Abstract
Background:
Programmed cell death protein 1 (PD-1) is associated with the immunopathology of rheumatoid arthritis (RA). JNJ-67484703 is a humanized IgG anti-PD-1 agonist antibody.
Objectives:
To report results of a phase Ib study evaluating the safety, tolerability, and disease-related outcomes of JNJ-67484703 in adults with active rheumatoid arthritis.
Design:
A randomized, double-blind, placebo-controlled, multiple-dose, phase Ib study in adults with active RA who were inadequately controlled on conventional synthetic disease-modifying antirheumatic drugs.
Methods:
JNJ-67484703 at doses of 2 or 3 mg/kg or placebo was administered by subcutaneous injection for 10 weeks (on weeks 0, 1, 2, 4, 6, 8, and 10). Safety was assessed for the first 6 participants in the 2 mg/kg group to determine if enrollment would proceed to the 3 mg/kg dose. The primary endpoint was the proportion of participants with treatment-emergent adverse events (TEAEs). Secondary endpoints included change from baseline at week 12 in Disease Activity Score 28 using C-reactive protein (DAS28-CRP) and circulating PD-1+ T-cell counts at the 3 mg/kg dose level only. Least squares (LS) mean difference between JNJ-67484703 groups and placebo were based on mixed-effect model for repeated measures for DAS28-CRP.
Results:
A total of 44 participants were randomized to receive JNJ-67484703 2 mg/kg (n = 5), 3 mg/kg (n = 25), or placebo (n = 14); TEAEs were reported in 4 (80%), 17 (68%), and 10 (71%), respectively. None were severe in intensity, and no deaths were reported. TEAEs of infection were reported in 21% of placebo participants and 20% of those in the combined JNJ-67484703 groups. No injection-site or hypersensitivity reactions were reported. At week 12, reduction in DAS28-CRP from baseline was numerically greater in the JNJ-67484703 2 mg/kg and 3 mg/kg groups versus placebo, with a difference in LS means between the 3 mg/kg group and placebo of −0.69 (95% CI −1.55, 0.18; p = 0.117). JNJ-67484703 decreased circulating PD-1+ T cells, with the greatest effect on those with the highest baseline expression of PD-1.
Conclusion:
JNJ-67484703 was well tolerated in participants with active RA and showed evidence for biologic and clinical activity in this small, proof-of-mechanism study.
Trial registration:
ClinicalTrials.gov, NCT04985812.
Plain language summary
This study tested a new drug, JNJ-67484703, in people with rheumatoid arthritis (RA). In RA, the immune system attacks the joints, causing pain and limiting mobility. In people with RA, activity of the programmed cell death protein 1 (PD-1) inhibitory pathway has been found to be reduced compared to patients without RA. JNJ-67484703 is a humanized antibody designed to target and activate PD-1 and restore normal immune function. In this study, JNJ-67484703 or placebo was injected under the skin of adults with RA whose RA did not respond to other treatments. JNJ-67484703 did not cause any severe side effects or injection-site reactions. Infections occurred in the same proportion of people who received JNJ-67484703 as placebo. JNJ-67484703 reduced symptoms of RA over 12 weeks compared with placebo. The results from this study support additional larger clinical studies of JNJ-67484703 to treat RA.
