Clinical Outcomes after Onset of Mirogabalin-Related Adverse Drug Reactions in Patients with Lumbar Spinal Stenosis: Post Hoc Analysis of the MiroTAS Study

Abstract

Introduction

The MiroTAS study demonstrated the pain-relieving effect and safety of mirogabalin in patients with lumbar spinal stenosis (LSS); however, whether mirogabalin-related adverse drug reactions (mrADRs) affect clinical courses, treatment continuation, or adherence remained unknown. We examined recovery outcomes and treatment status following mrADR onset and whether the analgesic effects of mirogabalin correlated with LSS and leg pain durations.

Methods

This post hoc analysis of the 12-week, randomized controlled MiroTAS study summarized mrADRs (dizziness, somnolence, peripheral edema, and edema) by severity and outcome. Spearman’s rank correlations between leg pain changes and duration of LSS and leg pain were examined.

Results

Data from 214 patients in the modified intention-to-treat set (mirogabalin add-on: 110; nonsteroidal anti-inflammatory drugs alone: 104) and 216 in the safety analysis set (110 and 106 patients, respectively) were analyzed. Regarding mrADRs, 27/33 dizziness, 25/35 somnolence, 3/6 peripheral edema, and 2/2 edema events recovered or improved. Median times to recovery for dizziness and somnolence were 8.0 and 33.5 days, respectively. Recovery or improvement of dizziness and somnolence occurred in most patients even without dose modification, with a higher proportion of recovery following dose reduction. Most events with moderate dizziness or somnolence (80.0% and 85.7%, respectively) recovered or improved. In the mirogabalin add-on group, Spearman’s rank correlation coefficients showed weak positive correlations between change in visual analogue scale (VAS) score for leg pain from baseline to week 12 and LSS duration (P = 0.0269) and no correlation between leg pain duration and change in VAS score.

Conclusions

Our data suggest that dizziness and somnolence are generally manageable under careful monitoring, with most events recovered or improved after onset. The analgesic effect of mirogabalin was weakly associated with LSS duration and not associated with leg pain duration. However, this was a post hoc analysis; given the limited sample size and statistical power, further research is needed.

Clinical Trial Registration

Japan Registry of Clinical Trials (jRCTs021200007).

Supplementary Information

The online version contains Supplementary Material available at https://doi.org/10.1007/s40122-026-00855-6.

Key Summary Points

Why carry out this study?

Lumbar spinal stenosis (LSS) is a leading cause of peripheral neuropathic pain, and pharmacological management often requires long-term treatment while carefully balancing efficacy and safety.

Although mirogabalin is an effective treatment option for neuropathic pain in LSS, uncertainty regarding the clinical course and impact of mirogabalin-related adverse drug reactions (mrADRs) may limit treatment continuation or adherence in routine clinical practice.

This post hoc analysis of the MiroTAS study examined outcomes after the onset of representative mrADRs, and whether the analgesic effects of mirogabalin on neuropathic pain correlated with the duration of LSS and leg pain.

What was learned from the study?

In patients with LSS and neuropathic pain, the most frequent duration of dizziness was 1–10 days and that for somnolence was 1–10 days and 61–70 days; 70–80% of events of moderate or lower severity resolved or improved after onset.

These findings suggest that dizziness and somnolence are manageable under careful monitoring, and even when mrADRs occur, physicians may consider continuing mirogabalin treatment, or adjusting its dose, anticipating recovery from mrADRs or improvement in pain.

The analgesic effect of mirogabalin add-on therapy to nonsteroidal anti-inflammatory drugs (NSAIDs) was weakly associated with LSS duration and not associated with leg pain duration, supporting its use in patients with neuropathic pain who do not achieve adequate pain relief with NSAID monotherapy.

Introduction

Lumbar spinal stenosis (LSS) is a common spinal disorder in which the spinal canal in the lumbar spine narrows because of age-related degeneration (thickening and protrusion) of the intervertebral discs, bones, and ligamentum flavum, potentially leading to chronic compression of the cauda equina and nerve roots [1]. Chronic nerve compression and ischemia due to the progression of LSS lead to nerve conduction disorders caused by demyelination and axonal degeneration. These may result in peripheral neuropathic pain, including leg pain, and gait disorders [2, 3].

