Pharmacokinetic Bridging Between an Autoinjector and a Prefilled Syringe Following Subcutaneous Administration of Mirikizumab in Healthy Participants

Abstract

Introduction

Mirikizumab, a humanized anti-interleukin-23p19 monoclonal antibody, is approved for the treatment of moderate-to-severe ulcerative colitis (UC) and Crohn’s disease (CD). Two phase 1, open-label, two-arm, randomized studies compared pharmacokinetics and safety of mirikizumab (200 mg, Study AMBW; 300 mg, Study AMBX) to establish bioequivalence. Mirikizumab was administered subcutaneously using an autoinjector or a prefilled syringe (PFS) as a single dose in healthy participants.

Methods

Participants (male and female) were randomized 1:1 to mirikizumab by a PFS (reference) and an autoinjector (test). Participants were sub-randomized (1:1:1) to one of three injection sites (abdomen, arm, thigh) and stratified into one of three weight groups. Primary endpoints were maximum observed drug concentration (C_max), and area under concentration versus time curves (time zero to infinity [AUC_0–∞]; time to last time point with measurable concentration [AUC_0–tlast]). Secondary objectives were safety and immunogenicity.

Results

In Study AMBW, 90% confidence intervals (CIs) for the ratios (autoinjector to PFS) of geometric least squares mean values for C_max, AUC_0–∞, and AUC_0–tlast were within prespecified bioequivalence limits (0.80–1.25). Similar results were observed in Study AMBX, after administration of mirikizumab 300 mg, with bioequivalence achieved for AUC_0–tlast and AUC_0–∞, while the upper bound of the 90% CI of the geometric least squares mean ratio for C_max was slightly above the bioequivalence upper threshold. Overall safety of mirikizumab was similar between devices in both studies. Immunogenicity was similar overall in Study AMBW, but slightly higher (autoinjector versus PFS) in Study AMBX.

Conclusion

Mirikizumab administered by autoinjector or PFS was considered bioequivalent at the 200-mg UC maintenance dose and the 300-mg CD maintenance dose. Safety and immunogenicity profiles were comparable between autoinjector and PFS. Availability of an autoinjector option may be preferred by some patients and may help improve patient adherence to treatment.

Clinical Trial Registration

ClinicalTrials.gov NCT04607733; NCT05069896.

Supplementary Information

The online version contains supplementary material available at 10.1007/s12325-025-03335-z.

Key Summary Points

Why carry out this study?

Two studies evaluated bioequivalence, safety, and immunogenicity of a single mirikizumab dose administered subcutaneously by autoinjector or prefilled syringe (PFS) in healthy participants, including comparison across three injection sites (abdomen, arm, thigh).

Use of an autoinjector to administer therapeutic biologics subcutaneously can improve patient adherence and dose accuracy compared with a PFS, but it is important to ensure that both devices deliver similar concentration given potential differences in the delivery devices.

What was learned from the study?

Administering mirikizumab by autoinjector was bioequivalent to PFS at the 200-mg subcutaneous (SC) dose used for maintenance treatment in patients with ulcerative colitis (UC), and bioequivalent for area under concentration versus time curve (AUC) parameters with a slight increase in maximum observed drug concentration (Cmax) at the higher 300-mg SC dose used for maintenance treatment in patients with Crohn’s disease (CD), with minimal impact on safety.

Availability of an autoinjector option for mirikizumab maintenance therapy may be preferred by some patients with UC or CD, owing to ease of use by patients or caregivers, dosing accuracy, and may encourage improved treatment adherence over long treatment durations.

Introduction

Biologic therapeutics tend to have poor oral bioavailability because of their large size, limited membrane permeability, and enzymatic degradation by the gastrointestinal tract [1, 2]. As such, biologics are commonly administered intravenously and subcutaneously. For subcutaneous (SC) administration, dosage formulations generally comprise injection of lyophilized powder or solution in a vial, injection of solution in a prefilled syringe (PFS), or use of an autoinjector [3]. Autoinjectors are designed to make administration easier, improve dose accuracy, and increase patient adherence to treatment [">4–7]. However, autoinjectors are often introduced later during clinical development and therefore it is important to establish bioequivalence between the different administration devices.

Mirikizumab is a humanized anti-interleukin-23p19 monoclonal antibody that is approved for the treatment of moderate-to-severe ulcerative colitis (UC) and Crohn’s disease (CD) [8]. The phase 3 studies of mirikizumab demonstrated rapid and sustained efficacy and safety in participants with moderate-to-severe UC [9] and CD [10] when mirikizumab was administered using a PFS as intravenous (IV) induction every 4 weeks (Q4W) for 12 weeks followed by SC maintenance doses Q4W for 40 weeks. An autoinjector developed by Eli Lilly and Company for other biologics in other indications [11, 12] is currently being adapted for use with mirikizumab, which would provide patients with a choice of injection device for maintenance therapy.

Pharmacokinetics (PK) analyses of mirikizumab administered by vial and syringe during the phase 2 AMAC trial and by PFS during the phase 3 LUCENT 1 + 2 trials demonstrated a linear exposure relationship with IV and SC doses in patients with UC [13] and CD [8, 14]. The bioavailability of mirikizumab administered SC by PFS was estimated to be approximately 40% in patients with UC or CD. Direct comparison of the PK of mirikizumab administered by the autoinjector and the PFS is required to establish whether the autoinjector delivers the required bioequivalent doses of mirikizumab in each indication. Two separate bioequivalence studies were conducted because of two types of device used: 1-mL autoinjector in UC and both 1- and 2-mL autoinjectors in CD.

