From my article on MDEdge, by Larry Robbins, M.D.
Editors’ note: For the April edition of Expert Perspectives, we asked two leading neurologists to present differing views on the use of calcitonin gene-related peptide (CGRPs) monoclonal antibodies (mAbs) for the treatment of migraine. Here, Larry Robbins, MD, of the Robbins Headache Clinic in Riverwoods, IL, discusses potential safety concerns associated with this drug class. To read a counterargument in which Jack D. Schim, MD, of the Neurology Center of Southern California, discusses the observed benefits of the CGRP mAbs, click here.
Safety Concerns with CGRP Monoclonal Antibodies
Lawrence Robbins, MD
Lawrence Robbins, MD is an Associate Professor of Neurology, Chicago Medical School, and in private practice in Riverwoods, Ill.
Dr. Robbins discloses Speaker’s bureau remuneration from AbbVie, Amgen, Biohaven, Impel NeuroPharma, Lundbeck, and Teva.
The calcitonin gene-related peptides (CGRP) monoclonal antibodies (mAbs) were introduced in 2018 as efficacious with few adverse effects (AE). Unfortunately, the phase 3 trials failed to indicate the large number of AEs that have since been identified. This is a common occurrence with new drugs and monoclonals.
There are few adverse events identified in the package insert (PI) for any of the 4 CGRP mAbs, but again that is not unique to this class. Post-approval, it frequently takes time to piece together the true AE profile. Reasons that the phase 3 studies may miss AEs include 1) The studies are powered for efficacy but are not powered for AEs in terms of both the number of patients and the length of the study; 2) Studies do not use a checklist of likely AEs; and 3) Adverse effects become “disaggregated:” one person states they have malaise, another tiredness, and another fatigue.
In the ensuing years after the launch of the CGRP mAbs, various lines of evidence point to the AEs attributed to these mAbs. The lines of evidence include the FDA/FAERS website, published studies, the collective experience of high prescribers, and filtered comments from patient chat boards. In addition, I have received hundreds of correspondences from providers and patients regarding serious AEs, which are detailed further in this article.
As of March 2022, the FDA/FAERS website has listed approximately 50,000 adverse events in connection with the mAbs. The number of serious events (hospitalization or life-threatening issues) was about 7,000. These large numbers, just 3.75 years after launch, are similar to those listed for onabotulinumtoxin A after 30 years! Most of the reports involve erenumab. This is most likely because erenumab was first out and is the most prescribed. I do not believe it is more dangerous than the others. Considering that the vast majority of adverse events go unreported these are staggering numbers. Regarding serious adverse events, only 1% to 10% are actually reported to the FDA. Nobody knows the true percentage of milder adverse events that are actually reported, but in my experience, it is extremely low.
Unfortunately, only adverse events are listed on the FDA/FAERS website, not adverse effects, which are just as important to discuss. In my small practice I have observed 4 serious adverse effects in women I believe are attributable to the CGRP mAbs. These include a cerebrovascular accident in a 21-year-old patient, a case of reversible cerebral vasoconstrictive syndrome in a 61-year-old patient, severe joint pain in a 66-year-old patient, and a constellation of symptoms that resembled multiple sclerosis in a 30-year-old patient. I have administered onabotulinumtoxinA to thousands of patients for 25 years, with no serious adverse effects.
There have been a number of post-approval articles, studies, and case reports published since the launch of the CGRP mAbs. Many “review” or “meta-analysis” articles tend to repeat the results of the pharma-sponsored studies. They typically characterize the CGRP monoclonals as safe, with few adverse events. Long-term safety extension studies almost always conclude that the drug is safe. In my opinion we cannot rely on the results of these studies.
However, other studies tell a different tale. One observational study of erenumab concluded that adverse effects contributed to 33% of the discontinuations. (10) Another study reported that 63.3% of patients taking the mAbs described at least one adverse effect. A study of patients who had been prescribed erenumab indicated that 48% reported a non-serious adverse event after 3 months.
Neurologists are generally unaware of the dangers posed by the CGRP monoclonals. In September 2021, I engaged in a debate on this topic during the International 15th World Congress on Controversies in Neurology (CONy). Prior to the debate the audience was polled, and 94% believed that the CGRP monoclonals to besafe. After our debate only 40% felt that these monoclonal antibodies to be safe. I think that the audience, primarily consisting of neurologists, was not informed as to the potential dangers of the CGRP monoclonals.
