Alpha-lipoic acid-associated insulin autoimmune syndrome: reproducibility package
# Alpha-lipoic acid–associated insulin autoimmune syndrome (IAS): reproducibility package **Project**: A case report of oral alpha-lipoic acid–induced IAS, an individual participant data (IPD) systematic review, and a FAERS pharmacovigilance analysis. **Deposited**: November 2026 (actual deposit date shown by OSF timestamp) ## Protocol summary (English) The study protocol (in Chinese) was documented before manuscript drafting and prespecifies: 1. **IPD systematic review**: search of PubMed and Chinese databases (cutoff Aug 27, 2026); case reports/series of alpha-lipoic acid (ALA)-associated IAS; single extraction with blinded independent re-appraisal using the JBI 8-item tool; main analysis set = records with JBI score ≥5. 2. **Recovery definitions (three tiers)**: symptom resolution / IAA negativization / follow-up floor for ongoing recovery. 3. **Statistics**: descriptive statistics prespecified as primary; exploratory Mann-Whitney U and Fisher exact tests flagged as hypothesis-generating only. 4. **FAERS analysis**: OpenFDA API snapshot (July 30, 2026); exposure = drug-name match ("lipoic acid"/"thioctic"); deduplication by case ID; 2×2 disproportionality with ROR/PRR and 95% CI; signal threshold = ROR lower 95% CI >1 with a ≥3; primary-suspect (PS) role restriction as the primary analysis; prespecified sensitivity analyses (F1: exclusion of concomitant glucose-lowering drugs; F2: PS-only; F3: by time period). 5. **Index case**: reported per CARE; causality by Naranjo scale; PEG precipitation test documented. The English translation of key protocol items is available from the corresponding author upon request. ## Contents - `protocol/Study_Protocol_v2.2.md` — analysis plan (Chinese), documented before manuscript drafting: 40-variable dictionary, JBI appraisal rules, recovery-time operational definitions, FAERS analysis plan (ROR/PRR thresholds, sensitivity analyses F1–F3). - `protocol/FAERS_feasibility_report.md` — FAERS feasibility verification (counts, exposure set construction). - `data/IPD_dataset_v2.4.csv` — IPD dataset (40 records × 45 fields, de-identified literature-derived data; index case de-identified). - `data/IPD_extraction_notes_v2.4.md` — extraction notes and version history. - `data/Raw_Data_Workbook_CN.xlsx` / `Raw_Data_Workbook_EN.xlsx` — 12-sheet workbook (IPD dataset, variable dictionary, JBI dual-review matrix, index-case raw data, IPD summary, FAERS 2×2 tables, sensitivity analyses, FAERS IAS report review, yearly trend, exposure records, references). - `code/` — Python scripts used for OpenFDA API download (`faers_download.py`), deduplication/counting (`faers_counts.py`), and disproportionality analysis (`faers_analyze.py`, `faers_analyze2.py`). ## Notes - FAERS raw data were obtained from the public OpenFDA API () and are not redistributed here; the exact query strings are embedded in `faers_download.py` and the protocol. - All patient data are de-identified. Written informed consent for publication was obtained from the index patient. - This study was not prospectively registered; the analysis plan was documented prior to manuscript drafting. - During preparation, AI-based tools were used for language polishing and assistance with data-processing code; all content and data were reviewed and verified by the authors, who take full responsibility for the publication. ## Contact Corresponding author: [to be completed upon publication]
Authors
- 王涛 (ORCID: https://orcid.org/0009-0007-1700-949X)
Publication Details
- Journal
- Open Science Framework
- Published
- 2026-11-15
- DOI
- https://doi.org/10.17605/osf.io/verqj
- Primary Topic
- Pharmacovigilance and Adverse Drug Reactions
- Type
- article
- Field-Weighted Citation Impact
- 0.00