The Japanese Orthopaedic Association guideline does not provide clear recommendations for choosing between surgical and conservative treatments, including pharmacotherapy, for neuropathic pain associated with LSS [4]. In clinical practice, nonsteroidal anti-inflammatory drugs (NSAIDs) and their combination with gabapentinoids such as pregabalin and mirogabalin are frequently used in Japan.

Mirogabalin besylate (mirogabalin), an oral gabapentinoid, is a selective ligand for the α2δ subunit of voltage-gated calcium channels and has been indicated in Japan for the treatment of neuropathic pain on the basis of the results of several phase 3 clinical trials [5–9]. The MiroTAS study demonstrated that adding mirogabalin to NSAID treatment resulted in greater improvements in leg pain and quality of life compared with NSAIDs alone in patients with neuropathic pain due to LSS [10].

While mirogabalin is an effective treatment option for neuropathic pain, mirogabalin-related adverse drug reactions (mrADRs) such as dizziness, somnolence, peripheral edema, and edema are frequently reported [6–10]. Therefore, managing neuropathic pain with mirogabalin requires careful balancing of efficacy and safety while maintaining continuous treatment. A recent post hoc analysis of the MiroTAS study further clarified this balance by examining the timing of onset of mrADRs and factors affecting the safety and efficacy of mirogabalin [11], and showed that mirogabalin was effective regardless of patient background. Furthermore, while specific patient characteristics were not associated with the onset of mrADRs, the analysis also indicated that caution is warranted regarding the occurrence of mrADRs during the initial dosing and titration phases [11]. However, despite these analyses, a clinical question remains unresolved regarding the appropriate course of management following the onset of mrADRs, including whether the outcomes of mrADRs differ by severity, and whether treatment should be continued. Furthermore, it is still unclear whether the analgesic efficacy of mirogabalin varies according to the duration of LSS and leg pain.

This additional post hoc analysis of the MiroTAS study aimed to examine outcomes (recovered, improved, and not recovered) following the onset of representative mrADRs and mirogabalin treatment status (no change, dose reduction, dose interruption, or discontinuation of mirogabalin treatment) following mrADR onset. Additionally, it analyzed whether the analgesic effects of mirogabalin on neuropathic pain correlate with the duration of LSS and leg pain.

Methods

Study Design

This was a post hoc analysis of the MiroTAS study, a 12-week, randomized, open-label, parallel-group study conducted at 32 study sites across Japan between June 2020 and October 2021 [10]. Participants were randomly assigned in a 1:1 ratio to receive either mirogabalin in addition to NSAIDs (mirogabalin add-on group) or NSAIDs alone (NSAIDs-alone group).

NSAIDs were administered in accordance with their respective Japanese package inserts, and dose modifications were not permitted during the study period in either group. Mirogabalin was administered at an initial dose of 5 mg twice daily (BID) during weeks 1 and 2 (days 1–15), followed by 10 mg BID during weeks 3 and 4 (days 15–29). From week 5 (day 29) to week 12, the dose was either maintained at 10 mg BID or increased to 15 mg BID, depending on patient tolerance. For patients with a creatinine clearance of 30 to < 60 mL/min, all doses were reduced by half.

The primary study was approved by the Clinical Research Review Committee, Fukushima Medical University Certified Review Board (CRB no. CRB2200002), which notified all participating centers of the approval. The primary study was conducted in accordance with the ethical principles, Clinical Trials Act, and relevant notifications stipulated in the Declaration of Helsinki (as revised in 2013). All patients provided written informed consent. The primary study was registered in the Japan Registry of Clinical Trials under identifier jRCTs021200007.

Patients

This post hoc analysis included patients who met the eligibility criteria of the primary MiroTAS study [10]. Briefly, the inclusion criteria were: patients aged ≥ 20 years with leg pain attributed to radicular-type LSS, as diagnosed by a physician at least 3 months prior to enrollment; continuous use of NSAIDs for ≥ 4 weeks without any change in dosage; and a visual analogue scale (VAS) score of ≥ 40 mm at both the time of informed consent and enrollment. Exclusion criteria have been previously reported [10].

