{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,18]],"date-time":"2026-08-18T04:38:23Z","timestamp":1787027903859,"version":"build-2736575974"},"reference-count":25,"publisher":"MDPI AG","issue":"6","license":[{"start":{"date-parts":[[2026,6,9]],"date-time":"2026-06-09T00:00:00Z","timestamp":1780963200000},"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>Large language models (LLMs) are increasingly explored for clinical use. However, the extent to which such models can reliably support physicians in reporting, staging, and the assessment of classification remains an active area of research. This study aimed to evaluate and compare multiple LLMs for automated PROMISE V2 classification for prostate cancer. A total of 126 unambiguous German-language PSMA PET\/CT text reports were retrospectively analyzed, with reference standards established by expert consensus based on image interpretation and the original report text. Five LLMs (GPT-5.4, DeepSeek-V3.2, Claude Sonnet 4.6, Gemini 3 Flash and Grok 4) were assessed using two English-language prompting strategies of varying complexity. Agreement with the reference standard served as the primary endpoint. Performance varied in the short-prompt setting (36.5\u201379.4%) but improved consistently with the long prompt (74.6\u201386.5%), with Gemini 3 Flash achieving the highest agreement. Across PROMISE V2 subcategories, agreement rates were high (miT: 81.0\u201392.1%, miN: 92.9\u201396.0%, miM: 92.9\u201395.2%), despite inter-model differences. In conclusion, contemporary LLMs demonstrate promising performance in deriving PROMISE V2 scores from unambiguous original report texts, particularly when guided by detailed prompts.<\/jats:p>","DOI":"10.3390\/curroncol33060349","type":"journal-article","created":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T00:58:12Z","timestamp":1781053092000},"page":"349","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":1,"title":["Automated PROMISE V2 Scoring from PSMA PET\/CT Reports Using Large Language Models: A Comparative Evaluation of Prompt Design and Model Performance"],"prefix":"10.3390","volume":"33","author":[{"given":"Tilman","family":"Speicher","sequence":"first","affiliation":[{"name":"Department of Nuclear Medicine, Saarland University-Medical Center, 66421 Homburg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-3744-8289","authenticated-orcid":false,"given":"Isa Ethem","family":"Demirkol","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Saarland University-Medical Center, 66421 Homburg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Arne","family":"Blickle","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Saarland University-Medical Center, 66421 Homburg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-2106-9548","authenticated-orcid":false,"given":"Moritz B.","family":"Bastian","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Saarland University-Medical Center, 66421 Homburg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1679-8080","authenticated-orcid":false,"given":"Stephan","family":"Maus","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Saarland University-Medical Center, 66421 Homburg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Andrea","family":"Schaefer-Schuler","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Saarland University-Medical Center, 66421 Homburg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0361-2833","authenticated-orcid":false,"given":"Mark","family":"Bartholom\u00e4","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Saarland University-Medical Center, 66421 Homburg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1522-8860","authenticated-orcid":false,"given":"Caroline","family":"Burgard","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Saarland University-Medical Center, 66421 Homburg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4110-3375","authenticated-orcid":false,"given":"Samer","family":"Ezziddin","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Saarland University-Medical Center, 66421 Homburg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2985-4099","authenticated-orcid":false,"given":"Florian","family":"Rosar","sequence":"additional","affiliation":[{"name":"Department of Nuclear Medicine, Saarland University-Medical Center, 66421 Homburg, Germany"},{"name":"Department of Nuclear Medicine, Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg and Universit\u00e4tsklinikum Erlangen, 91054 Erlangen, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,6,9]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"191","DOI":"10.1016\/j.eururo.2023.04.021","article-title":"2022 Update on Prostate Cancer Epidemiology and Risk Factors\u2014A Systematic Review","volume":"84","author":"Bergengren","year":"2023","journal-title":"Eur. Urol."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"596","DOI":"10.2967\/jnumed.120.257238","article-title":"Current and Emerging Clinical Applications of PSMA PET Diagnostic Imaging for Prostate Cancer","volume":"62","author":"Farolfi","year":"2021","journal-title":"J. Nucl. Med."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"1091","DOI":"10.1056\/NEJMoa2107322","article-title":"Lutetium-177\u2013PSMA-617 for Metastatic Castration-Resistant Prostate Cancer","volume":"385","author":"Sartor","year":"2021","journal-title":"N. Engl. J. Med."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"797","DOI":"10.1016\/S0140-6736(21)00237-3","article-title":"[177Lu]Lu-PSMA-617 versus cabazitaxel in patients with metastatic castration-resistant prostate cancer (TheraP): A randomised, open-label, phase 2 trial","volume":"397","author":"Hofman","year":"2021","journal-title":"Lancet"},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"1075","DOI":"10.1007\/s00259-021-05525-7","article-title":"177Lu-PSMA-617 radioligand therapy of metastatic castration-resistant prostate cancer: Initial 254-patient results from a prospective registry (REALITY Study)","volume":"49","author":"Khreish","year":"2021","journal-title":"Eur. J. Nucl. Med."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1016\/j.urology.2019.04.004","article-title":"The Impact of 68Ga-PSMA PET\/CT and PET\/MRI on the Management of Prostate Cancer","volume":"130","author":"Hoffmann","year":"2019","journal-title":"Urology"},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"107","DOI":"10.1016\/j.prnil.2020.07.008","article-title":"Performance of Ga-68 PSMA PET\/CT for diagnosis and grading of local prostate cancer","volume":"9","author":"Kwan","year":"2021","journal-title":"Prostate Int."