Introduction
Programmed cell death protein 1 (PD-1) is a receptor expressed on a limited variety of immune cells, including activated effector T cells, resting memory T cells, B cells, and thymocytes. When engaged with its ligands (PD-L1 and PD-L2), PD-1 suppresses T-cell function through multiple mechanisms that contribute to T-cell exhaustion and the suppressive tumor microenvironment.1,2 PD-1 also mediates T-cell suppression by promoting the development of regulatory T cells. 2
Therapeutic antibodies antagonizing PD-1 are effective against a variety of cancers by unleashing T cells suppressed by the tumor microenvironment. These PD-1 antagonists can also trigger immune-related adverse events (AEs), including rash, colitis, hypothyroidism, and arthralgia, suggesting a link between PD-1 pathway dysregulation and autoimmunity. 3
High expression of PD-1 (PD-1high) is found on chronically activated effector T cells and follicular helper and peripheral helper CD4+ T cells that assist in antibody production.4,5 These PD-1high T cells are believed to be pathogenic in autoimmune diseases, including rheumatoid arthritis (RA).4,6 In one study, patients with RA showed marked infiltration of PD-1high peripheral helper T cells in synovial tissue samples. 4 In another study, most (>82%) of the CD4+ and CD8+ T cells in the synovium of patients with RA were positive for PD-1; however, most samples (13 of 17) were negative for PD-L1 protein expression. 7 This suggests that PD-1 T-cell inhibitory activity is downregulated because PD-1 ligands are not available to interact with PD-1. Therefore, PD-1 agonist antibodies that restore T-cell inhibition may be effective for treating RA. 7
JNJ-67484703 is a humanized immunoglobulin G1 antibody that binds PD-1 but does not compete with either PD-L1 or PD-L2 for binding to PD-1. JNJ-67484703 targets the small proportion of T cells expressing PD-1 and requires both PD-1 and Fc gamma receptors to function (data not shown). JNJ-67484703 was designed to both agonize PD-1 to inhibit T-cell receptor signaling and deplete PD-1+ cells via antibody-dependent cellular cytotoxicity. JNJ-67484703 was developed as a treatment for systemic immune-mediated disorders, in which T cells, specifically follicular and peripheral helper T cells, play important roles in the initiation and progression of disease. Here, we report results from the first multiple-dose study of JNJ-67484703 in humans, conducted in participants with active RA.
Methods
Study design
This phase I, randomized, double-blind, placebo-controlled, multiple-dose, parallel, multicenter study consisted of a 6-week screening period, followed by 10 weeks of study treatment (weeks 0, 1, 2, 4, 6, 8, and 10) and 14 weeks of follow-up (Figure 1). Participants were randomly assigned to receive JNJ-67484703 or placebo (0.9% normal saline) by subcutaneous injection in a 5:1 ratio for the first 6 participants at each dose level and in a 2:1 ratio overall across the dose level used to assess efficacy.
Study design.
Safety was assessed through Day 29 for the first 6 participants receiving JNJ-67484703 2 mg/kg or placebo to determine whether enrollment would proceed to the 3 mg/kg dose level or a lower dose level. If dosing proceeded to the 3 mg/kg dose, cumulative safety and tolerability were assessed through Day 29 for the first 6 participants at this dose level; if there were no safety concerns, the remaining 30 participants were enrolled. Assessments for injection-site reactions were conducted for at least 30 min after subcutaneous administration of the study drug. During the treatment and follow-up periods, blood samples were collected for clinical laboratory testing and analysis of serum levels of JNJ-67484703 and for antibodies against JNJ-67484703. Flow cytometry for T-cell population analysis was conducted using peripheral blood mononuclear cells isolated from blood samples collected at weeks 0, 1, 2, 3, 4, and 8 during treatment and weeks 12 and 24 during follow-up.
CONSORT guidelines were followed for reporting of this randomized trial (Supplemental Material).
Participants
Eligible participants were 18–65-year-olds, inclusive, and had a body weight within the range of 50–120 kg. Participants had a diagnosis of RA based on American College of Rheumatology (ACR)/European League Against Rheumatism criteria (2010) 8 for ⩾6 months prior to screening. Active RA was defined by counts of ⩾6 swollen and ⩾6 tender joints at screening and Day 1; C-reactive protein (CRP) ⩾0.3 mg/dL at screening; and testing positive for rheumatoid factor or anti-cyclic citrullinated peptide antibody positive (>1 × upper limit of normal) at screening. Participants were required to have an inadequate response to ⩾12 weeks of one of the following conventional synthetic disease-modifying antirheumatic drugs (csDMARD): methotrexate, sulfasalazine, hydroxychloroquine or chloroquine, or leflunomide. Participants were expected to continue ⩾1 csDMARD (excluding methotrexate and leflunomide combinations) for the duration of the study, at a dosing stable for ⩾8 weeks prior to first study drug administration. Although initially prohibited, a protocol amendment during the study allowed prior exposure to up to two biologic DMARDs or targeted synthetic DMARDs.
