[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-352958-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-352958-en":131},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","benchmarking-large-language-models-and-prompt-engineering-strategies-in-microsatellite-instability-cancers-evaluation-study","Benchmarking Large Language Models and Prompt Engineering Strategies in Microsatellite Instability Cancers: Evaluation Study","","The reliability of general-purpose large language models (LLMs) for complex clinical tasks in microsatellite instability (MSI) cancers remains insufficiently characterized, creating unevaluated risks for patient safety. The study develops and validates MSIC-Bench, a two-tier MSI cancer benchmark covering both consensus and frontier knowledge. It evaluates multiple LLMs under distinct prompting strategies using accuracy, safety (honesty), error composition, and token usage.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/benchmarking-large-language-models-and-prompt-engineering-strategies-in-microsatellite-instability-cancers-evaluation-study/352958/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/benchmarking-large-language-models-and-prompt-engineering-strategies-in-microsatellite-instability-cancers-evaluation-study/352958.png","ImageObject",300,407,{"name":42,"@type":43},"Cipher","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-29","2026-09-22",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",12,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What gap does this study address in evaluating large language models for MSI cancer?","Question",{"text":63,"@type":64},"It targets the lack of a domain-specific benchmark to evaluate and optimize LLM performance across diverse MSI oncology tasks, which limits confidence in clinical safety and effectiveness.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"How is the MSIC-Bench benchmark constructed and what does it cover?",{"text":68,"@type":64},"MSIC-Bench is a 511-question two-tier benchmark derived from clinical guidelines and a curated knowledge base, covering both consensus and frontier knowledge.",{"name":70,"@type":61,"acceptedAnswer":71},"Which prompting and generation strategy performed best, and how did it affect safety?",{"text":72,"@type":64},"A domain-aligned retrieval-augmented generation (RAG) strategy improved accuracy substantially and shifted failures from knowledge deficits to retrieval failures. It also reduced high-risk fabrications by inducing more refusals, creating a safety-utility trade-off via false refusals.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},352958,1790372768,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,120,123,127],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":25,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":25,"category_name":129,"show_sort_weight":97,"slug":130},19,"General","general",{"code":4,"msg":82,"data":132},{"doc_id":79,"user_id":133,"nickname":42,"user_avatar":134,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":140,"language":141,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":12,"update_tm":145,"read_time":146,"preview_page_count":33,"preview_urls":147},687208528416,"https://ap-avatar.wpscdn.com/davatar_9964176cb1d06d4a9deccf72a44ae3dc","JOURNAL OF MEDICAL INTERNET RESEARCH Zhang et al  \nOriginal Paper  \nBenchmarking Large Language Models and Prompt  \nEngineering Strategies in Microsatellite Instability Cancers: Evaluation Study  \n\n| Yuxin Zhang* , BS; Jie Song* , BE; Cheng Bi, MSc; Xin Zheng, MSc; Zhichuan Xu, BE; Dan Cao, PhD; Bairong Shen, PhD |\n| --- |\n| Department of Medical Oncology, Institutes for Systems Genetics, Frontiers Science Center for Disease-Related Molecular Network, West China Hospital, Sichuan University, Chengdu, Sichuan, China\u003Cbr>*these authors contributed equally\u003Cbr>Corresponding Author:\u003Cbr>Bairong Shen, PhD\u003Cbr>Department of Medical Oncology, Institutes for Systems Genetics, Frontiers Science Center for Disease-Related Molecular Network\u003Cbr>West China Hospital, Sichuan University No 2222 Xinchuan Road, Gaoxin District Chengdu, Sichuan 610000\u003Cbr>China\u003Cbr>Phone: 86 15995854635\u003Cbr>Fax: 86 28 61528682\u003Cbr>[Email:](Email: Bairong.shen@scu.edu.cn)[ ](Email: Bairong.shen@scu.edu.cn)[Bairong.shen@scu.edu.cn](Email: Bairong.shen@scu.edu.cn)\u003Cbr>Abstract |\n\nBackground: The reliability of general-purpose large language models (LLMs) for complex clinical tasks in specialized domains such as microsatellite instability (MSI) cancers remains critically uncharacterized. The absence of a domain-specific benchmark to evaluate and guide the optimization of their capabilities across diverse clinical tasks poses unevaluated risks to patient safety.  \nObjective: This study aimed to develop and validate Microsatellite Instability Cancer Benchmark (MSIC-Bench), a novel, two-tiered benchmark for MSI cancer, covering both consensus and frontier knowledge. Using this framework, we aimed to systematically assess LLM performance across various prompting strategies, identify task-specific weaknesses, and reveal effective pathways for performance improvement.  \nMethods: We developed MSIC-Bench, a 511-question benchmark derived from clinical guidelines and a curated knowledge base. Three state-of-the-art LLMs (GPT-4o, [OpenAI], Gemini 2.5 Pro [Google], and Claude Opus 4 [Anthropic]) were evaluated using 4 prompting strategies, including vanilla, chain-of-thought, reflection of thoughts, and retrieval-augmented generation (RAG), under both multiple-choice and open-ended modalities. Performance was assessed on accuracy, safety (honesty), error composition, and token usage.  \nResults: LLMs demonstrated a significant “scaffolding effect,” with accuracy dropping substantially in open-ended scenarios. For non-RAG strategies, the primary failure mode was an internal knowledge deficit. The integration of RAG proved to be the most effective intervention. A domain-aligned RAG strategy not only significantly improved accuracy in complex decisionmaking tasks but also fundamentally shifted the system’s primary bottleneck from knowledge deficits to retrieval failures. In terms of safety, RAG induced a favorable shift from high-risk fabrications to safer refusals, though this introduced a safety-utility trade-off in the form of false refusals. Notably, our hybrid-RAG configuration, which combined both knowledge sources, demonstrated the most robust and generalizable performance across all tasks.  \nConclusions: Current LLMs lack the specialized knowledge required for reliable application in MSI oncology. A welldesigned RAG architecture is the pivotal intervention to address this gap. However, its success is not automatic; it transforms the nature of system failure, making retrieval precision and knowledge base quality the new critical determinants of performance and safety. Our findings establish a clear directive for developing trustworthy clinical artificial intelligence: focus must shift toward optimizing the retrieval component and curating high-quality, comprehensive knowledge sources . MSIC-Bench provides a robust framework to guide these future efforts.  \n[https://www.jmir.org/2026/1/e88614](https://www.jmir.org/2026/1/e88614) J Med Internet Res 2026 | vol. 28 | e88614 | p. 1  \n(p","cbCaieujlzMSE7yq","https://ap.wps.com/l/cbCaieujlzMSE7yq","pdf",3110849,17,"English","# Abstract\n# Introduction\n# Methods\n# Results\n# Conclusions","[{\"question\":\"What gap does this study address in evaluating large language models for MSI cancer?\",\"answer\":\"It targets the lack of a domain-specific benchmark to evaluate and optimize LLM performance across diverse MSI oncology tasks, which limits confidence in clinical safety and effectiveness.\"},{\"question\":\"How is the MSIC-Bench benchmark constructed and what does it cover?\",\"answer\":\"MSIC-Bench is a 511-question two-tier benchmark derived from clinical guidelines and a curated knowledge base, covering both consensus and frontier knowledge.\"},{\"question\":\"Which prompting and generation strategy performed best, and how did it affect safety?\",\"answer\":\"A domain-aligned retrieval-augmented generation (RAG) strategy improved accuracy substantially and shifted failures from knowledge deficits to retrieval failures. It also reduced high-risk fabrications by inducing more refusals, creating a safety-utility trade-off via false refusals.\"}]","Benchmarking Large Language Models and Prompt Engineering Strategies in Microsatellite Instability Cancers: Evaluation Study | PDF",1790102316,43,[148,152,156,160],{"page_no":22,"url":149,"width":150,"height":151},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/90e605a69da230c7019c235ac2fb232d.webp",983,1390,{"page_no":26,"url":153,"width":154,"height":155},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/b02233d98039ef19327de3c7a0fb1f5f.webp",982,1388,{"page_no":30,"url":157,"width":158,"height":159},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/12be03334fbc7831ffa3a1ee8fe971cf.webp",981,1387,{"page_no":33,"url":161,"width":162,"height":163},"https://res-academy.cache.wpscdn.com/upload/kdocshare/2026-11-16/29f8fc5b97faee32b4850064cb49e825.webp",980,1385]