WHAT IS ALREADY KNOWN ON THIS TOPIC•Immune checkpoint inhibitors have substantial efficacy in oncology, and immune checkpoint agonists have demonstrated initial evidence of efficacy in autoimmune diseases. The B and T lymphocyte attenuator (BTLA)/herpes virus entry mediator checkpoint receptor system has been shown to be dysregulated in systemic lupus erythematosus (SLE), suggesting that targeting inhibitory checkpoint receptors like BTLA, which regulate immune cell activation, could be a promising approach for the treatment of SLE.WHAT THIS STUDY ADDS•Venanprubart, an agonist antibody to BTLA, was assessed for safety and efficacy in inducing remission of rash and/or arthritis in patients with SLE. At an interim analysis, venanprubart did not display clinical efficacy compared to placebo, and the study was terminated.HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY•The lack of clinical benefit with venanprubart was not due to lack of target engagement or inadequate exposure.
Immune-related RNA transcript and protein changes were consistently modulated in patients treated with venanprubart, indicating some impact on the immune system. However, the molecular changes observed did not translate to clinical benefits.Alt-text: Unlabelled box dummy alt text
INTRODUCTION
Systemic lupus erythematosus (SLE) is a chronic autoimmune multisystem disease that can affect multiple organ systems, including mucocutaneous, musculoskeletal, cardiac, haematologic, pulmonary, renal, gastrointestinal, reproductive, and nervous system manifestations [1]. Targeted therapy that impacts numerous cell types but is not broadly immunosuppressive will likely be required to provide robust clinical outcomes for patients with SLE [[2], [3], [4]]. To this end, targeting the inhibitory checkpoint receptor B and T lymphocyte attenuator (BTLA), an Ig superfamily member and part of the family of checkpoint inhibitory receptors that negatively regulates immune cell activation, might have potential to fulfil this unmet medical need.
BTLA is expressed on multiple immune cells involved in SLE, including B cells, plasmacytoid dendritic cells, CD4+ T, CD8+ T, ƴdT cells, and myeloid-derived suppressor cells [5]. The natural ligand for BTLA is the tumour necrosis factor receptor superfamily member, herpes virus entry mediator (HVEM) [6], which induces tyrosine-phosphorylation of 2 conserved immunoreceptor tyrosine-based inhibitory motifs (ITIMs) on the cytoplasmic domain of BTLA [7]. This cascades down a signalling pathway that results in suppression of immune cell activation [8].
A recent study profiled the expression pattern of the BTLA/HVEM signalling system in patients with SLE [9]. Although BTLA expression was relatively unchanged between healthy volunteers and patients with lupus, HVEM expression (protein and RNA) was decreased in patients with SLE. Furthermore, lower HVEM expression correlated with disease activity markers in lupus, including Systemic Lupus Erythematosus Disease Activity Index (SLEDAI), interferon (IFN)-I signature genes, autoantibodies, complement, and leukopenia.
The findings in this study suggested a role of BTLA/HVEM signalling in regulating the immune-associated pathogenesis of SLE and a potential target for therapeutic intervention.
Venanprubart is an agonist antibody to BTLA that was developed as a potential treatment for autoimmune disorders, including SLE. The binding of venanprubart to BTLA induces tyrosine-phosphorylation of its cytoplasmic ITIM, mimicking the activity observed with its natural ligand HVEM. Venanprubart competes with HVEM in binding to BTLA.
It reduces lymphocyte activation, inhibits inflammatory cytokine production, and contributes to the resolution of inflammation in vitro and in vivo in mouse models [10]. This phase 2 (NCT05123586) study was conducted to evaluate the efficacy and safety of venanprubart in adult patients with at least moderately active SLE, but was terminated early due to futility.
METHODS
Study design
BT01 (NCT05123586) is a multinational, multicentre, randomised, double-blind, placebo-controlled, parallel-group, 2-arm phase 2 study designed to evaluate the efficacy and safety of venanprubart in addition to standard of care (SOC) in adult study participants with at least moderately active SLE. Randomised patients were assigned using a 1:1 ratio to receive either 450 mg venanprubart (N = 42) subcutaneously (SC) every 2 weeks plus SOC or placebo (N = 43) every 2 weeks plus SOC over a treatment period of 24 weeks. The study design is illustrated in Supplementary Figure S1, and the trial profile is illustrated in Supplementary Figure S2.
