Fremanezumab and its isotype slow propagation rate and shorten cortical recovery period but do not prevent occurrence of cortical spreading depression in rats with compromised blood brain barrier

Abstract

Most centrally-acting migraine preventive drugs suppress frequency and velocity of cortical spreading depression (CSD). The purpose of the current study was to determine how the new class of peripherally acting migraine preventive drug (i.e., the anti-CGRP-mAbs) affect CSD – an established animal model of migraine aura, which affects about 1/3 of people with migraine - when allowed to cross the blood brain barrier (BBB). Using standard electrocorticogram recording techniques and rats in which the BBB was intentionally compromised, we found that when the BBB was opened, the anti-CGRP-mAb fremanezumab did not prevent the induction, occurrence or propagation of a single wave of CSD induced by a pinprick, but that both fremanezumab and its isotype were capable of slowing down the propagation velocity of CSD and shortening the period of profound depression of spontaneous cortical activity that followed the spreading depolarization. Fremanezumab’s inability to completely block the occurrence of CSD in animals in which the BBB was compromised suggests that CGRP may not be involved in the initiation of CSD, at least not to the extent that it can prevent its occurrence. Similarly, we cannot conclude that CGRP is involved in the propagation velocity or the neuronal silencing period (also called cortical recovery period) that follows the CSD because similar effects were observed when the isotype was used. These finding call for caution with interpretations of studies that claim to show direct CNS effects of anti-CGRP-mAbs.

Introduction

A large body of evidence in both man and animals supports an important role for CGRP in the pathophysiology of migraine headache [12; 17; 37]. The strength and consistency of the evidence that support CGRP role in migraine has led to the development of CGRP receptor antagonists, humanized anti-CGRP monoclonal antibodies (anti-CGRP-mAb) and human anti-CGRP receptor monoclonal antibody (anti-CGRPr-mAb) for the treatment of chronic and high-frequency episodic migraine.

Because of their large size, monoclonal antibodies are unlikely to cross the blood brain barrier in large quantities, and so appear to accomplish their therapeutic effects by targeting peripheral structures outside the central nervous system (CNS). In support of this view, we showed recently that fremanezumab - an anti-CGRP-mAb - prevents the activation of certain classes of meningeal nociceptors after cortical spreading depression (CSD) and, consequently, activation of central trigeminovascular neurons whose activity is driven by input they receive from the nociceptors [24; 25]. While studying fremanezumab’s effects on activation of peripheral and central trigeminovascular neurons by CSD, we noted that despite its large size, fremanezumab alters some of the functional properties of CSD – a CNS event that takes place inside the blood brain barrier and is thought to underlie migraine aura [19; 30].

Two recent in vitro studies have shown that CGRP receptor antagonists reduce the magnitude of cortical (and retinal) spreading depression [44; 48], in the current study, we sought to explore this unexpected observation because if it holds true, it can define another mechanism by which this class of drugs may be able to reduce the overall impact of abnormal brain functions on the nociceptors. Because previous studies found that centrally-acting migraine preventive drugs such as topiramate, valproate, propranolol or amitriptyline slow down the propagation rate of CSD and suggested that slowing of CSD propagation can eventually lead to complete failure of CSD propagation [2; 4; 7; 45], we attempted to determine if a peripherally-acting migraine preventive drug such as fremanezumab can also slow down the propagation velocity of CSD if we intentionally let it enter the cortex. We also attempted to determine the effects of fremanezumab on the duration of neuronal silencing. Neuronal silencing usually follows the wave of depolarization and is marked by a period of time in which spontaneous and evoked activity of cortical neurons is profoundly depressed [20; 43].

Using standard electrophysiological techniques, in the current study we measured the amplitude of each CSD wave, its propagation rate between the occipital and parietal cortices, and the duration of the neuronal silencing period that followed each depolarization wave in fremanezumab-treated, isotype-treated, and untreated male and female rats in which the blood brain barrier (BBB) was compromised at the recording site in the occipital cortex and the CSD initiation site in the parietal cortex.

Materials and methods.

Experiments were approved by the Beth Israel Deaconess Medical Center and Harvard Medical School standing committees on animal care and were conducted in accordance with the U.S. National Institutes of Health’s Guide for the Care and Use of Laboratory Animals.