In our practice
For refractory patients I do prescribe these CGRP monoclonals. However, I feel they should only be prescribed after a number of other safer options have failed. These include the natural approaches (Petadolex, magnesium, riboflavin), several of the standard medications (amitriptyline, beta blockers, topiramate, valproate, ARBs), and onabotulinumtoxinA. With 27 years of demonstrated safety and efficacy, onabotulinumtoxinA is always a strong consideration. In addition, the newer gepants for prevention (rimegepant and atogepant) appear to be safer options than are the CGRP monoclonals, although we cannot say this definitively at this time.
In 2018, our clinic began a retrospective study that lasted until January 2020. We assessed 119 patients with chronic migraine who had been prescribed one of the CGRP monoclonals. (13) This study incorporated the use of a checklist of possible adverse effects. Each of these AEs had created a “signal”. We initially asked the patients, “Have you experienced any issues, problems, or side effects due to the CGRP monoclonal?” The patients subsequently were interviewed and asked about each possible AE included on the checklist. The patient and physician determined whether the AE was possibly due to the CGRP mAb. After discussing the checklist, 66% of the patients concluded they had experienced 1 additional AE, not originally disclosed with our initial question, that they attributed to the CGRP mAb. Most of these patients identified more than 1 additional AE through use of the checklist.
To determine the true AE profile post-approval, we rely upon the input of high prescribers, who often can provide this necessary feedback. I have assessed input from headache provider chat boards, private correspondence with many providers, and discussions with colleagues at conferences. The opinions do vary, with some headache providers arguing that there are not that many AEs arising from the mAbs. Many others believe, as I do, that there are a large number of AEs. In my observations, there is not a consensus among headache providers.
The CGRP patient chat boards are another valuable line of evidence. I have screened 2,800 comments from patients regarding AEs. I filtered these down into 490 “highly believable” comments. Among those, the AEs align very well with our other lines of evidence.
If we put all of the post-approval lines of evidence together, we come up with the following list of “adverse effect signals” attributable to the CGRP monoclonals. These include: constipation (may be severe: hence the warning in the erenumab PI), injections site reactions, joint pain, anxiety, muscle pain or cramps, hypertension or worsening hypertension (there is a warning in the erenumab PI), nausea (which may be severe: the “area postrema syndrome” has occurred), rash, increased headache, fatigue, depression, insomnia, hair loss, tachycardia (and other cardiac arrhythmias), stroke, angina and MI, weight gain or loss, irritability, and sexual dysfunction. There are also other AEs. In his review of the CGRP mAbs, Thomas Moore (a leading expert in AEs) cited the “sheer number of case reports, and it is likely that AEs of this migraine preventive were underestimated in the clinical trials.”
This discussion has focused on short-term AEs. We have no idea regarding long-term effects, but I suspect that we will encounter serious ones. Evolution has deemed CGRP to be vital for 450 million years. We ignore evolution at our peril.
CGRP is a powerful vasodilator and protects our cardiovascular and cerebrovascular systems. CGRP resists the onset of hypertension. Wound and burn healing, as well as tissue repair, require CGRP. Bony metabolism and bone healing are partly dependent upon CGRP. CGRP protects from GI ulcers and aids GI motility. CGRP mitigates the effects of sepsis.
CGRP is also involved with the following: flushing, thermoregulation, cold hypersensitivity, protecting the kidneys that are under stress, helping with the regulation of insulin release, and mediating the adrenal glucocorticoid response to acute stress (particularly in the mature fetus). Effects on the HPA axis are worrisome but unknown. CGRP is important as a vasodilator during stress, and these CGRP mAbs have not yet been tested under stress.
The CGRP mAbs have been terrific for many patients. Efficacy is on a par with onabotulinumtoxinA. However, in a short period of time we have witnessed a plethora of serious (and non-serious) adverse effects from short-term use. CGRP plays an important role in many physiologic processes. We have no idea as to the long-term consequences of blocking CGRP. We should proceed with caution.
From my article on MDEdge, by Larry Robbins, M.D.