Objective:

To conduct a systematic review and meta-analysis of clinical trials of biologics for Osteogenesis Imperfecta (OI) to evaluate their efficacy and safety, while implementing a large language model (LLM)-assisted workflow for the review process.

Approach:

• Search Strategy: Identified novel biologics through a scan of clinical trials and systematically searched PubMed, Embase, ScienceDirect, Web of Science, and the Cochrane Library for human studies published up to December 1, 2025.
• Eligibility Criteria: Included trials reporting preliminary efficacy outcomes and excluded those that were withdrawn, suspended, or conducted solely in animal models.
• LLM-Assisted Workflow: Implemented a workflow that simulates integrated human screening and appraisal to evaluate agreement with human reviewers.

Key Findings:

• Current therapies for OI are primarily palliative and less effective in adults or severe phenotypes.
• Emerging biologics offer targeted mechanisms of action but have uncertain clinical value due to limited data.
• Existing systematic reviews are limited by narrow drug scope and variability in study design.

Interpretation:

The study highlights the need for a comprehensive evaluation of biologics in OI, utilizing advanced AI methods to enhance the systematic review process.

Limitations:

• Heterogeneous study designs and inconsistent outcomes complicate the review process.
• Current AI applications are limited in task scope and methodological rigor.

Conclusion:

The study aims to provide an up-to-date synthesis of biologics for OI, addressing gaps in existing literature and improving the review process through AI integration.