The primary objective of these studies was to evaluate the PK of mirikizumab after SC administration of a solution formulation (Study AMBW, 200-mg dose; Study AMBX, 300-mg dose), administered using an autoinjector or a PFS, in healthy male and female participants. The secondary objective of the studies was to describe the safety and tolerability of mirikizumab. The studies included an exploratory analysis comparing PK and safety at the three injection sites (abdomen, arm, thigh) used in the phase 3 trials [9, 10] and their impact on the bioequivalence between devices.

Methods

Study Design

This analysis included two phase 1, open-label, two-arm, randomized, parallel-design, single-dose, multi-site studies in healthy participants. Study AMBW (mirikizumab 200 mg; ClinicalTrials.gov NCT04607733) was conducted at four sites in the USA between November 12, 2020, and May 17, 2021; Study AMBX (mirikizumab 300 mg; NCT05069896) was conducted at three sites in the USA between October 5, 2021, and June 22, 2022.

Ethical Approval

Both studies were conducted in accordance with the study protocols and with consensus ethics principles derived from international ethics guidelines, including the Declaration of Helsinki and Council for International Organizations of Medical Sciences International Ethical Guidelines, International Council for Harmonisation Good Clinical Practice Guidelines, and applicable laws and regulations. The study protocols were approved by the institutional review boards at each study site. All participants entered into the study provided written informed consent. Ethical review boards were in place for both studies: the WCG Independent Review Board was the overall (“master”) ethics committee for Study AMBW, and the Salus Institutional Review Board was the overall (“master”) ethics committee for Study AMBX. Informed consent forms, required to provide details about why the research was being done and what it would involve, were approved by the respective review boards and provided to participants. Subsequent dated written informed consent was required by each participant before the participant could participate in the study. The informed consent forms also noted that study data that did not directly identify participants could be published.

Endpoints

The primary endpoints of both studies were maximum observed drug concentration (C_max), the area under the concentration versus time curve from time zero to infinity (AUC_0–∞), and the AUC from time zero to time t, where t is the last time point with a measurable concentration (AUC_0–tlast).

The secondary endpoints of both studies were the incidence of treatment-emergent adverse events (TEAEs) and serious adverse events (SAEs), which were assessed at all study visits and by telephone between visits.

Exploratory endpoints of both studies were assessing the impact of the three injection sites on C_max, AUC_0–∞, and AUC_0–tlast, and the incidence of treatment-emergent anti-drug antibodies (TE-ADAs).

Study Population

Participants were required to be overtly healthy men or women (nonpregnant), aged between 18 and 65 years inclusive, and with a body mass index of 18.0 to 32.0 kg/m², inclusive, at screening.

Treatment

Eligible participants were stratified into one of three weight categories (< 70 kg, 70–80 kg, > 80 kg) since weight is a known factor that alters SC bioavailability and exposure of mirikizumab [13]. Within these categories, participants were randomized (computer-generated allocation code) 1:1 to mirikizumab administration with the PFS (reference) or autoinjector (test). Within each delivery device group, participants were sub-randomized (1:1:1) to one of the three injection sites (Fig. 1). All doses were administered by trained staff within a phase 1 clinical research unit. Details of the preparation and administration of doses with each device, together with an example of the autoinjector, are in the electronic supplementary materials section (Fig. S1). These mirikizumab formulations are not currently approved for use in CD and were not used in the phase 3 study [10].

Fig. 1.

Study designs. ADA anti-drug antibody, MIRI mirikizumab, PK pharmacokinetics, R randomization, SC subcutaneous

On day 1, participants received single doses of mirikizumab SC (100 mg/mL; Eli Lilly and Company) as follows. In Study AMBW, participants received a total of 200 mg mirikizumab SC, delivered into the assigned injection site as one injection of 1 mL in the left arm or thigh, or left side of the abdomen, followed by one injection of 1 mL on the right arm or thigh, or right side of the abdomen. In Study AMBX, participants received a total of 300 mg mirikizumab SC, delivered into the assigned injection site, as one injection of 2 mL on the left arm or thigh, or left side of the abdomen, followed by one injection of 1 mL on the right arm or thigh, or right side of the abdomen.

PK and Immunogenicity Sample Collection and Bioanalyses

Blood samples (approx. 3 mL) were collected for PK analyses pre-injection on day 1, and on days 3, 5, 6, 11, 15, 22, 29, 43, 57, 71, and 85 (or time of discontinuation) (Fig. 1). Samples for ADA assessment were collected pre-injection on days 1, 15, 29, and 85. Sampling times with visit intervals were measured in days, as per study protocols, and this approach was consistent with previous studies of pharmacokinetic comparability of mirikizumab formulations [15].

Serum mirikizumab was measured using a validated enzyme-linked immunosorbent assay by ICON Laboratory Services, Inc. (Whitesboro, NY). The lower limit of quantification was 100 ng/mL and the upper limit was 10,000 ng/mL. Inter-assay accuracy (% relative error) was − 5.6% to − 0.2% for Study AMBW and − 1.3% to 1.7% for Study AMBX. Inter-assay precision (% coefficients of variation [%CV]) was 5.9% to 6.6% for Study AMBW and 4.6% to 5.7% for Study AMBX.

