blood

Abstract

Patients with back pain comprise a large proportion of the outpatient practice among physiatrists. Diagnostic tools are limited to clinical history, physical examinations and imaging. Non-surgical treatments are largely empirical, encompassing medications, physical therapy, manual treatments and interventional spinal procedures. A body of literature is emerging confirming elevated levels of biomarkers including inflammatory cytokines in patients with back pain and/or radiculopathy, largely because the protein assay sensitivity has increased. These biomarkers may serve as tool to assist diagnosis and assess outcomes.

The presence of inflammatory mediators in the intervertebral disc tissues and blood helped confirming the inflammatory underpinnings of back pain related to intervertebral disc degeneration. Literature reviewed here suggests that biomarkers could assist clinical diagnosis and monitor physiological outcomes during and following treatments for spine related pain. Biomarkers must be measured in a large and diverse asymptomatic population, in the context of age and comorbidities to prevent false positive tests. These levels can then be rationally compared to those in patients with back disorders including discogenic back pain, radiculopathy and spinal stenosis. While studies reviewed here used “candidate marker” approaches, future non-biased approaches in clearly defined patient populations could uncover novel biomarkers in clinical management of patients.

BACKGROUND

Chronic back pain is a major health and economic problem leading to billions of dollars in healthcare expenditures in the United States.¹ Each year, millions of patients’ visits to hospitals, emergency departments, outpatient clinics, and physician offices entailed a diagnosis of low back pain.¹ Physical Medicine and Rehabilitation (PM&R) physicians often evaluate these patients clinically, depending primarily on a combination of history and physical examination, and imaging studies to make a diagnosis and determine the anatomical pain generator. Blood tests (serum or plasma) are performed only when discitis is suspected. However, intervertebral disc (IVD) degeneration detected by imaging does not consistently correlate with back/neck pain: many people with disc degeneration are not affected by chronic symptoms.^{2, 3} With the increased sensitivity of molecular methodologies, recent studies have shown elevated inflammatory markers and extracellular matrix degradation products in IVD tissues from patients with back pain who have IVD degeneration on imaging studies, compared with those with imaging evidence of disc degeneration who underwent surgeries for scoliosis deformity correction.^{4, 5} The blood and tissue-based molecular markers reviewed here may help to guide the diagnosis in a more refined manner than our current clinical approaches, thus may be justified for clinical usage in the near future.

Moreover, blood-based molecular markers when applied as outcome measures may guide clinical decision making during a recommended course of treatment, for example, deciding on whether to recommend non-surgical or surgical approaches. Current surgical treatments for refractory back pain associated with IVD degeneration are limited to procedures that disrupt, rather than restore, the disc structure, for example, discectomy with spinal fusion. The degree of symptomatic relief following surgery is unpredictable and often incomplete. As alternative to surgeries, conservative treatments include education, medications, exercise interventions, physical modalities and injection of anesthetic agents and steroids. Biomarkers, in combination with imaging, may guide the clinician and patients in making rational treatment decisions, particularly when a surgical intervention is being considered.

This is a narrative review article on biomarkers in relation to back pain. It will then discuss how, in the future, this knowledge could be applied diagnostically to identify the degenerate disc as the pain generator, and as outcome variables to determine the efficacy of specific treatments of back pain associated with IVD degeneration. Note that it may be easier to collect the blood than IVD tissues for biomarkers studies, thus blood biomarkers hold greater potential as diagnostic and prognostic tools. Since many tissues other than the spine could contribute to their elevation, specificity, validity, replicability, etc. must be addressed to prove clinical utility.

Inflammatory Mediators in the IVD tissue.