Participants were excluded from the study based on several criteria to ensure participant safety and the reliability of study outcomes. Individuals were excluded if they exhibited severe, progressive, or uncontrolled disturbances across major organ systems, including hepatic, renal, cardiac, vascular, pulmonary, gastrointestinal, endocrine, neurologic, hematologic, rheumatologic, psychiatric, or metabolic, as identified during screening or reported by the participant. Additionally, participants with known allergies, hypersensitivity, or intolerance to any biological medication or excipients of the study drug were excluded, along with those who had other inflammatory diseases that could confound the assessment of the study drug’s efficacy. Exclusion also applied to individuals with a history of malignancy within 5 years of screening or those with a history or active infection of HIV, hepatitis B virus, Epstein–Barr virus, or cytomegalovirus. Participants who had experienced a serious infection within 3 months prior to the first study intervention were not eligible. Furthermore, those who had received any investigational drug or used an invasive investigational medical device within 30 days or within five half-lives, whichever is longer, before the first dose of the study drug were excluded from participation.
Study assessments
The primary endpoint was the proportion of participants with treatment-emergent adverse events (TEAEs). Safety was also assessed through clinically relevant changes in physical examination findings, vital signs, clinical laboratory tests, and electrocardiograms.
Key secondary efficacy endpoints included the change from baseline in Disease Activity Index Score 28 using CRP (DAS28-CRP) at week 12 and the proportions of participants achieving ACR20, ACR50, and ACR70 responses at week 12, defined as ⩾20%, 50%, or 70% improvement in ACR score from baseline, respectively. 9 Efficacy was also assessed at week 12 by the proportions of participants achieving DAS28-CRP remission (<2.6) and DAS28-CRP low disease activity (⩽3.2). A key exploratory endpoint included the change from baseline in Clinical Disease Activity Index (CDAI) score.
Blood samples were collected to measure serum concentrations of JNJ-67484703 and detect antibodies to JNJ-67484703. The effects of JNJ-67484703 on subsets of PD-1+ and PD-1 negative (neg) T cells were evaluated by flow cytometry. PD-1 expression was determined relative to an isotype control with high density defined as >1000 molecules/cell and low density defined as <1000 molecules/cell. The absolute number of cells/L of blood was determined for each participant at baseline and at weekly intervals following treatment to calculate the median ratio to baseline for each group at each time point.
Statistical analyses
The data generated were descriptive in nature, and no formal statistical safety testing was planned for this study. The safety population consists of all participants who received the study intervention.
To detect a difference of −1 in the change from baseline in DAS28-CRP at week 12 between JNJ-67484703 3 mg/kg and placebo, a sample size of at least 30 participants (2 active: 1 placebo) was estimated to provide ⩾80% power using a 2-sided t-test at a significance level of 20%, based on assumptions of a change from baseline at week 12 in the placebo group of −0.8 and common standard deviation of 1.2. Efficacy analyses were conducted using the full analysis set, including all participants who were randomized and received study treatment, by randomized treatment allocation. The change from baseline in DAS28-CRP was analyzed using descriptive statistics, comparing JNJ-67484703 and placebo with mixed-effect models for repeated measures under the missing-at-random assumption, which was applied to handle any remaining missing data after treatment failure (TF) rules were applied. TF rules considered the following participants to have no change from baseline in efficacy assessments from the time TF occurred onward through week 12: participants who initiated DMARDs or increased the dose of csDMARDs above baseline, initiated or increased their dose of oral corticosteroid therapy, or discontinued study drug administration prematurely prior to a visit. Nominal p values and least squares (LS) means with 95% confidence interval (CI) were provided.
