{"slug": "gxp-agent-process-dag-topology-for-reliable-clinical-trial-programming-with-llm", "title": "GxP-Agent: Process-DAG Topology for Reliable Clinical Trial Programming with LLM Agents", "summary": "A new multi-agent system, GxP-Agent, achieved 100% structural match on the FDA pilot submission CDISCPilot01 (254 subjects, 49 ADSL variables) across three runs, compared to 59.2% for the best retrieval-augmented baseline and 0% for all single-agent and flat multi-agent approaches, according to a preprint on arXiv. The system encodes regulatory process ordering as a directed acyclic graph (DAG) with 15 domain-specific nodes, enabling reliable LLM-based clinical trial programming under CDISC standards.", "body_md": "arXiv:2608.16890v1 Announce Type: new\nAbstract: Clinical trial programming -- transforming study protocols into analysis-ready datasets under CDISC standards -- is a bottleneck in regulatory submissions, yet LLM-based code generation fails catastrophically on this task: across 11 single-shot attempts with five frontier models, none produces a valid subject-level analysis dataset. We introduce GxP-Agent, a multi-agent system that encodes regulatory process ordering as a directed acyclic graph (DAG), decomposing monolithic dataset generation into 15 domain-specific nodes executed by worker agents with pharmaverse skill context, validation gates, and conditional retry. On CDISC-Bench, a new execution-based benchmark built from the FDA pilot submission CDISCPilot01 (254 subjects, 49 ground-truth ADSL variables), GxP-Agent with Claude Sonnet 4.6 achieves 100% structural match (49/49 variables, 254 correct records) across three independent runs, compared to 59.2% for the best retrieval-augmented baseline and 0% for all single-agent and flat multi-agent approaches. The DAG topology also enables weaker models: GPT-4.1 achieves 59.2% mean structural match under the same DAG, where it scores 0% under every other architecture. The approach generalizes to ADAE (adverse events; 9-node branching DAG, 55 variables, 1,191 records), achieving 100% structural match on the first attempt. These results demonstrate that encoding domain process knowledge as graph topology -- rather than relying on LLM reasoning alone -- is a key enabler for reliable, GxP-compliant clinical trial programming.", "url": "https://wpnews.pro/news/gxp-agent-process-dag-topology-for-reliable-clinical-trial-programming-with-llm", "canonical_source": "https://arxiv.org/abs/2608.16890", "published_at": "2026-08-19 04:00:00+00:00", "updated_at": "2026-08-19 04:13:59.039409+00:00", "lang": "en", "topics": ["artificial-intelligence", "ai-agents", "ai-research"], "entities": ["GxP-Agent", "Claude Sonnet 4.6", "GPT-4.1", "CDISCPilot01", "CDISC", "pharmaverse", "FDA"], "alternates": {"html": "https://wpnews.pro/news/gxp-agent-process-dag-topology-for-reliable-clinical-trial-programming-with-llm", "markdown": "https://wpnews.pro/news/gxp-agent-process-dag-topology-for-reliable-clinical-trial-programming-with-llm.md", "text": "https://wpnews.pro/news/gxp-agent-process-dag-topology-for-reliable-clinical-trial-programming-with-llm.txt", "jsonld": "https://wpnews.pro/news/gxp-agent-process-dag-topology-for-reliable-clinical-trial-programming-with-llm.jsonld"}}