To screen for, confirm, and titer ADAs against mirikizumab and confirm neutralizing antibodies (NAbs) to mirikizumab, serum samples were analyzed using a validated up-front acid treatment affinity capture elution bridge assay at BioAgilytix Laboratories (Durham, NC, USA). The screening ADA assay was performed with a minimal required dilution of 1:10 with a sensitivity of 3.0 ng/mL and a drug tolerance of 229 μg/mL in the presence of 100 ng/mL affinity purified polyclonal anti-mirikizumab antibodies from hyperimmunized monkey serum.

Pharmacokinetic Analysis

PK parameters were calculated using noncompartmental methods using Phoenix WinNonlin Version 8.1 (Certara, Radnor, PA, USA). The primary parameters for analysis were C_max, AUC_0–∞, and AUC_0–tlast of mirikizumab. The secondary parameter for analysis was the time to maximum observed drug concentration (t_max) of mirikizumab. Other noncompartmental parameters, such as half-life associated with the terminal rate constant (t_1/2), apparent total body clearance of drug calculated after extravascular administration (CL/F), and apparent volume of distribution during the terminal phase after extravascular administration (Vz/F), were reported.

Statistical Analyses

The planned sample size for each study was 240 participants to ensure 216 participants completed the study, i.e., a sample size of 108 participants per delivery device group, which would provide approximately 90% power that the 90% confidence interval (CI) of the geometric least squares mean (LSM) ratio of C_max, AUC_0–∞, and AUC_0–tlast between groups would fall within an equivalence range of 0.80 to 1.25. This sample size calculation was based on the assumptions that the PK parameters have log-normal distribution, the geometric %CV for each parameter is approximately 40% (based on previous trials), the %CV is the same for participants from each delivery device group, and the expected ratio of geometric means between devices for each parameter is approximately 1.07. Arithmetic mean concentration versus time profiles were plotted using nominal PK sampling time points per the protocol. Mean concentrations were plotted for a given time if two-thirds of the individual data at that time point had quantifiable measurements within the sampling window (± 10%).

For the primary analysis, log-transformed C_max, AUC_0–∞, and AUC_0–tlast parameters were analyzed using a linear fixed-effects model with fixed effects for delivery device, injection location, and weight stratification. The dosing regimen differences between the autoinjector and PFS were back-transformed to present the ratios of geometric LSMs and the corresponding 90% CIs. The autoinjector and PFS were considered bioequivalent if the 90% CI of the ratio of geometric LSMs fell within 0.80 to 1.25. Similar analyses were conducted for subgroups of each device at each injection site, although these exploratory analyses were not sufficiently powered to establish bioequivalence. Data analysis was performed using SAS^® version 9.4. or greater (SAS Institute Inc., Cary, NC, USA).

Safety Analysis

TEAEs were classified using the Medical Dictionary for Regulatory Activities, version 23.0 (Study AMBW) or 24.0 (Study AMBX), by preferred term. Injection-site reactions (ISRs) were reported as TEAEs if the ISR was spontaneously reported by the participant during the study. Where multiple findings indicative of ISRs (comprising an event of either erythema, induration, pain, pruritus, or edema) were present at a single time point, these findings were reported as a single TEAE of ISR, and the severity recorded was the highest severity across all findings at that time point. Additional information on erythema, edema, pain, induration, and pruritis was collected for each ISR reported, regardless of the type of ISR reported as the TEAE, with each positive response in any category counted as an event; thus, more than one event could be counted for each TEAE. The intensity of injection-site pain was quantified by the participant using the 100-mm validated pain visual analog scale (VAS) [16]. AEs of special interest included infections and systemic allergic/hypersensitivity reactions.

Immunogenicity Analysis

The frequency and percentage of participants with preexisting ADAs and with TE-ADAs to mirikizumab were reported and listed. TE-ADAs were defined as those with a fourfold (two-dilution) increase in titer compared with baseline if ADAs were detected at baseline or a titer at least twofold (one-dilution) or greater than the minimum required dilution (1:10) if no ADAs were detected at baseline. The frequency and percentage of participants with NAbs were also reported for participants with TE-ADAs.

Results

Participant Disposition and Demographics

In Study AMBW, 240 participants were randomized to receive a single dose of mirikizumab 200 mg by autoinjector (n = 120) or PFS (n = 120) (Fig. S2). One participant in the autoinjector group and three in the PFS group discontinued before the end of the study. In Study AMBX, 237 participants were randomized to receive a single dose of mirikizumab 300 mg by autoinjector (n = 120) or PFS (n = 117) (Fig. S2). Two participants were enrolled but withdrew because of an adverse event (one of nausea, one of vomiting) before randomization and did not receive mirikizumab. Three participants in the autoinjector group and three in the PFS group discontinued before the end of the study. For Study AMBX, one participant was excluded from the PK analyses because only one post-dose sample was collected. The remaining randomized participants in both studies were included in both PK and safety analyses.

Demographic characteristics were generally similar across device groups and across studies, except for the number of white participants in Study AMBW (n = 83 in the PFS arm vs. n = 97 in the autoinjector arm) (Table 1). Approximately 60% of participants were female, and the mean age was 41.5 years, mean weight was 76.4 kg, and mean body mass index was 26.5 kg/m². Per the stratification scheme, approximately a third of participants in each device group were in each weight category and around one-third of participants were allocated to each of the three injection sites (abdomen, arm, or thigh).

Table 1.

