Safety, Tolerability, Efficacy, Pharmacokinetics, and Pharmacodynamics of Afimetoran, a Toll‐Like Receptor 7 and 8 Inhibitor, in Patients With Cutaneous Lupus Erythematosus: A Phase 1b Randomized, Double‐Blind, Placebo‐Controlled Study

Abstract

Objective

There is an unmet need for safe and effective oral treatments for cutaneous lupus erythematosus (CLE). Afimetoran is an investigational, first‐in‐class, orally bioavailable, selective small molecule inhibitor of Toll‐like receptors (TLRs) 7 and 8. We investigated the safety, tolerability, efficacy, pharmacokinetics, and pharmacodynamics of afimetoran in patients with CLE.

Methods

In this Phase 1b, randomized, double‐blind, placebo‐controlled study (NCT04493541), patients with CLE received afimetoran (30 mg once daily) or a placebo, in addition to select background medications, over 16 weeks. Safety and tolerability were mainly reported through adverse events (AEs). Pharmacokinetics were determined using plasma concentration–time data. Pharmacodynamic biomarkers and efficacy (CLE Disease Area and Severity Index‐Activity [CLASI‐A] scores) were exploratory end points.

Results

Thirteen patients were randomized (afimetoran, n = 8; placebo, n = 5), and 12 patients completed treatment (1 discontinued afimetoran [COVID‐19 infection]). Afimetoran demonstrated a favorable safety profile compared with placebo (patients with AEs: 62.5% vs 80.0%), with mainly mild‐to‐moderate AEs. Plasma concentrations exceeded the projected targeted 24‐hour 90% inhibition concentration, supporting once‐ daily dosing. Pharmacodynamic analyses showed a rapid response to afimetoran by week 1 maintained throughout and beyond treatment, with reduced expression of TLR7/8 pathway–associated cytokines. All patients who received afimetoran had a change in the interferon‐1 gene signature (P < 0.0001) at week 16, with five of eight patients demonstrating a >50% improvement in CLASI‐A scores.

Conclusion

These findings support afimetoran's potential as a once daily oral treatment for patients with CLE and suggest a possibly substantial therapeutic benefit, warranting further clinical investigation of afimetoran for lupus.

INTRODUCTION

Cutaneous lupus erythematosus (CLE) is a chronic inflammatory skin condition, occurring alone or as a manifestation of systemic lupus erythematosus (SLE). ¹ CLE can be debilitating, causing physical scarring and psychologic stress, ² and can have a significant and lasting effect on quality of life due to chronicity, visibility of lesions, and potential disfigurement. ³ , ⁴ The prevalence of CLE has been estimated at 70.4 per 100,000 persons, ⁵ and approximately 5% to 25% of patients who initially present with CLE will progress to SLE. ⁶ CLE is markedly sex‐biased, with a 3:1 female‐to‐male incidence rate in adults. ⁷ Recent translational data suggest that Toll‐like receptors (TLRs) 7 and 8 play a key role in this difference. TLR7/8 are X‐linked and have been found to escape X‐chromosome inactivation in a large portion of immune cells in women and female mouse models, resulting in increased responses of these cells to TLR7/8 ligands. ⁸

Treatment options for CLE are limited, as there is a lack of evidence supporting the use of the few therapeutic options that exist, which makes the management of CLE a significant challenge. ⁹ To date, no targeted medications for CLE have been approved by the US Food and Drug Administration or European Medicines Agency ¹⁰ ; however, plasma cell–, plasmacytoid dendritic cell–, and interferon (IFN)–targeted therapies that directly or indirectly inhibit IFN or IFN receptors are currently in development. ¹⁰ Common treatments for CLE range from preventive measures such as UV protection with sunscreen to topical prescription drugs, including glucocorticoids and calcineurin inhibitors, or systemic medication, which includes antimalarial cyclic guanosine monophosphate–AMP (GMP‐AMP) agents, such as hydroxychloroquine (HCQ), and glucocorticoids. ² , ¹⁰ , ¹¹ , ¹² , ¹³ , ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ Antimalarials and systemic steroids are recommended as first‐line systemic treatments for CLE, ¹¹ but treatment options remain limited due to adverse events (AEs), including skin atrophy, dermatitis, telangiectasias, and retinal toxicity. ¹⁰ , ¹¹ , ¹⁹ , ²⁰ , ²¹ , ²² , ²³ , ²⁴ Moreover, long‐term use of oral glucocorticoids or HCQ is known to be associated with serious AEs and is therefore not recommended per CLE treatment guidelines. ²⁵ Thus, there is an important unmet need for effective treatments that can replace long‐term antimalarial and systemic steroid use. ¹¹

