CAR T, obinutuzumab and telitacicept lead a new wave of lupus therapies

Systemic lupus erythematosus has long been one of the most stubborn targets in autoimmune medicine, a disease in which the innate and adaptive immune systems go awry simultaneously and produce a protean spectrum of symptoms that frustrate diagnosis and management alike. For a decade after belimumab became the first new lupus drug approved by the US Food and Drug Administration in 2011, the therapeutic pipeline was essentially frozen. That stagnation ended in 2021 with the approvals of voclosporin and anifrolumab, followed late in 2025 by obinutuzumab, and a new narrative review from rheumatologists at Walter Reed National Military Medical Center in Bethesda, Maryland, published in Frontiers in Lupus, now maps in detail which of the many candidate therapies are succeeding, which are failing, and why.

The authors, Michael F. Loncharich, Brenna E. Cafeo and Evan B. Noble, searched PubMed for lupus clinical trials published in the past five years, combed clinicaltrials.gov for phase III and phase IV studies with no date restrictions, and scanned the American College of Rheumatology Convergence 2025 abstract sessions for emerging literature. Phase I trials were excluded from the formal scope of the review, although the most eye-catching results in the field—chimeric antigen receptor T-cell therapy, or CAR T—come precisely from early-phase work and case series that the authors discuss as a rapidly expanding form of cell-based gene therapy migrating from oncology into autoimmune disease.

The technical logic behind CAR T in lupus is worth unpacking. T lymphocytes are engineered ex vivo to express a chimeric antigen receptor that redirects their cytotoxicity toward B cells, the source of the pathogenic autoantibodies that drive lupus pathology. The common target antigens are CD19, expressed broadly on B-lineage cells, and B-cell maturation antigen, or BCMA, which is prominent on plasma cells. This matters because conventional B-cell-depleting drugs have repeatedly failed in lupus, and the review offers two mechanistic explanations for why engineered T cells might succeed where monoclonal antibodies did not. First, B cells can remain hidden from biologics in tissues such as the synovial membrane, tonsils and lymph nodes, where rituximab has been shown not to deplete B cells in secondary lymphoid organs. Second, plasma cells and pro-B cells lack CD20, the target of rituximab and obinutuzumab, which shields them from depletion. CD19 or BCMA CAR T cells can penetrate these tissues and reach plasma cells and pro-B cells that antibodies cannot.

The clinical data are striking. In 2022, Mackensen and colleagues treated five patients with refractory lupus using autologous CD19 CAR T cells at a dose of one million cells per kilogram, following lymphodepleting chemotherapy with cyclophosphamide and fludarabine. All five achieved DORIS remission—the Definition of Remission in SLE—within three months, with durable responses of five to eighteen months, disappearance of anti-dsDNA and anti-Smith autoantibodies, and only one case of mild cytokine-release syndrome. In 2024, Muller and colleagues reported on eight patients, all of whom met both DORIS and lupus low disease activity state criteria after six months, with SLEDAI-2K scores remaining at zero for up to twenty-nine months and the same seroconversion of autoantibodies. More recently, Hu and colleagues completed an open-label trial of BCMA CAR T in seven patients with active lupus nephritis class III through V who had failed glucocorticoids and at least two immunosuppressants or biologics. The median SLEDAI improved from eighteen at baseline to zero at six months and stayed there for four years; five patients achieved DORIS remission and the remaining two met LLDAS criteria, with one grade 1 cytokine-release syndrome, no severe infections, and lymphodepletion resolving within four weeks.

Compound CAR T, carrying dual targets of BCMA and CD19, has also been tested in a phase 1 open-label trial of thirteen participants. Nine met DORIS remission and twelve met LLDAS criteria, eleven showed significant improvements in estimated glomerular filtration rate and proteinuria, and a repeat renal biopsy in one patient in drug-free remission revealed clearance of glomerular IgG immunocomplexes—direct histological evidence that the therapy removed the immune deposits driving nephritis. Neutropenia resolved within six weeks and B cells recovered within seven months, suggesting the deep depletion is not permanent. The authors caution that CAR T data remain limited to case series and phase I trials, but the early signal of sustained remission in refractory cases is among the most consequential developments in the field.

Among conventional biologics, the type II anti-CD20 antibody obinutuzumab stands out. Unlike type I agents such as rituximab and ocrelizumab, which failed in the EXPLORER, LUNAR and BELONG trials, obinutuzumab was glycoengineered for greater FcγRIIb affinity, permitting greater antibody-dependent cellular cytotoxicity with less reliance on complement, and it redistributes less to membrane lipid rafts, resulting in less internalization of the target. The phase II NOBILITY trial missed its primary endpoint of complete renal response at week 52 but showed significant improvements in key secondary endpoints including complete renal response at week 104 and partial renal remission rates. The phase III REGENCY trial then met its primary endpoint, with complete renal response at week 76 of 45 percent versus 33 percent for placebo, significantly favored secondary endpoints including treatment failure and rescue therapy need, no unexpected safety results, and fewer deaths or renal events in the treatment group. That profile carried obinutuzumab to FDA approval late in 2025.