Patients
The primary study population included male or female patients between 18 and 65 years old. Inclusion criteria are as follows: have received a clinical diagnosis of SLE at least 24 weeks before the screening visit, have a positive antinuclear antibody (titre ≥1:80) and/or a positive anti-dsDNA and/or positive anti-Smith antibody test at screening as assessed by the central laboratory, and have a SLEDAI 2000 (SLEDAI-2K) score ≥6 at screening. Inclusion criteria at randomisation included receiving at least 1 background SOC medication for SLE, having active rash and/or active arthritis as defined by the SLEDAI-2K at screening and randomisation, and having clinical SLEDAI-2K score ≥4 (not including any items related to laboratory values) at randomisation.
Patients were excluded from the study if any of the following key exclusion criteria were met: had received/were receiving concomitant treatment outside the limitations of the study’s prespecified limits, had an infection or medical condition that was included in the study’s exclusion criteria list, had an ‘excluded’ diagnosis or specific abnormality on screening laboratory tests. See the supplemental appendix for the full inclusion/exclusion criteria.
Patients were not involved in the design and conduct of this research.
Randomisation and masking
Assignment to treatment groups was determined by a computer-generated random sequence using the Interactive Web Response System. Stratification factors included disease activity at baseline (SLEDAI-2K <10; SLEDAI-2K ≥10), corticosteroid dose at baseline (<10 mg; ≥10 mg), and geographic region (United States/Europe/Rest of the world). Investigators and all individuals involved in administering the blinded treatment or performing assessments remained blinded throughout the course of the study.
Patients were also blinded to their assigned study intervention, and the blind was maintained by having an uninvolved party such as an unblinded pharmacist prepare and dispense all the study interventions. Blinded site personnel administered the study intervention to patients.
Procedures
Patients assigned to venanprubart or placebo were randomised at visit 2 and had study visits at baseline and at weeks 2, 4, 6, 8, 10, 12, 16, 18, 20, 22, and 24. Clinical laboratory tests were performed by a central laboratory or by a local laboratory. Laboratory test results that could unblind the study were not reported to investigative sites or to other blinded personnel.
The following lab tests were blinded: pharmacokinetics (PKs), antidrug antibodies (ADAs), receptor occupancy (RO), lymphocyte subsets, tryptase, complement (C3, C3a, C5a), cytokine panel, and N methylhistamine.
The PK analysis dataset included 202 serum samples collected from 41 venanprubart-treated participants at predose on nominal study days 29, 57, 85, 113, 141, and 169. Samples with time deviations >10% and prior doses of <10 days or >18 days were excluded from mean concentration–time plots and summary statistics.
Immunogenicity (ADAs) samples were collected at visits 2, 4, 8, and 14. A predose blood sample for PK analysis was collected at the same time points to aid in the interpretation of results.
Outcomes
The primary endpoint was the assessment of whether venanprubart was superior to placebo in inducing remission of rash and/or arthritis in patients with at least moderately active SLE. The primary efficacy endpoint was the proportion of participants who achieve remission of rash and/or arthritis, as defined by the SLEDAI-2K, at week 24. Remission is achieved if 1 of the following occurs: if only arthritis is present at baseline, then the primary endpoint is met if arthritis is absent at week 24; if only rash is present at baseline, then the primary endpoint is met if rash is absent at week 24; if both arthritis and rash are present at baseline, then the primary endpoint is met if either arthritis, or rash, or both arthritis and rash are absent at week 24.
Participants who fail to complete the 24-week treatment period or violate the concomitant medications rules were treated as nonresponders for the purpose of the primary endpoint analysis. The rationale for using the rash and arthritis domains of SLEDAI as the primary endpoint was to concentrate the primary endpoint on the clinical efficacy parameters with the highest probability of predicting success for composite registration endpoints in future trials, while minimising the statistical noise from SLEDAI domains that are substantially unpowered. The endpoint derived from a signal-to-noise analysis of the individual components making up the Systemic Lupus Erythematosus Responder Index-4 (SRI-4) composite registration endpoint which requires assessment of SLEDAI (24 items), British Isles Lupus Assessment Group (97 items), and Physician Global Assessment (1 item) for combined 122 total assessments, not including any additional assessments required for concomitant medications, etc [11].