Participant demographics and clinical characteristics

Variable	Study AMBW (mirikizumab 200 mg)		Study AMBX (mirikizumab 300 mg)
	PFS (N = 120)	Autoinjector (N = 120)	PFS (N = 117)	Autoinjector (N = 120)
Age, years, mean (SD)	40.1 (12.4)	43.3 (12.0)	41.4 (12.5)	41.3 (13.4)
Sex, female, n (%)	71 (59.2)	68 (56.7)	66 (56.4)	72 (60.0)
Ethnicity, n (%)
Not Hispanic or Latino	100 (83.3)	99 (82.5)	94 (80.3)	95 (79.2)
Hispanic or Latino	20 (16.7)	21 (17.5)	23 (19.7)	25 (20.8)
Race, n (%)
White	83 (69.2)	97 (80.8)	98 (83.8)	100 (83.3)
Black/African American	30 (25.0)	19 (15.8)	14 (12.0)	15 (12.5)
American Indian or Alaska Native	2 (1.7)	1 (0.8)	0 (0)	0 (0)
Asian	2 (1.7)	2 (1.7)	3 (2.6)	4 (3.3)
Native Hawaiian or Other Pacific Islander	1 (0.8)	0 (0)	0 (0)	0 (0)
Multiple	1 (0.8)	1 (0.8)	2 (1.7)	1 (0.8)
Not reported	1 (0.8)	0 (0)	0 (0)	0 (0)
BMI, kg/m2, mean (SD)	26.3 (2.8)	26.4 (3.2)	26.6 (3.3)	26.5 (3.4)
Weight, kg, mean (SD)	76.2 (12.9)	76.3 (13.4)	76.7 (13.7)	76.5 (13.5)
Weight category, n (%)
< 70 kg	40 (33.3)	40 (33.3)	38 (32.5)	39 (32.5)
70–80 kg	39 (32.5)	38 (31.7)	34 (29.1)	36 (30.0)
> 80 kg	41 (34.2)	42 (35.0)	45 (38.5)	45 (37.5)
Injection-site location, n (%)
Arm	41 (34.2)	39 (32.5)	39 (33.3)	40 (33.3)
Abdomen	40 (33.3)	40 (33.3)	39 (33.3)	40 (33.3)
Thigh	39 (32.5)	41 (34.2)	39 (33.3)	40 (33.3)

BMI body mass index, PFS prefilled syringe, SD standard deviation

Mirikizumab Pharmacokinetics (Primary Objective)

In both studies, the arithmetic mean concentration versus time profiles for mirikizumab were similar for the autoinjector and PFS (Fig. 2). The median time to t_max was approximately 4 days following administration of mirikizumab, regardless of delivery device or dose (Tables 2 and 3). The geometric mean t_1/2 was approximately 11 days across device groups and mirikizumab doses. For the 200-mg dose (Study AMBW), the geometric mean C_max was 15.2 μg/mL in the autoinjector group and 14.3 μg/mL in the PFS group, with corresponding geometric %CVs of 40% and 44%, respectively (Table 2). For the 300-mg dose (Study AMBX), the geometric mean C_max was 23.5 μg/mL in the autoinjector group and 20.0 μg/mL in the PFS group, with corresponding %CVs of 40% and 44%, respectively (Table 3).

Fig. 2.

Serum concentration versus time profile for a mirikizumab 200 mg and b mirikizumab 300 mg administered subcutaneously using prefilled syringe and autoinjector. AI autoinjector, PFS prefilled syringe, SD standard deviation

Table 2.

Statistical analysis of mirikizumab pharmacokinetic parameters after subcutaneous injection of mirikizumab 200 mg administered by prefilled syringe or autoinjector (Study AMBW)

Parameter	Device	Geometric mean (geometric %CV)a	Geometric LSM	Ratio of geometric LSM (AI:PFS) (90% CI)
Cmax (μg/mL)	PFS	14.3 (44)	14.4	1.06 (0.98–1.14)
	Autoinjector	15.2 (40)	15.2
AUC0–∞ (μg·day/mL)	PFS	246 (44)	247	1.06 (0.98–1.15)
	Autoinjector	262 (39)	262
AUC0–tlast (μg·day/mL)	PFS	244 (44)	245	1.05 (0.97–1.14)
	Autoinjector	257 (39)	257
tmax (day), median (min, max)	PFS	4.00 (1.81–10.23)	–	–
	Autoinjector	4.00 (1.83–7.05)	–	–
t1/2 (day), geometric mean (min, max)	PFS	10.9 (4.59–20.5)	–	–
	Autoinjector	11.1 (6.09–21.3)	–	–
CL/F (L/day)	PFS	0.813 (44)	–	–
	Autoinjector	0.764 (39)	–	–
Vz/F (L)	PFS	12.8 (37)	–	–
	Autoinjector	12.2 (35)	–	–

AI autoinjector, AUC_0–∞ area under the concentration versus time curve from time zero to infinity, AUC_0–tlast area under the concentration versus time from time zero to time t, where t is the last time point with a measurable concentration, CI confidence interval, CL/F apparent total body clearance of drug calculated after extravascular administration, C_max maximum observed drug concentration, CV coefficient of variation, LSM least squares mean, max maximum, min minimum, PFS prefilled syringe, t_1/2 half-life associated with the terminal rate constant in noncompartmental analysis, t_max time of maximum observed drug concentration, Vz/F apparent volume of distribution during the terminal phase after extravascular administration

^aExcept where noted

Table 3.

