Scholarlink

Objectives <p>Gastrointestinal stromal tumors (GISTs) are molecularly heterogeneous neoplasms whose management depends on individualized, multidisciplinary decision-making. While multidisciplinary tumor boards (MTBs) represent the standard of care, access remains limited in many clinical settings. This study evaluates the performance of two large language models in generating GIST MTB recommendations and assesses their agreement with expert MTB decisions using predefined clinical evaluation criteria.</p> Materials and methods <p>This retrospective single-center study included 99 GIST cases discussed at an institutional MTB. A structured prompt was developed to extract clinical variables and generate treatment recommendations. ChatGPT-5 and Qwen3 were independently evaluated across five predefined domains: diagnostic recommendations, therapeutic modalities, treatment sequence and timing, systemic therapy regimen selection, and clinical contextualization. Two expert reviewers scored all outputs in a blinded fashion. Normalized scores, inter-model comparisons, perfect-case rates, and inter-rater agreement were analyzed.</p> Results <p>Both models demonstrated high concordance with expert MTB recommendations, with mean total normalized scores of 0.901 for ChatGPT-5 and 0.875 for Qwen3, without a significant difference between models (<i>p &gt;</i> 0.05). Perfect agreement was observed in 52.5% of ChatGPT-5 cases and 48.5% of Qwen3 cases (<i>p &gt;</i> 0.05). Diagnostic recommendations scored significantly lower than all other domains in both models (all adjusted <i>p &lt;</i> 0.05). Overall inter-rater agreement was almost perfect (weighted Cohen’s kappa=0.978).</p> Conclusions <p>Both models demonstrated high agreement with expert GIST MTB recommendations, with no significant performance difference between them. Diagnostic reasoning represented the weakest domain, reflecting the challenge of reconstructing context-dependent workup decisions from tumor board documentation. These findings support a potential assistive role for LLMs in GIST MTB workflows, while underscoring the continued necessity of expert oversight.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Precision oncology meets Generative AI: assessing large language models in multidisciplinary GIST tumor boards

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年8月1日

查看全文
  • Reza Dehdab,
  • Judith Herrmann,
  • Fiona Mankertz,
  • Patrick Ghibes,
  • Nour Maalouf,
  • Sebastian Werner,
  • Jens Strohäker,
  • Katrin Benzler,
  • Lars Zender,
  • Saif Afat,
  • Konstantin Nikolaou,
  • Christoph K. W. Deinzer

摘要

Objectives

Gastrointestinal stromal tumors (GISTs) are molecularly heterogeneous neoplasms whose management depends on individualized, multidisciplinary decision-making. While multidisciplinary tumor boards (MTBs) represent the standard of care, access remains limited in many clinical settings. This study evaluates the performance of two large language models in generating GIST MTB recommendations and assesses their agreement with expert MTB decisions using predefined clinical evaluation criteria.

Materials and methods

This retrospective single-center study included 99 GIST cases discussed at an institutional MTB. A structured prompt was developed to extract clinical variables and generate treatment recommendations. ChatGPT-5 and Qwen3 were independently evaluated across five predefined domains: diagnostic recommendations, therapeutic modalities, treatment sequence and timing, systemic therapy regimen selection, and clinical contextualization. Two expert reviewers scored all outputs in a blinded fashion. Normalized scores, inter-model comparisons, perfect-case rates, and inter-rater agreement were analyzed.

Results

Both models demonstrated high concordance with expert MTB recommendations, with mean total normalized scores of 0.901 for ChatGPT-5 and 0.875 for Qwen3, without a significant difference between models (p > 0.05). Perfect agreement was observed in 52.5% of ChatGPT-5 cases and 48.5% of Qwen3 cases (p > 0.05). Diagnostic recommendations scored significantly lower than all other domains in both models (all adjusted p < 0.05). Overall inter-rater agreement was almost perfect (weighted Cohen’s kappa=0.978).

Conclusions

Both models demonstrated high agreement with expert GIST MTB recommendations, with no significant performance difference between them. Diagnostic reasoning represented the weakest domain, reflecting the challenge of reconstructing context-dependent workup decisions from tumor board documentation. These findings support a potential assistive role for LLMs in GIST MTB workflows, while underscoring the continued necessity of expert oversight.