Phase Ib, multicentre, open-label, dose-escalation study to evaluate the safety, tolerability, pharmacokinetics and pharmacodynamics of subcutaneously administered mosunetuzumab in participants with systemic lupus erythematosus Authors: Elsa Martins, Vishala L Chindalore, X Rebecca Sheng, Aaron V Schroeder, Vivian Chen, Hardik Mody, Audrey Arjomandi, Huiyan (Ashley) Mao, Imran Hassan, Iwona Dankiewicz-Fares, Svetlana Agachi, Theodore A Omachi, Jay P Garg, William F Pendergraft Published: 2026 ABSTRACT AbstractObjectiveTo determine the safety, tolerability, pharmacokinetics (PK) and pharmacodynamics (PD) of subcutaneously administered mosunetuzumab, a CD20xCD3 T-cell-engaging bispecific antibody, in participants with systemic lupus erythematosus (SLE).MethodsThis phase Ib, multicentre, open-label, dose-escalation study enrolled 15 participants diagnosed with active autoantibody-positive SLE, demonstrated by an SLE Disease Activity Index 2000 (SLEDAI-2K) score ≥4 at screening. The study included two fixed-dosing dose-finding cohorts (1.6 mg or 5 mg of mosunetuzumab on day 1) and three step-up dose-escalation cohorts (day 1 dose of mosunetuzumab established by dose finding, followed by 15 mg, 45 mg or 60 mg of mosunetuzumab on day 8). The primary objective was evaluation of safety; other key objectives were evaluating the PK and PD of mosunetuzumab. Peripheral B-cell counts and T-cell phenotyping were performed using flow cytometry.ResultsFifteen participants were enrolled from January 2022 through December 2023. Two participants discontinued treatment due to treatment-emergent adverse events (AEs). No immune effector cell-associated neurotoxicity syndrome events or high-grade cytokine release syndrome events were reported. The frequency of grade ≥3 AEs and serious AEs was 13.3% and 6.7%, respectively. Mosunetuzumab depleted peripheral B cells below the lower limit of quantification (0.441 cells/µL) in a dose-dependent manner. CD4+ and CD8+ T-cell activation occurred across all tested doses. SLEDAI-2K scores improved, especially among participants with higher baseline SLEDAI-2K scores and those receiving higher doses. The PK profile of mosunetuzumab was consistent with that observed in the relapsed or refractory follicular lymphoma population.ConclusionsMosunetuzumab demonstrated dose-dependent peripheral B-cell depletion and a safety profile consistent with prior studies in follicular lymphoma. Based on these data, mosunetuzumab is currently being evaluated in a phase II study in severe SLE (NCT07598396).Trial registration numberNCT05155345. WHAT IS ALREADY KNOWN ON THIS TOPIC Monoclonal B-cell-targeting antibodies have been investigated as therapies for systemic lupus erythematosus (SLE) and other B-cell-driven autoimmune diseases. Monoclonal B-cell-targeting antibodies have been investigated as therapies for systemic lupus erythematosus (SLE) and other B-cell-driven autoimmune diseases. T-cell-engaging bispecific antibodies have the potential to achieve greater tissue B-cell depletion and potentially improve clinical efficacy. T-cell-engaging bispecific antibodies have the potential to achieve greater tissue B-cell depletion and potentially improve clinical efficacy. WHAT THIS STUDY ADDS This is the first report of clinical study results of a CD20xCD3 bispecific antibody in SLE. This is the first report of clinical study results of a CD20xCD3 bispecific antibody in SLE. This phase Ib dose-escalation study showed that mosunetuzumab, a bispecific B-cell-depleting antibody, demonstrated dose-dependent B-cell depletion with a safety profile consistent with prior studies in follicular lymphoma. This phase Ib dose-escalation study showed that mosunetuzumab, a bispecific B-cell-depleting antibody, demonstrated dose-dependent B-cell depletion with a safety profile consistent with prior studies in follicular lymphoma. HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY These data, together with the safety and efficacy data accumulated in follicular lymphoma studies, support additional, larger clinical studies to assess the safety and efficacy of mosunetuzumab in patients with moderate-to-severe SLE. These data, together with the safety and efficacy data accumulated in follicular lymphoma studies, support additional, larger clinical studies to assess the safety and efficacy of mosunetuzumab in patients with moderate-to-severe SLE. Introduction Systemic lupus erythematosus (SLE) is a chronic, debilitating autoimmune rheumatic disease predominantly affecting women of childbearing age, characterised by diverse and often severe multisystem involvement.1 