The IVD could produce multiple inflammatory mediators in response to insults or with aging (Figure 1). The IVD is considered an immune privileged site which is naturally less subject to immune responses than most other areas of the body. However, leukocytes could infiltrate the IVD injured tissue when annulus fibrosus (AF) tear or nucleus pulposus (NP) herniation occurs. In response to such injuries, the origin of inflammatory mediators may be diverse: herniated NPs attract leukocytes, and then leukocytes migrate into injured IVDs (Table 2),^6–8 and produce inflammatory mediators. Furthermore, the disc cells are capable of directly producing inflammatory mediators.⁵ Moreover, it is recognized that back pain related to IVD degeneration increases with age. During immunosenescence, an imbalance between inflammatory and anti-inflammatory networks is augmented in favor of inflammation, which results in the low grade chronic pro-inflammatory status among older individuals (inflammaging).^{9, 10} Furthermore, a combination of factors may contribute to the inflammatory responses when the IVD is breached: disc cell senescence, and exposure to damage-associated molecular patterns (DAMPs), such as fibronectin fragments¹¹ generated by disc tissue breakdown. There also may be some evidence that low grade bacterial infection contributes to low back pain.¹² The pathogen-associated molecular patterns (PAMPS), small molecular motifs conserved within a class of microbes, may induce inflammation in the IVDs of these patients. The IVD have a higher extracellular osmolarity than most other tissues; and their osmolarity changes by around 25% during the course of the day. Inappropriate osmolarity or biomechanical forces can induce inflammation by activating innate immunity.^{13, 14}

Figure 1. Intervertebral discs produce inflammatory cytokines/chemokines with degeneration.

Aging, mechanical insults, Damage-associated molecular patterns (DAMPs) and Pathogen-associated molecular patterns (PAMPs) accelerate disc degeneration. Left: intact mouse tail IVD stained with Alcian blue, with hematoxylin and eosin counter stain; Right: injury-induced degenerative IVD. NP: nucleus pulposus; AF: annulus fibrosus.

Table 2.

Leukocytes Infiltrating the Intervertebral Disc Tissue.

Leukocytes	Diagnosis	References
CD68+ and T cells	Degenerative/herniated disc	Shamji et al.6
CD68+ cells	Herniated disc/radiculopathy	Park et al.7
CCR7+, CD163+, and CD206+	Degenerative disc	Nakazawa et al.8

Multiple inflammatory markers have been described in human degenerative, painful, or herniated IVDs (Table 1). Both disc cells as well as invading macrophages (Table 2) could be the source of the detected cytokines. Importantly, occurrence of inflammatory cytokines/chemokines in the disc may contribute to the progression of degeneration by inducing the expression of matrix-degrading enzymes as well as by inhibiting extracellular matrix synthesis. In addition, inflammatory mediators may play a crucial role in generating back pain as well as related radicular symptoms. It is likely that only a subset of the inflammatory mediators is readily and reliably measurable in the blood. Measurement of these inflammatory mediators could be informative in diagnosis and of the disease, therefore are valued as inflammatory markers. Some biomarkers may have prognostic value. For example, multiple studies have shown that levels of serum biomarkers at baseline and after treatments correlated with pain reduction (Table 3).^8–11 Selected inflammatory markers and their potential diagnostic associations are summarized below (Table 1).

Table 1.

Biomarkers in the Intervertebral Disc Tissue.

Inflammatory Markers	Diagnosis	References
IL8, IL7, IL10; RANTES and IL1b	Discogenic back pain confirmed with discography	Zhang et al.; Kepler et al.4, 5
GMCSF, ENA78, IL1b and TNFa	Back pain with Modic changes	Schroeder et al.15
MMPs, NO, PGE2 and IL6	Herniated disc	Kang et al.16
TRPC6, IL6, IL8, IL15 and type I IFN	Degenerative/herniated disc, back pain	Sadowska et al.9
IL17, 4, 6, 12 and IFNγ	Degenerative/herniated disc	Shamji et al.6
IL1b and TNFa	Degenerative/herniated disc	Le Maitre et al.17
IL6, IL8, TNFa, GMCSF, MCP1, eotaxin	Spondylolisthesis	Sutovsky et al.18
IL8, TNFa, TGFb, VEGF and NGF	Degenerative compared with herniated disc	Lee et al.19
TNFa, TNFR1 and 2	Aging/degenerative disc	Bachmeier et al.20

Table 3.