Results
Participants
The study was conducted from October 2021 to May 2023. Of 92 individuals screened, 44 were enrolled and randomized to JNJ-67484703 2 mg/kg (n = 5), JNJ-67484703 3 mg/kg (n = 25), or placebo (n = 14). Among those who failed screening (n = 48), most did not meet eligibility criteria (n = 47, 97.9%); one participant was not enrolled due to the investigator’s decision that a regional crisis could impact the ability to provide adequate treatment and follow-up care. A total of 39 (88.6%) participants completed study treatment through week 10, and 5 (11.4%) discontinued treatment early (two due to AEs, two for personal reasons, and one left the country).
Demographics and baseline characteristics were generally comparable across the JNJ-67484703 and placebo groups (Table 1). The median age of participants was 55.5 years, with a range of 22–65 years. Most participants were female (84.1%), White (97.7%), and not Hispanic/Latino (97.7%). The median body mass index was 27.1 kg/m2, and the median duration of RA was 6 years overall. For the five participants in the JNJ-67484703 2 mg/kg group, the median duration of RA was longer at 12.7 years. Overall, median tender and swollen joint counts were 37.5 and 23.0, respectively, and median DAS28-CRP score was 6.3. Both the patients’ and physicians’ median global assessment of disease activity score was 7.0. All participants were receiving methotrexate, 72.7% were receiving oral corticosteroids, and 40.9% were receiving NSAIDs at stable dosing during the study. No participants had prior exposure to biologic therapy.
Demographics and baseline characteristics.
BMI, body mass index; CCP, cyclic citrullinated peptide; CRP, C-reactive protein; csDMARD, conventional synthetic disease-modifying antirheumatic drug; DAS28-CRP, Disease Activity Index Score 28 using C-reactive protein; HAQ, Health Assessment Questionnaire; NSAID, nonsteroidal anti-inflammatory drug; RA, rheumatoid arthritis; RF, rheumatoid factor; VAS, visual analog scale, y, year.
Safety
The median duration of treatment was 71 days in all treatment groups, and the mean duration of follow-up was 164 days. TEAEs were reported in 4 (80.0%) participants receiving JNJ-67484703 2 mg/kg, 17 (68.0%) participants receiving JNJ-67484703 3 mg/kg, and 10 (71.4%) participants receiving placebo (Table 2). None of the reported TEAEs were considered severe in intensity, and no deaths were reported during the study. Three serious TEAEs were reported in two participants: one participant who received JNJ-67484703 2 mg/kg had a SARS-COV-2 infection followed by pneumonia and one participant who received JNJ-67484703 3 mg/kg developed community-acquired pneumonia. Both participants experiencing serious TEAEs were hospitalized and treated with antibiotics, and the study intervention was discontinued.
Summary of TEAEs through week 24.
Preferred term.
System organ class.
CMV, cytomegalovirus; EBV, Epstein–Barr virus; TEAE, treatment-emergent adverse event; UTI, urinary tract infection.
Infections occurred in 20.0% of participants in each JNJ-67484703 treatment group and in 21.4% of participants in the placebo group. Urinary tract infections (UTIs) were most common, occurring in 10.0% of participants in the combined JNJ-67484703 group and 7.1% in the placebo group. All patients who developed UTIs had a history of recurrent UTIs and no clinical signs of active infection at enrollment. Other frequently reported TEAEs, occurring in more than two participants, were positive tests for Epstein–Barr virus (EBV) and cytomegalovirus (CMV), and anemia. Positive tests for EBV and CMV occurred in 5 (16.7%) and 3 (10.0%) participants in the combined JNJ-67484703 group and in 4 (28.6%) and 2 (14.3%) participants in the placebo group, respectively. The positive viral test results were transient and not accompanied by symptoms suggestive of a viral infection. TEAEs of anemia were reported in 4 (13.3%) participants, all in the JNJ-67484703 3 mg/kg treatment group; none of these TEAEs were considered related to study treatment.
No injection-site reactions or hypersensitivity reactions were reported.
Efficacy
At week 12, a greater reduction in DAS28-CRP from baseline was observed in the JNJ-67484703 2 mg/kg (−1.62 (−2.84, −0.41)) and 3 mg/kg (−1.96 (−2.48, −1.44)) groups compared with the placebo group (−1.27 (−1.96, −0.58)). The LS mean difference (95% CI) between the JNJ-67484703 3 mg/kg and placebo groups was −0.69 (−1.55, 0.18) with a nominal p = 0.117 (Figure 2(a)).
