Assessment of Large Language Models in Colorectal Cancer Multidisciplinary Tumor Board Decision-Making: A Retrospective Single-Center Comparison of Guideline-Integrated General-Purpose vs. Domain-Specialized Models
Current Oncology · 26.05.2026 · 33 (6) · 309
DOI: 10.3390/curroncol33060309 ↗Kurz gefasst
Vergleichsstudie in der Zeitschrift Current Oncology, online erschienen am 26.05.2026. Verfasst von Aydan Farzaliyeva und 4 weiteren Autor:innen. Beteiligte Einrichtung: Division of Medical Oncology, Department of Internal Medicine, Faculty of Medicine, Baskent University, 06490 Ankara, Türkiye. Hauptthemen laut MEDLINE: Colorectal Neoplasms, Clinical Decision-Making. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 27 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Background
Large language models (LLMs) are emerging as clinical decision-support tools in oncology, yet their ability to generate reliable treatment recommendations in real-world multidisciplinary tumor board (MTB) settings remains uncertain, particularly for complex colorectal cancer (CRC).
Methods
In this retrospective study, 300 consecutive adult CRC cases discussed at a tertiary MTB were evaluated. Standardized de-identified case summaries were independently submitted to Gemini 2.5 (general-purpose, guideline-integrated) and MedGemma 27B (domain-specialized; T = 0.0 and T = 1.0). Concordance with MTB decisions was assessed using weighted Cohen’s kappa (κ), accuracy, F1 score, and recall. Safety was adjudicated by blinded senior MTB members using a three-tier risk framework. Importantly, Gemini 2.5 was evaluated in a guideline-integrated setting, whereas MedGemma operated without external guideline retrieval, introducing a predefined asymmetry in knowledge augmentation.
Results
Gemini 2.5 demonstrated substantial agreement (κ = 0.792, p < 0.001), the highest exact concordance (85.0%), and a clinical safety rate of 83.7%, significantly outperforming MedGemma (p < 0.001). MedGemma showed moderate agreement at T = 0.0 (κ = 0.566; accuracy = 70.3%) and modest improvement at T = 1.0 (κ = 0.610; accuracy = 73.5%; p = 0.038), with lower safety rates (62.3% and 62.7%; p < 0.001). Discordance predominantly occurred in clinically complex scenarios, including surgical/interventional planning, active surveillance, recurrent disease management, and patients with compromised performance status.
Conclusions
A guideline-integrated general-purpose LLM demonstrated superior concordance and safety compared with a domain-specialized model operating without external retrieval, supporting its adjunctive use within MTBs while preserving expert clinical judgment.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Stichwörter der Autor:innen
Autor:innen und Institutionen
- Division of Medical Oncology, Department of Internal Medicine, Faculty of Medicine, Baskent University, 06490 Ankara, Türkiye.
- Department of Medical Informatics, Faculty of Medicine, Baskent University, 06490 Ankara, Türkiye.
ORCID-Profile: Aydan Farzaliyeva, Mehmet Nezir Ramazanoglu, Zafer Akcali
Zitieren
APA
Farzaliyeva, A., Ramazanoglu, M. N., Oguz, A., Altundag, O., & Akcali, Z. (2026). Assessment of Large Language Models in Colorectal Cancer Multidisciplinary Tumor Board Decision-Making: A Retrospective Single-Center Comparison of Guideline-Integrated General-Purpose vs. Domain-Specialized Models. Current Oncology, 33(6), 309. https://doi.org/10.3390/curroncol33060309
Vancouver
Farzaliyeva A, Ramazanoglu MN, Oguz A, Altundag O, Akcali Z. Assessment of Large Language Models in Colorectal Cancer Multidisciplinary Tumor Board Decision-Making: A Retrospective Single-Center Comparison of Guideline-Integrated General-Purpose vs. Domain-Specialized Models. Current Oncology. 2026;33(6):309. doi:10.3390/curroncol33060309
BibTeX
@article{Farzaliyeva2026,
author = {Farzaliyeva, Aydan and Ramazanoglu, Mehmet Nezir and Oguz, Arzu and Altundag, Ozden and Akcali, Zafer},
title = {Assessment of Large Language Models in Colorectal Cancer Multidisciplinary Tumor Board Decision-Making: A Retrospective Single-Center Comparison of Guideline-Integrated General-Purpose vs. Domain-Specialized Models},
journal = {Current Oncology},
year = {2026},
volume = {33},
number = {6},
pages = {309},
doi = {10.3390/curroncol33060309}
}
EUVIMED