The BAFF and APRIL cytokine axis, which promotes survival and maturation of immature B cells and expands autoreactive populations, has produced a mixed record. Belimumab, the first anti-BAFF agent approved in 2011, now has six-to-eight-year extension data showing no new or unexpected safety signals, discontinuation rates due to adverse events as low as nine percent, and flare rates as low as fifteen percent over seven years. Blisibimod failed its phase II PEARL-SC and phase III CHABLIS-SC1 primary endpoints despite biomarker improvements and corticosteroid sparing. Tabalumab failed when dosed monthly in the ILLUMINATE trials but showed better SRI-5 response when dosed biweekly, without differences in adverse events. Ianalumab, a BAFF receptor inhibitor, met its phase II primary endpoint with significantly higher SRI-4 response and sustained corticosteroid reduction, with no increased serious adverse events or infections. Atacicept, a TACI fusion protein, was complicated by two deaths in its high-dose arm that were adjudicated non-treatment related, but the closely related telitacicept met its phase III primary endpoint in China with a modified SRI-4 response of 67 percent versus 33 percent for placebo, a difference detectable as early as week 4, and no difference in serious infections. Telitacicept was approved for SLE in China in 2021.

Intracellular kinase inhibitors tell a more nuanced story. Baricitinib, a selective JAK1 and JAK2 inhibitor, produced conflicting results in the SLE-BRAVE trials: the 4-milligram dose outperformed placebo on SRI-4 in SLE-BRAVE I, but no dose beat placebo in SLE-BRAVE II, and a pooled post-hoc analysis failed while noting an unusually high placebo SRI-4 response rate of 46 percent. Deucravacitinib, a TYK2 inhibitor that binds the inactive regulatory domains of its target kinases rather than the active catalytic sites, theoretically permitting a more durable response, showed better SRI-4 responses than placebo across all doses in the phase II PAISLEY study—58, 50 and 45 percent versus 34 percent—at 32 weeks, with better BICLA, CLASI, LLDAS and joint count outcomes, though with higher rates of infections and cutaneous events; phase III trials are ongoing. Elsubrutinib, a Bruton’s tyrosine kinase inhibitor, failed its primary endpoint, and the review offers a mechanistic explanation: in plasmacytoid dendritic cells, BTK inhibition blocks interferon-alpha production via TLR-9 but not via TLR-7, and TLR-7 stimulation enhances interferon-alpha production by disinhibiting the BDCA-2 pathway, potentially worsening the interferon signature that drives lupus.

Anifrolumab, the first-in-class global type I interferon receptor inhibitor approved in 2021, now has long-term data confirming durable efficacy: a three-year open-label extension of the TULIP trials found greater SLEDAI improvement, greater corticosteroid dose reduction, and a fifty percent lower annualized flare rate than placebo, with marginally lower exposure-adjusted adverse event rates and no significant differences in nonopportunistic infections. Higher rates of influenza and herpes zoster were noted in the original trials. However, anifrolumab failed in lupus nephritis: neither the TULIP-LN trial nor its two-year extension found significant differences versus placebo, although the IRIS phase III trial is expected to complete in September 2026. Voclosporin, the calcineurin inhibitor approved for lupus nephritis in 2021, achieved complete renal remission rates of 41 to 71 percent versus 23 percent for placebo in its pivotal trials, and a three-year safety study with repeat renal biopsy found no change in chronicity, no vascular lesions and no tubular compartment changes, indicating genuine long-term tolerability.

Elsewhere in the pipeline, litifilimab, which blocks blood dendritic cell antigen 2 on plasmacytoid dendritic cells, met its endpoints in both cutaneous lupus and SLE trials with better CLASI responses, joint counts and SRI-4 responses than placebo, and in the SLE trial more adverse events occurred in the placebo group. Iberdomide, a cereblon ligand that degrades the transcription factors Ikaros and Aiolos, thereby inhibiting BAFF- and CD40L-induced proliferation, showed a tolerable safety profile and, in a trial powered for efficacy, significantly outperformed placebo on SRI-4, SLEDAI and CLASI responses. Cenerimod, a selective S1P1 receptor modulator that traps lymphocytes in lymphoid organs to prevent end-organ damage, showed a 1.43-fold greater likelihood of response than placebo, with phase III trials ongoing. By contrast, ustekinumab, secukinumab, dapirolizumab and low-dose interleukin-2 all failed to beat placebo in their initial analyses, though the LUPIL-2 trial found a roughly thirty percent estimated response rate once two sites with a hundred percent placebo response rate were excluded. The Walter Reed authors conclude that obinutuzumab and telitacicept have demonstrated both safety and efficacy in phase III trials, that litifilimab, ianalumab, deucravacitinib and iberdomide are showing promise in phase II, and that CAR T, though still early, is delivering sustained remission in the most refractory lupus cases ever studied.

Subject of Research: Safety and efficacy of novel therapeutics in lupus: a narrative review

Article Title: Safety and efficacy of novel therapeutics in lupus: a narrative review

Article References: Loncharich, M. F., Cafeo, B. E., & Noble, E. B. (2026). Safety and efficacy of novel therapeutics in lupus: a narrative review. Frontiers in Lupus, 4, Article 1800946. https://doi.org/10.3389/flupu.2026.1800946

Image Credits: AI Generated

DOI: 10.3389/flupu.2026.1800946

Keywords: Safety, efficacy, novel, therapeutics, lupus, narrative, review, scientific research