Secondary efficacy endpoints included the proportion of participants who achieved SRI-4 response and SLEDAI-4 (defined as a ≥4-point reduction in SLEDAI-2K score from baseline) response at week 24.
Other secondary endpoints included PKs, immunogenicity, and biomarker analyses.
RO of BTLA is quantified with median fluorescence intensity of BTLA using Alexa fluor 647 assay with complete antibody 22B3 on B cells (CD19+/CD3− cells), CD4+ T cells (CD3+/CD4+/CD8− cells), and CD8+ T cell (CD3+/CD8+/CD4− cells). %RO is calculated using baseline sample as 0% control and saturated samples at the given time point as 100% control.
Safety outcomes were reported and included all adverse events (AEs), AEs of special interest, vital signs, and were compared to placebo at the time of analysis.
Statistical analysis
The primary endpoint for this study was the proportion of patients who achieve remission of rash and/or arthritis at week 24 and did not violate any concomitant medication rules and did not permanently discontinue from the study intervention between venanprubart 450 mg vs placebo. The primary endpoint analysis used the modified intent-to-treat population, and nonresponder imputation was used to account for the missing data.
The primary analysis method for treatment comparisons of binary efficacy and health outcome variables was g-computation using a logistic regression model as a working model with treatment group (venanprubart or placebo), baseline disease activity (SLEDAI-2K <10 or SLEDAI-2K ≥10), corticosteroid dose at baseline (<10 mg/d or ≥10 mg/d), and geographic region (USA, Europe, or rest of the world) in the model. For treatment group and placebo group comparison, an estimate of the unconditional difference with corresponding 90% CI, the unconditional odds ratio with corresponding 90% CI, and P value is presented. The P value for all other explanatory variables is presented.
If the working model for g-computation does not converge, the difference in the simple average will estimate the unconditional difference between different doses of venanprubart and placebo, and Fisher’s exact test will calculate the P value. The treatment response rates, treatment differences vs placebo, and their corresponding 90% CIs are provided.
Secondary efficacy endpoints included the proportion of participants who achieved SRI-4 response and SLEDAI-4 response at week 24. Participants who failed to complete the 24-week treatment period or violated the concomitant medications rules were imputed to nonresponse SRI-4, and SLEDAI-4 response at treatment completion was analysed using the model specified for the primary analysis.
Biomarker section
Exploratory analyses included pharmacodynamic changes in gene expression and protein markers associated with venanprubart, which were assessed using bulk RNA sequencing (95 patients in the venanprubart group and 98 in the placebo group) and the Olink Explore 5400 panels (83 patients in the venanprubart group and 90 in the placebo group) at weeks 12 and 24. The data were log-transformed before analysis, and changes were calculated as differences from baseline using a mixed model with repeated measurements and an AR1 covariance structure. A threshold of a 1.5-fold change and a P value of .05 were applied.
The Benjamini-Hochberg (BH) procedure was used to adjust the P values for multiple comparisons. Differential expression between patients with SLE and healthy controls (HCs) (59 patients at baseline for bulk RNA sequencing, 65 for Olink, and 30 HCs) were assessed using linear models with log-transformed data, a 1.5-fold change threshold and a P value of .05, with the BH procedure applied to adjust P values for multiple comparisons.
RESULTS
Baseline demographics and disease characteristics
Demographics were balanced across treatment groups (Table 1). The mean age was 44 years, with a mean body mass index of 27.8 kg/m2, and 94% of participants were female. Patients had a baseline mean SLEDAI-2K score of 9.81 (SD 2.98), where 42.4% had a SLEDAI-2K status <10 and 57.6% ≥10 (Table 1).
