Etanercept is no better than placebo for the treatment of subacute sciatica


A study published in the Annals of Internal Medicine (online) has found that the tumour necrosis factor -α (TNF-α) inhibitor etanercept (Enbrel, Amgen/Pfizer) does not improve leg pain compared with placebo in patients with lumbosacral radiculopathy.

Lead investigator Steven P Cohen, Johns Hopkins School of Medicine, Baltimore, USA, and Walter Reed National Military Medical Center, Bethesda, USA, and colleagues reported that the lack of reliable treatment is a “major contributor” to the burden of neuropathic back pain. They added that epidural steroidal injections were increasingly used to treat back pain because of the uncertain long-term effects of surgery and the high rates of adverse effects with analgesic medications, but these treatments were “mired in controversy” as data on their long-term efficacy is mixed and there are concerns about their safety. Therefore, alternatives are being sought, which include the epidural administration of TNF-α inhibitors. Cohen et al wrote: “A major obstacle in evaluating the literature on epidural administration is that no randomised study has compared treatments. To address this gap, we conducted at multicentre, placebo-controlled study comparing epidural steroids, etanercept, and saline for lumbosacral radiculopathy.”

Of the 84 patients enrolled in the study (mean age 42.29 years; average pain duration 2.7 months), 30 were randomised to saline epidural injections, 28 were randomised to epidural steroidal injections, and 26 were randomised to receive epidural etanercept.

After one month, both patients in the saline and steroid groups saw a significant improvement in their Oswestry Disability Index (ODI) scores, but patients in the etanercept group did not (p=0.006). There was a significant difference in ODI scores between the etanercept and steroid groups (p=0.002) and the saline and etanercept groups (p=0.04), but none between the steroid and saline groups (p=0.23). Also at the one-month mark, patients in the steroid group had a lower, although not significantly, decrease mean adjusted leg pain score than either the saline group or the etanercept group (2.54 vs. 3.56 vs. 3.78, respectively; p=0.24)

After three months, of the patients still in the study, Cohen et al noted: “The proportion of satisfied patients at three months was 48% in the saline group, 65% in the steroid group, and 50% in the etanercept group. These proportions remained stable at six months at 48% in the saline group but decreased slightly to 63% in the steroid group and 45% in the etanercept group.”

According to Cohen, there are several explanations why their results “neither exclude nor prove a modest benefit for steroids” and that etanercept did not appear to be any more beneficial that placebo.  They wrote: “We believe that the most probable explanation is that the analgesic effects of steroids are short-lived and that spontaneous improvement resulted in sustained beneficial effects in all treatment groups. That is the conceptual appeal of epidural steroid and TNF inhibitors (that is, to alleviate pain until the body heals itself), because the natural course of a herniated disc is resorption and symptom resolution.” They added that a second less likely explanation was that all treatments were equally efficacious (or equally “non-efficacious”) in the long-term, but the steroid effect is realised earlier.

As to why etanercerpt appeared to be no better than placebo, they wrote that a higher epidural dose or a long-acting TNF inhibitor may have had additional benefit, but said that this was uncertain.

Cohen told Spinal News International that the implications of the study were that “we still have a lot of work to do in terms of finding out whether these drugs will work and if so, in what dosages, and whether they will be more effective or safer than the default treatment, which is epidural steroids.” He added: “The conceptual appeal of etanercept is that because of its non-particulate formulation, epidural injections involving the transforaminal approach (which is more effective than the conventional ‘interlaminar’ approach used to provide pain relief for pregnancy and anaesthesia) may be safer than with steroids- especially in the neck and mid-back, and in patients who have already undergone surgery.  Similar to what is done when opioid (narcotic) medications are administered epidurally or spinally, we reduced the dose because of a very auspicious pilot study showing that low doses might be effective, since low doses may minimise some of the side effects of these types of drugs, such as depressing the immune system.”