Inflammatory Markers in Blood.

Inflammatory Markers	Diagnosis	References
RANTES, NPY, serotonin	Axial back pain; high NPY and serotonin correlates with favorable response to steroid injection	Sowa et al., Schaaf et al.21, 22
SCGFb, GMCSF	Disc herniation; correlate with pain reduction after steroid injection	Weber et al.23
IL6	Low back pain; level correlates with BMI, age, and duration	Weber et al.24
IL6, TNFa	Elevated in patients with disc degeneration	Divi et al.25
CRP	Correlates with radicular and back pain	Gebhardt et al.26
IL6, TNFa, IL10	Radiculopathy; elevated pro- and anti-inflammatory cytokine	Wang et al.27
IL23, Th/Treg	Chronic low back pain	Luchting et al.28
IL8	Herniation; correlate with extrusion or sequestration	Brisby et al.29
AITR, AITRL, IL2, 6, 8, and TNFa	Herniated disc/radiculopathy	Park et al.7
IL6, IL8	Herniated disc/radiculopathy; correlate with more pain at 12- month follow up	Pedersen et al.30
PGE2, IL17, Th17	Herniated disc; correlate with pain	Cheng et al.31
IL6	Herniated disc/radiculopathy; correlates with more pain at 1 year follow up	Schistad et al.32
CCL17 etc.	Radiculopathy	Moen et al.33

Leukocytes in the IVD tissue.

The release of chemokines from degenerating discs promotes the infiltration and activation of immune cells, further amplifying the inflammatory cascade. Despite the abundance of chemokines described above, only macrophages and T cells have been described in the diseased IVDs (Table 2). Note that CD68 (macrosialin)-expressing cells may include macrophages and dendritic cells. Nakazawa et al. further distinguished disc tissue macrophages into proinflammatory (M1; CCR7⁺), remodeling (M2c; CD163⁺) and anti-inflammatory (M2a; CD206⁺) phenotypes,⁸ suggesting their role in modulating inflammation and tissue repair.

Inflammatory markers in blood.

These small molecules could diffuse from the degenerative IVD tissues. Other aging tissues such as arthritic joints and muscles may contribute to the inflammatory markers in the blood. Multiple inflammatory markers were found elevated in serum or plasma from patients with disc degeneration, back pain, and/or radiculopathy (Table 3).

Discussion.

Treatments for back pain by physiatrists with anti-inflammatory drugs, and/or steroid injections, are based on a biochemical point of view of the disease. However, few tests are available to examine the biochemical factors related to back pain. The biomechanical aspects are addressed by empirically physical therapists, often in collaboration with physicians. Treatments by surgeons, on the other hand, are largely based on a mechanical understanding of the spine: for example, laminectomy and foraminectomy to alleviate pressure on the nervous tissues, and spinal fusion to stabilize the motion segments. Imaging studies (plain radiographs, magnetic resonance imaging, computerized tomography) are widely available to guide surgical approaches and injections. Since the prognostic value have been shown for some biomarkers,^8–11 they may become an especially useful tool to guide patient selection and outcome evaluations for non-invasive treatments of back pain.

Inflammatory mediators found in the IVD tissues are intriguing and may help in understanding the disease mechanisms and identify the inflammatory component of pain (summarized in Table 1). Only a few studies have described infiltrating leukocytes in the disc tissue (Table 2). Although it is difficult to reconstruct the precise location of surgically removed tissues, Nakazawa et al. used cadaveric human IVDs and found more macrophages in unhealthy areas of the IVDs and near endplate defects than healthy appearing areas.⁸ Animal model is helpful in identifying the precise location of infiltrating leukocytes. In the well-established mouse tail IVD injury model,³⁴ 1 week post injury, F4/80⁺ macrophages penetrate only the outermost layers of the AF (Figure 2A). There were very few cells stained for the F4/80 marker in the adjacent intact discs (Figure 2B). Under high magnification, some of the F4/80⁺ cells assumed a polarized shape, while some assumed an elongated shape (blue arrows, Figure 2C). There are areas seemingly stained positive for F4/80⁺ near the endplates, but no macrophage-like cells have been identified (Figure 2D), suggesting the resident cells in endplate and adjacent bone may be capable of expressing the F4/80⁺ antigen, EMR1. Although disc cells and leukocytes may contribute to the cytokines/chemokines found in the disc tissues, inflammatory mediators produced by injured or senescent resident disc cells are likely the initiating event. These cytokines/chemokines may attract leukocyte migration to the injured AF or endplate, causing additional inflammation.

