The recent article by Cohen et al.  1uses intradiscal administration of the anti–tumor necrosis factor (TNF) biologic etanercept for the treatment of two subsets of patients with chronic disc-related pain. The rationale for the selection of an anti-TNF biologic for these patients, including the 42% of the study subjects with chronic lumbosacral radiculopathy, has a substantial scientific basis, which includes both basic science and clinical evidence that excess TNF-α is centrally involved in the pathogenesis of disc-related pain.2–5In the case of lumbosacral radiculopathy, the anatomical site of inflammation and neuronal dysfunction is well delineated by the clinical presentation establishing dysfunction of a particular nerve root.

In addition, there is a reasonable scientific basis to study delivery of etanercept, which is anatomically targeted to the nerve root in those patients with clinically defined chronic radiculopathy. This scientific basis is supported not only by the evidence cited above implicating TNF-α in the initiation, amplification, and maintenance of disc-related pain, but also by the long history of the use of anatomically targeted delivery of corticosteroids for the treatment of sciatica.6But it should be pointed out that each of these studies, and established medical practice, involves the use of perispinal extradiscal  administration of antiinflammatories, rather than intradiscal  administration, as used by the study authors.

It is therefore puzzling that Cohen et al.  chose to include patients with lumbosacral radiculopathy in a study of minute doses of etanercept delivered into the intervertebral disc, where the anti-TNF biologic would be surrounded by a thick, fibrous capsule, the annulus fibrosus. It seems reasonable that the annulus fibrosus, in view of its structure, might impede delivery of etanercept to the nerve root, the expected primary site of TNF-α–mediated pathology.

The intradiscal design of the study of Cohen et al.  is all the more puzzling in view of the published work of my colleagues and me, some of which is cited in the article of Cohen et al.  2,3The use of etanercept for disc-related pain was first described in 1999.7Since that time, my colleagues and I have successfully treated a large number of patients with severe and intractable disc-related pain using perispinal extradiscal  etanercept, a method designed to deliver etanercept in therapeutic concentration to the site of neuronal pathology, including the nerve root.2,3,8–12 

To the credit of the authors, they point out that the low doses of etanercept studied, ranging from 0.1 to 1.5 mg, may have contributed to the lack of therapeutic effect in their study. The intradiscal  doses they used ranged from 0.4% to 6% of the extradiscal  etanercept dose in our studies.2,3,8Although it is certainly reasonable to be cautious in choosing the appropriate dose, selection of a subtherapeutic dose may doom a study to failure. In the case of etanercept for neuropathic pain, there is basic science evidence that high local concentrations of etanercept may be necessary for an optimal therapeutic effect, a rationale supporting targeted extradiscal etanercept.13,14In their discussion, the authors hypothesize whether systemic delivery may be superior. One would argue, based on the above, that it was the subtherapeutic doses of etanercept selected, along with the choice of intradiscal rather than extradiscal perispinal administration, that resulted in the lack of efficacy observed.

It seems that Cohen et al.  may not disagree with the above analysis, because they presently are conducting a trial of perispinal extradiscal etanercept, delivered epidurally, for treatment of sciatica* at doses ranging from 2 to 6 mg. The concept of epidural etanercept for treatment of sciatica has been previously described,9–11but we currently are uncertain whether this more invasive delivery method will be as efficacious as larger doses of etanercept injected superficial to the ligamentum flavum.2,3,8,12We look forward to the results of this clinical trial.

University of California, Los Angeles, California.

Cohen SP, Wenzell D, Hurley RW, Kurihara C, Buckenmaier CC III, Griffith S, Larkin TM, Dahl E, Morlando BJ: A double-blind, placebo-controlled, dose–response pilot study evaluating intradiscal etanercept in patients with chronic discogenic low back pain or lumbosacral radiculopathy. Anesthesiology 2007; 107:99–105
Tobinick EL, Britschgi-Davoodifar S: Perispinal TNF-alpha inhibition for discogenic pain. Swiss Med Wkly 2003; 133:170–7
Tobinick E, Davoodifar S: Efficacy of etanercept delivered by perispinal administration for chronic back and/or neck disc-related pain: A study of clinical observations in 143 patients. Curr Med Res Opin 2004; 20:1075–85
Sommer C, Schafers M: Mechanisms of neuropathic pain: The role of cytokines. Drug Discovery Today Dis Mech 2004; 1:441–8
Aoki Y, Takahashi K, Ohtori S, Moriya H: Neuropathology of discogenic low back pain: A review. Internet J Spine Surg 2005; 2 (1)
Arden NK, Price C, Reading I, Stubbing J, Hazelgrove J, Dunne C, Michel M, Rogers P, Cooper C: A multicentre randomized controlled trial of epidural corticosteroid injections for sciatica: The WEST study. Rheumatology (Oxford) 2005; 44:1399–406
Tobinick E, Tobinick A, inventors; TACT IP, LLC, assignee: Tumor necrosis factor antagonists for the treatment of neurological disorders. US patent 6,015,557. January 18, 2000
Tobinick EL: Targeted etanercept for discogenic neck pain: Uncontrolled, open-label results in two adults. Clin Ther 2003; 25:1211–8
Tobinick E, inventor; TACT IP, LLC, assignee: Cytokine antagonists for the treatment of localized disorders. US patent 6,419,944. July 16, 2002
Tobinick E, inventor; TACT IP, LLC, assignee: Cytokine antagonists for treatment of localized disorders. US patent 6,537,549. March 25, 2003
Tobinick E, inventor; TACT IP, LLC, assignee: Cytokine antagonists for neurological and neuropsychiatric disorders. US patent 6,982,089. January 3, 2006
Tobinick E: Spinal delivery of p38: TNF-alpha inhibitors. PLoS Med 2006; 3:e511
Sommer C, Schafers M, Marziniak M, Toyka KV: Etanercept reduces hyperalgesia in experimental painful neuropathy. J Peripher Nerv Syst 2001; 6:67–72
Wacnik PW, Eikmeier LJ, Simone DA, Wilcox GL, Beitz AJ: Nociceptive characteristics of tumor necrosis factor-alpha in naive and tumor-bearing mice. Neuroscience 2005; 132:479–91