The clearest result
The randomized evidence on telitacicept found a substantially better lupus response at 48 to 52 weeks. The main response measure, called SRI-4, had a risk ratio of 2.04, with a 95% confidence interval, the study's uncertainty range, from 1.66 to 2.50. On an absolute scale, that was an estimated 343 additional responses per 1,000 people, with a range from 218 to 496. The evidence was rated moderate certainty.
At the same follow-up, any adverse event was probably slightly more common with telitacicept: the risk ratio was 1.09, with a 95% confidence interval from 1.02 to 1.17. That corresponds to 78 more events per 1,000 people, with a range from 17 to 143. Serious adverse events went in the opposite direction in the point estimate, at 0.68, but the interval from 0.35 to 1.30 included both fewer and more events; the absolute estimate was 48 fewer per 1,000, ranging from 96 fewer to 44 more. Upper respiratory tract infection had a risk ratio of 1.31, but its interval from 0.83 to 2.06 also left the result uncertain. Certainty was moderate for any adverse event, low for serious adverse events and very low for upper respiratory tract infection.
How the evidence was assembled
Those figures came from a review that kept different kinds of evidence separate. It synthesized placebo-controlled randomized trials, other comparative studies and single-arm cohorts. Eligibility required adults aged 18 or older with systemic lupus erythematosus, or SLE; adult lupus nephritis was prespecified as a subgroup, and the intervention was subcutaneous telitacicept, usually with standard therapy. The review followed PRISMA 2020 and registered its protocol in PROSPERO as CRD420261464193.
Thirty-nine reports were included, and 22 contributed to quantitative synthesis. All 39 reports came from China. The review assessed randomized trials with RoB 2, comparative observational studies with NOS and uncontrolled cohorts with JBI tools; two reviewers assessed all studies independently, with a third reviewer resolving disagreements when needed. GRADE found no outcome to be high certainty: SRI-4 and any adverse event were moderate, serious adverse events were low, and upper respiratory tract infection, comparative models and single-arm models were very low.
Results beyond the randomized trials
Outside the randomized trials, the picture was less settled. At 24 to 26 weeks, a nonstandard clinical total effective response favored telitacicept-containing regimens, with a risk ratio of 1.19 and a 95% confidence interval from 1.18 to 1.21. The SLEDAI endpoint was lower by 1.81 points, while C3 and C4 were higher by 0.19 g/L and 0.062 g/L. No clear differences appeared for any adverse event or infection. Because the response definitions varied and were not equivalent to SRI-4, the comparative evidence was rated very low certainty.
Comparisons with belimumab were also observational. They showed no clear differences in any adverse event or infection, with risk ratios of 1.11, with a 95% confidence interval from 0.84 to 1.49, and 1.10, with an interval from 0.75 to 1.60. At 24 to 26 weeks, C3 was higher with telitacicept by 0.14 g/L, while C4 was uncertain. The unadjusted observational endpoints cannot establish comparative efficacy, and the evidence was rated very low certainty.
Kidney findings need caution
Lupus nephritis comparisons showed a more pronounced signal, but it was not a secure treatment effect. Complete renal response, or CRR, had a risk ratio of 2.32, with a 95% confidence interval from 1.30 to 4.16, with add-on telitacicept. Primary efficacy renal response, or PERR, had a risk ratio of 1.67, with an interval from 1.15 to 2.42. The corresponding absolute estimates were 227 and 272 additional responses per 1,000 people. Certainty was very low because the evidence was observational, and residual or unmeasured confounding could explain part of the association.
Single-arm cohorts offered descriptive snapshots rather than comparisons. Their pooled SRI-4 proportions were 69.1% at 12 weeks, 76.2% at 24 to 26 weeks and 80.4% at 48 to 52 weeks. LLDAS attainment at 24 to 26 weeks was 44.8%. The review kept those time windows separate because the denominators differed, so the figures do not show that response rose within one cohort over time. Single-arm certainty was very low and descriptive only.
Single-arm lupus nephritis cohorts reported CRR proportions of 62.7% at 24 to 26 weeks and 66.8% at 48 to 52 weeks. Pooled proportions for any adverse event and infection were 15.7% and 13.3%. These figures are not comparative safety effects. High heterogeneity and inconsistent event ascertainment limited interpretation, so the findings remain very low certainty.
What the evidence can and cannot say
Taken together, the evidence supports a narrower conclusion than a general claim that telitacicept has settled the treatment question. In the Chinese randomized populations studied, the review found a probable benefit on SRI-4 at 48 to 52 weeks and a slight increase in any adverse events. The randomized safety analysis covered that follow-up and did not provide high-certainty evidence, so it does not settle longer-term risk. The observational belimumab comparison also cannot establish comparative efficacy. Because every included report came from China, generalizability beyond those populations remains unresolved.
The authors reported financial support from Henan Provincial Traditional Chinese Medicine research and academic-school programs, including grants 2025LHZX2178 and cz0422.
Paper data and sources
Original title: Efficacy and safety of telitacicept in adults with systemic lupus erythematosus: a systematic review and stratified meta-analysis
Authors: Wang Z, Li H, Li J et al.
Journal/Repository: Frontiers in medicine
Status: Peer-reviewed
First online: 2026-10-10
DOI: Not available
Original paper · Full text