Statistical analysis of mirikizumab pharmacokinetic parameters after subcutaneous injection of mirikizumab 300 mg administered by prefilled syringe or autoinjector (Study AMBX)

Parameter	Device	Geometric mean (geometric %CV)a	Geometric LSM	Ratio of geometric LSM (AI:PFS) (90% CI)
Cmax (μg/mL)	PFS	20.0 (44)	20.3	1.17 (1.07–1.27)
	Autoinjector	23.5 (40)	23.7
AUC0–∞ (μg·day/mL)	PFS	359 (44)	363	1.15 (1.06–1.24)
	Autoinjector	412 (39)	415
AUC0–tlast (μg·day/mL)	PFS	355 (44)	359	1.16 (1.07–1.25)
	Autoinjector	411 (38)	415
tmax (day), median (min, max)	PFS	4.02 (1.88–9.86)	–	–
	Autoinjector	4.02 (1.91–10.33)	–	–
t1/2 (day), geometric mean (min, max)	PFS	11.0 (5.75–19.0)	–	–
	Autoinjector	11.3 (6.20–17.5)	–	–
CL/F (L/day)	PFS	0.836 (44)	–	–
	Autoinjector	0.728 (39)	–	–
Vz/F (L)	PFS	13.3 (38)	–	–
	Autoinjector	11.9 (31)	–	–

AI autoinjector, AUC_0–∞ area under the concentration versus time curve from time zero to infinity, AUC_0–tlast area under the concentration versus time from time zero to time t, where t is the last time point with a measurable concentration, CI confidence interval, CL/F apparent total body clearance of drug calculated after extravascular administration, C_max maximum observed drug concentration, CV coefficient of variation, LSM least squares mean, max maximum, min minimum, PFS prefilled syringe, t_1/2 half-life associated with the terminal rate constant in noncompartmental analysis, t_max time of maximum observed drug concentration, Vz/F apparent volume of distribution during the terminal phase after extravascular administration

^aExcept where noted

For the 200-mg mirikizumab dose (Study AMBW), the 90% CIs for the ratios (autoinjector to PFS) of the geometric LSM values for C_max, AUC_0–∞, and AUC_0–tlast were within the prespecified limits of 0.80 and 1.25 for bioequivalence (Table 2; Fig. 3a). For the 300-mg mirikizumab dose (Study AMBX), the 90% CI of the geometric LSM ratio was within the prespecified limits for bioequivalence for AUC_0–∞ and AUC_0–tlast, and slightly outside of the limit for C_max (Table 3; Fig. 3b). There was no significant difference in t_max between device groups for both Study AMBW and AMBX.

Fig. 3.

Geometric LSM ratios (90% CI) of C_max, AUC_0–∞, and AUC_0–tlast for mirikizumab a 200 mg and b 300 mg delivered in the abdomen, arm, or thigh by AI versus PFS. The devices were considered bioequivalent if the 90% CI of the ratio of geometric LSM fell within 0.80 to 1.25. AI autoinjector, AUC_0–∞ area under the concentration versus time curve from time zero to infinity, AUC_0–tlast area under the concentration versus time curve from time zero to time t, where t is the last time point with a measurable concentration, CI confidence interval, C_max maximum observed drug concentration, LSM least squares mean, PFS prefilled syringe

Mirikizumab Pharmacokinetics (Exploratory Objective)

Both Study AMBW and AMBX were not powered for comparisons by injection-site locations. In Study AMBW, the 90% CIs were within the prespecified limits for all three primary PK parameters following administration in the abdomen or thigh; however, the geometric LSM ratio was up to 12% higher for the autoinjector compared with PFS following administration in the arm (Fig. 3a). Compared with injection in the abdomen, across exposure parameters exposure was lower after administration in the arm (autoinjector group, up to 9%; PFS group, up to 18%) and was higher after administration in the thigh (autoinjector group, up to 22%; PFS group, up to 15%) (Table S1 in the electronic supplementary material).

In Study AMBX, the 90% CIs were within the prespecified limits for all three primary PK parameters following administration in the arm or thigh; however, the geometric LSM ratios of the AUC values and C_max were approximately 40% higher and 53% higher, respectively, for the autoinjector compared with PFS following administration in the abdomen (Fig. 3b). For the PFS, there was similar exposure following administration into the arm compared with the abdomen (Table S2). However, AUCs were approximately 30% higher and C_max was 37% higher following administration into the thigh compared with the abdomen. For the autoinjector, there was similar exposure following administration into the thigh compared with the abdomen (Table S2). However, AUCs were approximately 25% lower, and C_max was 34% lower, following administration into the arm compared with the abdomen.

Safety and Tolerability

In Study AMBW, 39 (32.5%) and 43 (35.8%) participants reported TEAEs in the autoinjector and PFS groups, respectively (Table 4). Of these, 21 (17.5%) participants (autoinjector) and 18 (15.0%) participants (PFS) reported TEAEs considered related to mirikizumab, while a similar proportion (autoinjector, 11 [9.2%]; PFS, 8 [6.7%]) reported device-related TEAEs. Most TEAEs were mild with no events leading to discontinuation, and there were no deaths. In Study AMBX, 45 (37.5%) and 39 (33.3%) participants reported TEAEs in the autoinjector and PFS groups, respectively (Table 4). Of these, treatment-related TEAEs occurred in 15 (12.5%) participants (autoinjector) and 12 (10.3%) participants (PFS). Device-related TEAEs were reported by 11 (9.2%) participants (autoinjector) and 2 (1.7%) participants (PFS) (Table 4). The majority of TEAEs were mild: two participants in the PFS group discontinued because of COVID-19-related adverse events, and there were no deaths.

Table 4.