The chronic pathologic cycle of CLE is fueled by continuous reactivation of innate immune pathways through adaptive effector mechanisms involving a polyclonal, autoantigen‐specific expansion of Teff and B cells. ²⁶ Further attention to the underlying pathology of CLE is critical to the development of targeted treatment options. For example, the sensing of self‐RNA by the endosomal TLRs 7/8 initiates pathogenic mechanisms underlying autoimmunity in lupus ²⁷ and stimulates production of inflammatory cytokines including chemokines, particularly type I IFNs, that play a central role in the pathogenesis of CLE. ²⁷ , ²⁸ This drives multiple responses across cell types, with autoantibody production and immune complex formation completing the feed‐forward loop to accelerate disease activity (Figure 1). In addition, the activation of TLR7/8 in multiple immune cell types is known to result in steroid resistance. ²⁹ As such, the inhibition of TLR7/8 signaling may offer a therapeutic option that both improves manifestations of lupus ²⁷ and directly reduces the need for steroids.

Figure 1.

Mechanism of disease in cutaneous lupus erythematosus. ¹ , ²⁷ , ²⁸ , ⁵³ IC, immune complex; IFN, interferon; IL‐6, interleukin‐6; IRAK, interleukin‐1 receptor‐associated kinase; IRF, interferon regulatory factor; pDC, plasmacytoid dendritic cell; ssRNA, single‐stranded RNA; TLR7, Toll‐like receptor 7; TNF‐a, tumor necrosis factor α.

Afimetoran is an investigational, first‐in‐class, orally bioavailable, equipotent, small molecule selective inhibitor of TLR7/8. Consistent with the involvement of TLR7/8 in the pathobiology of lupus, preclinical studies of afimetoran have demonstrated robust efficacy in lupus animal models and substantial steroid‐sparing potential. ³⁰ , ³¹ , ³² , ³³ In an NZB/W mouse model with advanced lupus, treatment with afimetoran, including in combination with low‐dose prednisolone, improved survival, reversed proteinuria and glomerular IgG deposition, and reduced levels of circulating cytokines. ³⁰ In vitro analyses also showed that afimetoran combined with low‐dose steroids synergistically blocked TLR7/8‐mediated production of cytokines and suppressed TLR7/8‐mediated B cell production of antibodies in an additive manner. ³² Furthermore, dosing with afimetoran resulted in significant improvement in disease markers in mice with moderate disease, which was even greater when afimetoran was combined with low‐dose prednisolone. ³¹ Importantly, afimetoran reversed TLR7‐mediated resistance to steroid‐induced apoptosis of plasmacytoid dendritic cells and B cells in vitro. ³¹ , ³² Given the strength of the preclinical evidence, clinical studies to assess the potential of afimetoran treatment in lupus are warranted. Here, we describe the first clinical data of the once daily oral TLR7/8 inhibitor, afimetoran, when added to standard therapy in patients with CLE.

PATIENTS AND METHODS

Study design

This was a Phase 1b, randomized, double‐blind, placebo‐controlled single‐site study (NCT04493541) that investigated the safety, tolerability, exploratory efficacy, pharmacokinetics (PK), and pharmacodynamics (PD) of afimetoran versus placebo, plus select background medications, in patients with CLE. The study was conducted at a single site (Charité – Universitätsmedizin, Berlin, Germany). Patients with CLE were screened for up to 4 weeks, after which those who qualified were randomized in a masked manner 2:1 to receive oral afimetoran (30 mg once daily) or placebo for 16 weeks and then monitored for an additional 4 weeks after stopping treatment (Figure 2). There was a minimum washout period before randomization for biologics and immunosuppressants (excluding glucocorticoids or antimalarial agents) that ranged from 4 weeks to 52 weeks. Further detailed information on medication classes and associated washout periods is listed in Supplemental Material S1.

Figure 2.

Study design. EULAR/ACR 2019 classification criteria. ³⁵ ACR, American College of Rheumatology; CLASI‐A, Cutaneous Lupus Erythematosus Disease Area and Severity Index‐Activity; CLE, cutaneous lupus erythematosus; GC, glucocorticoid; QD, once daily; R, randomization; SLE, systemic lupus erythematosus.

Informed consent was obtained from all study participants before enrollment and included permission for consent for anonymized photography of patients’ skin for comparison before and after treatment. The study complied with the applicable International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use and Good Clinical Practice guidelines. Additionally, the study was conducted in accordance with the principles of the Declaration of Helsinki and received appropriate approval by the German Bundesinstitut für Arzneimittel und Medizinprodukte health authority, the Landesamt für Gesundheit und Soziales Geschaftsstelle der Ethik Kommission des Landes Berlin, and Ethikkommission Der Charité Universitätsmedizin Berlin ethics committees. The laws and regulatory requirements of the study country (Germany) were followed.

