Lupus endpoints are built from several disease activity indices, lab values and steroid doses, combined into responder definitions. Capture records every component as structured data, so SRI-4, BICLA and steroid tapering can be derived and checked.
Free sandbox · No credit card · 21 CFR Part 11 aligned
Musculoskeletal and skin descriptors
SLEDMSKImmunology descriptors
SLEDIMMRenal descriptors
SLEDRENWhat matters in lupus trials
Study design
SRI-4 and BICLA are derived from three or more instruments assessed at the same visit. If any component is missing or recorded inconsistently, the participant cannot be classified as a responder, and non-responder imputation often follows. That makes completeness at the visit level the single most important data quality goal in lupus trials.
Record each SLEDAI-2K descriptor as present or absent rather than a total, each BILAG organ system grade, and the physician global assessment on a defined scale. Then derive totals and responder status in analysis, where the rules are written once. The same principle applies to steroid doses: record dose, frequency and dates on the concomitant medications form, and derive daily prednisone-equivalent dose in analysis.
The same approach applies across autoimmune disease: component-based activity indices in Sjögren's disease, vasculitis and myositis, joint counts and composite scores in rheumatoid arthritis. See the ACR response and DAS28 template.
| Criterion | Met | |
|---|---|---|
| SLEDAI-2K reduction of 4 or more | 12 to 6 | |
| No new BILAG A, at most one new B | None new | |
| Physician global not worse | +0.1 | |
| Steroid taper per protocol | 7.5 mg/day |
Laboratory data
Lupus activity depends on laboratory values as much as clinical signs. Lab tables with analytes as rows and unit, result and reference range columns can be drafted by AI from a lab manual and split by sex and age where the source has them. Range edit checks raise queries at entry.
Haematology
| Row | Code | Unit | Range | Stratum |
|---|---|---|---|---|
| Haemoglobin | HGB | g/dL | 13.0-17.0 | Male 18-65 y |
| Haemoglobin | HGB | g/dL | 12.0-15.5 | Female 18-65 y |
| Platelets | PLAT | 10^9/L | 150-400 | All |
| ALT | ALT | U/L | 7-56 | All |
| Site QC flag | SITEQC | - | - | All |
Set up component-level forms in the free sandbox.
Endpoints
| Endpoint | Components |
|---|---|
| SRI-4 | SLEDAI-2K change, BILAG-2004 new grades, physician global assessment |
| BICLA | BILAG-2004 improvement and no worsening, SLEDAI-2K, physician global, no treatment failure |
| Lupus low disease activity state | SLEDAI-2K, no new activity, physician global, prednisone dose, standard therapy |
| Complete renal response | Urine protein-to-creatinine ratio, eGFR, rescue therapy |
| Flare | SLEDAI flare index or BILAG-defined flare |
| Fatigue and quality of life | FACIT-Fatigue, SF-36, lupus-specific QoL measures |
Definitions vary between trials. Write the exact rules in the protocol and analysis plan.
Assessors
BILAG-2004 in particular requires training: the grade for each organ system depends on detailed definitions and on whether features are new, worse, the same or improving. Inconsistent grading across sites is a known source of noise in lupus trials. Sponsors usually train assessors before the first participant and review grading patterns centrally during the study.
Fatigue is often what participants care about most. Collecting FACIT-Fatigue at home between visits, alongside clinical indices, gives a fuller picture of treatment effect.
Study build
SLEDAI-2K descriptors, BILAG grades and physician global.
SRI-4, BICLA or LLDAS rules in the analysis plan.
Dose, frequency and dates for every course.
Complement, anti-dsDNA, urinalysis and UPCR with ranges.
BILAG and SLEDAI training before first participant.
Scheduled at home between visits.
The SLE Responder Index: a reduction of at least 4 points in SLEDAI-2K, no new BILAG A and no more than one new BILAG B organ score, and no worsening of the physician global assessment.
SLEDAI-2K scores weighted descriptors into a total. BILAG-2004 grades each of nine organ systems from A to E.
Store each descriptor so the total can be derived and checked. Totals and responder status are derived in analysis.
Record dose, frequency and dates on the concomitant medications form, and derive daily prednisone-equivalent dose in analysis.
Yes. AI can draft a lab table from a lab manual or file, for your review. Nothing is saved without review.
Yes, in the free sandbox with every feature.
Every component structured and checked. Free sandbox.