Restless legs-like syndrome as an emergent adverse event of CGRP monoclonal antibodies: A report of two cases

Abstract

Abstract

Background

One of the advantages of CGRP monoclonal antibodies is their excellent safety and tolerability. However, postmarketing surveillance, is essential to detect potential rare emergent adverse events.

Objectives

To report two patients who developed restless legs syndrome symptoms after treatment with CGRP antibodies.

Methods and results

Two women with chronic refractory migraine, with no significant medical antecedents, developed typical restless legs syndrome symptoms 1.5 and 4 months after starting erenumab 140 mg, respectively. In case 1 symptoms resolved when erenumab was stopped for two months but reappeared on galcanezumab. In both patients migraine attacks had dramatically decreased and no iron deficiency was found.

Conclusions

Even though caution is needed before establishing a causal relationship, these cases suggest that restless legs-like symptoms might be an emergent adverse event of CGRP antibodies, regardless of the mechanism of action. We propose that plastic changes in CGRP sensory fibers, which are very abundant in legs, induced by CGRP monoclonal antibodies could be the reason for restless legs syndrome development.

Introduction

Calcitonin gene-related peptide (CGRP) monoclonal antibodies are the first treatments specifically developed for migraine prevention. Both controlled trials and short real life experience have shown that CGRP antibodies are effective even for chronic migraineurs who previously failed the standard treatments with a clean safety and tolerability profile, similar to placebo in controlled studies (1). As with each new medication, postmarketing surveillance allows monitoring of the rare adverse events of treatments for the long-term and provide further information on benefit-risk profile of these antibodies in real-world settings.

Here we report on two chronic migraine patients who developed de novo restless legs syndrome (RLS) symptoms during treatment with CGRP antibodies.

Case reports

Case 1. This 47-year-old housewife, with no medical or psychiatric antecedents, referred to a history of migraine without aura since adolescence. There was no personal or family history of RLS. She was sent to our Headache Unit 12 years ago. For the last two years she suffered between 20 and 25 headache days per month, most of them meeting migraine without aura criteria and needing symptomatic treatment with NSAIDs plus triptans. Beta-blockers, flunarizine and amitriptyline, at therapeutic dosages, were not effective. General and neurological examination and brain MRI were unremarkable. We began topiramate with some apparent benefit, but this drug was interrupted by intolerance due to loss of appetite and weight. Candesartan was well tolerated, but no clear efficacy was noticed. Ten years ago, quarterly treatment with onabotulinumtoxinA, following PREEMPT protocol, was initiated with moderate response: headache frequency decreased to 10–16 days per month and she felt that her remaining attacks usually responded better to rizatriptan. No adverse events were reported.

Monthly subcutaneous erenumab 140 mg was started in February 2020 and onabotulinumtoxinA was stopped. Already during the first month her headaches dramatically improved to ≤4/month. Two weeks after the second injection she complained of moderate-severe burning and painful symptoms in both legs, more pronounced late in the afternoon and/or when relaxing, and worsening when she lay down. These symptoms urged her to move her legs, as walking improved these symptoms. She had a body-mass index of 19.4 kg/m² and the examination of both legs was unremarkable. Laboratory determinations, with complete biochemistry, CK, immunologic profile (including antinuclear, anti-smooth muscle, anti-mitochondrial and anti-dermatomyositis/scleroderma antibodies together with C3, C4, IgG4, IgA and rheumatoid factor levels) and biochemical markers of bone turnover and metabolism (calcium, parathyroid hormone (PTH), vitamin D, procollagen type I N-terminal propeptide (PINP) peptide, and β-Cross-Laps) offered normal results, EMG, densitometry, Doppler ultrasound exam and MRI of the lower legs disclosed no abnormalities. Her blood count (hemoglobin 14.5 g/dL), as well as her complete iron profile, were unremarkable, but, as she had ferritin levels (25 μg/L; normal range in our laboratory 10–291 μg/L) in the low range and referred to heavy menstrual periods, she was put on oral iron supplements for 3 months with no relief. Treatment with clobazam was attempted, with slight efficacy, and ropinirol was not tolerated due to persistent nausea even at low dosages. At month 4, erenumab was reduced to 70 mg. Her legs symptoms remained, and headache frequency increased to 7 headache days/month. Erenumab was stopped, onabotulinumtoxinA treatment was resumed and leg symptoms disappeared after 20 days, but her headache frequency progressively increased up to 19 days at month 8, when treatment with subcutaneous galcanezumab, 240 mg the first month and then 120 mg monthly, was initiated. Her headaches clearly improved again to ≤5 attacks/month, but identical legs symptoms reappeared two weeks after galacanezumab was begun (Figure 1). Physical exam and laboratory determinations were, again, unremarkable. For the moment she has decided to remain on CGRP antibodies taking into account the relevant improvement on her migraine headaches.

Figure 1.

Relationship in time between the administration of CGRP antibodies and the development of RLS symptoms in Case 1.