Treatment-emergent adverse events following mirikizumab administered with either prefilled syringe or autoinjector

Participants with event, n (%)	Study AMBW (mirikizumab 200 mg)		Study AMBX (mirikizumab 300 mg)
	PFS (N = 120)	Autoinjector (N = 120)	PFS (N = 117)	Autoinjector (N = 120)
All TEAEs	43 (35.8)	39 (32.5)	39 (33.3)	45 (37.5)
Treatment-related TEAEs	18 (15.0)	21 (17.5)	12 (10.3)	15 (12.5)
Device-related TEAEs	8 (6.7)	11 (9.2)	2 (1.7)	11 (9.2)
SAEs	0	0	0	1 (0.8)
Deaths	0	0	0	0
AEs leading to study discontinuation	0	0	2 (1.7)	0
Infection TEAEs	7 (5.8)	11 (9.2)	10 (8.5)	13 (10.8)
Systemic allergic/hypersensitivity TEAEs	0	0	0	0

AE adverse event, PFS prefilled syringe, SAE serious adverse event, TEAE treatment-emergent adverse event

Infection rates were low and comparable across autoinjector and PFS groups in both studies (range, 5.8–10.8%) (Table 4). No systemic allergic/hypersensitivity TEAEs were reported in either study. TEAEs by frequency are shown in Table S3 (in the electronic supplementary material). The frequencies were generally comparable between the autoinjector and PFS groups across the two studies, with ISRs and headache being the most common TEAEs.

In Study AMBW, ISRs that triggered further assessment were reported in 11 (9.2%) participants (autoinjector) and 10 (8.3%) participants (PFS) (Table 5). In the PFS group, ISR incidence was similar across injection sites, while in the autoinjector group, ISR incidence was higher in the thigh (12.2%) versus the abdomen (7.5%) or arm (7.7%). All ISRs were considered related to study treatment, with most also related to the relevant study device. While most ISRs were mild or moderate in severity as assessed in the ISR follow-up form, 2 (1.7%) participants in the autoinjector group self-reported VAS pain score > 70 mm, which is considered severe pain [17].

Table 5.

ISRs following mirikizumab administered with either prefilled syringe or autoinjector

Participants with event, n (%)	Study AMBW (mirikizumab 200 mg)		Study AMBX (mirikizumab 300 mg)
	PFS (N = 120)	Autoinjector (N = 120)	PFS (N = 117)	Autoinjector (N = 120)
Participants with any ISRa	10 (8.3)	11 (9.2)	2 (1.7)	12 (10.0)
Number of ISRs	30	33	3	34
Erythema
Noticeable but very mild redness	3 (10)	5 (15.2)	0 (0)	8 (23.5)
Clearly red	2 (6.7)	0 (0)	0 (0)	0 (0)
Bright red	2 (6.7)	0 (0)	0 (0)	0 (0)
Dark with ulceration, or necrosis	0 (0)	0 (0)	0 (0)	0 (0)
Pruritus
Mild	3 (10)	0 (0)	0 (0)	7 (20.6)
Moderate	2 (6.7)	0 (0)	0 (0)	0 (0)
Severe	0 (0)	0 (0)	0 (0)	0 (0)
Induration
Barely noticeable	2 (6.7)	4 (12.1)	0 (0)	2 (5.9)
Slight	0 (0)	0 (0)	0 (0)	0 (0)
Moderate	0 (0)	0 (0)	1 (33.3)	0 (0)
Severe	0 (0)	0 (0)	0 (0)	0 (0)
Edema
Mild	2 (6.7)	2 (6.1)	0 (0)	4 (11.8)
Moderate	1 (3.3)	0 (0)	0 (0)	0 (0)
Severe	0 (0)	0 (0)	0 (0)	0 (0)
Painb	7 (5.8)	11 (9.2)c	2 (1.7)	7 (5.8)

ISR injection-site reaction, PFS prefilled syringe, VAS visual analog scale

^aISRs that triggered further assessment (AMBW, n = 21; AMBX, n = 14)

^bOverall data for left or right side of body (arm, thigh, or abdomen)

^cTwo (1.7%) participants self-reported a VAS pain score > 70 mm. Additional information on erythema, edema, pain, induration, and pruritis was collected for each ISR reported, regardless of the type of ISR reported as the TEAE, with each positive response in any category counted as an event; thus, more than one event could be counted for each TEAE. Pain categories are defined as follows: no pain, VAS pain score = 0; mild pain, VAS pain score > 0 and ≤ 30; moderate pain, VAS pain score > 30 and ≤ 70; severe pain = VAS pain score > 70

In Study AMBX, ISRs requiring further assessment occurred in 12 (10.0%) participants (autoinjector) and 2 (1.7%) participants (PFS) (Table 5). In the PFS group, ISR incidence was similar across injection sites, whereas in the autoinjector group, the incidence was lower in the thigh (2.5%) versus the abdomen (15.0%) or arm (12.5%). All ISRs were mild in severity, except for one event of moderate induration in the PFS group.

Immunogenicity

In Study AMBW, all 240 participants were evaluable for TE-ADAs (Table S4 in the electronic supplementary material). Frequencies of TE-ADAs were similar overall and across injection sites when mirikizumab was administered by autoinjector or PFS (overall range, 10.3–19.5%). There was no temporal relationship between the presence of TE-ADAs and reports of ISRs or potential hypersensitivity TEAEs.