Baseline disease characteristics were also generally well balanced across treatment groups.Table 1Baseline characteristics and demographics.Table 1 dummy alt textVariableVenanprubartPlaceboTotalN = 42N = 43N = 85Age (y) mean (SD)44.2 (12.1)43.3 (11.1)43.8 (11.5)Body mass index, mean (SD)27.8 (6.9)27.8 (6.6)27.8 (6.7)Female, n (%)42 (97.7)38 (90.5)80 (94.1)Race, n (%) American Indian or Alaska7 (16.7)7 (16.3)14 (16.5)Native Asian3 (7.1)3 (7.0)6 (7.1) Black or African American1 (2.4)3 (7.0)4 (4.7) White30 (71.4)30 (69.8)60 (70.6) Multiple1 (2.4)01 (1.2)Country, n (%) Argentina7 (16.7)8 (18.6)15 (17.6) Mexico8 (19.0)7 (16.3)15 (17.6) Poland6 (14.3)7 (16.3)13 (15.3) Taiwan3 (7.1)2 (4.7)5 (5.9) United States18 (42.9)19 (44.2)37 (43.5)Time since onset of lupus (y), mean (SD)9.97 (8.11)9.07 (8.77)9.52 (8.41)SLEDAI-2K at baseline Avg. score9.6 (2.8)10.1 (3.2)9.8 (3.0) Status ≥10 (%)24 (57.1)25 (58.1)49 (57.6)BICLA eligibility At least 1 A, yes (%)10 (23.8)12 (27.9)22 (25.9) No A and at least 2 B, yes (%)18 (42.9)21 (48.8)39 (45.9) Total28 (66.7)33 (76.7)61 (71.8)Corticosteroid baseline dose (mg/d), mean (SD)6.3 (5.8)5.9 (6.3)6.1 (6.0)Corticosteroid use Avg. dose6.3 (5.8)5.9 (6.3)6.21 (6.0) <10 mg/d (%)26 (61.9)32 (74.4)58 (68.2) ≥10 mg/d (%)16 (38.1)11 (25.6)27 (31.8)Corticosteroid use, n (%) No13 (31.0)15 (34.9)28 (32.9) Yes29 (69.0)28 (65.1)57 (67.1)Immunosuppressant use, n (%) No15 (35.7)16 (37.2)31 (36.5) Yes27 (64.3)27 (62.8)54 (63.5)Antimalarial use, n (%) No24 (57.1)19 (44.2)43 (50.6) Yes18 (42.9)24 (55.8)42 (49.4)No. of joint count Swollen6.9 (4.3)6.4 (3.7)6.7 (4.0) Tender11.0 (5.5)11.2 (6.3)11.1 (5.9)CLASI Total activity score6.3 (5.4)5.3 (4.4)5.8 (4.9) Total damage score0.7 (2.0)0.9 (3.0)0.8 (2.5) PGA score5.9 (1.2)5.4 (1.4)5.7 (1.4)BICLA, BILAG-based Composite Lupus Assessment; BILAG, British Isles Lupus Assessment Group; CLASI, Cutaneous Lupus Erythematosus Disease Area and Severity Index; n, number of subjects; N, number of subjects in population; PGA, Physician Global Assessment; SLEDAI-2K, Systemic Lupus Erythematosus Disease Activity Index 2000.
Efficacy
The trial was terminated early for lack of efficacy at an interim analysis that was conducted on data from the first 54 patients (26 on venanprubart; 28 on placebo) who had completed 24 weeks of treatment. At this time, the treatment dropout rate was ∼9%. Of the patients who entered the study, 92.9% (79/85) completed the 24 weeks of treatment (Supplementary Fig S2).
Results presented here reflect all data collected through the end of the study.
At week 24, after accounting for intercurrent events and missing data, venanprubart (11.9%; 5/42) did not separate from placebo (20.9%; 9/43) for remission of rash and/or arthritis in patients with at least moderately active SLE (P = .452) (Fig 1).Figure 1Remission of SLEDAI-2K rash and/or arthritis response rates through week 24 in patients treated with venanprubart or placebo. Nonresponder imputation was used. Placebo, N = 43; venanprubart, N = 42.
