Introduction: Early diagnosis of pancreatic cancer remains a significant clinical challenge due to lack of recommended screening and vague and non-specific symptoms. The vast majority of patients present with advanced disease, with survival rates less than 5% at 2 years. Leveraging Electronic Medical Records (EMRs) may help identify high-risk patients for early intervention. This study aims to assess the clinical performance of the cancer case finding functionality of the clinical decision support platform, C the Signs. Methods: A retrospective analysis was conducted using data from the Mayo Data Platform between 1st January 2002 and 31st December 2021, comprising 895,361 patients, of whom 3,918 were diagnosed with pancreatic cancer. C the Signs was utilized to identify patients at risk of pancreatic cancer based on EMR data. Sensitivity and specificity analyses were performed to evaluate the platform's performance in identifying high-risk patients. Additionally, we assessed how many patients were diagnosed with pancreatic cancer earlier by C the Signs compared to the diagnoses made by primary care physicians. Results: The analysis revealed a sensitivity of 55.2% and a specificity of 71.6% for C the Signs in identifying patients at risk of pancreatic cancer. Notably, 23.0% of patients with pancreatic cancer were diagnosed up to 5 years before the diagnosis made by primary care physicians, highlighting the potential of early detection facilitated by the platform. Conclusions: This study highlights the potential of clinical decision support platforms like C the Signs in addressing these challenges. Despite modest sensitivity and specificity, given the starting blocks, the platform demonstrated the ability to identify a significant proportion of patients at risk of pancreatic cancer earlier than traditional diagnostic methods. Early identification of high-risk individuals holds promise for improving patient outcomes and reducing the burden of pancreatic cancer.
BIO25-027: Addressing Early Diagnosis Challenges: Utilizing C the Signs Clinical Decision Support Platform for Pancreatic Cancer Risk Assessment
Authors:
,
,
,
,
, and
Sana Raoof
MD, PhD
Sana Raoof
Memorial Sloan Kettering Cancer Center, New York, NY
Search for other papers by Sana Raoof in
Current site
Google Scholar
PubMed
Search for other papers by Sana Raoof in
Current site
Google Scholar
PubMed
Seema Dadhania
MBBS, MRCP
Seema Dadhania
Cancer Research UK, London, United Kingdom
Computational Oncology Group, Institute for Global Health Innovation, Imperial College London, London, United Kingdom
Search for other papers by Seema Dadhania in
Current site
Google Scholar
PubMed
Computational Oncology Group, Institute for Global Health Innovation, Imperial College London, London, United Kingdom
Search for other papers by Seema Dadhania in
Current site
Google Scholar
PubMed
Tushar Patel
MB, ChB
Tushar Patel
Division of Gastroenterology & Hepatology, Department of Internal Medicine, Mayo Clinic, Jacksonville, FL
Search for other papers by Tushar Patel in
Current site
Google Scholar
PubMed
Search for other papers by Tushar Patel in
Current site
Google Scholar
PubMed
Brian Herrick
MD
Brian Herrick
Tufts School of Medicine, Boston, MA
Harvard Medical School, Boston, MA
Search for other papers by Brian Herrick in
Current site
Google Scholar
PubMed
Harvard Medical School, Boston, MA
Search for other papers by Brian Herrick in
Current site
Google Scholar
PubMed
Bea Bakshi
MBBS, MRCGP, BSc
Bea Bakshi
C the Signs, London, United Kingdom
Search for other papers by Bea Bakshi in
Current site
Google Scholar
PubMed
Search for other papers by Bea Bakshi in
Current site
Google Scholar
PubMed
Miles Payling
MBBS, BSc, MRCP
Miles Payling
C the Signs, London, United Kingdom
Search for other papers by Miles Payling in
Current site
Google Scholar
PubMed
Search for other papers by Miles Payling in
Current site
Google Scholar
PubMed
© 2019-2024 National Comprehensive Cancer Network