(a) Change from baseline in DAS28-CRP score and (b) proportion of participants achieving ACR20 response by visit through week 12.
At week 12, the proportions of participants achieving ACR20, ACR50, and ACR70 were generally comparable across treatment groups (Table 3). Analysis by visit showed numerically greater ACR20 responses in the JNJ-67484703 3 mg/kg group versus placebo between weeks 4 and 10 (Figure 2(b)).
Efficacy endpoints at week 12.
Nominal p = 0.117 versus placebo.
Nominal p = 0.089 versus placebo.
ACR, American College of Rheumatology; CDAI, Clinical Disease Activity Index; CI, confidence interval; CRP, C-reactive protein; DAS28-CRP, Disease Activity Index Score 28 using C-reactive protein; HAQ-DI, Health Assessment Questionnaire Disability Index; LS, least squares; SD, standard deviation.
No differences were observed in DAS28-CRP remission rates across treatment groups at week 12 (Table 3). More participants in the JNJ-67484703 3 mg/kg group (28.0%) achieved DAS28-CRP low disease activity compared with those receiving placebo (14.3%) at week 12. At week 12, participants in the JNJ-67484703 3 mg/kg group had a numerically greater reduction in CDAI score from baseline (−23.83 (−29.99, −17.66)) compared with those in the placebo group (−14.96 (−23.13, −6.79)) with a nominal p = 0.089.
Numerical improvements were observed for the JNJ-67484703 3 mg/kg group versus the placebo group for ACR components of tender joint count, swollen joint count, Physician’s Global Assessment of Disease Activity, and CRP at week 12 (Table 3). Patient’s Global Assessment of Disease Activity, Patient’s Global Assessment of Pain, and Health Assessment Questionnaire Disability Index were comparable between the JNJ-67484703 groups and the placebo group at week 12 (Table 3).
Pharmacokinetics and immunogenicity
Steady state was achieved at approximately week 4 (data not shown). Treatment-emergent antibodies to JNJ-67484703 were detected in 2 of 30 (6.7%) participants through week 24. Due to the low number of participants who were positive for treatment-emergent antibodies to JNJ-67484703, the effect on pharmacokinetics could not be determined.
Pharmacodynamics
CD4+ PD-1high T cells showed a sustained decrease in both the JNJ-67484703 2 mg/kg and 3 mg/kg dose groups with a median ratio to baseline at week 12 of 0.12 and 0.33, respectively (Figure 3). The median ratio to baseline at week 12 in the placebo group was 1.40. Similar results were observed for PD-1high follicular and peripheral T cells, with median ratios at week 12 of 0.28 for the JNJ-67484703 2 mg/kg group, 0.57 for the 3 mg/kg group, and 1.59 for the placebo group. CD4+ low expressing PD-1 (PD-1low) T cells decreased from baseline but to a lesser extent than the PD-1high T cells, with median ratios to baseline at week 12 of 0.30, 0.68, and 1.30, respectively. In contrast, CD4+ PD-1neg T cells did not show a sustained decrease from baseline, with median ratios at week 12 of 1.19, 0.94, and 1.19, respectively.
Time profile for median ratio to baseline of (a) PD-1high, (b) PD-1low, and (c) PD-1neg CD4+ T cells by treatment group.
Discussion
In this phase Ib safety and tolerability study, administration of JNJ-67484703 at two dose levels in participants with active RA was safe and well tolerated. All TEAEs were mild to moderate in severity and occurred at similar frequencies in the JNJ-67484703 dose groups and the placebo group. Two of 30 (6.7%) participants developed antibodies to JNJ-67484703; the analysis of antibodies to JNJ-67484703 on pharmacokinetics was not feasible due to the small number of participants with treatment-emergent antibodies.