2 Despite current standard-of-care treatments, including glucocorticoids, antimalarials and immunosuppressants, many patients continue to experience disease flares and progressive organ damage.1 2 This results in a considerably increased mortality risk.36 The heterogeneous nature of SLE further complicates diagnosis and management, and despite advances, the proportion of patients achieving complete clinical response or sustained remission remains inadequate under existing treatment options. As such, there remains an urgent need for more effective and safer therapeutic options, along with improved tools for disease monitoring and established biomarkers to support disease management. This unmet clinical need is rooted in the complex underlying pathophysiology of the disease. SLE presents with exacerbated autoantibody production and immune complex deposition, ultimately leading to tissue inflammation and damage. Cytokines also play a key role in the disease aetiology by regulating cellular differentiation, maturation and activation as well as participating in the inflammatory processes that result in tissue injury. However, targeting cytokines has proven helpful in the treatment of inflammatory joint, bowel and skin diseases, but has largely failed in SLE, despite the established cytokine dysregulation in this condition.7 B cells are central to SLE pathogenesis, acting as autoantigen presenters, autoantibody producers, cytokine secretors and T-cell co-stimulators.8 This makes them an attractive therapeutic target; however, early attempts to target B cells in SLE, such as trials with the anti-CD20 monoclonal antibody rituximab, largely failed, potentially due to B-cell dysregulation and resistance to depletion in patients with SLE.9 10 Obinutuzumab, a type 2 anti-CD20 antibody, has recently shown improved response rates both in SLE and lupus nephritis (LN).1113 Nevertheless, there remains room for improved clinical outcomes, and it currently remains unknown whether a more profound B-cell depletion might be required to achieve superior efficacy. Chimeric antigen receptor (CAR) T-cell therapies, which involve engineering T cells to specifically recognise and eliminate B cells, have shown unprecedented efficacy in B-cell malignancies and have been approved for B-cell lymphomas, leukaemias and multiple myeloma.1416 This success highlighted their potential for the treatment of B-cell-driven non-malignant conditions. Recently, use of CAR T-cell therapy in SLE has demonstrated B-cell depletion in blood and in secondary lymphoid tissues resulting in clinical efficacy, as well as potential elimination of autoreactive B cells (an ‘immune reset’) and durable clinical remission.1720 However, cellular therapies currently require intensive lymphodepleting chemotherapy and specialised inpatient monitoring for severe toxicities such as cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS).21 Management is further complicated by the high cost, as well as a prolonged manufacturing process in the case of autologous T cells.22 Bispecific antibodies have also proven to be effective in the depletion of malignant B cells in oncology settings, including relapsed/refractory (R/R) diffuse large B-cell lymphoma and follicular lymphoma.23 These are now being explored for autoimmune conditions like SLE, offering the potential for a more accessible, ‘off-the-shelf’ treatment option that might represent a more scalable approach for a broader SLE patient population. Both CAR T-cell therapy and bispecific antibodies offer the potential for more potent B-cell depletion compared with monoclonal antibodies; however, their safety and efficacy in SLE require further evaluation.14 22 23 Mosunetuzumab is a humanised IgG1 CD20xCD3 bispecific antibody currently approved for the treatment of R/R follicular lymphoma after two or more prior therapies.2427 It functions by simultaneously binding CD3 on T cells and CD20 on B cells, creating an immunological synapse that triggers potent T-cell activation and targeted B-cell lysis.28 The unique design of mosunetuzumab, including an N297G amino acid substitution in the Fc region, minimises Fc-dependent effector functions, ensuring its action is primarily T-cell mediated. This T-cell-engaging mechanism is hypothesised to enable more extensive killing of tissue-infiltrating B cells and efficient elimination of cells with low CD20 expression. By targeting depletion of B cells within lymphoid niches, mosunetuzumab could achieve a profound ‘reset’ of the B-cell compartment, disrupt the cytokine milieu that maintains chronic inflammation and address