In Study AMBX, all 237 participants were evaluable for TE-ADAs (Table S4 in the electronic supplementary material). The incidences of TE-ADAs were numerically higher in the autoinjector group compared with the PFS group. TE-ADA incidence was comparable across injection sites within the autoinjector group (overall range, 17.5–20.0%), while rates varied across injection sites within the PFS group (abdomen, 17.9%; arm, 12.8%; thigh, 7.7%). There were two reports of ISRs or potential hypersensitivity TEAEs that had a possible temporal association with the presence of TE-ADAs. Of these, one participant with mild pruritus between day 4 and day 5 had an ADA titer of 1:160 on day 15, and one participant with two mild ISRs that occurred from day 1 through day 7 had an ADA titer of 1:80 on day 16.

Discussion

Bioequivalence studies are considered a surrogate for the clinical evaluation of therapeutic equivalence of drug products [18], and are required by worldwide regulatory agencies to provide evidence of achieving bioequivalence in drug exposure [18, 19]. In Study AMBW, the bioequivalence of mirikizumab SC administered with an autoinjector and with a PFS was demonstrated for the 200-mg dose approved for UC. Similar results were observed in Study AMBX, after administration of mirikizumab 300 mg, with bioequivalence achieved for AUC_0–tlast and AUC_0–∞, while the upper bound of the 90% CI of the geometric LSM ratio for C_max was slightly above the bioequivalence upper threshold.

Study AMBW had a numerical difference in the number of white participants between the PFS (n = 83) and autoinjector (n = 97) arms. However, this difference is not likely to affect the bioequivalence of drug exposure between the two groups. This is because ethnicity was not included as a covariate in the UC population pharmacokinetics model for mirikizumab and was not deemed to have an impact on PK [13].

Subcutaneously administered drugs can enter the systemic circulation via blood capillaries or the lymphatic system [20]. Monoclonal antibodies are absorbed through lymphatic capillaries, influenced by interstitial pressure and extracellular tissue matrix composition [21]. A computational model showed that factors such as autoinjector base diameter and skin pinching affect fluid pressure near the injection site, impacting drug dispersion and retention in adipose layers [21]. Other factors affecting monoclonal antibodies bioavailability include skin thickness, SC adipose tissue, injection depth, and lymphatic circulation [21, 22]. Skin and adipose tissue thickness correlate with waist circumference and body mass index (BMI) [23]. Higher BMI can lead to higher drug exposure due to poor lymphatic drainage [21], while higher body weight may result in lower drug exposure for some proteins [22]. In the current studies, body weight-based stratification was used, as body weight has previously been shown to significantly affect mirikizumab pharmacokinetics [13]. The current results showed similar mean body weight across device groups and studies, ensuring that the comparison between devices was not confounded by body weight.

The main differences between the autoinjector and PFS are in injection technique and depth. PFS requires a skin pinch-up and 45° injection to avoid intramuscular injection due to its longer needle, while autoinjectors need no pinch-up and are placed vertically on the skin, ensuring a shallow injection depth of 5.5 mm. The 1-mL and 2-mL PFS used in the mirikizumab phase 3 studies had needle lengths of 12.7 mm and 8.0 mm, respectively, with 45° injection angles, resulting in vertical injection depths of 9.0 mm and 5.7 mm. In contrast, the marketed 1-mL and 2-mL autoinjectors have a needle extension length of 5.5 mm and are injected vertically. As a result, the vertical injection depth is similar between the 2-mL PFS and 2-mL autoinjector, but slightly longer for the 1-mL PFS compared with the 1-mL autoinjector. Autoinjectors have a consistent injection duration of less than 10 s, while PFS injection duration varies. Bioequivalence studies confirmed that these differences do not significantly affect the PK of mirikizumab, aligning with a review suggesting that autoinjectors with shallower injection depths (< 7 mm) are more likely to achieve bioequivalence with PFS [4].

The results of the present analysis showed that while all three primary PK parameters in Study AMBW and AUCs in Study AMBX met the bioequivalence criterion, for C_max in Study AMBX, the upper bound of the 90% CI of the geometric LSM ratio was slightly above the bioequivalence upper threshold of 1.25. This is not surprising as C_max is more closely affected by the absorption process than AUC, and similar findings have been previously observed for the PK bridging for other therapeutic proteins. However, despite the differences in C_max observed between devices in Study AMBX, this should not negatively affect efficacy since the autoinjector resulted in a higher exposure. The slightly higher C_max is also not expected to have a clinically meaningful impact on the safety profile because of the small effect size, Changed from i.e., 27% higher C_max at the mean level. Furthermore, exposure following administration of the 300-mg mirikizumab maintenance dose by autoinjector was substantially lower than exposure with the recommended IV induction dose (900 mg) in the phase 3 CD trial [14], and there was no apparent exposure–response relationship for TEAEs with either IV or SC administration in the phase 2 and phase 3 trials in CD [14] or in UC [24].

The higher exposure with autoinjector compared with PFS administration in Study AMBX was mainly driven by data from the abdomen injection site. The geometric LSM ratios (90% CI) of C_max and AUC_0–∞ are 1.53 (1.32–1.78) and 1.40 (1.22–1.61) for the abdomen. This may be a statistical anomaly given the small sample size with multiple comparisons of interest. Alternatively, injection-site location may also affect the rate and extent of SC absorption [20, 22]. Differences in SC uptake are generally attributed to differences in blood flow to those areas and/or to regional variations in lymph flow [25]. The risk of pre-systemic degradation is reduced following femoral injection as lymph flow in the thigh and in the arm is relatively fast, and therefore transit time for proteins such as monoclonal antibodies to the systemic circulation is reduced. By comparison, lymphatic tissue is less abundant—and lymphatic flow is slower—in the abdomen than in the thigh or arm [26]. These physiological differences may have potentiated the difference in the PK of mirikizumab caused by delivery devices. It was also noted that although the studies were not powered to establish injection site-specific bioequivalence, the bioequivalence criteria were met for injection of 200 mg into the abdomen and thigh, and for injection of 300 mg into the arm and thigh.

