{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,14]],"date-time":"2026-09-14T21:08:17Z","timestamp":1789420097824,"version":"build-2803163510"},"reference-count":27,"publisher":"MDPI AG","issue":"6","license":[{"start":{"date-parts":[[2026,5,26]],"date-time":"2026-05-26T00:00:00Z","timestamp":1779753600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Current Oncology"],"abstract":"<jats:p>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\u2019s kappa (\u03ba), 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 (\u03ba = 0.792, p &lt; 0.001), the highest exact concordance (85.0%), and a clinical safety rate of 83.7%, significantly outperforming MedGemma (p &lt; 0.001). MedGemma showed moderate agreement at T = 0.0 (\u03ba = 0.566; accuracy = 70.3%) and modest improvement at T = 1.0 (\u03ba = 0.610; accuracy = 73.5%; p = 0.038), with lower safety rates (62.3% and 62.7%; p &lt; 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.<\/jats:p>","DOI":"10.3390\/curroncol33060309","type":"journal-article","created":{"date-parts":[[2026,5,26]],"date-time":"2026-05-26T11:02:51Z","timestamp":1779793371000},"page":"309","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"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"],"prefix":"10.3390","volume":"33","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8504-4798","authenticated-orcid":false,"given":"Aydan","family":"Farzaliyeva","sequence":"first","affiliation":[{"name":"Division of Medical Oncology, Department of Internal Medicine, Faculty of Medicine, Baskent University, 06490 Ankara, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-4976-182X","authenticated-orcid":false,"given":"Mehmet Nezir","family":"Ramazanoglu","sequence":"additional","affiliation":[{"name":"Division of Medical Oncology, Department of Internal Medicine, Faculty of Medicine, Baskent University, 06490 Ankara, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Arzu","family":"Oguz","sequence":"additional","affiliation":[{"name":"Division of Medical Oncology, Department of Internal Medicine, Faculty of Medicine, Baskent University, 06490 Ankara, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ozden","family":"Altundag","sequence":"additional","affiliation":[{"name":"Division of Medical Oncology, Department of Internal Medicine, Faculty of Medicine, Baskent University, 06490 Ankara, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2473-4431","authenticated-orcid":false,"given":"Zafer","family":"Akcali","sequence":"additional","affiliation":[{"name":"Division of Medical Oncology, Department of Internal Medicine, Faculty of Medicine, Baskent University, 06490 Ankara, T\u00fcrkiye"},{"name":"Department of Medical Informatics, Faculty of Medicine, Baskent University, 06490 Ankara, T\u00fcrkiye"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,5,26]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"51","DOI":"10.1016\/S1470-2045(24)00600-4","article-title":"Colorectal cancer incidence trends in younger versus older adults: An analysis of population-based cancer registry data","volume":"26","author":"Sung","year":"2025","journal-title":"Lancet Oncol."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"365","DOI":"10.1007\/s12029-023-00967-8","article-title":"The Fidelity of Artificial Intelligence to Multidisciplinary Tumor Board Recommendations for Patients with Gastric Cancer: A Retrospective Study","volume":"55","author":"Park","year":"2024","journal-title":"J. Gastrointest. Cancer"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"Umihanic, S., Osmanovic, H., Selak, N., Kopric, D., Huseinbasic, A., Sehic-Kozica, E., Babic, B., and Umihanic, F. (2025). Evaluating the Concordance Between ChatGPT and Multidisciplinary Teams in Breast Cancer Treatment Planning: A Study from Bosnia and Herzegovina. J. Clin. Med., 14.","DOI":"10.3390\/jcm14186460"},{"key":"ref_4","doi-asserted-by":"crossref","unstructured":"Schmidl, B., H\u00fctten, T., Pigorsch, S., St\u00f6gbauer, F., Hoch, C.C., Hussain, T., Wollenberg, B., and Wirth, M. (2024). Assessing the role of advanced artificial intelligence as a tool in multidisciplinary tumor board decision-making for recurrent\/metastatic head and neck cancer cases\u2014The first study on ChatGPT 4o and a comparison to ChatGPT 4.0. Front. Oncol., 14.","DOI":"10.3389\/fonc.2024.1455413"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"248","DOI":"10.1007\/s00432-025-06304-9","article-title":"The imitation game: Large language models versus multidisciplinary tumor boards: Benchmarking AI against 21 sarcoma centers from the ring trial","volume":"151","author":"Li","year":"2025","journal-title":"J. Cancer Res. Clin. Oncol."},{"key":"ref_6","first-page":"e79570","article-title":"A Narrative Review on the Role of Artificial Intelligence (AI) in Colorectal Cancer Management","volume":"17","author":"Babu","year":"2025","journal-title":"Cureus"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Yin, Z., Yao, C., Zhang, L., and Qi, S. (2023). Application of artificial intelligence in diagnosis and treatment of colorectal cancer: A novel Prospect. Front. Med., 10.","DOI":"10.3389\/fmed.2023.1128084"},{"key":"ref_8","unstructured":"Sellergren, A., Kazemzadeh, S., Jaroensri, T., Kiraly, A., Traverse, M., Kohlberger, T., Xu, S., Jamil, F., Hughes, C., and Lau, C. (2025). Medgemma technical report. arXiv."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"8","DOI":"10.4274\/balkanmedj.2017.0966","article-title":"Patient Privacy in the Era of Big Data","volume":"35","author":"Kayaalp","year":"2018","journal-title":"Balk. Med. J."},{"key":"ref_10","unstructured":"Ki\u015fisel Verileri Koruma, K. (2025, December 01). Ki\u015fisel Verilerin Silinmesi, Yok Edilmesi veya Anonim Hale Getirilmesi, Available online: https:\/\/www.kvkk.gov.tr\/Icerik\/2038\/kisisel-verilerin-silinmesi-yok-edilmesi-veya-anonim-hale-getirilmesi."