Fremanezumab and isotype injections and surgical preparation:

Male and female Sprague-Dawley rats weighing 250–340 g were anesthetized with urethane (0.9 –1.2 g/kg i.p.), fitted with an endotracheal tube to allow artificial ventilation (0.1 L/min of O2) and an intra-femoral vein cannula for later infusion of fluids and drugs. Rats were placed in a stereotaxic apparatus and core temperature was kept at 37°C using a heating blanket. End-tidal CO2 was monitored continuously and kept within physiological range (3.5–4.5 pCO2). Once stabilized, rats received a bolus intravenous injection of fremanezumab (30 mg/kg, diluted in 0.7 ml saline), isotype (30 mg/kg diluted in 0.7 ml saline), or saline (0.7 ml saline) followed by continuous infusion of 0.9% saline at a rate of 1.0 ml/h. Isotype control antibodies are used in preclinical models to distinguish between effects of an antibody binding to a specific antigen (e.g. CGRP) from that of non-antigen binding effects caused by interactions with Fc receptors or other proteins and the antibody. As preclinical animal models are used routinely to determine mechanism of therapeutic antibodies this is a very important control to ensure that you are examining the mechanism attributed to the targeted antigen, in our case binding to CGRP and thereby blocking its interaction with CGRP receptors. For placing the electrocorticogram (ECoG) electrodes, a large craniotomy (6 × 6 mm) was performed between Lambda and Bregma and the exposed dura was kept covered and moist.

CSD induction and electrocorticogram recording:

Four hours after infusion of fremanezumab, isotype, or saline, we induced CSD in the occipital cortex by pinprick and recorded its propagation using two glass micropipettes (7μm tip, ∼ 1 MΩ, filled with 0.9% saline, at a depth of ∼ 100 μm), one placed in the occipital cortex and the second, 4 mm anteriorly within the parietal cortex (Fig. 1). This configuration allowed us to measure the amplitude, recovery period and spreading velocity of each wave of CSD. Using Spike 2 software (CED), the ECoG was captured and filtered off-line in 2 different ways to allow accurate measurements of the different parameters. A partial DC removal filter (time constant of 30s) was used to measure CSD amplitude and the propagation velocity, whereas full DC removal filter (time constant of 0.07s) was used to measure the neuronal silencing period – defined here as the recovery period.

Fig 1.

A CSD trace illustrating the parameters measured for the study. Amplitude is defined as the peak-to-peak positive-to-negative DC shift expressed in mV. Spreading velocity is calculated as the time between the appearance of the CSD wave at the occipital and parietal electrodes, divided by the distance between them, and expressed in mm/min. Neuronal silencing period is measured between the last signal of cortical activity at the onset of CSD and the first signal of cortical activity after the CSD wave had passed the recording electrode, and expressed in seconds. B. Schematic representation of the rat’s brain that showing electrodes placement and site of CSD induction by pinprick.

Experimental protocol:

All experiments included continuous recording of cortical activity for 30 minutes prior to induction of a single wave of CSD (baseline period) and 30 minutes after the CSD wave was recorded in both the occipital and parietal cortices.

Data analyses:

Included in the data analyses were CSD amplitude (defined as the peak-to-peak positive-to-negative DC shift expressed in mV), CSD spreading velocity (calculated as the time between the appearance of the CSD wave at the occipital and parietal electrodes, divided by the distance between them, and expressed in mm/min), and the neuronal silencing period (representing the depression phase of CSD) that followed the CSD wave (measured between the last signal of cortical activity at the onset of CSD and the first signal of cortical activity after the CSD wave had passed the recording electrode, and expressed in seconds). Median values calculated in each of the 3 experimental groups were compared using non-parametric statistics (Wilcoxon signed-ranks test). Two tailed level of significance was set at 0.05. Results are expressed as median [interquartile range, IQR].

Results:

Female vs. male (Fig. 2):

Fig 2.

Box plots (median and IQR) of spreading velocity (A), amplitude (B), and recovery (C), showing no difference between male vs female control rats.

Experiments were conducted in 50 female and 52 male rats. The data obtained from the female and male rats did not differ and thus were combined. Spreading velocity was 3.9 mm/min [IQR: 3.7 to 5.4] in males and 4.5 mm/min [IQR: 4.0 to 5.0] in females (p= 0.747). CSD Amplitude was 24.3 mV [IQR: 19.9 to 27.9] in males and 24.9 mV [IQR: 20.3 to 27.1] in females (p= 0.681). Neuronal silencing (recovery) period was 392 seconds [IQR: 287 to 492] in males and 402 seconds [IQR: 305 to 566] in females (p= 0.681).

The propagation rate of CSD (spreading velocity) was slower in animals treated with fremanezumab or isotype (Fig. 3):

Fig 3.

A. Examples of CSD traces that show different spreading velocities among saline-, isotype- and fremanezumab-treated groups. B. Box plots (median and IQR) showing differences in the CSD spreading velocity between the saline-, isotype- and fremanezumab-treated groups. Note that the propagation velocity was slower in the isotype- and fremanezumab-treated animals, and that there was no significant difference between the fremanezumab- and isotype-treated groups (p>0.05).