N, number of patients in the treatment group or placebo; SLEDAI-2K, Systemic Lupus Erythematosus Disease Activity Index 2000.Figure 1 dummy alt text
The proportion of patients who achieved SLEDAI-4 response through week 24 was not statistically significant (P = .520) between venanprubart (9.5%; 4/42) and placebo (16.3%; 7/43) (Fig 2A). The proportion of patients who achieved SRI-4 response through week 24 was also not statistically significant (P = .520) between venanprubart (9.5%; 4/42) and placebo (16.3%; 7/43) (Fig 2B).Figure 2Response rates for SLEDAI-4 (A) and SRI-4 (B) through week 24 in patients treated with venanprubart or placebo. Nonresponder imputation was used.
Placebo, N = 43; venanprubart, N = 42. N, number of patients in the treatment group or placebo; SLEDAI, Systemic Lupus Erythematosus Disease Activity Index; SRI-4, Systemic Lupus Erythematosus Responder Index-4.Figure 2 dummy alt text
A summary of the primary and secondary endpoints at week 24, including the CIs, is in Supplementary Table S1.
The RO percentage of venanprubart was demonstrated by a rapid increase of BTLA on B cell, CD4+ T cell, and CD8+ T cell starting at week 4 and remained maximum occupancy over time (Fig 3).Figure 3Venanprubart receptor occupancy of BTLA over 24 weeks. % RO of venanprubart was calculated using the baseline sample as 0% control, with saturated samples at a given timepoint as 100% control. RO of BTLA was measured with MESF of BTLA using Alexa fluor 647 assay with complete antibody 22B3 on B cells (CD19+/CD3– cells), CD4+ T cells (CD3+/CD4+/CD8– cells), and CD8+ T cell (CD3+/CD8+/CD4– cells). BTLA, B and T lymphocyte attenuator; MESF, median fluorescence intensity; RO, receptor occupancy.Figure 3 dummy alt text
PK and immunogenicity assessments
Mean Ctrough concentrations by study day show PK results were variable (Fig 4; Supplementary Table S2). At baseline, of the 82 participants evaluable for treatment-emergent antidrug antibodies (TE ADAs), 3 (3.7%) were positive for pre-existing ADA against venanprubart. Postbaseline, of the 40 participants in the venanprubart-treated group that were evaluable for TE ADA, 19 (47.5%) were positive for TE ADA with titre ranging from 1:20 to 1:10,240 (median 1:160).
Of the 19 participants classified as TE ADA+, only 2 exhibited notable decreases in venanprubart concentrations relative to other participants. No participant in the placebo group was positive for TE ADA. Venanprubart concentrations that were below the lower level of quantitation (LLOQ = 0.2 pg/mL) at baseline were set to 0, whereas all postbaseline concentrations that were below LLOQ were imputed as LLOQ/2 = 0.1 pg/mL (Supplementary Fig S3).Figure 4Mean Venanprubart Ctrough by study day, log-linear (A), linear ± SD (B).
Ctrough, predose concentration.Figure 4 dummy alt text
Biomarker analyses
An IFN transcriptomic signature based on genes DDX60, LY6E, MX, OAS3, IFI44, and USP18 showed differential gene expression in patients with SLE vs HCs (adjusted P value <.05). In total, 49.3% of patients had higher expression of these genes than the median and between 43% and 50% had higher expression than the mean (Fig 5A). Using an fold change (FC) threshold of 1.5 and P value <.05, transcriptomics data showed that 43 genes were modulated by venanprubart at week 12 and 32 genes at week 24, and 6 genes were modulated at weeks 12 and 24, 1 downregulated (SMIM10) and 5 upregulated (HCAR1, LURAP1, RPS3AP40, SLC7A9, SPTBN4) (Fig 5B).
Using an FC threshold of 1.5 and an adjusted P value <.05, we determined 22 proteins were changed from baseline at both weeks 12 and 24 in the venanprubart treatment arm (Fig 5C; Supplementary Tables S3 and S4). Several of the downregulated proteins in the venanprubart arm, eg, ARHGAP42, CXCL13, CXCL9, IFNB1, IL2RA, KRT18, MAPK4, MTG2, PDCD1, RERG, RRM2, and TNFRSF8, are involved in regulating immune cell migration, activation, and survival through cytokine signalling, T cell differentiation and exhaustion, as well as cell proliferation and apoptosis. The upregulated proteins in the venanprubart arm included ALDH1A3, CCDC148, CPSF2, EIF4B, IQSEC2, MIDN, MMP12, MUC2, SERPINA9, and TCEAL1, which are involved in the regulation of inflammation, protection of mucosal surfaces, and immune response modulation.
