MCW Cancer Center Pioneers AI-Driven Clinical Trial Matching with OncoLLM
核心洞察
The Medical College of Wisconsin (MCW) Cancer (搜索) Center is piloting OncoLLM (搜索), an AI platform, to automate and improve patient-trial matching, potentially enhancing access to clinical trials.
OncoLLM (搜索), developed by Triomics (搜索), uses natural language processing to analyze unstructured clinical data, matching or exceeding the accuracy of medical professionals in identifying suitable trials.
MCW's secure, HIPAA-compliant implementation of OncoLLM (搜索) ensures patient privacy while streamlining trial matching across various Disease Oriented Teams (DOTs).
The Medical College of Wisconsin (MCW) Cancer (搜索) Center is leading the way in using artificial intelligence (AI) to transform clinical trial matching. A pilot study at MCW leverages OncoLLM (搜索), a platform developed by Triomics (搜索), to automate the complex process of connecting cancer patients with appropriate clinical trials. This initiative aims to improve treatment outcomes by ensuring that no patient is overlooked for potentially life-saving cancer treatments.
AI-Powered Precision in Trial Matching
OncoLLM (搜索) employs a large language model (LLM) similar to ChatGPT but is specifically trained on clinical data. It uses advanced natural language processing to interpret and extract meaningful information from unstructured data sources such as clinician notes, lab results, and imaging reports. This data is then translated into a structured format that can be systematically analyzed to match patients with relevant clinical trials.
According to Anai Kothari, MD, MS, Assistant Professor of Surgery and Co-Director of the Center’s Geospatial, Epidemiology and Outcomes Shared Resource, early results are promising. "Since implementing the OncoLLM (搜索) platform in July, we’ve been able to evaluate all the patients in the Disease Oriented Teams (DOTs) that are part of our pilot project," said Dr. Kothari. "This means that every patient who has an upcoming appointment will be rigorously assessed for eligibility for open clinical trials at the MCW Cancer (搜索) Center."
Ensuring Privacy and Accuracy
Brad Taylor, MBA, FAMIA, Chief Research Informatics Officer in the CTSI, emphasized the importance of privacy in handling sensitive clinical data. "Instead of using ChatGPT or some other commercial source, we train OncoLLM (搜索) in a secure, privacy-preserving manner. This method ensures that the LLM is isolated from the internet, in a HIPAA-secure private platform," said Taylor. A recent research paper co-authored by Dr. Kothari and Taylor demonstrated that OncoLLM not only outperforms existing AI models like GPT-3.5 but also matches the accuracy of qualified medical professionals in patient-trial matching.
Comprehensive Patient Support
Recognizing that matching patients to trials is only the first step, the MCW Cancer (搜索) Center has also launched a new Patient Navigation Program. This program, led by Adrena Luckett, helps patients navigate the logistical challenges of enrolling in a clinical trial, such as understanding trial requirements, coordinating appointments, managing transportation, and addressing financial concerns. This holistic approach ensures that patients have both access to trials and the resources needed to participate fully.
Expanding the Initiative
The pilot study is currently testing OncoLLM (搜索) in the gastrointestinal, genitourinary, breast, and thoracic DOTs. Based on the positive early results, the team plans to implement the platform across all of the Cancer (搜索) Center’s DOTs and eventually make it accessible to investigators across the institution. This expansion is made possible by MCW's robust infrastructure for organizing and structuring clinical data, which allows researchers to effectively harness data for impactful research.