Endpoints

Among the reported adverse drug reactions (ADRs), dizziness, somnolence, peripheral edema, and edema were defined as mrADRs. Regarding safety, this analysis evaluated the durations of mrADRs in the mirogabalin add-on group according to outcome (recovered, improved, and not recovered), as well as the number of mrADR events in the mirogabalin add-on group stratified by outcome after mrADR onset, mirogabalin treatment status after mrADR onset (no change, dose reduction, dose interruption, or discontinuation of mirogabalin treatment), and severity of mrADR (mild, moderate, and severe), including their combined effects.

Regarding efficacy, this analysis evaluated the relationships between changes in VAS score for leg pain from baseline to week 12 and (1) LSS duration and (2) leg pain duration in both treatment groups. VAS scores were assessed on a 100-mm scale, with 0 mm indicating no pain and 100 mm indicating the worst pain imaginable. LSS duration was defined as the period from the time the physician made a definitive diagnosis of LSS, and leg pain duration was defined as the period from the time the patient first reported leg pain.

Statistical Analysis

This post hoc analysis was conducted using the same predefined analysis populations as the primary analysis [10].

The duration of mrADRs (in days) was calculated as follows: date of mrADRs outcome confirmation − date of onset + 1. For events classified as recovered, the recovery date was used as the outcome date. For events classified as improved or not recovered, the outcome date was defined as the date on which the attending physician confirmed the outcome, therefore reflecting the date on which the attending physician confirmed the mrADR outcome rather than the actual date on which the outcome occurred. Adverse events were coded using the Medical Dictionary for Regulatory Activities (MedDRA/J version 24.1).

For the change in VAS scores for leg pain from baseline to week 12, the estimated difference in adjusted means and its 95% confidence interval (CI) and P-value were calculated using a linear mixed model for repeated measures, which included the presence/absence of mrADRs, week (time point), and their interaction as fixed effects, baseline VAS score for leg pain as a covariate, and patients as random effects.

For the relationship between the change in VAS score for leg pain from baseline to week 12 and LSS disease duration and leg pain duration, scatter plots were generated, and Spearman’s rank correlation coefficients and P-values were calculated. Correlation coefficients closer to 1 indicate stronger linear relationships, and if positive, represent that pain improvement increases with longer disease duration. Conversely, if the correlation coefficient is negative, pain improvement decreases with longer disease duration. This analysis excluded data from patients who had missing VAS score data for leg pain, disease duration, or leg pain duration.

Statistical significance was evaluated at the 5% level with two-sided testing. All statistical analyses were conducted using SAS software version 9.4 (SAS Institute Inc., Cary, NC, USA).

Results

Patients

This post hoc analysis used data from 214 patients in the modified intention-to-treat set (mirogabalin add-on group, n = 110; NSAIDs-alone group, n = 104), and 216 in the safety analysis set (mirogabalin add-on group, n = 110; NSAIDs-alone group, n = 106) from the primary analysis [10]. The background characteristics of patients were similar between the two treatment groups [10]. The mean ages were 67.8 and 70.9 years; proportions of female patients, 54.5% and 49.0%; mean creatinine clearance, 81.5 and 70.7 mL/min; proportions of patients with creatinine clearance 30 to < 60 mL/min, 27.3% and 33.7%; and mean VAS scores, 63.8 and 62.8 mm in the mirogabalin add-on group and NSAIDs-alone group, respectively.

mrADR Frequency and Duration

Overall, there were 33 mrADRs of dizziness, 35 of somnolence, 6 of peripheral edema, and 2 of edema. Of these, a respective 27 (81.8%), 25 (71.4%), 3 (50.0%), and 2 (100.0%) of these events recovered or improved (Fig. 1). For dizziness, the most frequent duration was 1–10 days, whereas for somnolence, the most frequent duration categories were 1–10 days and 61–70 days (Fig. 1). No distinct duration trends were observed for peripheral edema or edema (Supplementary Fig. S1). The median (quartile 1, quartile 3) durations from mrADR onset to recovery were 8.0 (2.0, 20.5) days for dizziness and 33.5 (14.0, 62.0) days for somnolence (Fig. 2 and Supplementary Table S1).

Fig. 1.