},{"key":"ref_11","unstructured":"(2025, December 01). National Cancer, I. Colon Cancer Treatment (PDQ\u00ae)\u2013Health Professional Version, Available online: https:\/\/www.cancer.gov\/types\/colorectal\/hp\/colon-treatment-pdq."},{"key":"ref_12","unstructured":"(2025, December 01). National Cancer, I. Rectal Cancer Treatment (PDQ\u00ae)\u2013Health Professional Version, Available online: https:\/\/www.cancer.gov\/types\/colorectal\/hp\/rectal-treatment-pdq."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"159","DOI":"10.2307\/2529310","article-title":"The measurement of observer agreement for categorical data","volume":"33","author":"Landis","year":"1977","journal-title":"Biometrics"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"112980","DOI":"10.1016\/j.isci.2025.112980","article-title":"Comprehensive application of artificial intelligence in colorectal cancer: A review","volume":"28","author":"Sun","year":"2025","journal-title":"iScience"},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"832","DOI":"10.1093\/jamia\/ocaa334","article-title":"Comparison of an oncology clinical decision-support system\u2019s recommendations with actual treatment decisions","volume":"28","author":"Suwanvecho","year":"2021","journal-title":"J. Am. Med. Inform. Assoc. JAMIA"},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Aikemu, B., Xue, P., Hong, H., Jia, H., Wang, C., Li, S., Huang, L., Ding, X., Zhang, H., and Cai, G. (2020). Artificial Intelligence in Decision-Making for Colorectal Cancer Treatment Strategy: An Observational Study of Implementing Watson for Oncology in a 250-Case Cohort. Front. Oncol., 10.","DOI":"10.3389\/fonc.2020.594182"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"356","DOI":"10.1111\/ans.18749","article-title":"Conversational artificial intelligence (chatGPTTM) in the management of complex colorectal cancer patients: Early experience","volume":"94","author":"Choo","year":"2024","journal-title":"ANZ J. Surg."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"zraf040","DOI":"10.1093\/bjsopen\/zraf040","article-title":"Concordance of ChatGPT artificial intelligence decision-making in colorectal cancer multidisciplinary meetings: Retrospective study","volume":"9","author":"Chatziisaak","year":"2025","journal-title":"BJS Open"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"41","DOI":"10.1097\/DCR.0000000000003504","article-title":"Comparing the Management Recommendations of Large Language Model and Colorectal Cancer Multidisciplinary Team: A Pilot Study","volume":"68","author":"Horesh","year":"2025","journal-title":"Dis. Colon Rectum"},{"key":"ref_20","doi-asserted-by":"crossref","unstructured":"Zhou, S., and Yu, S. (2025). High-throughput biomedical relation extraction for semi-structured web articles empowered by large language models. BMC Med. Inform. Decis. Mak., 25.","DOI":"10.1186\/s12911-025-03204-3"},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Bhatnagar, A., Scherer, R., Samico, G.A., Muralidhar, R., Gutkind, N.E., Palazoni, V., Medeiros, F.A., and Swaminathan, S.S. (2025). Identification of Risk Factors for Glaucoma Progression in Free-Text Clinical Notes Using a Local Large Language Model. medRxiv.","DOI":"10.1101\/2025.09.26.25336746"},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Zhao, X., Blotske, K., Cargile, M., Tilley, A., Murray, B., Gao, Y., Henry, K., Smith, S.E., Barreto, E.F., and Bauer, S. (2025). Rx-LLM: A benchmarking suite to evaluate safe large language model performance for medication-related tasks. medRxiv.","DOI":"10.64898\/2025.12.01.25341004"},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Blotske, K., Zhao, X., Henry, K., Gao, Y., Tilley, A., Cargile, M., Murray, B., Smith, S.E., Barreto, E.F., and Bauer, S. (2025). Drug-drug interaction identification using large language models. medRxiv.","DOI":"10.64898\/2025.12.03.25341549"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"e87193","DOI":"10.2196\/87193","article-title":"Multimodal Large Language Models for Cystoscopic Image Interpretation and Bladder Lesion Classification: Comparative Study","volume":"28","author":"Shih","year":"2026","journal-title":"J. Med. Internet Res."},{"key":"ref_25","unstructured":"Modi, M., Krull, J.E., Johnson, D., Wang, X., Gauntner, T.D., Li, M., Cheng, H., Ma, A., Zhang, P., and Stover, D.G. (2026). Why Large Language Models\u2019 Clinical Reasoning Fails: Insights from Explainable Deep Learning. medRxiv."},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Chitoran, E., Rotaru, V., Gelal, A., Ionescu, S.O., Gullo, G., Stefan, D.C., and Simion, L. (2025). Using Artificial Intelligence to Develop Clinical Decision Support Systems-The Evolving Road of Personalized Oncologic Therapy. Diagnostics, 15.","DOI":"10.3390\/diagnostics15182391"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"107342","DOI":"10.1016\/j.lungcan.2023.107342","article-title":"Clinical score for colorectal cancer patients with lung-limited metastases undergoing surgical resection: Meta-Lung Score","volume":"184","author":"Ziranu","year":"2023","journal-title":"Lung Cancer"}],"container-title":["Current Oncology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1718-7729\/33\/6\/309\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,26]],"date-time":"2026-05-26T11:50:28Z","timestamp":1779796228000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1718-7729\/33\/6\/309"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,26]]},"references-count":27,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,6]]}},"alternative-id":["curroncol33060309"],"URL":"https:\/\/doi.org\/10.3390\/curroncol33060309","relation":{},"ISSN":["1718-7729"],"issn-type":[{"value":"1718-7729","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,26]]}}}