Case 2. This 38-year-old female, with the only antecedents of episodic migraine (without aura) since >20 years and non-severe anxiety-depressive syndrome, came to our Headache Unit in 2010 due to chronic migraine starting two years earlier. Her general and neurological exams and a brain MRI were normal. The following preventive treatments have been either non-tolerated or non-efficacious: nadolol, topiramate, amitriptyline, valproate, venlafaxine and candesartan. She takes either eletriptan or zolmitriptan as acute treatments with inconsistent efficacy. Since 2011 she is on quarterly onabotulinumtoxinA (PREEMPT protocol) and since July 2019 her preventive treatment includes duloxetine (60 mg/24h). In spite of these combined treatments she was suffering between 25 and 30 headache days per month, most of them containing clear migrainous characteristics. Six months ago, she began monthly subcutaneous treatment with erenumab 140 mg. A dramatic and consistent improvement, with a reduction higher than 80%, was observed since the first month of injection. At the third month visit she referred no side effects. In her recent sixth month visit she explained us that four months after the beginning of erenumab she noticed crawling and creeping sensation of both legs with a need to move them. These symptoms are more pronounced late in the afternoon and/or when relaxing and worsen when she lay down and improve on walking. She has remained on quarterly 200 U onabotulinumtoxinA injections since erenumab was begun. Physical exam of both lower extremities and laboratory determinations, including CK, iron, immunologic and bone turnover profiles, were unremarkable. Even though these symptoms appear daily, they are really unpleasant just about one-third of the days and, taking into account her excellent response to erenumab, she preferred to remain on this treatment and try first a few lifestyle changes, local treatment and oral iron supplements together with a benzodiazepine at night if necessary.

Discussion

We describe the cases of two women with refractory chronic migraine who developed symptoms in their lower limbs fulfilling current diagnostic criteria for RLS after beginning treatment with erenumab 140 mg. Even though symptoms such as numbness or spasms have been (rarely) noticed with CGRP antibodies in clinical practice (2), to our knowledge, the development of RLS-like symptoms has not been reported as an emergent adverse event of CGRP inhibitors (1,3). Considering the prevalence of the two diseases, migraine and RLS, we cannot rule out a mere coincidence, but a direct relationship between treatment with these antibodies and the development of RLS-like symptoms could also be an explanation, especially in Case 1, where RLS symptoms came out in the second treatment month with erenumab and disappeared three weeks after the CGRP antibody was stopped. Furthermore, in this case the symptoms reappeared after the initiation of galcanezumab, which, on the one hand, reinforces the role of CGRP antibodies in the RLS development and, on the other, shows that this emergent adverse event is shared by two CGRP antibodies with different mechanisms of action.

RLS is a common movement disorder, with an estimated prevalence of 2–15%, which typically affects the legs and can induce sleep disturbance. Recent studies have shown that RLS is a complex disorder in which genetic and environmental factors contribute to the phenotype. Well-demonstrated RLS associated comorbidities include iron deficiency anemia, renal insufficiency (when leading to dialysis) or pregnancy (4). Almost thirty studies have evaluated the prevalence of RLS in patients with migraine or headaches (4,5). RLS as a comorbid disease is increased in migraine with or without aura, ranging from 8.7% to 39.0%, though the degree of association appears to be strongly related to the study design (5). If these two comorbid entities would share pathophysiological mechanisms, successful treatment in this case of migraine, as happened in our two patients, could theoretically have prevented but not induced the appearance of RLS symptoms. Interestingly, RLS cases have also been reported as an adverse event of migraine preventatives, such as topiramate or amitriptyline (6–8).

The cause of RLS is unknown (4) and, accordingly, we do not have a definite explanation for our clinical observations. Using rat experimental models, however, Barry et al demonstrated that leg muscles are predominantly innervated by peptidergic sensory fibers, and that a large proportion of these fibers contain only CGRP and not other sensory peptides, such as substance P as occurs in other parts of the body (9). It is also known that plastic changes in these peripheral CGRP sensory nerves that supply muscles contribute to muscle pain and these changes have been linked to hyperalgesia, allodynia, sensitization of dorsal horn neurons and increases in their receptive field (9,10). Intriguingly, these CGRP positive fibers in the gastrocnemius are of the Aδ type (9), the same kind of fibers where CGRP monoclonal antibodies seem to exert their antimigraine action (11), and exhibit properties consistent with nociceptors (12). It is therefore tempting to suggest that plastic changes in the homeostasis of this peripheral CGRP sensory system induced by the CGRP antibodies could be the reason for RLS development and, according to this reasoning, it could be possible that, as happens with topiramate (13), CGRP antibodies could, in other cases, hypothetically improve RLS symptoms. The observation in Case 1 that RLS-like symptoms were seen with two CGRP antibodies with different targets, the ligand or its receptor, could indicate that it is the antagonism over the CGRP and not their actions on other potential targets, such as the amylin receptor, that could explain a hypothetical sensitization of the peripheral nociceptive system, though this is a matter of speculation. The fact that the development of RLS symptoms was not immediate, with an interval of few months in both cases, could support this hypothesis. A final interesting point, that is worth mentioning, is how purely peripherally acting drugs, such as the CGRP antibodies, are able to improve migraine or develop in this case RLS, two entities in which different areas of the central nervous system play a key role in their pathophysiological mechanisms.

In conclusion, these cases suggest that RLS-like symptoms could be an emergent adverse event of CGRP monoclonal antibodies, regardless of the mechanism of action. We propose that changes in the homeostasis of CGRP sensory fibers, which are very abundant in legs, could be the reason for RLS development.

Clinical implications

RLS-like symptoms could be an emergent adverse event of CGRP monoclonal antibodies, regardless of its target.

The delayed appearance of these symptoms could indicate that changes in the homeostasis of the peripheral nociceptive system inducing sensitization of the CGRP sensory fibers in the legs could be the explanation for RLS development.