In reporting our study, we followed the outline recommended by the Consolidated Standards of Reporting Trials statement. ³⁴ Randomization envelopes were shipped directly to a pharmacist or other individual(s) who were responsible for dispensing of the masked study treatment. The randomization envelopes were maintained in a secure location with access limited to authorized personnel.

Inclusion and exclusion criteria

Patients were aged 18 to 65 years and of either sex. Key disease‐associated entry criteria included (1) either European Alliance of Associations for Rheumatology/American College of Rheumatology criteria for the diagnosis of SLE with cutaneous manifestations ³⁵ or, if no SLE diagnosis was made, biopsy‐proven acute CLE, subacute CLE, or discoid lupus erythematosus CLE (qualifying skin biopsies must have demonstrated interface changes with apoptotic keratinocytes or at least two of the following: mononuclear cellular infiltrate, edema of the upper dermis, or focal liquefaction degeneration of the epidermal basal cell layer); (2) active cutaneous lupus disease, defined as a modified CLE Disease Area and Severity Index‐Activity (mCLASI‐A) score ≥6; and (3) active cutaneous lupus skin lesions amenable to biopsy. The mCLASI‐A score was used to determine eligibility; scoring of mucocutaneous ulcers was not included, and scores for alopecia were only included if they were ≥2 points. The purpose of this modification was to minimize confounding by chronic noninflammatory changes and other nonlupus causes of mucocutaneous ulcers and alopecia. Patients were also required to undergo punch biopsies and tape stripping of the skin at least at two time points during the study and to be willing to have lesions photographed. Further inclusion criteria were absence of retinal toxicity on ophthalmologic evaluation and positivity for autoantibodies such as antinuclear antibodies (ANA), anti‐Smith, anti–double‐stranded DNA (anti‐dsDNA), and anti‐Ro antibodies. Stable doses of oral glucocorticoids and/or antimalarials at baseline were permitted. The oral glucocorticoid dosage was restricted to ≤20 mg/day. Further detailed inclusion and exclusion criteria are listed in Supplemental Material S2.

Study end points

Safety and tolerability were the primary end points of the study. AEs (including serious AEs [SAEs]), clinical laboratory values, physical examination findings, vital signs, and electrocardiography (EKG) parameters were evaluated to assess safety. The secondary end points were PK measures. The PK parameters were calculated from plasma concentrations obtained using a liquid chromatography/tandem mass spectrometry assay. The PK parameters included maximal plasma concentration (C_max), time to C_max (T_max), plasma concentration at the end of dosing interval (C_trough), and area under the plasma concentration–time curve over the dosing interval (AUC_tau) of afimetoran.

PD biomarkers and efficacy were measured as exploratory end points. PD biomarkers assessed included autoantibodies, flow cytometry‐defined cell populations, and target engagement cytokines. In a separate analysis for PD assessments, data from age‐, race and ethnicity–, and sex‐matched healthy volunteers (HVs) were used for comparison with patients with CLE. Levels of biomarkers associated with TLR7/8 pathway activation were measured, including messenger RNA (mRNA) signatures of IFN and inflammation pathways, TLR7 signaling, TLR8 signaling, and immune cell populations such as immature and activated dendritic cells. In total, levels of 16 cytokines were measured at each time point to determine the effects of afimetoran on disease and target pathways. Transcriptomics sequencing and cytokine analyses were performed using peripheral whole blood and serum, respectively, collected at each study visit, that is, baseline and weeks 1, 2, 4, 8, 12, and 16 (end of treatment). For cytokine analysis, the whole‐blood samples were stimulated with either gardiquimod (TLR7, InvivoGen) or TL8‐506 (TLR8, InvivoGen). Type‐1 IFN gene signature was analyzed using the DxTerity platform (DxTerity). Exploratory efficacy was assessed via change in CLASI‐A scores from baseline to weeks 4, 8, 12, and 16 (end of treatment), and week 20 follow‐up.

Statistical analyses

A sample size of 24 patients was initially planned and was not based on formal sample size calculations or statistical power considerations but rather on general considerations regarding preliminary effect estimation in pilot studies ³⁶ and probability calculation for AE occurrence. Probability calculations were performed to help ensure that the planned sample size of 24 patients provided sufficient information about AE occurrence rates.