Particularly, the molecular expression of HVEM (TNFRSF14), the ligand for BTLA, showed nominal expression level differences compared to healthy volunteers with statistical P value differences of .49 for bulk RNA-seq and .16 for OLINK (Supplementary Fig S4).Figure 5Differential expression analysis of gene and protein markers between patients with SLE (Olink n = 65, Bulk RNA-seq n = 59) and healthy controls (HCs n = 30). (A) Violin plot of gene expression for the IFN gene signature, based on genes DDX60, LY6E, MX, OAS3, IFI55, and USP18. (B) Volcano plot of bulk RNA-seq data comparing patients with SLE and healthy volunteers at baseline.
(C) Volcano plot of the 5400 Olink Explore Panel comparing patients with SLE and healthy volunteers at baseline. In the volcano plots, the y-axis represents the negative logarithm (base 10) of the P value, and the x-axis shows the log2 fold change (FC) between 2 groups. Genes and proteins labelled in both volcano plots were differentially modulated by venanprubart at weeks 12 and 24 (Bulk RNA-seq P value <.05, Olink adjusted P value <.05).
IFN, interferon; SLE, systemic lupus erythematosus.Figure 5 dummy alt text
Safety
Among the 85 patients randomised to the study drug, treatment-emergent AEs (TEAEs) were similar across treatment groups (venanprubart: 69.0% [29/42]; placebo: 62.8% [27/43]) (Table 2). The majority of TEAEs were mild (venanprubart: 35.7% [15/42]; placebo: 30.2% [13/43]), then moderate (venanprubart: 31.0% [13/42]; placebo: 27.9% [12/43], and then severe (venanprubart: 2.4% [1/42]; placebo: 4.7% [2/43]). The rate of serious AEs (SAEs) and AEs leading to discontinuation of venanprubart was 9.5% (4/42) and 2.4% (1/42), respectively, whereas it was 2.3% (1/43) and 0% for placebo, respectively.
SAEs in the 4 venanprubart-treated patients were anaemia, joint dislocation, abnormal blood glucose, and SLE flare. No SAE was considered related to the study drug by the investigators (Table 2).Table 2Safety data.Table 2 dummy alt textVariableVenanprubartN = 42n (%)PlaceboN = 43n (%)Subjects ≥1 TEAE29 (69.0)27 (62.8)TEAE29 (69.0)27 (62.8)Mild15 (35.7)13 (30.2)Moderate13 (31.0)12 (27.9)Severe1 (2.4)2 (4.7)Infections and Infestations21 (50.0)17 (39.5)Bronchitis3 (7.1)0Influenza3 (7.1)1 (2.3)Oral herpes3 (7.1)0General disorders and administration site conditions9 (21.4)4 (9.3)Injection-site pain6 (14.3)3 (7.0)Injection-site erythema5 (11.9)1 (2.3)Fatigue3 (7.1)0Injection-site bruising2 (4.8)0Injection-site reaction2 (4.8)0Nervous system disorders7 (16.7)4 (9.3)Headache4 (9.5)1 (2.3)SAE4 (9.5)1 (2.3)Death00Discontinue due to AE1 (2.4)0AE, adverse event; N, number of participants in the treatment group or placebo; n, number of participants within each specified category; SAE, serious adverse event; TEAE, treatment-emergent adverse event.Treatment or placebo was administered every 2 week subcutaneously.
No serious infections and no opportunistic infections were reported during the study. Systemic hypersensitivity was reported in 14.3% (6/42) of patients treated with venanprubart and 4.7% (2/43) of patients treated with placebo. Injection-site reactions (ISRs) were reported in 19.0% (8/42) patients treated with venanprubart and 7.0% (3/43) patients treated with placebo.
No difference was observed in the number of ISRs between TE ADA+ (n = 4) and TE ADA− (n = 4) participants in the venanprubart-treated group. Further details on ADAs are listed in Supplementary Table S5. Changes in vital signs (including blood pressure, temperature, and weight) were not considered clinically meaningful across treatment groups and were not associated with AEs.