Exploratory analyses revealed minor differences in mirikizumab PK when administered at different injection sites using the autoinjector or PFS. While differences in exposure between injection sites were observed, these differences were within the inter-individual variabilities estimated in population PK analyses of the phase 3 trials [13, 14] and are therefore not considered clinically relevant. Overall, these results suggest that mirikizumab can be administered at either the abdomen, arm, or thigh injection site via either autoinjector or PFS.

The overall safety profile of mirikizumab was similar between devices in both studies. TEAEs occurred in approximately one-third of participants in each group in each study, with no notable differences between doses or devices. This result was anticipated as, within each study, participants received the same drug and dose. The frequency of ISRs was similar for the autoinjector groups in AMBW and AMBX. In the PFS groups, the frequency of ISRs was comparable in AMBW but lower in AMBX. The majority of ISRs were mild or moderate. In Study AMBW, mirikizumab was delivered as two 1-mL injections (one per side), whereas in Study AMBX, it was delivered as a 2-mL injection (left side) and 1-mL injection (right side). Previous studies did not find a significant impact of device type (PFS vs. autoinjector) in ISRs or TEAEs, which are often driven by impurities (e.g., host cell proteins) or the drug itself (pharmacological effect). However, the injection-site analyses were underpowered and so no definitive conclusions can be made.

The frequency of TE-ADAs was similar overall and by injection site in Study AMBW but was numerically higher in the autoinjector versus PFS group in Study AMBX, where ADA incidence rates also varied across injection sites within the PFS group. As a result of the short duration of this study (85 days) and the small group sizes, results of anti-mirikizumab antibodies from these trials cannot be generalized.

In terms of limitations, the exploratory analyses of PK by injection-site location were not powered to establish bioequivalence of the autoinjector compared with PFS. In addition, the study was conducted in healthy participants rather than in patients with UC or CD, although previous studies have reported no difference in the PK of mirikizumab between healthy participants and patients with UC [13]. Further, this was a single-dose study, and therefore did not assess the effects of multiple injections and drug accumulation that would be measured in clinical trials of patients with UC or CD.

Conclusion

The results of these studies indicate that administration of mirikizumab using an autoinjector is bioequivalent to injection with a PFS in Study AMBW, at the 200-mg dose that is used for patients with UC [8]. The results observed in Study AMBX following administration of 300 mg mirikizumab were similar to those in Study AMBW, with bioequivalence achieved for AUC_0–tlast and AUC_0–∞, while the upper bound of the 90% CI of the geometric LSM ratio for C_max was slightly above the bioequivalence upper threshold. The availability of an autoinjector option for maintenance therapy with mirikizumab may be preferred by some patients with UC or CD, and may assist in improving adherence to treatment.

Supplementary Information

Below is the link to the electronic supplementary material.

Author Contributions

All authors participated in the interpretation of study results, and in the drafting, critical revision, and approval of the final version of the manuscript. Xin Zhang, Christopher D. Payne, Nathan J. Morris and Rodrigo Escobar were involved in the study design. Xin Zhang, Yuki Otani, Nathan Morris and Laiyi Chua were involved in the statistical analyses.

Funding

This study, and the Rapid Service and Open Access Fees for this publication, was funded by Eli Lilly and Company, Indianapolis, Indiana, USA, the manufacturer of mirikizumab. Eli Lilly and Company was involved in the study design, data collection, data analysis, and preparation of the manuscript.

Data Availability

The datasets generated during and/or analyzed during the current study are available in the Vivli repository, www.vivli.org. Lilly provides access to all individual participant data collected during the trial, after anonymization, with the exception of pharmacokinetic or genetic data. Data are available to request 6 months after the indication studied has been approved in the US and EU and after primary publication acceptance, whichever is later. No expiration date of data requests is currently set once data are made available. Access is provided after a proposal has been approved by an independent review committee identified for this purpose and after receipt of a signed data sharing agreement. Data and documents, including the study protocol, statistical analysis plan, clinical study report, blank or annotated case report forms, will be provided in a secure data sharing environment. For details on submitting a request, see the instructions provided at www.vivli.org.

Declarations

Conflict of Interest

Xin Zhang, Yuki Otani, Christopher D. Payne, Nathan J. Morris, Laiyi Chua, Rodrigo Escobar, Sihe Wang, and Galen Shi are employees and shareholders of Eli Lilly and Company.

Ethical Approval

Both studies were conducted in accordance with the study protocols and with consensus ethics principles derived from international ethics guidelines, including the Declaration of Helsinki and Council for International Organizations of Medical Sciences International Ethical Guidelines, International Council for Harmonisation Good Clinical Practice Guidelines, and applicable laws and regulations. The study protocols were approved by the institutional review boards at each study site. All participants entered into the study provided written informed consent. Ethical review boards were in place for both studies: the WCG Independent Review Board was the overall (“master”) ethics committee for Study AMBW, and the Salus Institutional Review Board was the overall (“master”) ethics committee for Study AMBX. Informed consent forms, required to provide details about why the research was being done and what it would involve, were approved by the respective review boards and provided to participants. Subsequent dated written informed consent was required by each participant before the participant could participate in the study. The informed consent forms also noted that study data which did not directly identify participants could be published.