AEs and SAEs were summarized using counts and percentages of patients experiencing an event, by system organ class, preferred term, and treatment group, to characterize severity and causal relationship. Vital signs, clinical laboratory test results, and EKG test results were summarized as changes from baseline using descriptive statistics for continuous variables and frequency distributions for categorical variables.

All patients with at least one evaluable PK parameter were included in the statistical analysis. To calculate PK parameters, predose concentrations below the lower level of quantification (LLOQ) and concentrations before the first quantifiable concentration that were below the LLOQ were set to zero but were treated as missing for the calculation of summary statistics and generation of individual and mean plasma concentration–time plots. All other concentrations below the LLOQ were set as missing for the calculation of PK parameters, summary statistics, and generation of individual and mean plasma concentration–time plots. Individual patient PK parameter values were derived by a noncompartmental analysis method using plasma afimetoran concentration and time data. Actual times of sample collection based on elapsed time from time of dose administration were used for calculation of PK parameters, and nominal times were used for generation of mean serum concentration–time plots and summaries. Post hoc statistical analyses were performed to pool and compare data from patients with CLE and matched HVs for the PD parameters. Treatment responses were evaluated longitudinally in each treatment arm. Gene Set Variation Analysis scores were calculated using Gene Set Variation Analysis package in R, followed by normalization to placebo or HV. Mean, SE, and percent change from baseline metrics were calculated for cytokine data.

No formal statistical testing was performed for the efficacy analysis. CLASI scores and change from baseline in CLASI‐A scores were summarized and listed by treatment and time point. The minimum CLASI‐A score improvement (>4 points from baseline) for this study was based on the analyses and conclusions from a study by Klein et al. ³⁷ The treatment differences in adjusted mean change from baseline were estimated using a mixed‐effect model of repeated measures with treatment, visit, and treatment‐by‐visit interaction as the fixed effects; measurements for each patient as the repeated measurements; and baseline value as a covariate. Data will be shared on request; Bristol Myers Squibb policy on data sharing may be found at https://www.bms.com/researchers-and-partners/independent-research/data-sharing-request-process.html.

RESULTS

Study population

A total of 32 patients provided consent for screening in the study. Because of COVID‐19–related recruitment challenges prompting early study termination, 13 patients (40.6%) were randomized: 8 patients to the afimetoran arm and 5 patients to the placebo arm (Supplementary Figure S1). Twelve patients completed 16 weeks of treatment, and one patient discontinued afimetoran treatment due to COVID‐19 infection.

Baseline patient characteristics were generally balanced. The majority (61.5%; n = 8) of patients had moderate CLE (CLASI‐A scores 10–20); ³⁷ patients receiving afimetoran had more active disease at baseline (mean CLASI‐A 15.0) than patients receiving placebo (mean CLASI‐A 11.6), and the afimetoran group had more patients with CLASI‐A scores >20. Patients in both arms were positive for ANA, anti‐dsDNA, and anti‐Ro autoantibodies, with a lower proportion positive for anti‐dsDNA antibodies in the afimetoran arm. The afimetoran arm also showed a higher proportion of current smokers (Table 1).

Table 1.