the fundamental immunological dysregulation and cytokine imbalances that drive SLE pathogenesis more effectively than currently available therapies. This phase Ib, open-label, dose-escalation study (NCT05155345) evaluated the safety, tolerability, pharmacokinetics (PK) and pharmacodynamics (PD) of subcutaneously administered mosunetuzumab in adults with active SLE. Methods Study objectives The primary objective of this phase Ib study was to evaluate the safety of subcutaneously administered mosunetuzumab in participants with SLE. This included evaluating the tolerability of the dosing schedule(s) and dose(s) and characterising dose-limiting toxicities (DLTs) on the basis of (i) incidence and severity of adverse events (AEs), including DLTs; (ii) change from baseline in targeted vital signs; and (iii) change from baseline in targeted clinical laboratory test results. The PK objective was to characterise the PK profile of step-up and fixed dosing of mosunetuzumab based on serum drug concentrations at specified timepoints and estimation of the PK parameters of mosunetuzumab. The PD objective was to characterise the PD effects of mosunetuzumab based on peripheral CD19+ B-cell counts at specified timepoints and the duration of B-cell depletion. The immunogenicity objective was to evaluate the immune response to mosunetuzumab based on the prevalence of antidrug antibodies (ADAs) at baseline and the incidence of ADAs during the study. Exploratory objectives further assessed the clinical response of mosunetuzumab in participants with SLE based on Patient Global Impression of Severity (PGI-S), Physician Global Assessment (PGA), change from baseline in anti-double-stranded DNA (dsDNA) antibodies and change from baseline in complement C3 and C4. Study design and oversight This was a multicentre, open-label, dose-escalation phase Ib study. Participants were enrolled at three sites in three countries (Republic of Moldova, Poland and the USA). An internal monitoring committee reviewed safety, PK and PD data throughout the trial. The study consisted of the following periods: screening, dose finding (comprising two fixed-dosing single-dose cohorts), dose escalation (comprising three step-up dosing cohorts) and safety follow-up periods (figure 1). The two fixed-dosing dose-finding cohorts of the study comprised one participant allocated to receive 1.6 mg mosunetuzumab on day 1 and at least three participants who would receive 5 mg mosunetuzumab on day 1 (figure 1). Patient dosing was staggered by at least 72 hours and participants were required to be hospitalised for 72 hours for acute toxicity monitoring. In the three step-up dose-escalation cohorts, the day 1 dose for all participants (at least three participants per cohort) was 5 mg of mosunetuzumab. The day 8 dose was either 15 mg, 45 mg or 60 mg (figure 1). Day 1 doses were staggered by at least 72 hours and participants were required to be hospitalised for 72 hours after administration on both day 1 and day 8. Study design. Study design schematic for the phase Ib evaluation of mosunetuzumab in SLE. The study uses a sequential cohort design (total n=15) divided into two phases. (Left) The dose-finding phase employs fixed-dosing cohorts receiving a single infusion on day 1 (1.6 mg or 5 mg), with a 21-day DLT assessment period. (Right) The dose-escalation phase employs ‘step-up’ dosing cohorts receiving infusions on day 1 and day 8 to mitigate toxicity. In these cohorts, the day 1 dose is determined by preceding dose-finding data, while the day 8 dose represents the target escalation level (15 mg, 45 mg or 60 mg); the DLT assessment period extends to 28 days. Safety protocols include a mandatory 72-hour hospitalisation post-infusion and a 72-hour staggering interval between participants during escalation. *Day 1 dose determined during the dose-finding period. BTK, Bruton’s tyrosine kinase; DLT, dose-limiting toxicity; JAK, Janus kinase; PD, pharmacodynamics; PK, pharmacokinetics; SLEDAI-2K, SLE Disease Activity Index 2000; TYK2, tyrosine kinase 2; ULN, upper limit of normal. The minimum mosunetuzumab dose to be tested (1.6 mg) and the maximum step-up dose of 5+60 mg were previously determined based on dosing in R/R follicular lymphoma trials.27 Participants had to permanently discontinue study treatment for the following reasons: any medical condition considered by the investigator as jeopardising the participant’s safety; investigator or sponsor decision; pregnancy; clinical indication requiring treatment with prohibited therapy; significant worsening of SLE; DLTs or unacceptable toxicity or a dose delay of >3 days. The safety follow-up had a minimum duration of 12 months after the last administration of mosunetuzumab, depending on the participant’s time to B-cell repopulation. DLTs were assessed during a 21-day period for the fixed-dosing cohorts and a 28-day period (21 days after day 8) for the step-up dosing cohorts. DLTs included grade ≥3 CRS, any grade ≥3 AE (unless unrelated to study drug), Hy’s Law29 criteria for liver injury and significant worsening of SLE. Following the dose-escalation phase, additional participants could be enrolled in dose-expansion cohorts to gather further safety, tolerability, PK and PD data for selected doses (the dose-expansion period). Participant population and eligibility criteria Enrolled participants were aged 18–75 years with a diagnosis of SLE, according to the 2019 European Alliance of Associations for Rheumatology (EULAR)/American College of Rheumatology classification criteria, at least 12 weeks prior to screening. Enrolment required the presence of one or more SLE autoantibodies (positive anti-nuclear antibodies ≥1:160, anti-dsDNA above upper limit of normal (ULN) or anti-Smith above ULN) documented within 12 months prior to or during screening; active SLE disease, defined by an SLE Disease Activity Index 2000 (SLEDAI-2K) total score of ≥4 at screening and receiving stable doses of one or more standard SLE therapies (details in online supplemental methods). Other key inclusion and exclusion criteria are listed in the online supplemental methods. Study treatment and administration Premedications (dexamethasone 10 mg predose, 24-hour±4-hour post-mosunetuzumab injection and 48-hour±4-hour post-mosunetuzumab injection; acetaminophen 500–1000 mg predose; diphenhydramine 50–100 mg predose) were required for each mosunetuzumab injection to mitigate the risk of CRS. Tocilizumab was available for the management of CRS. Assessments Safety assessments included AEs, serious AEs and AEs of special interest (including CRS, ICANS, drug-induced liver toxicity or transmission of an infectious agent by the study drug), vital signs and electrocardiograms. CRS severity was determined according to the 2019 American Society for Transplantation and Cellular Therapy CRS Consensus Grading Criteria.30 Clinical and laboratory assessments, including SLEDAI-2K scores, were conducted at study visits. dsDNA was measured using an IMT1851 anti-dsDNA, immunoglobulin G INOVA QUANTA Lite subcutaneous ELISA, run at LabCorp. Flow cytometry Peripheral whole blood was collected for conventional flow cytometry analysis of B-cell, T-cell and NK-cell markers at (i) days 1 (predose, 4 hours postdose), 2, 8 (predose; 4 hours postdose), 9, 15 and 29, and months 3, 6, 9 and 12 and every 6 months thereafter for step-up dosing cohorts; (ii) days 1 (predose; 4 hours postdose), 2, 8 and 29 and study months 3, 6, 9 and 12 and every 6 months thereafter for fixed-dosing cohorts. Peripheral CD19+ B-cell counts were analysed using a high-sensitivity flow cytometry assay (lower limit of quantification: 0.441 cells/µL)31 at LabCorp (online supplemental table 5), with samples collected at days 1 (predose), 8 (predose), 15 and 29 and months 2, 3, 4, 5, 6, 9 and 12 for all cohorts. T-cell markers were measured with a validated custom panel at ICON (online supplemental methods and online supplemental table 5), with samples collected at (i) days 1 (predose, 4 hours/10 hours postdose), 2, 8 (predose, 4 hours/10 hours postdose), 9 and 15 and month 3 for step-up dosing cohorts; (ii) days 1 (predose, 4 hours/10 hours postdose), 2, 8 and 15 and month 3 for fixed-dosing cohorts. T-cell activation was measured in both CD4+ and CD8+ T-cell populations using the early activation marker CD69 surface expression. Data were analysed as a percentage of the parent population over time. PK analysis The concentration of mosunetuzumab in human serum was measured using a validated ELISA (online supplemental methods). The PK analysis was carried out using non-compartmental methods with Phoenix WinNonlin (V.8.3) software to estimate the mosunetuzumab PK parameters. Immunogenicity analysis The presence of ADAs to mosunetuzumab in human serum was detected using a validated ELISA-based bridging assay (online supplemental methods). Statistical analyses Approximately 13–25 participants were planned based on the dose-escalation rules. All participants who received any amount of study drug were included in the safety-evaluable set for primary safety objectives. Safety data were summarised descriptively by dose level and overall in all participants. Participants were grouped according to the