Discussion
The aim of this preliminary phase 1 study was to evaluate the potential efficacy of interventions that might be used to mitigate ISRs following the SC administration of 60 mg elamipretide in healthy subjects as well as any potential impact on PK and safety. We focused on treatments that target mast cells or factors upstream or downstream to their activation, due to our preclinical data that strongly suggest that elamipretide-associated ISRs are caused by direct activation of mast cells through MRGPRX2. Overall, mometasone appeared to favorably mitigate the presumed pseudo-allergic reactions observed following SC elamipretide, specifically reducing pruritus and induration/swelling with no effect on PK. Ice application ameliorated early injection site pain and itching, but reduced the absorption of elamipretide, while diphenhydramine demonstrated potential in reducing induration/swelling but caused sedation in some cases. Tacrolimus and doxepin, as utilized in this study, demonstrated little impact on reported ISRs compared to elamipretide alone and are not likely to be recommended for ISR mitigation, though additional investigation could be considered. There were no significant changes in elamipretide or M1 and M2 metabolite plasma exposures with any of the treatments except for ice application, which reduced exposure to elamipretide and its metabolites.
Although ice application and diphenhydramine showed some reduction in ISR signs and symptoms, mometasone revealed the most promise. The reduction in pruritus and induration/swelling demonstrated by mometasone enabled patients to be more comfortable with the elamipretide treatment. In addition, mometasone may further improve elamipretide tolerability by reducing scratching and subsequent scratching-related skin damage. In some subjects, the hydrocolloid occlusive dressing applied over the mometasone ointment resulted in redness of the covered area (i.e. the injection site area), confounding the ability to identify erythema due to ISRs. The capture of ISR photographs proved useful in deciphering erythema grading in subjects where the pattern of redness appeared to be related to the use of the occlusive dressing. However, it is possible that erythema as an ISR was overreported in the mometasone treatment arm of this study.
Despite ice application and diphenhydramine reducing some ISR signs and symptoms, both interventions presented undesirable effects. Specifically, the simple intervention of ice application reduced early injection site pain and itching, but reduced the Cmaxand AUC0-6h of elamipretide and its metabolites (M1 and M2), possibly due to vasoconstriction. Neither M1 nor M2 are biologically active and the potential reduction in plasma concentrations of these metabolites is not therefore anticipated to have a clinical impact [data on file, Stealth BioTherapeutics Inc.]. However, while the potential ramification on efficacy of the disruption in absorption of the active parent drug is unclear, lowering plasma exposure of elamipretide is not desirable, making ice application a less appropriate mitigation option. Similarly, diphenhydramine showed some potential in reducing induration/swelling, but a significant incidence (50%) of mild somnolence was reported in this treatment arm. While second generation antihistamines were not included in this study, the observations with diphenhydramine suggest that other antihistamines that are less sedating could provide utility in mitigating ISRs.
Based on the outcomes of mometasone and antihistamine use, further investigation is warranted to include separate and combined interventions with oral second-generation antihistamines (such as fexofenadine) and topical mometasone without the use of an occlusive dressing. Fexofenadine is a selective peripheral H1 receptor antagonist that does not readily cross the blood–brain barrier, causing less drowsiness in comparison to the first-generation antihistamines, such as diphenhydramine. Several international dermatology and allergy organizations issued a common guideline on chronic urticaria management, recommending the regular use of second-generation antihistamines as first-line treatment. Fexofenadine appears safe and well tolerated and daily doses can be titrated upwards in case of no improvement (Zuberbier, 2018). Chronic urticaria could be considered as the closest model to chronic ISRs in terms of available data on therapeutic management. With regard to mometasone, given that elamipretide is injected daily at alternating sites, applications with mometasone twice a day (one prior and one post-injection) appear to be a more practical alternative to the use of mometasone with occlusive dressing.
Our focus on mast cells arose from preclinical experiments which established that elamipretide, in addition to its intended function, also acts as an agonist of the human G protein-coupled receptor MRGPRX2. MRGPRX2 is primarily expressed by mast cells, which are constitutive residents of the skin and other tissues, trigger rapid tissue inflammation, and mediate many of the symptoms of allergic diseases. We reasoned that MRGPRX2 might be involved because it can be activated by many cationic peptides and small molecules with properties similar to elamipretide and which also cause ISRs and other pseudo-allergic reactions (McNeil 2021a). We have several pieces of evidence to support the hypothesis that elamipretide-associated ISRs are due to MRGPRX2 activation. First, elamipretide triggers intracellular signaling pathways, as measured by calcium mobilization, in cell lines forced to express MRGPRX2 or its mouse ortholog Mrgprb2, but not in unmodified cell lines that do not natively express the receptors (Figures 2 A, B). Notably, this could be blocked by a molecule reported to inhibit MRGPRX2 signaling (Figure 2C). MRGRPX2 activation by elamipretide is physiologically relevant, as the calculated EC50 is over 1,000-fold lower than the injection concentration in our clinical trial. Second, elamipretide causes intracellular signaling, again as measured by calcium flux, in primary mouse mast cells from wild type but not Mrgprb2 knockout mice (Figure 3). Third, SC elamipretide injection into mouse hindpaws triggered rapid swelling via fluid extravasation from the bloodstream, similar to human ISRs, in wild type mice, while extravasation was nearly absent in mice lacking Mrgprb2 (Figure 4). These effects almost certainly are due to the parent drug, as elamipretide’s metabolites have no effect on MRGPRX2 signaling (Figures 2 D, E).
While itch produced in humans by injection of MRGPRX2 agonists can be blocked by antihistamines (Hasbak et al., 2006), development of inhibitors is still in its infancy with no candidates in clinical trial yet (McNeil, 2021a). Indeed, a combination of H1 and H2 histamine receptor antagonists have demonstrated efficacy in blocking MRGPRX2-driven systemic and local reactions (McNeil 2021b) and could be considered in future elamipretide-driven ISR studies. Although the topical steroid mometasone reduced pruritus and induration/swelling following SC elamipretide in this study, its anti-inflammatory mechanism is unclear but is thought to act by inhibition of the arachidonic acid pathway (Spada, et al., 2018; ELOCON® Prescribing Information, 2018). In addition to anti-inflammatory effects, topical steroids, such as mometasone, possess anti-mitotic, immunosuppressive, and vasoconstrictive effects (Gabros et al., 2021), which may have played a role in mitigating ISR signs in this study.
Overall, the data collected in this study support prior findings that SC elamipretide is generally safe and well-tolerated (Karaa et al., 2020; Thompson et al., 2021; Mettu et al., 2022). With the exception of data relating to ISRs, very few TEAEs were identified in this study. The only TEAE seen in more than one subject was that of somnolence in the diphenhydramine treatment arm. Given that somnolence is a well-known side effect of diphenhydramine (Sicari & Zabbo 2021), this adverse event was considered likely related to diphenhydramine and not to elamipretide treatment.
A limitation of the study was the relatively low number of subjects evaluated (N=10), which impacted the potential to demonstrate statistically significant results with respect to efficacy of mitigation. This study was not powered to show a statistically significant difference in the ISR profile between treatment arms but was meant to identify signals that could warrant further investigation in other clinical settings.