Distribution of symptom duration for (a) dizziness and (b) somnolence by outcome category. Recovery was based on the recovery date in the electronic data capture system, whereas improvement/no recovery was based on the date of physician confirmation. Therefore, the recorded dates for improvement do not necessarily represent the actual timing of clinical improvement

Fig. 2.

Duration from onset of (a) dizziness and (b) somnolence to recovery. The lower edge of the box represents quartile 1, the central line indicates the median, and the upper edge represents quartile 3. The lower and upper whiskers correspond to the minimum and maximum values, respectively

Outcomes of mrADRs, Treatment Status, and mrADR Severity

Figure 3 and Supplementary Fig. S2 show the number of mrADR events according to outcome and mirogabalin treatment status after mrADR onset. Regarding outcomes, the numbers of events classified as recovered, improved, and not recovered were 20 (60.6%), 7 (21.2%), and 6 (18.2%), respectively, for dizziness and 18 (51.4%), 7 (20.0%), and 10 (28.6%), respectively, for somnolence (Fig. 3a, b). After the onset of dizziness, 76.2% (16/21) of events with no change in mirogabalin dose and 88.9% (8/9) with mirogabalin dose reduction were recovered or improved (Fig. 3a). After the onset of somnolence, 70.8% (17/24) of events with no change in the mirogabalin dose and 75.0% (6/8) with mirogabalin dose reduction were recovered or improved (Fig. 3b). Corresponding results for peripheral edema and edema are shown in Supplementary Fig. S2.

Fig. 3.

Number of (a) dizziness and (b) somnolence events according to outcome and mirogabalin treatment status following the onset of mirogabalin-related adverse drug reactions. Data are n (%)

Figure 4 shows the number of events for dizziness and somnolence according to severity and outcome after mrADR onset (Fig. 4a, b) as well as according to outcome, severity, and treatment status (Fig. 4c, d). No notable differences in outcomes between mild and moderate severity were observed for dizziness and somnolence. Following the onset of dizziness, 80.0% (4/5) of events with moderate symptoms were recovered. Following the onset of somnolence, 85.7% (6/7) of events with moderate symptoms were recovered or improved.

Fig. 4.

Number of (a) dizziness and (b) somnolence events according to severity and outcome following the onset of mirogabalin-related adverse drug reactions, and number of (c) dizziness and (d) somnolence events according to outcome, severity, and treatment status. Data are n (%). Discont. discontinuation

For dizziness and somnolence of mild severity, the most common treatment status was no change in the mirogabalin dose, regardless of the outcome (21/28; 75.0% and 23/28, 82.1%, respectively; Fig. 4c, d). For dizziness of moderate severity, the proportions of dose reduction (3/5 [60.0%]), interruption (1/5 [20.0%]), and discontinuation (1/5 [20.0%]) were higher than those of no change in the mirogabalin dose (0/5 [0.0%]). Dose reduction (3/7 [42.9%]) and discontinuation (2/7 [28.6%]) also occurred more frequently than no change in the mirogabalin dose (1/7 [14.3%]) for somnolence of moderate severity, whereas interruption occurred at the same frequency (1/7 [14.3%]). Corresponding data for peripheral edema and edema are shown in Supplementary Fig. S3.

Relationships between VAS Score for Leg Pain and Duration of LSS and Leg Pain

In both treatment groups, there was a weak positive correlation between LSS duration and change in VAS score for leg pain from baseline to week 12 (mirogabalin add-on group: n = 85, r = 0.2401, P = 0.0269; NSAIDs-alone group: n = 75; r = 0.3078; P = 0.0072; Fig. 5a). A weak positive correlation between leg pain duration and change in VAS score for leg pain from baseline to week 12 was observed in the NSAIDs-alone group (n = 75; r = 0.2515, P = 0.0295), whereas no statistically significant correlation was observed in the mirogabalin add-on group (n = 85; r = 0.1835; P = 0.0928; Fig. 5b). Similar results were obtained in the per protocol set (Supplementary Fig. S4).

Fig. 5.