dose level they were originally assigned. PD objectives were evaluated in the same safety-evaluable set. Exploratory biomarker and efficacy data were summarised descriptively by dose level, where participants were grouped based on the dose they received. Changes from baseline at months 3 and 6 were presented for PGA, SLEDAI-2K and PGI-S. There was no statistical testing performed for any of the end points. Patient and public involvement The study was validated with patient experts from the Roche Global Lupus Patient Council: Amy Somers (Lupus EU), Elfriede Wijisma (Lupus EU), Jeanette Andersen (Lupus EU), Sibongile Komati (Lupus Foundation of South Africa), Huiji Ling (SLE Patients Club China) and June Alikhan (Lupus Ontario). Results A total of 30 participants were screened, 15 of whom were enrolled in the study from January 2022 through December 2023. Participants were assigned to treatment cohorts in order of enrolment. All 15 participants received their initial dose in their respective cohort, but two participants (13.3%), one in the 5+15 mg cohort and one in the 5+45 mg cohort, discontinued treatment due to a treatment emergent AE prior to receiving their day 8 injection, so data for these two participants were included in the 5 mg dose cohort instead. One additional participant was recruited into the 5+15 mg and 5+45 mg cohort each. Therefore, the number of participants in each dose cohort was as follows: 1.6 mg (n=1), 5 mg (n=5), 5+15 mg (n=3), 5+45 mg (n=3) and 5+60 mg (n=3). No participants withdrew from the study. Baseline characteristics Baseline characteristics for the 15 participants are shown in table 1. The median age was 51.0 years (range 23–68), and the majority were female (14 participants, 93.3%) and white (13 participants, 86.7%). The mean baseline SLEDAI-2K score was 7.47 and the majority of participants had negative anti-dsDNA status at baseline (nine participants, 64.3%). Table 1Demographics and baseline characteristics of the study populationDose cohortAll(n=15)1.6 mg(n=1)5 mg(n=3)5+15 mg(n=4)5+45 mg(n=4)5+60 mg(n=3)Age—median (range), years51.0 (23–68)68.0 (68)55.0 (46–67)47.5 (34–67)53.0 (39–57)24.0 (23–29)Sex—n (%) Female14 (93.3)1 (100)3 (100)4 (100)3 (75.0)3 (100) Male1 (6.7)0001 (25.0)0Race—n (%) Black/African American2 (13.3)1 (100)01 (25.0)00 White13 (86.7)03 (100)3 (75)4 (100)3 (100)Ethnicity—n (%) Not Hispanic or Latino14 (93.3)NA*3 (100)4 (100)4 (100)3 (100)Geographical region—n (%) Europe6 (40.0)001 (25.0)2 (50.0)3 (100) North America9 (60.0)1 (100)3 (100)3 (75.0)2 (50.0)0Total SLEDAI-2K score—mean (SD)7.47 (2.97)6.00 (NE)5.33 (2.31)7.50 (3.00)8.00 (3.65)9.33 (3.06)Anti-dsDNA antibody status†—n (%) Positive3 (21.4)001 (33.3)1 (25.0)1 (33.3) Negative9 (64.3)1 (100)3 (100)2 (66.7)2 (50.0)1 (33.3) Borderline2 (14.3)0001 (25.0)1 (33.3)C3 complement level‡—mean (SD)1.10 (0.36)1.51 (NE)1.33 (0.32)1.15 (0.49)0.87 (0.33)0.98 (0.09)Low C3 levels (<90 mg/dL at baseline)—n (%)5 (33)–––––C4 complement level‡—mean (SD)0.18 (0.13)0.16 (NE)0.21 (0.08)0.21 (0.24)0.15 (0.07)0.15 (0.06)Low C4 levels (<10 mg/dL at baseline)—n (%)5 (33)–––––Disease duration (years)—median (IQR)5 (2–15) (n=11)–––––Patients were grouped according to the dose level to which they were originally assigned.*The ethnicity of this participant was unknown.†The following cutoffs were used for anti-dsDNA antibody categories: negative: <30 IU/mL; borderline: 30–75 IU/mL; positive: >75 IU/mL.‡Central laboratory results.dsDNA, double-stranded DNA; NA, not available; NE, not estimable; SLEDAI-2K, SLE Disease Activity Index 2000. Demographics and baseline characteristics of the study population Patients were grouped according to the dose level to which they were originally assigned. The ethnicity of this participant was unknown. The following cutoffs were used for anti-dsDNA antibody categories: negative: <30 IU/mL; borderline: 30–75 IU/mL; positive: >75 IU/mL. Central laboratory results. dsDNA, double-stranded DNA; NA, not available; NE, not estimable; SLEDAI-2K, SLE Disease Activity Index 2000. The medical history was representative of the SLE population. The number of participants previously treated with SLE medications was as follows: antimalarials (hydroxychloroquine sulfate or chloroquine phosphate; n=12, 80%), mycophenolate mofetil (MMF) (n=3, 20%), methotrexate (n=4, 26.7%) and corticosteroids (n=10, 66.7%). Safety and tolerability All 15 participants (100%) in the safety-evaluable population experienced at least one AE (online supplemental table 1). A total of 40 AEs, reported in 13 participants (86.7%), were considered related to mosunetuzumab