No deaths occurred during this study.
DISCUSSION
At the prespecified interim analysis, results of the primary endpoint analysis for Study BT01 showed that venanprubart did not separate from placebo in remission of rash and/or arthritis in participants with at least moderately active SLE at week 24. Study BT01 was therefore terminated early due to futility. The numbers of TEAEs were similar across treatment groups, whereas SAEs were reported in a greater proportion of venanprubart participants compared with placebo.
One patient taking placebo experienced an SAE characterised by the gastrointestinal disorder known as volvulus. The remainder of SAEs occurred in 4 patients treated with venanprubart 450 mg SC every 2 weeks. The SAEs included worsening anaemia due to a flare of SLE, shoulder dislocation due to a fall, elevated blood glucose, and flare of SLE disease activity.
No death occurred during this study, and 1 participant receiving venanprubart discontinued treatment due to an AE of SLE flare that required hospitalisation.
The pharmacodynamic data suggest that lack of efficacy was not related to exposure, target engagement, or ADA. The dose of 450 mg Q2W was selected from phase 1 studies to provide very high target engagement. This dose provided BTLA RO on B cells, CD4+ T cells, and CD8+ T cells of nearly 100% at all timepoints after the initial exposure over the 24 week study duration.
Although approximately half the patients in the venanprubart treatment group had TE ADA, titres were generally low and only 2 of the 19 patients with TE ADA+ showed a notable loss of concentrations over time.
The rationale for targeting the BTLA/HVEM axis in SLE was based on the prominent role of B cells and autoantibodies in SLE, as well as an inverse association between gene expression and disease activity for HVEM [9]. Directly targeting HVEM is challenging due to complex cis and trans interactions in immune cells and broad expression across immune cell types [12]. The current approach targeted BTLA agonism due to specificity of expression, which is mostly concentrated in lymphocytes, and the direct effect of BTLA agonism to suppress immune cell function.
However, BTLA agonism did not improve disease outcomes in this clinical trial.
Low HVEM expression levels on lymphocytes have been linked to disease activity markers in SLE in large phase 3 studies [9]. However, in this study, protein and transcript levels of HVEM did not appear to separate in patients with lupus compared to healthy volunteers (Supplementary Fig S4). Type-I IFN signature genes were elevated in patients with lupus in this study, as expected.
However, the distribution between IFN-I high and IFN-I low patients was not as pronounced as seen in other lupus trials [13]. These differences in expression profiles of both HVEM and IFN-stimulated genes relative to previous observations are potentially due to natural distribution and random sampling of patients selected in this limited phase 2 Proof of Concept trial. As such, the lack of efficacy of venanprubart is likely due to targeting the wrong epitope for BTLA agonism in the patients selected in this study.
One of the limitations of the study is the insensitivity of SLEDAI for detecting partial improvement in any specific disease manifestation, as each item is scored as present or absent. However, the SLEDAI instrument is used to measure efficacy for the SRI-4 registration endpoint, ie, improvement from baseline to primary.
In conclusion, venanprubart was not superior to placebo on the primary endpoint or secondary clinical measures (see summary Supplementary Fig S5). We showed that 450 mg Q2W venanprubart was not effective for treating patients with moderate to severe SLE. The lack of efficacy is unlikely to be due to inadequate exposure or target engagement.
There were immune-related RNA transcripts and proteins that were consistently modulated in the venanprubart treatment group at weeks 12 and 24, indicating some impact on the immune system. However, these molecular changes did not translate to clinical efficacy. Venanprubart was well tolerated, and this study was well controlled with a low placebo response rate (∼25%) compared to other published SLE clinical trials.
Contributors
All authors contributed to the conception and design. IG, JM-B, and C-YC contributed to the acquisition, analysis, and interpretation of data. All authors contributed to the analysis and interpretation of the data and drafting of the manuscript.
All authors contributed to the critical revision of the manuscript for important intellectual content and participated sufficiently in the work to agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All authors had full access to all the data in the study, gave final approval of the version to be published, and had final responsibility for the decision to submit for publication.