The evidence review contains the synthesis; the specification contains the proposed study. Source checks were performed on 2026-09-30. This is a targeted literature review, not a systematic search or meta-analysis. No patient-level data were analysed.
Status Markers
- ✅ Checked. The stated claim was checked in the relevant source text. The note identifies the section and access level; it does not mean every result or supplement was verified.
- ⚠️ Partial. Abstract, indexed publisher text or selected passages were available, but the full methods/results needed for a stronger claim were not checked. Use only the limited claim recorded here.
- ❌ Not checked. A reading lead only, not evidence supporting the review.
Some publisher pages blocked direct retrieval while their indexed abstracts or article passages remained readable. Such checks are labelled partial. Values transcribed from a report have not been recalculated from raw data.
Reading Queue
Recommended Overview
- Junt et al. 2025 — Defining immune reset: achieving sustained remission in autoimmune diseases. ⚠️ Nature Reviews Immunology. Question: Which clinical, cellular and molecular meanings should be separated? The publisher abstract confirms this framework. Full-text reading remains outstanding; no detailed framework or threshold is attributed to the unread sections.
Primary Studies to Read Next
- CASTLE, Müller et al. 2026. ✅ Read the B cell reconstitution results and Extended Data Figure 4 legend to distinguish the paired analysis from everyone contributing a sample. Source 7 below.
- Md Yusof et al. 2017. ⚠️ Read the plasmablast relapse analysis and its supplementary validation figure. The reported threshold and accuracy values were checked in indexed article text; the supplement and raw data were not checked. Source 5.
- Jenks et al. 2018. ⚠️ Read the subset definitions and functional experiments to understand why CD27-negative cells need subdivision. Methods/discussion passages were available through the article index. Source 10.
- Li et al. 2026. ⚠️ Obtain the full report to determine the timing and number of relapse samples and whether the molecular findings precede relapse. Abstract-level findings alone cannot establish prediction. Source 8.
Sources
1. Junt et al. 2025 — Definitions ⚠️
Defining immune reset: achieving sustained remission in autoimmune diseases. Nature Reviews Immunology 25:528–541. DOI.
Used for: The distinction among clinical, cellular and molecular reset concepts. Checked in: Publisher abstract. Limit: Perspective; not a validation dataset. Detailed full-text recommendations remain unread.
2. Anolik et al. 2007 — Delayed Memory Recovery ⚠️
Delayed memory B cell recovery in peripheral blood and lymphoid tissue in systemic lupus erythematosus after B cell depletion therapy. PubMed abstract; DOI.
Used for: Prolonged memory-cell delay accompanying sustained response in a subset. Checked in: Abstract Methods and Results, including cohort size and mean follow-up. Limit: Individual trajectories and detailed gates were not extracted; no validated decision rule is claimed.
3. Vital et al. 2011 — Return Before Relapse ⚠️
B cell biomarkers of rituximab responses in systemic lupus erythematosus. Arthritis & Rheumatism 63:3038–3047. Publisher abstract.
Used for: Faster memory/plasmablast return among earlier relapsers. Checked in: Indexed publisher abstract Methods and Results. Limit: No independent reanalysis of the longitudinal association; concurrent steroids and observational treatment must remain part of interpretation.
4. Lazarus et al. 2012 — Heterogeneity at Relapse ⚠️
B-cell numbers and phenotype at clinical relapse following rituximab therapy differ in SLE patients according to anti-dsDNA antibody levels. Rheumatology 51:1208–1215. Article; DOI.
Used for: Variation in relapse timing and phenotype with anti-DNA status. Checked in: Indexed abstract and Discussion. Limit: Phenotyping subgroup; full subgroup extraction and original figures remain unchecked. The review does not turn observations at relapse into validated prediction.
5. Md Yusof et al. 2017 — Plasmablast Prediction ⚠️
Predicting and managing primary and secondary non-response to rituximab using B-cell biomarkers in systemic lupus erythematosus. Annals of the Rheumatic Diseases 76:1829–1836. Article; Publisher PDF.
Used for: The month-6 plasmablast rule in the review. Checked in: Indexed Results, “Plasmablast repopulation as a biomarker of relapse,” including validation denominator, timing, threshold, confidence intervals and area under the curve. Limit: Supplementary Figure S1 not inspected. The biomarker subset is smaller than the overall clinical cohort; cross-centre or cross-therapy validation is not established here.
6. Mackensen 2022 and Müller 2024 — Early CAR-T Reports ⚠️
Mackensen et al. Anti-CD19 CAR T cell therapy for refractory systemic lupus erythematosus. Nature Medicine 28:2124–2132. Article.
Used for: Remission continuing after return of naïve, non-class-switched B cells. Checked in: Publisher abstract. Limit: Small compassionate-use series; full methods and correction not checked for detailed phenotyping.