treatment. The most frequent AEs by preferred term (≥2 participants) are shown in table 2; these were mainly grade 1–2. Two participants (13.3%) experienced grade 3–5 AEs: one participant in the 5+60 mg cohort had a grade 3 non-serious neutropenia, which was considered related to study treatment and resolved, and one participant in the 5+45 mg cohort had a fatal event due to bilateral bacterial pneumonia. This death occurred at study day 66, which was 37 days after the end of the DLT assessment period. The patient experienced peripheral B-cell depletion to below the limit of quantification (BLOQ) during the study but had detectable levels, although low, at the last study visit (study day 60) prior to this event. The patient’s absolute neutrophil count at baseline was 6.35×109/L and remained within the normal range for the duration of the study, with the last result prior to the event being 2.11×109/L. The study investigator assessed the event as unrelated to the study drug. No DLTs were reported in this study. Table 2AEs in the safety-evaluable population: most common AEs* and CRS or ICANSAll patients(n=15)1.6 mg(n=1)5 mg(n=3)5+15 mg(n=4)5+45 mg(n=4)5+60 mg(n=3)Most common AEs, n (%) Injection-site reaction, any grade12 (80.0)1 (100)2 (66.7)4 (100)2 (50.0)3 (100) Grade 111 (73.3)02 (66.7)4 (100)2 (50.0)3 (100) Grade 21 (6.7)1 (100)0000 Infections and infestations†, any grade11 (73.3)03 (100)3 (75.0)3 (75.0)2 (66.7) Grade 12 (13.3)01 (33.3)001 (33.3) Grade 29 (60.0)02 (66.7)3 (75.0)3 (75.0)1 (33.3) Headache, any grade5 (33.3)1 (100)02 (50.0)2 (50.0)0 Grade 14 (26.7)1 (100)02 (50.0)1 (25.0)0 Grade 21 (6.7)0001 (25.0)0 Fever, any grade3 (20.0)001 (25.0)1 (25.0)1 (33.3) Grade 13 (20.0)001 (25.0)1 (25.0)1 (33.3) Rash‡, any grade3 (20.0)01 (33.3)1 (25.0)1 (25.0)0 Grade 12 (13.3)001 (25.0)1 (25.0)0 Grade 21 (6.7)01 (33.3)000 Neutropenia, any grade1 (6.7)0001 (25.0)0 Grade 21 (6.7)0001 (25.0)0 Neutrophil count decreased, any grade1 (6.7)00001 (33.3) Grade 31 (6.7)00001 (33.3) Diarrhoea, any grade2 (13.3)00002 (66.7) Grade 12 (13.3)00002 (66.7) Dyspepsia, any grade2 (13.3)001 (25.0)1 (25.0)0 Grade 11 (6.7)001 (25.0)00 Grade 21 (6.7)0001 (25.0)0 Foot fracture, any grade2 (13.3)01 (33.3)1 (25.0)00 Grade 22 (13.3)01 (33.3)1 (25.0)00 Pain, any grade2 (13.3)001 (25.0)1 (25.0)0 Grade 22 (13.3)001 (25.0)1 (25.0)0 Interstitial lung disease, any grade1 (6.7)0001 (25.0)0 Grade 51 (6.7)0001 (25.0)0CRS, n (%)4 (26.7)01 (33.3)1 (25.0)1 (25.0)1 (33.3) CRS associated with day 1 injection, any grade4 (26.7)01 (33.3)1 (25.0)1 (25.0)1 (33.3) Grade 13 (20.0)01 (33.3)01 (25.0)1 (33.3) Grade 21 (6.7)001 (25.0)00 CRS associated with day 8 injection, any grade1 (11.1)0001 (33.3)0 Grade 1000000 Grade 21 (11.1)0001 (33.3)0ICANS, any grade, n (%)000000The COVID-19 Preferred Term was ‘COVID-19’ only. There were no other events of COVID-19 pneumonia, COVID-19 test positive or other COVID-19-related events.*Most common AEs are defined as AEs occurring in two or more participants; this excludes incidence of CRS.†Most common infections include COVID-19 (three participants; 20%), urinary tract infections (three participants; 20%), nasopharyngitis (three participants; 20%) and upper respiratory tract infections (two participants; 13.3%).‡Rash terms include ‘Rash’ and ‘Dermatitis contact’.AE, adverse event; CRS, cytokine release syndrome; ICANS, immune effector cell-associated neurotoxicity syndrome. AEs in the safety-evaluable population: most common AEs* and CRS or ICANS The COVID-19 Preferred Term was ‘COVID-19’ only. There were no other events of COVID-19 pneumonia, COVID-19 test positive or other COVID-19-related events. Most common AEs are defined as AEs occurring in two or more participants; this excludes incidence of CRS. Most common infections include COVID-19 (three participants; 20%), urinary tract infections (three participants; 20%), nasopharyngitis (three participants; 20%) and upper respiratory tract infections (two participants; 13.3%). Rash terms include ‘Rash’ and ‘Dermatitis contact’. AE, adverse event; CRS, cytokine release syndrome; ICANS, immune effector cell-associated neurotoxicity syndrome. Two study participants, allocated to receive a step-up dose of mosunetuzumab on days 1 and 8, experienced an AE leading to discontinuation of the study treatment after receiving 5 mg of mosunetuzumab on day 1; as such, dose administration on day 8 did not occur. Both AEs were considered related to the study treatment (grade 2 sinus bradycardia and grade 2 CRS). CRS and ICANS: Four study participants (26.7%) experienced a total of five CRS events (table 2); all were grade 1 (n=3) or