JNJ-67484703 demonstrated biological activity in adults with RA, inducing a decrease in circulating PD-1+ T cells. The extent of decrease was dependent on the cellular surface density of PD-1, as the median ratio to baseline observed at week 12 was lower for PD-1high T cells than for PD-1low T cells, and there was no sustained decrease in PD-1neg T cells. The extent of depletion appeared similar between the 2 and 3 mg/kg doses, but the results must be interpreted with caution due to the limited number of samples for analysis in the 2 mg/kg dose group.
Clinical efficacy measures were explored in this phase Ib trial, which was designed to assess for a potential efficacy signal similar to that seen with other biologic therapies. The primary comparison was between the 3 mg/kg JNJ-67484703 treatment (n = 25) and placebo (n = 14) groups. The JNJ-67484703 2 mg/kg treatment group (n = 6; five active and one placebo) was included only for safety evaluation to facilitate dose escalation. Although the efficacy endpoints for the 2 mg/kg group are reported in this manuscript, this cohort did not have sufficient sample size to draw any conclusions regarding a treatment dose response.
Evidence for a clinical efficacy signal with JNJ-67484703 3 mg/kg treatment was supported by several endpoints. Compared to placebo, participants treated with JNJ-67484703 3 mg/kg had numerically greater reductions from baseline in DAS28-CRP and CDAI at week 12 (nominal p = 0.117 and p = 0.089, respectively). Although the proportions of participants who achieved ACR20 response at week 12 were comparable between the JNJ-67484703 3 mg/kg and placebo groups (approximately 60%), a large increase in the placebo response rate was unexpectedly observed between weeks 10 and 12 (35.7% vs 57.1%), confounding interpretation of this endpoint. In contrast, improvements in the ACR20 response were primarily observed between weeks 4 and 10 after initiating JNJ-67484703 treatment. In addition, numeric improvements in several ACR component scores (tender joint count, swollen joint count, CRP, and physician global assessment) were observed at week 12, and a higher proportion of participants achieved DAS28-CRP low disease activity with JNJ-67484703 treatment. No differences were observed in ACR50, ACR70, DAS-28 CRP remission, or HAQ-DI responses between the JNJ-67484703 3 mg/kg and placebo treatment groups. However, this study was underpowered to detect differences in these endpoints, based on the smaller effect sizes commonly observed for these endpoints in RA trials.
Conclusion
JNJ-67484703 was well tolerated in participants with active RA and showed evidence for biologic and clinical activity in this small, proof-of-mechanism study. Study limitations include a small sample size and testing in a relatively homogeneous advanced therapy-naïve, seropositive population. Despite these limitations, the results of this phase Ib study support further investigation of the efficacy and safety of JNJ-67484703 in a larger trial of individuals with RA.
Supplemental Material
sj-doc-1-tab-10.1177_1759720X251385857 – Supplemental material for Safety, tolerability, and efficacy of the PD-1 agonist antibody JNJ-67484703 in adults with active rheumatoid arthritis: results of a multicenter, double-blind, placebo-controlled, randomized, multiple-dose phase Ib study
Supplemental material, sj-doc-1-tab-10.1177_1759720X251385857 for Safety, tolerability, and efficacy of the PD-1 agonist antibody JNJ-67484703 in adults with active rheumatoid arthritis: results of a multicenter, double-blind, placebo-controlled, randomized, multiple-dose phase Ib study by Irving T. C. Ling, Stanley J. Marciniak, Stephen H. Clarke, Vani Lakshminarayanan, Matthew J. Loza, Sophia G. Liva, Tong Wang, Ashley Orillion, Ilia Tikhonov and Erika H. Noss in Therapeutic Advances in Musculoskeletal Disease
Footnotes
Acknowledgements
We would like to thank all the investigators who were part of the study and contributed to recruiting participants into the trial. We would also like to thank Erin Doyle, PhD, of J&J Immunology Translational Science for support with sample evaluation and data analysis. This study was funded by Johnson & Johnson. Medical writing assistance was provided by Michelle McDermott, on behalf of Lumanity Communications Inc., and was funded by Johnson & Johnson.
Declarations
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