Figure 2. F4/80 (macrophage marker) in the injured mouse intervertebral disc.

F4 A&B: Injured and Intact discs; red arrow indicates direction of needle injury; C&D: magnification of green outlined areas in the outer annulus fibrosus; blue arrows indicate polarizing or elongated shaped F4/80⁺ cells; E&F: magnification of blue outlined areas in the cartilaginous endplate. scale bar: 100 μm.

Biomarkers in the blood have immense potential for clinical use, to confirm diagnosis or predict treatment outcomes of inflammation-mediated back pain. Multiple inflammatory markers in blood (serum or plasma) have been described in patients with back pain and/or radiculopathies (Table 3). These developments are made possible by recent improvements in sensitivities and specificity of immunoassays, and have tremendous potential in assisting diagnosis and predicting outcomes for chronic back pain. For example, Weber et al. confirmed that serum IL6 was elevated with aging and in patients with back pain related to disc herniation, degeneration or spinal stenosis, compared with age matched controls without pain.²⁴ Elevated serum IL6 levels have also been shown to correlate with poorer outcomes in patients with radiculopathy.³² Interestingly, high NPY and serotonin correlates with favorable response to steroid injection.^{21, 22} Blood biomarkers will help differentiate inflammation-mediated from psychosocial factors mediated back pain,³⁵ and guide treatment approaches.

Despite the potential clinical utility of biomarkers, caution must be used as we do not have precise sensitivity and specificity for these biomarkers across large and diverse enough samples of patients to provide a firm foundation upon which to use this information for clinical decisions. Much like magnetic resonance imaging (MRI) which is very sensitive that it identifies many nonspecific/irrelevant findings unrelated to spine pain,^{2, 3} caution must be used before considering these biomarkers as clinically meaningful tests. It would be unfortunate if they are used inappropriately to justify lumbar spinal fusions in a similar fashion as MRI. In addition, these molecules in the blood could be secreted from any senescent, injured, or even normal tissues, the findings must be thoroughly evaluated in appropriate clinical settings. Finally, these highly sensitive assays have great potential to overcall “inflammation” which may not be directly related to a patient’s pain.

Among the serum biomarkers, high levels of serotonin and NPY before epidural steroid injection corresponded to a favorable response to treatment.²² Lower SCGFb and GMCSF levels correlated to pain reduction after steroid injection.²³ Elevated levels of IL6 and IL8 correlated to persistent pain at 1 year follow up.^{30, 32} Finally, 13 cytokines were found elevated among the largest array studied to date, consisting of 92 cytokines.³³ With the advance of detection methodology and introduction of non-biased approaches, novel markers may emerge.

Conclusion.

Inflammatory biomarkers are potentially valuable in assisting clinical decision making, especially when such factors as age and comorbidities are considered. Another potential application is to predict outcomes for targeted non-surgical treatments of back pain related to IVD degeneration. There is a need for identification of inflammatory biomarkers that are specific to the painful degenerating discs. With the advance of detection methodology and introduction of non-biased approaches, novel markers may emerge that are suitable for clinical use. Finally, thorough assessment of the prevalence of biomarkers in a large and diverse asymptomatic population is necessary to fully delineate ranges of normal values, to prevent false positive clinical tests which may lead to overcall of discogenic pain and subsequent excessive interventions such as spine surgeries.