Müller et al. CD19 CAR T-Cell Therapy in Autoimmune Disease — A Case Series with Follow-up. New England Journal of Medicine 390:687–700. Article.
Used for: Cohort composition and the longitudinal SLE phenotyping subset. Checked in: Indexed publisher recruitment text and B cell phenotype figure legend. Limit: No clinical outcome rates extracted here from the blocked full report. Patient overlap with earlier publications has not been mapped; these reports are not summed as independent cohorts.
7. Müller et al. 2026 — CASTLE ✅
CD19 CAR-T cells for treatment-refractory autoimmune diseases: the phase 1/2 CASTLE basket trial. Nature Medicine 32:1142–1151. Article.
Used for: Reconstitution pattern and its coexistence with clinical benefit. Checked in: Abstract, B cell results, long-term efficacy paragraph and Extended Data Figure 4 legend in publisher HTML. Limit: Individual-level prognostic analysis was not performed here. Different disease endpoints and phenotyping denominators are retained in the review.
8. Li et al. 2026 — Autoimmune Hemolytic Anemia ⚠️
CD19 CAR T-Cell Therapy for Autoimmune Hemolytic Anemia. New England Journal of Medicine 394:253–267. Article.
Used for: Clinical cohort description and the reported remission/relapse molecular observations. Checked in: Complete indexed publisher abstract. Limit: Full report blocked. Relapse count, tissue sampling, temporal ordering and causal interpretation of cell interactions were not verified. The phrase “relapse-associated niche” describes the reported mechanistic finding, not a validated predictive signature.
9. Fedak et al. 2026 — BCMA-Directed Therapy in Myasthenia ⚠️
BCMA-directed mRNA CAR-T cell therapy for myasthenia gravis: exploratory biomarker analysis of a placebo-controlled phase 2b trial. Article; PubMed.
Used for: Treatment-associated immune changes extending beyond bulk B cell reconstitution. Checked in: Abstract and indexed Discussion. Limit: Exploratory biomarker analyses; assay-specific denominators and multiplicity were not extracted. Randomization supports treatment comparison, not automatic validation of each proposed biomarker.
10. Jenks et al. 2018 — Double-Negative Subsets ⚠️
Distinct Effector B Cells Induced by Unregulated Toll-like Receptor 7 Contribute to Pathogenic Responses in Systemic Lupus Erythematosus. Immunity. Article.
Used for: IgD/CD27 classification and refinement of DN2 cells. Checked in: Indexed Methods, Discussion and figure descriptions. Limit: Mechanistic disease study, not post-depletion prediction. The harmonized panel proposed here has not been validated as a complete assay.
11. COMPARE 2026 — Rheumatoid Arthritis ⚠️
CD19 CAR T cell therapy for treatment-refractory seropositive rheumatoid arthritis: a phase 1 trial. Nature Medicine; published 27 August 2026. Article.
Used for: Distinguishing clinical improvement, remission and humoral change. Checked in: Publisher abstract, including cohort size, follow-up and clinical response endpoints. Limit: Full text requires subscription; no detailed returning-subset or relapse predictor is inferred.
12. Wang et al. 2019 — Transitional Cells and Autoreactivity ⚠️
High TLR7 Expression Drives the Expansion of CD19+CD24hiCD38hi Transitional B Cells and Autoantibody Production in SLE Patients. Frontiers in Immunology 10:1243. Article.
Used for: Transitional phenotype does not guarantee tolerance. Checked in: Indexed primary abstract and Discussion. Limit: Mechanistic SLE study rather than a reconstitution cohort; no relapse prediction claimed.
Search Scope and Remaining Checks
Searches combined “B cell reset,” “immune reset,” “reconstitution,” “naïve,” “memory,” “plasmablast,” “relapse,” “rituximab” and “CAR T” with autoimmune disease terms. Primary publisher and PubMed/PMC records were prioritized; 2026 CASTLE, hemolytic anemia, myasthenia and rheumatoid arthritis reports were included. This selection supports the two project questions but does not establish exhaustive coverage of every disease or intervention.
The remaining checks are specific:
- Obtain full text and supplements for the partial entries before reusing their assay gates or effect estimates in a clinical protocol.
- Map patient overlap among related CAR-T reports before any quantitative evidence synthesis.
- Extract event-level timing and nonresponder data for a formal assessment of prediction, particularly from the newer molecular studies.
- Extend the comparison to prospective T-cell-engager reconstitution cohorts and other diseases as suitable longitudinal primary evidence is identified.
The proposed study addresses the missing prediction evidence. Completing the source checks would improve precision of the review, but would not substitute for validating a biomarker in new patients.
Back to top