grade 2 (n=2) and resolved. These four participants experienced CRS on day 1 after a 5 mg injection of mosunetuzumab (grade 1, n=3; grade 2, n=1). One participant in the 5+45 mg cohort experienced an additional CRS event after injection of 45 mg on day 8 (CRS grade 1 following day 1 injection and CRS grade 2 following day 8 injection). These events were managed with supportive care; no participant required tocilizumab. No ICANS events were observed in this study. Pharmacokinetics Mosunetuzumab serum concentration reached the maximum level around 3–8 days postdose and declined in a bi-exponential fashion (figure 2). Mosunetuzumab concentrations generally increased as the dose increased across the dose range tested. Largely overlapping exposures were observed for the 5+45 mg and 5+60 mg step-up doses. The mean t1/2 of mosunetuzumab after subcutaneous administration across the dose levels ranged from approximately 13 to 17 days. PK profile of subcutaneous mosunetuzumab: mean (SD) concentration-time profile for (A) fixed-dosing or single-dose cohorts and (B) step-up dose cohorts with doses on day 1 and day 8. Patients are grouped as per actual dose received, regardless of their original cohort assignment. One patient assigned to the 5+15 mg cohort and one patient assigned to the 5+45 mg cohort received only the day 1 injection (5 mg) and did not receive their respective day 8 injections. The horizontal line represents the LLOQ of 0.01 µg/mL. Results below LLOQ were imputed as LLOQ/2 for the mean calculation. LLOQ, lower limit of quantification; PK, pharmacokinetic. Pharmacodynamics Dose-dependent peripheral CD19+ B-cell depletion was observed (figure 3). Higher doses were associated with a greater proportion of participants exhibiting CD19+ B-cell depletion to BLOQ (0.441 cells/µL) for longer durations. CD19+ B cells in all nine participants from the 5+15 mg, 5+45 mg or 5+60 mg dose cohorts were depleted to BLOQ by approximately day 15 and remained either low or BLOQ until around the 6-month timepoint. Of note, all study participants in the 5+60 mg cohort were receiving daily MMF, which did not appear to have an impact on B-cell depletion. Peripheral B-cell depletion in patients with SLE treated with mosunetuzumab. Patients are grouped as per actual dose received, regardless of their original cohort assignment. One patient assigned to the 5+15 mg cohort and one patient assigned to the 5+45 mg cohort received only the day 1 injection (5 mg) and did not receive their respective day 8 injections. Results below LLOQ were imputed as LLOQ/2 or 0.2205 cells/μL. BLOQ, below limit of quantification; LLOQ, lower limit of quantification; SC, subcutaneous. T-cell activation, indicated by increased CD69 expression, occurred across all dose levels with no evidence of a dose-response relationship (online supplemental figure 1). Immunogenicity No ADAs were detected at baseline in any participant. Postbaseline ADAs were detected in three of 15 participants (20%); one each in the 1.6 mg, 5 mg and 5+15 mg cohort. However, ADAs did not appear to have any impact on PK or safety. Clinical activity and biomarkers While this study was not powered for efficacy assessments, positive trends in clinical responses were noted. Most participants treated with ≥5 mg mosunetuzumab showed disease improvement based on reductions in the PGA score at both month 3 (maximum reduction ranging from –0.42 to –0.99) and at month 6 (maximum reduction ranging from –0.93 to –1.5, online supplemental table 2). PGA/PGI-S assessment at baseline and over the study period for each participant is shown in online supplemental table 3. SLEDAI-2K data at months 3 and 6 were available for 14 out of the 15 participants enrolled. Overall, there was a mean reduction in SLEDAI-2K score from baseline at both months 3 and 6 (−2.29 and −2.71 points, respectively); in the 5+60 mg cohort (n=3), the mean change in SLEDAI-2K from baseline was −4.67 and −6 points at months 3 and 6, respectively. For participants with a SLEDAI-2K score >6 at baseline (all treated with ≥5 mg), a mean change of −4.4 points was observed at month 3, and a mean change of −6 points at month 6 (online supplemental table 2). A decrease of ≥4 points in SLEDAI-2K score was observed in 4/6 participants with a baseline score of ≥8 points, irrespective of dosing level (online supplemental figure 2). All available SLEDAI-2K assessments are shown in online supplemental table 4. A general trend towards improvement was observed in anti-dsDNA antibodies for patients with borderline/positive levels at baseline (see online supplemental figure 3), and