Cardio- and cerebrovascular safety of erenumab,a monoclonal antibody targeting CGRP receptors – important studies on human isolated arteries

The monoclonal antibody erenumab (AMG 334) is a new hope for patients suffering from frequent episodic or chronic migraine. It was approved by the FDA in May 2018, soon followed by approval in several European countries (1). Erenumab is a human IgG antibody binding the canonical calcitonin gene-related peptide (CGRP) receptor (2). Like the newly developed monoclonal antibodies binding both α- and β-CGRP, eptinezumab (ALD403), fremanezumab (TEV-48125) and galcanezumab (LY2951742), it has a favorable safety profile superior to all former migraine prophylactics (3–8). In particular, cardiovascular side effects were registered as not more frequent than in the placebo arms. Since all these phase II and phase III studies excluded migraine patients with cardiovascular risks, there was a need to look especially to this patient group. A separate multicenter, randomized, placebo-controlled phase II study was launched in a patient group with stable angina pectoris and documented coronary artery disease (9). The total exercise time until the patients reported angina or showed clear ST-segment depression was not different between the patients treated with erenumab compared to the placebo arm. However, serious pharmacological concerns about the validity of this specific study remain, because no evidence was presented on whether effective CGRP receptor blockade was achieved at the time of the treadmill test (10). Also, the study population needs further attention. Patients with stable angina pectoris in most cases suffer from stenosis of the epicardial conducting portions of the coronary artery, where the role of CGRP is limited, in contrast to the coronary microarteries, where particularly women get their coronary problems when suffering from myocardial ischemia (11). Whereas most patients using the antibodies will be female, the mentioned treadmill study included 78% males. Thus, the question remains open whether blockade of the CGRP receptor would have an effect on coronary artery function in women with microvascular disease.

Indeed, preclinical evidence has demonstrated that CGRP is one of the most potent vasodilatory substances at cranial and coronary arteries (12,13). Moreover, the triptans, classical antimigraine drugs which block CGRP release, are contraindicated in patients with cardio- and cerebrovascular risks (14). Several preclinical studies implicated CGRP as an important signal molecule that is protective against cardiovascular ischemia (15). For example, in the isolated perfused rat heart (Langendorff preparation), the increase in endothelin-induced perfusion pressure as a measure of constriction of small coronary arteries and the concomitant lowering in pH were significantly ameliorated by CGRP (16). In a similar preparation, coronary artery occlusion followed by reperfusion was applied as a model of myocardial infarction (17). Ischemic preconditioning by short occlusion, which increased CGRP release, reduced the infarct size to a similar extent as exogenous CGRP application, while the CGRP receptor antagonist olcegepant (BIBN4096BS) abolished this protective effect. Following chronic ventricular infarction in a pig model, the number of dorsal root ganglion neurons immunoreactive to CGRP and neuronal NO synthase (nNOS) was increased (18). Additional evidence for the important role of CGRP in cardiovascular functions results from experiments with mice with genetically deleted CGRP. In a mouse model of chronic transverse aortal constriction, fibrosis of cardiac tissues was significantly higher in CGRP knockout mice compared to wild type animals, and their lifetime was reduced (19). Similarly, aortal hypertrophia and hypertension induced by angiotensin was more severe in CGRP knockout animals (20). These data strongly suggest that CGRP is protective against cardiac infarction and hypertrophia. Also, CGRP knockout mice have been reported to develop higher arterial pressure, increased aortic stiffness and elevated plasma levels of norepinephrine compared to wild-type mice (21).

Why then does the CGRP receptor targeting antibody erenumab up to now show no cardio- or cerebrovascular side-effects? Important contributions to clarify these questions provide the current Cephalalgia publications from the groups of A MaassenVanDenBrink (22) and L Edvinsson (23), in which vasoactive responses to various vasodilatory and vasoconstrictory substances at isolated human cerebral, meningeal, mammary and coronary arteries were examined in the presence of erenumab. The antibody showed no direct contractile or relaxant effect but, as expected, it antagonized the vasorelaxation induced by CGRP. The contraction induced by sumatriptan or dihydroergotamine or the relaxation induced by substance P, acetylcholine, sodium nitroprusside, pituitary adenylate cyclase-activating peptide (PACAP), vasoactive intestinal polypeptide (VIP) and nicardipine was not significantly influenced. These results confirm the specificity of erenumab in inhibiting the relaxation of human arteries induced by CGRP without interaction with other vasoactive compounds. The results are also consistent with those of older preclinical experiments, which showed that the blockade of CGRP receptors with the competitive antagonists CGRP_8-37 and olcegepant inhibits arterial dilatation and meningeal blood flow increases evoked by electrical field stimulation but has nearly no impact on the basal arterial diameter and blood flow (24–26). Also, scavenging CGRP by a mirror-image CGRP binding oligonucleotide (Spiegelmer) did not change arterial diameter and basal blood flow (25–27), whereas 5-HT₁ receptor agonists like sumatriptan decreased both the basal meningeal blood flow and the evoked flow (28,29).Taken together, the decisive difference between CGRP or CGRP receptor binding substances and triptans is that only the latter cause direct arterial vasoconstriction through 5-HT_1B receptor activation, which decreases cyclic monophosphate (cAMP) in smooth arterial vessels (30).

If there is no constitutive effect on arterial vessels of substances interacting with the CGRP mediator system, we can conclude that the basal CGRP release does not reach concentrations that cause significant vasodilatation. Thus, the effect of these CGRP inhibiting substances is limited to situations with high CGRP release, as it probably occurs in migraine (31) but also in cardio- or cerebrovascular pathological events (13). A further difference between attack treatments and the use of preventive antibodies is that acute drugs usually have a short plasma half-life and thus their potential cardiovascular risk will be of short duration, while the antibodies have a long plasma half-life and thus remain present during several weeks or months. However, apart from the evidence provided by the papers of Rubio Beltran and Ohlsson discussed here, there are other reasons that might explain the lack of cardiovascular events caused by erenumab (and probably other substances interacting with the CGRP system) thus far in clinical trials. First, CGRP can bind to other receptors of the calcitonin family, namely to adrenomedullin and amylin receptors, to the amylin-1 receptor even with the same affinity as to the canonical CGRP receptor (32,33). Second, plasma levels of adrenomedullin, which activates the same intracellular cAMP pathway as CGRP causing vasodilatation, is found to be significantly increased during myocardial infarction (35), and adrenomedullin is involved in a widespread network of additional neurohumoral and endothelial factors that are all cardioprotective (36). Taken together, coronary vascular regulation seems to be a highly secured multifunctional system, which may explain the apparent lack of cardiovascular side-effects of erenumab (and CGRP-binding monoclonal antibodies) observed thus far. Nevertheless, we would urge that studies be performed in relevant models under cardiovascular and cerebrovascular ischemic conditions to obtain essential knowledge on the use of these drugs in patients that may particularly be at potential risk.