Relationships between change in visual analogue scale (VAS) score for leg pain from baseline to week 12 and (a) LSS duration and (b) leg pain duration (modified intention-to-treat set). Spearman’s rank correlation coefficient (r). The change from baseline to week 12 in VAS score was estimated using the mixed model for repeated measures. LSS lumbar spinal stenosis; NSAIDs nonsteroidal anti-inflammatory drugs

Discussion

Mirogabalin is associated with ADRs such as dizziness, somnolence, peripheral edema, and edema. In our previous post hoc analysis, most initial ADR onsets were observed at the start of administration or at dose escalation [11]. However, the duration of these ADRs, that is, the time to recovery or improvement, has not been previously clarified. Although the findings from this study are limited to patients with LSS, they provide novel and clinically relevant information regarding the clinical course, including outcomes and resolution of mrADRs, for all patients taking mirogabalin. Additionally, they may enhance the confidence of clinicians and patients who have previously avoided mirogabalin initiation or dose escalation owing to concerns about tolerability.

The present study clarified the duration of adverse effects and outcomes following their onset in patients receiving mirogabalin. A previous study of pregabalin reported the onset times and durations of adverse effects in patients with central neuropathic pain [12]. However, that study provided only summary statistics (e.g., median or mean values) and did not include a detailed analysis as performed in the present study. The novelty of the present study is to provide detailed evidence regarding the duration, outcomes, and treatment status following the onset of adverse effects. Among the mrADRs, dizziness (25.5%) and somnolence (30.0%) are most frequent [10]. Compared with somnolence, dizziness is considered a symptom that patients rarely experience outside the context of drug-related adverse effects. Our findings showed that most events of dizziness following mirogabalin administration were recovered within 1–10 days. In contrast, somnolence showed a broader distribution of recovery duration, with most events recovered or improved within 1–10 days or 61–70 days. A previous pregabalin study reported that the duration of dizziness was 70.0 days and that of somnolence was 166.5 days [12]. Even when mirogabalin treatment was continued without dose modification, most dizziness and somnolence were recovered or improved, and a higher proportion of events were recovered or improved following dose reduction. The outcomes did not differ according to the severity of these mrADRs. Although some patients did not recover from mrADRs during the study period, our findings suggest that immediate discontinuation of mirogabalin may not be necessary when mild or moderate mrADRs occur. It should be noted that further evaluation is warranted regarding peripheral edema and edema, for which the number of events was limited.

A previous analysis of the MiroTAS study suggested that mrADRs may not substantially affect analgesic efficacy, quality of life, or patient treatment satisfaction [11]. The present study further characterized the duration and outcomes of mrADRs and demonstrated that most events were recovered. Taken together with previous findings, these results suggest several clinical advantages of mirogabalin use. Physicians may initiate treatment after informing patients about the expected timing and duration of dizziness and somnolence. Furthermore, explaining that these adverse events do not compromise treatment efficacy, quality of life, or treatment satisfaction may help improve adherence and enhance therapeutic benefits. A previous database study reported that patients who were titrated to an effective dose of mirogabalin within 45 days had lower treatment discontinuation rates and higher medication adherence compared with those who were not titrated [13].

The primary MiroTAS study demonstrated that adding mirogabalin to NSAID treatment resulted in greater improvements in leg pain and quality of life compared with NSAIDs alone [10]. In addition to LSS, mirogabalin has shown improvements in neuropathic pain due to other orthopedic conditions such as cervical spondylotic radiculopathy and lumbar disc herniation [14, 15]. Although there is considerable evidence for mirogabalin, there are limited data on whether the pain-relieving effects of mirogabalin differ according to disease duration and pain duration. Our previous post hoc analysis of the MiroTAS study demonstrated no significant difference in the percentage of patients with reduced leg pain (a clinically meaningful improvement in VAS score for leg pain of ≥ 20 mm) at week 12, regardless of the duration of LSS and leg pain (< 6 versus ≥ 6 months) [11]. However, these previous results partly differed between the analysis populations (modified intention-to-treat and per protocol sets), and the analysis categorized data into two groups using a cutoff point without performing correlation analysis using continuous variables. These limitations suggest a need for further detailed analysis. To address this issue, the present analysis examined the relationships between changes in VAS score and the durations of LSS and leg pain, revealing weak correlations between changes in VAS score and leg pain duration. This weak positive correlation indicates a trend toward greater reductions in VAS score (i.e., more pain improvement) with longer leg pain duration; however, this association was not clinically meaningful. Given the results of our previous post hoc analysis [11], these findings suggest that mirogabalin as an add-on to NSAIDs may provide pain relief in patients with insufficient pain relief (limited improvement in VAS scores for leg pain) following NSAID monotherapy, irrespective of LSS disease duration or leg pain duration; however, this interpretation should be considered exploratory. Overall, LSS disease duration and leg pain duration were not identified as strong prognostic factors for pain outcomes in this study.