C3 and C4 levels below the reference ranges at baseline. Discussion This phase Ib study, the first clinical report for a T-cell engager CD20xCD3 bispecific antibody in SLE, provides insights into the safety, tolerability, PK and PD of the bispecific antibody, mosunetuzumab, in patients with active SLE. Monoclonal antibodies that deplete B cells primarily via antibody-dependent cellular cytotoxicity and antibody-dependent cellular phagocytosis have been studied in SLE for almost two decades.32 The type I anti-CD20 antibody rituximab was investigated in two phase III trials, the Exploratory Phase II/III Systemic Lupus Evaluation of Rituximab (EXPLORER) trial in non-renal SLE9 and the LN Assessment with Rituximab (LUNAR) trial in LN,10 neither of which met their primary end points. However, the type II anti-CD20 monoclonal antibody obinutuzumab, which has been shown to have greater B-cell depletion compared with type I anti-CD20 antibodies, demonstrated superiority over placebo when added to standard therapy in patients with active LN and SLE in phase III trials.12 13 It is hypothesised that mosunetuzumab, through its bispecific CD3/CD20 mechanism of action, potentially may provide greater efficacy than type II anti-CD20 antibodies, such as obinutuzumab, by inducing greater B-cell depletion, including tissue B-cell depletion. This possibility was recently demonstrated in preclinical mouse models where mosunetuzumab showed greater depletion of splenic germinal centre and age-associated B cells as well as bone marrow mature and immature B cells than GA101, the research analogue to obinutuzumab.33 The greater B-cell depletion demonstrated by CAR T-cell therapies suggests the possibility of even greater responses, which supports the hypothesis that the potential greater B-cell depletion that can be achieved with mosunetuzumab may translate into superior clinical results with a manageable safety profile.20 23 The safety profile of mosunetuzumab observed in this study was consistent with prior studies in follicular lymphoma, for which mosunetuzumab received accelerated approval by the US Food and Drug Administration and conditional approval from the European Medicines Agency, with the specific indication being for treatment of adult patients with R/R follicular lymphoma after two or more lines of systemic therapy.25 26 Successful mitigation of the cytokine storm that can occur following administration of T-cell engaging therapies is key to allow administration in an outpatient setting, without requirement for specialised inpatient infrastructure. In a recent systematic review of 47 patients with SLE treated with CAR T-cell therapy, CRS occurred in 87% of cases, although mostly mild (grades 1–2), with nine patients requiring usage of tocilizumab; ICANS occurred in one patient (grade 1).34 In our study, no ICANS were observed and the incidence and severity of CRS was low, none of which required tocilizumab use. This highlights the potential for mosunetuzumab as a scalable outpatient B-cell depletion therapy for the broader SLE population, particularly for those with moderate disease activity, where the risk-to-benefit ratio of intensive cell therapy may be less favourable. A number of limitations are associated with interpretation of the findings from this study. The small number of participants enrolled lacked specific disease manifestations that may be of interest, including moderate-to-severe LN. Additionally, participants could have been taking a wide range of concomitant medicines, which could affect findings. Only a fixed-dosing single dose (on day 1) or step-up doses (on days 1 and 8) of mosunetuzumab were administered; additional doses will likely be necessary to achieve a sustained level of B-cell depletion to induce greater clinical effects. Although B-cell depletion was substantial, only peripheral B-cell levels were assessed without any tissue-based measurements and B-cell subtype profiling was not performed. While the observed AE profile of mosunetuzumab was manageable in this study, larger studies are required to fully characterise the safety profile in more diverse and severe SLE populations. Future studies, including those that incorporate tissue-based assessments of B-cell depletion, can build on these findings to establish optimal dosing regimens, characterise efficacy and assess the long-term safety and immunogenicity of mosunetuzumab in patients with SLE. The safety, PK and PD results from this phase Ib study are critical for advancing its clinical development for B-cell-driven autoimmune diseases, including SLE. Supplementary material