(Discussions at ASCO Breakthrough, Singapore 2026)
Translating clinical knowledge into effective bedside care means turning theory into practical action. This involves observing patients closely, listening with empathy, and using clinical reasoning to solve health problems in real time. Implementing effective bedside care depends on several essential actions:
- Active Listening and Empathy: Clinicians should listen carefully, acknowledge patient feelings, and use clear language to help patients understand their care. Focused Examination: Clinicians should tailor physical exams to each patient rather than simply following standard routines. Integrating Technology: Using bedside tools like ultrasound helps check findings and improve exam skills in real time. Role Modeling: Teachers should explain their thinking and show good bedside manner and consent practices to trainees. In liver cancer care, artificial intelligence (AI) is a helpful tool for doctors and patients. AI supports decision-making but does not replace the doctor’s Viewpoint. AI models help provide more precise solutions for liver cancer.
Diagnostic Accuracy
- Imaging and Detection: AI can analyze CT and MRI images, detecting small tumors that might otherwise be missed. Some tools can even flag uncertain results for doctors to review. Early Screening: AI can help find liver cancer early in high-risk patients, even when tumors are very small. Treatment and Prognosis
- Personalized Care: AI can predict how patients will respond to different treatments, including new drug combinations. Radiation Targeting: AI tracks liver tumors during radiation, helping focus treatment on cancer and protect healthy tissue. Drug Discovery: AI speeds up the search for new drug targets and treatments.
Considerations for the Responsible Use of AI
AI is powerful, it has limits. For example, some AI does well with simple liver cancer cases but struggles with more complex ones. AI should help the cancer team, but doctors make the final decisions. This information will enable the provision of more tailored details regarding applicable AI tools or clinical trials.
Problem Statement: Clinical research teams currently manually match patients to trials, a process that takes approximately 25 minutes per trial. This approach restricts patient options and reduces clinical trial enrollment rates.
Proposed Solution: In 2023, a collaborative initiative between PCC and Massive Bio was launched to identify more eligible patients for biomarker-directed clinical trials. This was achieved by pre-screening all patients with next-generation sequencing (NGS) test results against all available cancer trials simultaneously. The American Society of Clinical Oncology (ASCO) has accepted the results generated by this AI technology for online publication.*
Is Now the Appropriate Time to Consider Clinical Trial Matching?
Transitioning from manual screening to AI-driven, patient-centric trial matching aims to address low enrollment rates. Integrating real-time biomarker data and genomic profiling enables the rapid identification of the most viable, targeted treatment protocols for individual patients.
Efforts to improve patient matching at ASCO Meetings focus on specific technological advancements and precision medicine platforms:
- AI-Driven Prescreening: Manual chart review requires approximately 25 minutes per trial. Recent studies indicate that AI and natural language processing (NLP) can reduce screening time by over 40 percent and increase potential patient eligibility for trials by up to twofold.
- Blue-Button Integration: Tools developed by organizations such as the American Cancer Society Cancer Action Network (ACS CAN) are designed to enable automated, single-step clinical trial matching directly within Electronic Health Records (EHR).
- Genomically Matched Therapies: ASCO highlights the importance of early predictive biomarker testing (such as PD-L1, HER2, KRAS, and RET mutations) to enable patients to receive highly specific targeted agents and antibody-drug conjugates (ADCs) instead of standard chemotherapy.
To identify current open trials for a specific patient in San Francisco, the official ClinicalTrials.gov registry may be utilized.
To further narrow down specific trials, the following patient information is required:
- Cancer type and stage
- Biomarker or genetic mutation status (e.g., EGFR, HER2, KRAS)
- Willingness to travel to academic cancer centers (such as UCSF) versus community clinics
Problem Statement: Clinical research teams manually match patients to trials one at a time, a process that takes approximately 25 minutes per trial. This limits patient options and reduces clinical trial enrollment rates.
Proposed Solution: In 2023, a collaborative project between PCC and Massive Bio was initiated to identify more eligible patients for biomarker-directed clinical trials by pre-screening all patients with next-generation sequencing (NGS) test results against all cancer trials simultaneously. The American Society of Clinical Oncology (ASCO) has accepted the results generated by this AI technology for online publication. *
A video resource is available that details the nearly twofold increase in potential patient eligibility for trials across all tumor types, as well as the significant reduction in manual effort.
*You can access the abstract on the ASCO website – abstract number e13501