Does Pregabalin Still Have a Role in Treating Cancer-Induced Bone Pain?

Abstract

Bone metastases are common in patients with advanced cancer and often cause cancer-induced bone pain (CIBP), which can significantly affect patients’ quality of life. Although radiotherapy is effective for CIBP, many patients do not attain complete relief of their symptoms after treatment. In the article that accompanies this editorial, Fallon et al conducted a randomized controlled trial (RCT) evaluating the efficacy of pregabalin in patients receiving radiotherapy for CIBP. A total of 233 patients were randomized to either pregabalin or placebo. There was no statistically significant difference in treatment response, defined as reduction in worst pain with stable or reduced opioid dose, average pain, pain interference, and quality of life between the two arms. Fallon et al concluded that the routine use of pregabalin in patients receiving radiotherapy for CIBP is not warranted. We applaud the investigators for their efforts but caution readers about two important issues in the design of the trial that may limit the generalizability of the study results. One issue is whether the study targeted the right patients. Pregabalin is a drug that binds to the a2-d subunit of calcium channels and there is evidence from reported RCTs supporting its use in neuropathic pain caused by diabetic peripheral neuropathy (DPN), postherpetic neuralgia (PHN), spinal cord injury, and fibromyalgia. Per the pregabalin product monograph, these conditions are the recommended indications for its use as an analgesic. Given its proven effectiveness in treating the aforementioned conditions, pregabalin is also prescribed for neuropathic and nonneuropathic CIBP as an off-label indication. Although there is preclinical evidence in animal models that CIBP is responsive to neuropathic agents, the only RCT investigating the efficacy of pregabalin in CIBP was terminated early because an interim analysis revealed that the study was underpowered to show the intended treatment effect. Thus, the role of pregabalin in the treatment of all neuropathic and nonneuropathic CIBP remains to be determined. The Fallon et al trial excluded patients who were already receiving pregabalin or gabapentin; this would eliminate patients with neuropathic pain features, because many of these patients would be started on neuropathic agents already. These excluded patients are, in fact, the ideal target group who would potentially benefit the most from pregabalin. It is unclear how many of the enrolled patients had a neuropathic component to their CIBP, because this was not formally assessed or documented. The estimated incidence of neuropathic pain features in patients receiving palliative radiotherapy to bonemetastases is 17% to 20% and the incidence of neuropathic pain in the Fallon et al study would be even lower. If pregabalin were more effective in CIBP with neuropathic features, its true treatment effect would have been underestimated as a result of potential exclusion from this trial. Of 1,970 patients screened in the study, 1,737 were excluded because they did not meet the inclusion criteria. Only 233 patients were randomized and the trial failed to meet its target accrual of 260 patients. Given that 88% of the screened patients were excluded and the difficulty with accrual, leading to early termination, the generalizability of the trial result is limited. The second issue is whether the study captured the right end points. The primary end point of the trial was a reduction of two or more points on a 0 to 10 numerical rating scale (NRS) for worst pain measured at 4 weeks, accompanied by a stable or reduced opioid medication dose, compared with baseline. However, previous studies have shown that a significant portion of patients with CIBP will respond to radiotherapy after 4 weeks. The Bone Pain Trial from the United Kingdom showed that 1 month after randomization, only 54% of patients in the single 8-Gy arm had an improvement in pain relief. However, 8 weeks and 12 weeks following randomization, 68% and 75% of patients had an improvement in pain relief, respectively. Similarly, the Dutch Bone Metastasis Study showed that 4 weeks after randomization, the mean pain score in the single 8-Gy arm decreased from 7.2 to 5.1. However, 8 weeks and 12 weeks following randomization, the mean pain score further decreased to 4.2 and 3.9, respectively. Therefore, the end point at 4 weeks does not appear to capture the full treatment effect of radiotherapy. In the only radiation trial that selectively included patients with neuropathic pain, a fractionated course of radiotherapy (20 Gy in five fractions) might have been more effective than a single-fraction treatment, although the results were not statistically significant. The investigators suggested a dose-escalation trial of 30 Gy versus 8 Gy for further investigation. In the Fallon et al study, both 8 Gy in one fraction and 20 Gy in five fractions were allowed. Given the uncertainty in the optimal radiation prescription parameters for neuropathic pain, radiation dose should ideally be stratified in the analysis.