Wansu Chen1, Rebecca K Butler1, Eva Lustigova1, Suresh T Chari2, Anirban Maitra3, Jo A Rinaudo4, Bechien U Wu5. 1. Kaiser Permanente Southern California Research and Evaluation, Pasadena. 2. Department of Gastroenterology, Hepatology, and Nutrition. 3. Sheikh Ahmed Center for Pancreatic Cancer Research, University of Texas MD Anderson Cancer Center, Houston, TX. 4. Division of Cancer Prevention, National Cancer Institute, Bethesda, MD. 5. Department of Gastroenterology, Center for Pancreatic Care, Los Angeles Medical Center, Southern California Permanente Medical Group, Los Angeles, CA.
Abstract
BACKGROUND: New-onset diabetes (NOD) has been suggested as an early indicator of pancreatic cancer. However, the definition of NOD by the American Diabetes Association requires 2 simultaneous or consecutive elevated glycemic measures. We aimed to apply a machine-learning approach using electronic health records to predict the risk in patients with recent-onset hyperglycemia. MATERIALS AND METHODS: In this retrospective cohort study, health plan enrollees 50 to 84 years of age who had an elevated (6.5%+) glycated hemoglobin (HbA1c) tested in January 2010 to September 2018 with recent-onset hyperglycemia were identified. A total of 102 potential predictors were extracted. Ten imputation datasets were generated to handle missing data. The random survival forests approach was used to develop and validate risk models. Performance was evaluated byc-index, calibration plot, sensitivity, specificity, and positive predictive value. RESULTS: The cohort consisted of 109,266 patients (mean age: 63.6 y). The 3-year incidence rate was 1.4 (95% confidence interval: 1.3-1.6)/1000 person-years of follow-up. The 3 models containing age, weight change in 1 year, HbA1c, and 1 of the 3 variables (HbA1c change in 1 y, HbA1c in the prior 6 mo, or HbA1c in the prior 18 mo) appeared most often out of the 50 training samples. Thec-indexes were in the range of 0.81 to 0.82. The sensitivity, specificity, and positive predictive value in patients who had the top 20% of the predicted risks were 56% to 60%, 80%, and 2.5% to 2.6%, respectively. CONCLUSION: Targeting evaluation at the point of recent hyperglycemia based on elevated HbA1c could offer an opportunity to identify pancreatic cancer early and possibly impact survival in cancer patients.
BACKGROUND: New-onset diabetes (NOD) has been suggested as an early indicator of pancreatic cancer. However, the definition of NOD by the American Diabetes Association requires 2 simultaneous or consecutive elevated glycemic measures. We aimed to apply a machine-learning approach using electronic health records to predict the risk in patients with recent-onset hyperglycemia. MATERIALS AND METHODS: In this retrospective cohort study, health plan enrollees 50 to 84 years of age who had an elevated (6.5%+) glycated hemoglobin (HbA1c) tested in January 2010 to September 2018 with recent-onset hyperglycemia were identified. A total of 102 potential predictors were extracted. Ten imputation datasets were generated to handle missing data. The random survival forests approach was used to develop and validate risk models. Performance was evaluated byc-index, calibration plot, sensitivity, specificity, and positive predictive value. RESULTS: The cohort consisted of 109,266 patients (mean age: 63.6 y). The 3-year incidence rate was 1.4 (95% confidence interval: 1.3-1.6)/1000 person-years of follow-up. The 3 models containing age, weight change in 1 year, HbA1c, and 1 of the 3 variables (HbA1c change in 1 y, HbA1c in the prior 6 mo, or HbA1c in the prior 18 mo) appeared most often out of the 50 training samples. Thec-indexes were in the range of 0.81 to 0.82. The sensitivity, specificity, and positive predictive value in patients who had the top 20% of the predicted risks were 56% to 60%, 80%, and 2.5% to 2.6%, respectively. CONCLUSION: Targeting evaluation at the point of recent hyperglycemia based on elevated HbA1c could offer an opportunity to identify pancreatic cancer early and possibly impact survival in cancer patients.
Authors: Ben Boursi; Brian Finkelman; Bruce J Giantonio; Kevin Haynes; Anil K Rustgi; Andrew D Rhim; Ronac Mamtani; Yu-Xiao Yang Journal: Gastroenterology Date: 2016-12-05 Impact factor: 22.682
Authors: Gregory A Nichols; Emily B Schroeder; Andrew J Karter; Edward W Gregg; Jay Desai; Jean M Lawrence; Patrick J O'Connor; Stanley Xu; Katherine M Newton; Marsha A Raebel; Ram D Pathak; Beth Waitzfelder; Jodi Segal; Jennifer Elston Lafata; Melissa G Butler; H Lester Kirchner; Abraham Thomas; John F Steiner Journal: Am J Epidemiol Date: 2014-12-16 Impact factor: 4.897
Authors: Ayush Sharma; Thomas C Smyrk; Michael J Levy; Mark A Topazian; Suresh T Chari Journal: Gastroenterology Date: 2018-04-30 Impact factor: 22.682
Authors: Suresh T Chari; Cynthia L Leibson; Kari G Rabe; Lawrence J Timmons; Jeanine Ransom; Mariza de Andrade; Gloria M Petersen Journal: Gastroenterology Date: 2007-10-26 Impact factor: 22.682
Authors: Douglas K Owens; Karina W Davidson; Alex H Krist; Michael J Barry; Michael Cabana; Aaron B Caughey; Susan J Curry; Chyke A Doubeni; John W Epling; Martha Kubik; C Seth Landefeld; Carol M Mangione; Lori Pbert; Michael Silverstein; Melissa A Simon; Chien-Wen Tseng; John B Wong Journal: JAMA Date: 2019-08-06 Impact factor: 56.272