},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Wang, R., Shen, G., Huang, M., and Tian, R. (2021). The Diagnostic Role of 18F-Choline, 18F-Fluciclovine and 18F-PSMA PET\/CT in the Detection of Prostate Cancer with Biochemical Recurrence: A Meta-Analysis. Front. Oncol., 11.","DOI":"10.3389\/fonc.2021.684629"},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"240","DOI":"10.1038\/s41391-022-00544-3","article-title":"The impact of PSMA PET on the treatment and outcomes of men with biochemical recurrence of prostate cancer: A systematic review and meta-analysis","volume":"26","author":"Pozdnyakov","year":"2023","journal-title":"Prostate Cancer Prostatic Dis."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"242","DOI":"10.1016\/j.euf.2024.11.006","article-title":"Better Oncological Outcomes After Prostate-specific Membrane Antigen Positron Emission Tomography-guided Salvage Radiotherapy Following Prostatectomy","volume":"11","author":"Zamboglou","year":"2025","journal-title":"Eur. Urol. Focus"},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"405","DOI":"10.1016\/j.eururo.2023.02.002","article-title":"Second Version of the Prostate Cancer Molecular Imaging Standardized Evaluation Framework Including Response Evaluation for Clinical Trials (PROMISE V2)","volume":"83","author":"Seifert","year":"2023","journal-title":"Eur. Urol."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"1930","DOI":"10.1038\/s41591-023-02448-8","article-title":"Large language models in medicine","volume":"29","author":"Thirunavukarasu","year":"2023","journal-title":"Nat. Med."},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Sallam, M. (2023). ChatGPT Utility in Healthcare Education, Research, and Practice: Systematic Review on the Promising Perspectives and Valid Concerns. Healthcare, 11.","DOI":"10.3390\/healthcare11060887"},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"299","DOI":"10.1007\/s10462-024-10921-0","article-title":"Large language models in medical and healthcare fields: Applications, advances, and challenges","volume":"57","author":"Wang","year":"2024","journal-title":"Artif. Intell. Rev."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"2792","DOI":"10.7150\/ijms.111780","article-title":"Large Language Models in Medicine: Applications, Challenges, and Future Directions","volume":"22","author":"Yu","year":"2025","journal-title":"Int. J. Med. Sci."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"1013","DOI":"10.1007\/s10115-024-02232-1","article-title":"Automatic regex synthesis methods for english: A comparative analysis","volume":"67","author":"Tariq","year":"2025","journal-title":"Knowl. Inf. Syst."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"598","DOI":"10.1016\/j.jpurol.2023.05.018","article-title":"ChatGPT and large language model (LLM) chatbots: The current state of acceptability and a proposal for guidelines on utilization in academic medicine","volume":"19","author":"Kim","year":"2023","journal-title":"J. Pediatr. Urol."},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"e251844","DOI":"10.1148\/radiol.251844","article-title":"A Prospective Controlled Trial of Large Language Model-based Simplification of Oncologic CT Reports for Patients with Cancer","volume":"317","author":"Prucker","year":"2025","journal-title":"Radiology"},{"key":"ref_19","doi-asserted-by":"crossref","unstructured":"Mergen, M., Spitzl, D., Eiber, M., Braren, R.F., and Steinhelfer, L. (2025). LLM-powered TNM staging of neuroendocrine tumors from PET\/CT reports. BMC Med. Imaging, 26.","DOI":"10.1186\/s12880-025-02092-3"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"106053","DOI":"10.1016\/j.ijmedinf.2025.106053","article-title":"LLM-powered breast cancer staging from PET\/CT reports: A comparative performance study","volume":"204","author":"Spitzl","year":"2025","journal-title":"Int. J. Med. Inform."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"93","DOI":"10.1007\/s11307-025-02072-7","article-title":"Staging Prostate Cancer with AI: A Comparative Study of Large Language Models and Expert Interpretation on PSMA PET-CT Reports","volume":"28","author":"Ismayilov","year":"2025","journal-title":"Mol. Imaging Biol."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"51","DOI":"10.1007\/s11604-024-01643-y","article-title":"Preliminary assessment of TNM classification performance for pancreatic cancer in Japanese radiology reports using GPT-4","volume":"43","author":"Suzuki","year":"2024","journal-title":"Jpn. J. Radiol."},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"324","DOI":"10.1055\/a-2707-4584","article-title":"Performance of a large language model in lymphoma stage assignment based on written PET\/CT reports","volume":"64","author":"Voltin","year":"2025","journal-title":"Nuklearmedizin"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"565","DOI":"10.1200\/JCO.2026.44.7_suppl.565","article-title":"Artificial Intelligence (AI) as a practical decision support tool for oncologists: Genitourinary (GU) cancer cases","volume":"44","author":"Jahanzep","year":"2026","journal-title":"J. Clin. Oncol."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"ooaf128","DOI":"10.1093\/jamiaopen\/ooaf128","article-title":"A comparative performance analysis of regular expressions and a large language model-based approach to extract the BI-RADS score from radiological reports","volume":"8","author":"Lerch","year":"2025","journal-title":"JAMIA Open"}],"container-title":["Current Oncology"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1718-7729\/33\/6\/349\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,10]],"date-time":"2026-06-10T02:13:24Z","timestamp":1781057604000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1718-7729\/33\/6\/349"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,9]]},"references-count":25,"journal-issue":{"issue":"6","published-online":{"date-parts":[[2026,6]]}},"alternative-id":["curroncol33060349"],"URL":"https:\/\/doi.org\/10.3390\/curroncol33060349","relation":{},"ISSN":["1718-7729"],"issn-type":[{"value":"1718-7729","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,9]]}}}