The CSD spreading velocity was 4.2 mm/min [IQR: 3.8 to 5.2] in the control group (n=12), 3.4 mm/min [IQR: 3.1 to 3.6] in the isotype group (n=11), and 3.0 mm/min [IQR: 2.4 to 3.7] in the fremanezumab group (n=12). These values yielded significant difference between the fremanezumab-treated and saline-treated groups (p=0.003), between the isotype-treated and saline-treated groups (p=0.012), but not between the fremanezumab-treated and isotype treated groups (p=0.373)

The cortical recovery period (period of neuronal silencing) that follows CSD was shorter in animals treated with fremanezumab or isotype (Fig. 4):

Fig 4.

A. Examples of CSD traces (DC removal filter applied) showing the neuronal silencing period of a saline, isotype and fremanezumab treated animal. The neuronal silencing period was shorter in both isotype and fremanezumab treated animals as compared to the saline treated animal.

B. Box plot that shows the difference in the neuronal silencing period between the Saline, Isotype and Fremanezumab treated groups. The saline treated group was significantly longer than the Isotype (p=0.038) and Fremanezumab (p=0.007) treated groups (the neuronal silencing period induced by CSD lasted 398 [IQR: 293 to 509] in the control group (n=24), 300 sec [IQR: 234 to 271] in the isotype group (n=22), and 312 sec [IQR: 223 to 348] in the fremanezumab group (n=21)). There was no significant difference between the Fremanezumab and isotype treated groups (p>0.05).

The neuronal silencing period induced by CSD lasted 398 sec [IQR: 293 to 509] in the control group (n=24), 300 sec [IQR: 234 to 271] in the isotype group (n=22), and 312 sec [IQR: 223 to 348] in the fremanezumab group (n=21). These values yielded significant difference between the fremanezumab-treated and saline-treated groups (p=0.007), between the isotype-treated and saline-treated groups (p=0.038), but not between the fremanezumab-treated and isotype treated groups (p=0.741)

CSD amplitude was unaffected by fremanezumab or its isotype (Fig. 5):

Fig 5.

A. Examples of CSD traces of the saline, isotype and fremanezumab treated groups showing no differences in the amplitude between them. B. Box plot showing no differences in the amplitude of the CSD wave between the saline treated group and the Isotype (p=0.54) and Fremanezumab (p=0.35) treated groups. (CSD amplitude was 24.7 mV [IQR: 20.1 to 27.4] in the control group (n=24), 24.7 mV [IQR: 20.4 to 29.7] in the isotype group (n=22), and 26.6 mV [IQR: 20.0 to 28.5] in the fremanezumab group (n=20). There was no significant difference between the Fremanezumab and isotype treated groups (p>0.05).

CSD amplitude was 24.7 mV [IQR: 20.1 to 27.4] in the control group (n=24), 24.7 mV [IQR: 20.4 to 29.7] in the isotype group (n=22), and 26.6 mV [IQR: 20.0 to 28.5] in the fremanezumab group (n=20). The p value for the control vs. isotype is p=0.54. For the control vs. fremanezumab it is p=0.35. These values yielded no significant difference between the fremanezumab-treated and saline-treated groups (p=0.35), between the isotype-treated and saline-treated groups (p=0.58), and between the fremanezumab-treated and isotype treated groups (p=0.528).

Discussion

In the current study, we used standard electrocorticogram recording techniques to test the effects of a new class of migraine preventive drugs, namely the anti-CGRP-mAbs, on different properties of CSD in rats with locally-compromised BBB. We found that fremanezumab did not prevent the induction, occurrence or propagation of a single wave of CSD induced by a pinprick, but that both fremanezumab and its isotype were capable of slowing down the propagation velocity of CSD and shortening the periods of profound depression of spontaneous and cortical activity that follows the spreading depolarization. Because our experimental set up allows fremanezumab to reach cortical areas that surround the cortical recording electrode and the pinprick insertion path [29], its failure to prevent the initiation and propagation of a single wave of CSD is unexpected. It is unexpected because fremanezumab, which is too large to cross the BBB and reach the cortex in rats with uncompromised BBB [29; 41], reached the cortical area where recording of CSD was made [29], because CGRP and its receptors are densely distributed throughout the cortex [49], and because in cortical slices CGRP receptors blockade inhibited CSD [44]. Given the unequivocal evidence for fremanezumab presence in the cortex hours before induction of CSD, our findings suggest that CGRP may not be importantly involved in the initiation or propagation of CSD in animals or aura in humans. While unable to prevent the occurrence of CSD, fremanezumab slowed down the propagation rate and shortened the cortical recovery period that followed the CSD. But because these effects were also observed in rats treated with isotype, the interpretation of these findings require caution. One logical way to interpret these findings is that the observed altered properties of CSD may be unrelated to the mechanism of action by which fremanezumab prevents migraine (prevention of CGRP binding to its receptor) – a possibility that is also supported by our recent reports of clear distinction between fremanezumab and isotype effects on the activation and sensitization of central and peripheral trigeminovascular neurons by CSD [23; 25]