Baseline demographics and clinical characteristics^*

Characteristic	Afimetoran (n = 8)	Placebo (n = 5)	Total (N = 13)
Age, mean (SD), y	46.0 (11.1)	51.2 (12.4)	48.0 (11.4)
Female sex, n (%)	7 (87.5)	4 (80.0)	11 (84.6)
Body mass index, mean (SD)	26.6 (4.8)	25.4 (4.1)	26.1 (4.4)
Weight, mean (SD), kg	77.0 (16.6)	74.3 (14.5)	76.0 (15.3)
Race, n (%)
White	8 (100.0)	5 (100.0)	13 (100.0)
Smoking status (tobacco), n (%)
Current	6 (75.0)	2 (40.0)	8 (61.5)
Former	1 (12.5)	1 (20.0)	2 (15.4)
Primary diagnosis, n (%)
CLE	5 (62.5)	2 (40.0)	7 (53.8)
SLE	3 (37.5)	3 (60.0)	6 (46.2)
Time from first CLE/SLE diagnosis to randomization, mean (SD), y	11.9 (8.8)	8.2 (4.6)	10.4 (7.5)
Median (min–max)	10.6 (1.7–26.2)	9.1 (2.2–14.4)	9.1 (1.7–26.2)
SLE per EULAR/ACR 2019 classification criteria, a n (%)
Subacute or discoid lupus	6 (75.0)	5 (100.0)	11 (84.6)
Acute cutaneous lupus	5 (62.5)	2 (40.0)	7 (53.8)
Nonscarring alopecia	3 (37.5)	2 (40.0)	5 (38.5)
Joint involvement	2 (25.0)	4 (80.0)	6 (46.2)
Fever	1 (12.5)	0	1 (7.7)
Low C3 or C4 level	1 (12.5)	2 (40.0)	3 (23.1)
Low C3 and C4 level	0	1 (20.0)	1 (7.7)
Pleural or pericardial effusion	0	3 (60.0)	3 (23.1)
Oral ulcers	0	2 (40.0)	2 (15.4)
Leukopenia	0	1 (20.0)	1 (7.7)
Thrombocytopenia	0	1 (20.0)	1 (7.7)
Autoimmune hemolysis	0	1 (20.0)	1 (7.7)
Seizure	0	1 (20.0)	1 (7.7)
Proteinuria (>0.5 g/24 h)	0	1 (20.0)	1 (7.7)
Renal biopsy class III or IV lupus nephritis	0	1 (20.0)	1 (7.7)
mCLASI‐A score, mean (SD)	14.9 (7.3)	11.0 (3.0)	13.4 (6.1)
Median (min–max)	13.5 (6.0–25.0)	9.0 (9.0–16.0)	12.0 (6.0–25.0)
CLASI‐A score, mean (SD)	15.0 (7.2)	11.6 (2.7)	13.7 (6.0)
Median (min–max)	13.5 (6.0–25.0)	11.0 (9.0–16.0)	12.0 (6.0–25.0)
CLASI‐A score: category 1, n (%)
<10	2 (25.0)	1 (20.0)	3 (23.1)
≥10	6 (75.0)	4 (80.0)	10 (76.9)
CLASI‐A score: category 2, n (%)
Mild (0–9)	2 (25.0)	1 (20.0)	3 (23.1)
Moderate (10–20)	4 (50.0)	4 (80.0)	8 (61.5)
Severe (21–70)	2 (25.0)	0	2 (15.4)
CLASI‐D score, mean (SD)	7.6 (9.3)	5.6 (1.6)	6.8 (7.0)
Median (min–max)	2.0 (0–25.0)	6.0 (4.0–8.0)	5.0 (0–25.0)
Presence of autoantibodies, n (%) b
NA	8 (100.0)	5 (100.0)	13 (100.0)
Anti‐Ro antibodies	4 (50.0)	1 (20.0)	5 (38.5)
Anti‐dsDNA antibodies	3 (37.5)	3 (60.0)	6 (46.2)
Anti‐Smith antibodies	0	1 (20.0)	1 (7.7)
Anti‐cardiolipin antibodies, anti‐β2GP1 antibodies, or lupus anticoagulant	0 (0)	0 (0)	0 (0)

^*ACR, American College of Rheumatology; ANA, antinuclear antibodies; anti‐β2GP1, anti–β2‐glycoprotein 1; CLASI‐A, Cutaneous Lupus Erythematosus Disease Area and Severity Index‐Activity; CLASI‐D, Cutaneous Lupus Erythematosus Disease Area and Severity Index‐Damage; CLE, cutaneous lupus erythematosus; dsDNA, double‐stranded DNA; max, maximum; mCLASI‐A, modified Cutaneous Lupus Erythematosus Disease Area and Severity Index‐Activity; min, minimum; NA, not available; SLE, systemic lupus erythematosus.

^a Patients can be counted in more than one domain/criterion.

^b Autoantibody test was considered positive at the following titers: ANA, ≥1:80; anti‐Smith, ≥1:7; anti‐dsDNA, ≥1:25; anti‐Ro, ≥1:7.

The treatment duration and compliance rates were balanced between the afimetoran and placebo arms. The median duration of treatment was 113 days (afimetoran: median 113.0 [minimum–maximum 56–113] days; placebo: median 113.0 [minimum–maximum 113–113] days), with a high treatment compliance rate (>99%) for both arms despite the operational constraints caused by the COVID‐19 pandemic.

At baseline, use of both oral glucocorticoids and antimalarials was higher in the placebo arm (4 patients [80%]) than the afimetoran arm (2 patients [25%]). In the afimetoran arm, none of the patients received oral glucocorticoids, whereas two (25%) patients received systemic antimalarials at day 1. In the placebo arm, three patients (60%) received oral glucocorticoids and four patients (80%) received systemic antimalarials at day 1. None of the patients in either arm required oral glucocorticoid rescue during the treatment period.

