| アブストラクト | PURPOSE: To examine the interval between stable public safety reporting patterns and subsequent substantive label strengthening in targeted oncology therapies, and assess comparator effects on signal onset. METHODS: DailyMed label histories were linked to quarterly FDA Adverse Event Reporting System (FAERS) counts from openFDA. The primary analysis included 27 products across antibody-drug conjugates, immune checkpoint inhibitors, and tyrosine kinase inhibitors. Three predefined themes were assessed: interstitial lung disease/pneumonitis, hepatotoxicity, and dermatologic toxicity. Stable signal onset required at least three themed reports, reporting odds ratio >/= 2, proportional reporting ratio >/= 2, and persistence into the next quarter. Baseline and pre-signal status were recorded; post-signal changes were retained only after direct comparison with preceding usable archived versions confirmed substantive safety strengthening. RESULTS: The label archive contained 52 correctly mapped product histories and 1022 archived version records. Among 81 product-theme pairs, 52 met the stable signal rule under the oncology proxy background. Twenty-four had substantive label strengthening by 31 December 2025, 23 had no qualifying event, 1 was excluded after manual review, and 4 were source-indeterminate. The median interval among retained events was 388 days (interquartile range 79-697). Sensitivity analysis identified 62 pairs stable under at least one background; among 52 stable under both backgrounds, 24 had differing detection dates. CONCLUSION: The timing between public reporting signals and later label strengthening can be reconstructed from pharmacovigilance and label-history records. The observed interval is a timing measure between public records and should not be interpreted as incidence, causality, or regulatory response time. |
|---|
| 組織名 | Department of Pharmacy, The First Affiliated Hospital of Zhengzhou University,;Zhengzhou, Henan, 450052, China. zxg_gcp@126.com.;Zhengzhou, Henan, 450052, China. |
|---|