Although not well reported in the literature, the few studies in which aura was followed in patients undergoing different brain scanning [6; 11; 31; 32; 38] suggest that in most cases, only one wave of CSD is formed and propagated. As fremanezumab does not prevent the occurrence of a single wave of CSD or its ability to propagate, it may be tempting to suggest that fremanezumab cannot completely block CSD because CSD properties are not CGRP-mediated. Along this line, not even centrally-acting migraine preventive drugs appear to block the induction and propagation of a single wave of CSD [2; 4; 7; 14; 18; 45]. If, as suggested before, CSD provides a therapeutic target for migraine preventive drugs, it may be more appropriate to try set a goal for such drugs to completely eliminate our ability to induce CSD.

Previous studies in animal models of migraine aura found that centrally-acting migraine preventive drugs such as topiramate, valproate, propranolol, amitriptyline, methysergide, lamotrigine, gabapentin, tonabersat, ketamine and propofol slow down the propagation rate of CSD and suggested that slowing of CSD propagation can eventually lead to complete failure of CSD propagation [2; 4; 7; 14; 18; 35; 45]. In agreement with these studies, we found that cortical presence of fremanezumab also slowed down the propagation velocity of CSD. The difficulty in discussing how fremanezumab slows down the propagation velocity is that neither the significance of CSD propagation rate nor the exact mechanism that determines it is well known – likely a result of technical inability to image rapid changes in ion and neurotransmitter levels in vivo with subcellular resolution [8]. While it is generally accepted that during CSD neuronal depolarization spreads in a wave-like manner that involves Ca²⁺ [13; 27], K⁺ [10; 47] and glutamate [9], it is unclear whether it is the increase in intracellular Ca²⁺, or the massive release of K⁺ or glutamate into the interstitial space that is at the forefront of the propagating CSD wave [34; 43]. It is also unclear whether it is the astrocytes [5; 21; 33] or the neuronal Ca²⁺ influx [13] that determine the propagation properties of CSD [8]. In the absence of such knowledge, one can only speculate that the slowing of CSD propagation rate observed in the current study involves a decrease in Ca²⁺ influx and/or interstitial K+ and/or glutamate level. Theoretically, the decrease in interstitial K+, currently thought to appear before the neuronal Ca2+ increase or the glutamate release [8], could reflect transient increase in astrocytic ability to clear the K+ and glutamate from the extracellular space [28; 36], or reduced neuronal swelling which consequently slows down interstitial fluid flow into the glymphatic system [40] – a major passage for clearance of molecular waste [16].

To the best of our knowledge, the current study provides the first documentation of a migraine preventive drug (i.e., fremanezumab) effect on the duration of the neuronal silencing period after CSD. As stated before, profound depression of spontaneous and evoked cortical activity, attributed to neuronal silencing, is one of the most notable characteristics of spreading depolarization [20; 43]. The few studies about factors that contribute to the period of time in which cortical neurons are silent attribute it to massive inhibition of synaptic transmission resulting from decreased neuronal firing and presynaptic release, postsynaptic inhibitory shift in excitatory/inhibitory ratio [39] and alterations in the metabolic demand of repolarizing molecules such as adenosine [22] and K_ATP channels activation level. But because the neuronal silencing period was also shortened by the isotype, we cannot conclude that neutralization of cortical CGRP mediated these effects. Since the isotype, which is a human immunoglobulin G2Δa that lacks the ability to inhibit CGRP signaling, it may be reasonable to consider the possibility that its effects on CSD properties may be attributed to the overall anti-inflammatory and immunomodulatory effects of IVIG therapy [1; 42]. IVIG therapy has been shown to reduce seizure frequency and normalize EEG recording in several randomized and non-randomized controlled trials [3; 15; 26; 46; 50; 51].

In summary, our findings suggest that even when anti-CGRP-mAbs are allowed to reach the cortex, they cannot abolish the initiation and propagation of CSD. These findings provide further evidence for the view that the mechanisms by which this class of drugs prevent migraine is mainly through their ability to directly alter headache-related peripheral functions in meningeal nociceptors, cerebral and meningeal blood vessels and possible immune cells which indirectly alter excitability and responsiveness of neurons in brain areas involved in migraine pathophysiology.