Safety

Afimetoran was well tolerated compared with the placebo (Table 2). No deaths, SAEs, or drug‐related SAEs were reported in the study. Overall, AEs were mild or moderate and resolved by the study end without intervention. The overall proportion of patients with AEs was lower in the afimetoran arm than the placebo arm (n = 5 of 8 [62.5%] vs n = 4 of 5 [80.0%]) except for infections and infestations including nasopharyngitis or bronchitis and COVID‐19 infection, which were higher in the afimetoran arm (37.5% vs 20.0% in the placebo arm). Further, there were no serious AEs or safety signals observed in laboratory tests, vital signs, or EKG findings.

Table 2.

Safety summary for afimetoran versus placebo^*

Event, n (%)	Afimetoran (n = 8)	Placebo (n = 5)
Death	0	0
AEs	5 (62.5)	4 (80.0)
Serious AEs	0	0
AEs leading to discontinuation a	1 (12.5)	0
Drug‐related AEs b	2 (25.0)	2 (40.0)
Nonserious AEs	5 (62.5)	4 (80.0)
Infections and infestations c	3 (37.5)	1 (20.0)
Nervous system disorders	1 (12.5)	2 (40.0)
Injury, poisoning, and procedural complications	1 (12.5)	2 (40.0)
General and administration site disorders	2 (25.0)	0
Gastrointestinal disorders	1 (12.5)	1 (20.0)
Psychiatric disorders	1 (12.5)	0
Respiratory, thoracic, and mediastinal disorders	1 (12.5)	0
Blood and lymphatic system disorders	0	1 (20.0)
Musculoskeletal and connective tissue disorders	0	1 (20.0)

^*Data (n) represent the number of patients within the afimetoran or placebo arm with an event. AE, adverse event.

^a Patient discontinued afimetoran at week 6 (study discontinuation visit delayed to week 8 due to COVID‐19 pandemic), attributed to symptomatic COVID‐19.

^b Drug‐related AEs were investigator‐assessed at the study site. Drug‐related AEs were reported during the masking treatment period; however, after unmasking, those patients were found to be in the placebo group.

^c Infections and infestations included nasopharyngitis/bronchitis and COVID‐19 infection. One patient discontinued from the afimetoran arm at week 6 for moderate COVID‐19 infection.

PK profile

Near maximum exposure was reached by week 1, substantially preceding the first efficacy timepoint at week 4, and was consistent over the course of active treatment (Figure 3A). Geometric mean C_max increased approximately four‐fold from 47.6 ng/mL at day 1 to 200.0 ng/mL at week 16. Similarly, geometric mean AUC_tau increased four‐fold from 883 ng×h/mL on day 1 to 3,599 ng×h/mL at week 16. Median T_max was 6 hours (range 2–8 hours) on day 1 and 4 hours (range 2–6 hours) at week 16. Plasma concentrations of afimetoran exceeded the projected target inhibition concentration over 24 hours.

Figure 3.

(A) Pharmacokinetic profile of afimetoran and (B) inhibition of interferon‐1 (IFN‐1) gene signature. Gene module score represents the mean percent inhibition of IFN‐1, in which the mean of placebo from baseline is subtracted. Data are reported after the last dose and normalized to HV samples. Lower limit of quantification = 0.1 ng/mL. CI, confidence interval; C_trough, plasma concentration at the end of dosing interval.

PD biomarkers

PD biomarkers were compared between 8 patients with CLE and 5 age‐, race and ethnicity–, and sex‐matched HVs. Expression of the drug target TLR7/8 pathway genes (TLR7, TLR8, ANXA3, SUNCNR1, INHBA, SMIM3, and PLAUR; Supplementary Figure S2A and B) was reduced in the afimetoran group, demonstrating its potent and sustained PD effect. Gene set variation analysis showed a PD impact of afimetoran on immune cell populations, including immature and activated dendritic cells (Supplementary Figure S2C and D) and macrophages (data not shown). Furthermore, potent inhibition of the IFN‐1 gene signature (interferon‐inducible protein 6 [IFI6], HERC5, IFIT1, MX1, and TNF receptor superfamily 21) was observed early (week 1), maintained through active treatment (week 16), persisted 4 weeks after stopping treatment (week 20) (Figure 3B), and was significantly reduced compared to placebo (area under the curve 0.77; P < 0.0001 [data not shown]).

In the afimetoran arm, TLR7/8‐stimulated samples showed >50% decreased secretion for most cytokines and chemokines, which was not seen in the placebo arm (Supplementary Figure S3). For TLR7 stimulation, complete inhibition (>90%) of cytokine secretion by afimetoran was observed at week 1 through week 16 for IFNγ, interleukin‐6 (IL‐6), CXCL8/IL‐8, CCL3/ macrophage inflammatory protein 1α (MIP‐1α), CCL4/MIP‐1β, and tumor necrosis factor α (TNFα). Partial inhibition (from week 1 to week 4) was seen for IL‐18 (>25%), IL‐3 (>20%), and lymphotoxin α (LTA) (>40%). Approximately 5% inhibition was observed for IL‐2 at week 1 followed by 20% inhibition from week 2 to week 4.

