OncoAgentic-RAG: An evidence-grounded multi-agent hybrid retrieval-augmented system for patient-centric clinical oncology question answering
Abstract
Objective General-purpose large language models (LLMs) show promise for medical question answering (QA) but face oncology-specific challenges, including hallucination, weak evidence attribution, and limited handling of sensitive patient queries. This study develops an evidence-grounded, patient-oriented oncology QA system integrating hybrid retrieval and specialised multi-agent processing to improve evidence attribution, safety, and empathetic communication. Methods The proposed OncoAgentic-RAG is a collaborative multi-agent hybrid retrieval-augmented generation framework for oncology QA. Its knowledge base comprises standard oncology textbooks and clinical guidelines. Dense ChromaDB and BM25 sparse retrieval are combined for evidence retrieval and fusion in a LangGraph-based multi-agent architecture that incorporates crisis detection, intent orchestration, medical knowledge retrieval, treatment guidance, empathy-aware generation, clinical verification and patient-friendly response synthesis. The system was evaluated on the 1,015-question OncoIntentQA benchmark using retrieval, semantic, faithfulness, relevance, safety, completeness, empathy, and user-friendliness measures, with comparisons against NoRAG, BaselineRAG, and AgenticRAG. Generated responses were additionally assessed by clinical experts for safety and appropriateness. Results OncoAgentic-RAG achieved Precision@5 of 0.8120, Recall@5 of 0.7980, BERTScore F1 of 0.7634, Faithfulness of 0.8146, and Answer Relevance of 0.8625. It achieved a weighted S.C.O.P.E. (Safety, Completeness, Originality, Precision, and Efficiency) score of 4.68/5.0, Empathy Composite of 0.9180, and User-Friendliness of 0.9580. Clinical expert assessment provided additional evaluation of response safety and content. Conclusions Integrating hybrid retrieval with specialised multi-agent processing improved evidence-grounded, source-traceable, and patient-friendly oncology QA within the evaluated benchmark. The findings demonstrate the technical utility of OncoAgentic-RAG for patient-oriented oncology information, while prospective clinical and safety validation remains necessary.
Related
- BanglaDiaHallu: Hallucinations of large language models and the effect of guideline-grounded retrieval-augmented generation in Bangla diabetes question answering
- Real-world evaluation of a multilingual agentic retrieval-augmented clinical AI framework for evidence-grounded medical decision support
- Evidence-based clinical question answering using retrieval-augmented generation (RAG) and open-source large language models
- LitBench: A Benchmark for Retrieval-Grounded Multi-Paper Evidence Synthesis in Epilepsy
- Retrieval-augmented generation in biomedicine: a survey of technologies, datasets, and clinical applications
BibTeX
@article{jayasrilakshmi2026oncoagenticrag,
title = {{OncoAgentic-RAG: An evidence-grounded multi-agent hybrid retrieval-augmented system for patient-centric clinical oncology question answering}},
author = {S Jayasrilakshmi and Ansuman Mahapatra and Ankit Kujur and Choudhari Aniket Baban and Aditya Kumar Singh and Sachin Nandan Mohanty},
journal = {Digital Health},
year = {2026},
month = oct,
doi = {10.1177/20552076261500454},
url = {https://doi.org/10.1177/20552076261500454},
note = {CC BY},
}
RIS
TY - JOUR
TI - OncoAgentic-RAG: An evidence-grounded multi-agent hybrid retrieval-augmented system for patient-centric clinical oncology question answering
AU - S Jayasrilakshmi
AU - Ansuman Mahapatra
AU - Ankit Kujur
AU - Choudhari Aniket Baban
AU - Aditya Kumar Singh
AU - Sachin Nandan Mohanty
PY - 2026
DA - 2026/10/23
JO - Digital Health
DO - 10.1177/20552076261500454
UR - https://doi.org/10.1177/20552076261500454
AB - Objective General-purpose large language models (LLMs) show promise for medical question answering (QA) but face oncology-specific challenges, including hallucination, weak evidence attribution, and limited handling of sensitive patient queries. This study develops an evidence-grounded, patient-oriented oncology QA system integrating hybrid retrieval and specialised multi-agent processing to improve evidence attribution, safety, and empathetic communication. Methods The proposed OncoAgentic-RAG is a collaborative multi-agent hybrid retrieval-augmented generation framework for oncology QA. Its knowledge base comprises standard oncology textbooks and clinical guidelines. Dense ChromaDB and BM25 sparse retrieval are combined for evidence retrieval and fusion in a LangGraph-based multi-agent architecture that incorporates crisis detection, intent orchestration, medical knowledge retrieval, treatment guidance, empathy-aware generation, clinical verification and patient-friendly response synthesis. The system was evaluated on the 1,015-question OncoIntentQA benchmark using retrieval, semantic, faithfulness, relevance, safety, completeness, empathy, and user-friendliness measures, with comparisons against NoRAG, BaselineRAG, and AgenticRAG. Generated responses were additionally assessed by clinical experts for safety and appropriateness. Results OncoAgentic-RAG achieved Precision@5 of 0.8120, Recall@5 of 0.7980, BERTScore F1 of 0.7634, Faithfulness of 0.8146, and Answer Relevance of 0.8625. It achieved a weighted S.C.O.P.E. (Safety, Completeness, Originality, Precision, and Efficiency) score of 4.68/5.0, Empathy Composite of 0.9180, and User-Friendliness of 0.9580. Clinical expert assessment provided additional evaluation of response safety and content. Conclusions Integrating hybrid retrieval with specialised multi-agent processing improved evidence-grounded, source-traceable, and patient-friendly oncology QA within the evaluated benchmark. The findings demonstrate the technical utility of OncoAgentic-RAG for patient-oriented oncology information, while prospective clinical and safety validation remains necessary.
SN - 2055-2076
ER -