Limitations

This study has several limitations. First, the recorded onset dates for mrADRs represented the dates entered in the electronic data capture system; however, when the exact onset date could not be verified, the date on which outcomes were confirmed by the attending physicians was used as the onset date. This may have led to discrepancies between recorded and true mrADR durations. Additionally, regarding outcomes, recovery was based on the recovery date in the electronic data capture system, whereas improvement/no recovery was based on the date of physician confirmation. Therefore, the recorded dates for improvement do not necessarily represent the actual timing of clinical improvement. Another limitation is that recurrent adverse events of the same type could be reported either as separate events or as a single event, at the discretion of the treating physician. Because the primary endpoint of the primary study was designed to evaluate the efficacy of mirogabalin add-on to NSAIDs, the protocol did not provide detailed guidance on the handling of recurrent adverse events, which may have introduced variability in adverse event reporting. Of note, only mrADRs of dizziness and somnolence were examined in detail in this study, whereas other adverse events could not be adequately evaluated owing to the small number of events. Finally, as the study population was limited to patients with LSS, it is unclear whether these findings can be generalized to other diseases.

Conclusions

This post hoc analysis of the MiroTAS study suggests that the most frequent mrADRs (dizziness and somnolence) are generally manageable under careful monitoring, with most events recovered or improved after onset. However, this was a post hoc descriptive analysis; given the limited sample size and statistical power, further research is needed to draw definitive conclusions. Furthermore, the pain-relieving effect of mirogabalin add-on therapy was weakly associated with LSS duration and not associated with leg pain duration, suggesting that mirogabalin may have consistent efficacy in patients with insufficient pain relief following NSAID monotherapy, irrespective of these characteristics.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship for this article, take responsibility for the integrity of the work as a whole, and have given their approval for this version to be published. Conceptualization, methodology, and writing—review and editing: Takuya Nikaido, Kazuhito Shiosakai, and Hiroshi Sakoda; formal analysis: Takuya Nikaido; validation: Takuya Nikaido; and writing—original draft: Takuya Nikaido.

Funding

This study was funded by Daiichi Sankyo, Co., Ltd., Tokyo, Japan, which also provided funding for fees associated with medical writing support and the journal’s Rapid Service Fee. Daiichi Sankyo Co., Ltd. was involved in the study design, planning of the data analysis, data interpretation, and development of the manuscript, but was not involved in data management and statistical analysis. EPS Co., Ltd. was directly involved in data management and statistical analysis, which was funded by Daiichi Sankyo, Co., Ltd. The study results were not influenced by the commercial sponsor of this study (Daiichi Sankyo Co., Ltd.). The corresponding author had final responsibility for the decision to submit.

Data Availability

The anonymized data underlying the results presented in this manuscript may be available to researchers upon submission of a reasonable request to the corresponding author. The decision to disclose the data will be made by the corresponding author and the funder, Daiichi Sankyo Co., Ltd. The data disclosure can be requested for 36 months from the article publication.

Declarations

Conflict of Interest

Takuya Nikaido has received lecture fees from Daiichi Sankyo Co., Ltd. Kazuhito Shiosakai and Hiroshi Sakoda are employees of Daiichi Sankyo Co., Ltd.

Ethical Approval

The primary study was approved by the Clinical Research Review Committee, Fukushima Medical University Certified Review Board (CRB no. CRB2200002), which notified all participating centers of the approval. The primary study was conducted in accordance with the ethical principles, Clinical Trials Act, and relevant notifications stipulated in the Declaration of Helsinki (as revised in 2013). All patients provided written informed consent. The primary study was registered in the Japan Registry of Clinical Trials under identifier jRCTs021200007.