No inhibition was seen for granulocyte–macrophage colony‐stimulating factor (GM‐CSF), IL‐4, IL‐5, and IL‐7. A rebound in IL‐2, IL‐3, and IL‐5 cytokine production was seen as early as week 8 as the inhibition percentages dropped. However, for IFNγ, IL‐6, CXCL8/IL‐8, CCL3/MIP‐1α, CCL4/MIP‐1β, and TNFα, a 90% or higher inhibition was still observed at week 16. The placebo‐treated group showed IFNγ, IL‐6, CCL3/MIP‐1α, and CCL4/MIP‐1β cytokine production starting week 1.

For TLR8 stimulation, a complete inhibition of cytokine secretion by afimetoran was observed at week 1 through week 16 for IFNγ, IL‐18, IL‐6, CXCL8/IL‐8, CCL3/MIP‐1α, CCL4/MIP‐1β, and TNFα. LTA (>50%), GM‐CSF (>30%), IL‐3 (>30%), IL‐2 (>30%), and IL‐4 (>20%) showed inhibition at week 4. There was no reported inhibition of IL‐5 and IL‐7, and only rebound in IL‐2 and IL‐5 production by week 8 was seen. For IFNγ, IL‐6, CXCL8/IL‐8, CCL3/MIP‐1α, CCL4/MIP‐1β, and TNFα, a 90% or higher inhibition was still observed at week 16.

Efficacy

Compared with placebo, patients treated with afimetoran showed a greater mean reduction in CLASI‐A scores as early as the first assessment at week 4, which continued through the 16 weeks of active treatment and persisted to the week 20 follow‐up (Figure 4A). Fifty percent of patients treated with afimetoran achieved at least the minimum CLASI‐A score improvement (>4 points from baseline ³⁷ ) at week 16 compared with 0% of patients receiving placebo. At week 16, there was a mean reduction of 42.6% in CLASI‐A from baseline with afimetoran compared with 3.4% with placebo (Figure 4B). Fifty percent of patients treated with afimetoran achieved CLASI‐A50 response (a 50% change from baseline in CLASI score) at week 16 compared with 0% of patients receiving placebo. As a supplemental efficacy marker, improvement in the molecular disease profile, which resulted from a transcriptomics comparison and was defined by the differential of gene set variation analysis enrichment scores between patients with CLE and HVs, was rapid (week 1), sustained throughout the treatment period (week 16), and maintained up to 4 weeks after treatment (week 20) (data not shown). No impact on the disease profile was seen in placebo‐treated patients (data not shown).

Figure 4.

(A) Mean change and (B) mean percent change in CLASI‐A scores from baseline for afimetoran versus placebo over 20 weeks. Baseline is the last nonmissing results with a collection date/time less than the date/time of the first dose of study medication. In (A), values next to number of patients represent adjusted mean change from baseline (SE) for afimetoran or placebo. In (B), values next to number of patients represent adjusted mean percentage change from baseline (SE) for afimetoran or placebo. The blue‐shaded area represents the posttreatment follow‐up period. BL, baseline; CLASI‐A, Cutaneous Lupus Erythematosus Disease Area and Severity Index‐Activity; QD, once daily.

DISCUSSION

Afimetoran, a first in class, orally bioavailable, potent, selective small molecule inhibitor of TLR7/8, demonstrated a favorable safety profile without apparent safety concerns and was well tolerated compared with a placebo in patients with CLE, thereby meeting the primary objective of the study. This study also preliminarily suggests potential efficacy with associated PD effects of afimetoran in patients with CLE.

PK analyses showed that afimetoran reached near maximal concentration after 1 week of treatment, with an approximate four‐fold increase in exposure (C_max and AUC_tau) over the course of treatment, similar to afimetoran PK in HVs. ³⁸ PD analyses showed a rapid response to afimetoran by week 1, with reduced expression of various disease‐associated and TLR7/8 pathway–associated cytokines and mRNA signatures. Afimetoran also demonstrated treatment efficacy early, throughout, and beyond the treatment period, showing early and sustained clinical response as measured by CLASI‐A. Importantly, the CLASI‐A correlated with changes in key biomarkers, most notably the IFN‐1 gene expression signature, which is associated with disease characteristics, activity, and response to certain treatments. ³⁹ , ⁴⁰ , ⁴¹ , ⁴²

The aim of CLE treatment is to reduce disease activity and minimize cosmetic damage. ⁴³ Therapeutic options are currently limited for CLE, with antimalarials and systemic steroids being recommended as first‐line systemic treatment. ¹¹ The antimalarial HCQ is often the first treatment prescribed for patients with CLE with severe lesions. ¹⁹ However, its mechanism of action is broad; HCQ inhibits TLRs 7, 8, and 9 and the cyclic GMP‐AMP synthase‐stimulator of IFN gene pathway, and as such it can have many different effects on cellular processes. ⁴⁴ , ⁴⁵ , ⁴⁶ Systemic treatments such as thalidomide and lenalidomide have suggested high response rates in patients with CLE, ⁴⁷ and Phase 3 trials of biologic therapies in CLE are ongoing. ⁴⁸ , ⁴⁹ Despite these options, gaps in therapy persist, including parenteral administration, ⁵⁰ incomplete characterization and off‐label use, ¹² and toxicity with long‐term use ²⁰ , ²¹ ; therefore, better alternatives are needed. Afimetoran selectively targets TLR7 and 8, and, considering the preliminary tolerance and safety profile demonstrated in our study, may be a potential treatment alternative for patients with CLE. Further investigation is required to confirm the efficacy and safety of afimetoran over longer‐term periods in larger samples of patients with CLE and to identify the optimal use of afimetoran alone or with existing treatments such as HCQ.

Previously, a receiver operating characteristic analysis found a reduction of 4 points in CLASI‐A score (or percentage decrease of 20%) from baseline correlated with a clinical improvement in CLE severity. ³⁷ In our study, the adjusted mean reduction in adjusted CLASI‐A scores from baseline to week 16 was 7.6 points—more than a 40% mean reduction from baseline at week 16 with afimetoran treatment. The PK and PD results demonstrated that afimetoran can be administered orally in a convenient once daily dosing schedule, which could potentially overcome the patient adherence issues seen with other recommended CLE treatments. ⁵¹ Plasma concentrations of afimetoran also exceeded the targeted inhibition concentration, supporting a 30‐mg once daily dosage. Further, PD analyses demonstrated potent and durable inhibition of levels of several cytokines, suggesting a sustained PD effect. TLR7/8 target engagement data also showed potent cytokine inhibitions with afimetoran when compared to a placebo. These PD effects correlated with exploratory efficacy response (CLASI). Overall, our results support the potential efficacy of once daily dosing of afimetoran in patients with CLE.

The limitations of the study include a small sample size, single‐center conduct, lack of racial and ethnic diversity (White‐only population), and relatively variable antimalarial use (25% of patients on day 1). The mCLASI‐A score used for determining eligibility was used to minimize confounding for chronic noninflammatory changes and other nonlupus inflammatory changes; however, the modified scale is not validated for CLE trials. Additionally, as most autoantibody levels were below the LLOQ of the assay, the full impact of afimetoran on autoantibodies could not be assessed. Although TLR7 has been implicated in the genetic epidemiology of severe COVID‐19, particularly in men, ⁵² the small sample size of this study precludes speculation regarding any mechanistic relationship between TLR7 inhibition and the single AE of nonsevere COVID‐19 in a female participant. During the study conduct, the background rate of SARS‐CoV‐2 infection was significant and the pandemic in Germany was sufficiently intense to interfere with recruitment.

In summary, this study introduces a potential new CLE treatment with a novel mechanism of action, potent PD effect, favorable safety profile, and selectivity for TLR7/8, accompanied by clinical and biomarker data, which may prove to be significant in changing the landscape of CLE. Our results build on the previous preclinical data on the effects of afimetoran and will contribute to the continued clinical investigation of afimetoran in lupus, including the ongoing Phase 2b study of afimetoran in SLE (NCT04895696). Together, these findings demonstrate the possibility of substantial therapeutic benefit for patients with CLE and warrant longer‐term studies.

AUTHOR CONTRIBUTIONS

All authors contributed to at least one of the following manuscript preparation roles: conceptualization AND/OR methodology, software, investigation, formal analysis, data curation, visualization, and validation AND drafting or reviewing/editing the final draft. As corresponding author, Dr Hosein confirms that all authors have provided the final approval of the version to be published and takes responsibility for the affirmations regarding article submission (eg, not under consideration by another journal), the integrity of the data presented, and the statements regarding compliance with institutional review board/Declaration of Helsinki requirements.

Supporting information

Appendix S1: Supplementary Information