P2125 - Prevalence of Non-Pancreatic Incidental Findings Noted on MRI and EUS Performed for Pancreatic Cancer Screening

Andy Silva-Santisteban, MD¹, Ryan B. Perumpail, MD¹, Ishani Shah, MD¹, Vaibhav Wadhwa, MD², Katharine Germansky, MD¹, Saad Alrajhi, ¹, Mohammad Bilal, MD³, Jack F. Mlabasati, MD¹, Cinthana Kandasamy, MD¹, Dora Huang, MD¹, Tyler M. Berzin, MD, FACG¹, Douglas Pleskow, MD¹, Abraham F. Bezuidenhout, MD⁴, Leo Tsai, MD¹, Mandeep S. Sawhney, MD^1
1Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA; ²Center for Advanced Endoscopy, Beth Israel Deaconess Medical Center, Harvard Medical School, Boston, MA; ³Minneapolis VA Health Care System, Minneapolis, MN; ⁴Beth Israel Deaconess Medical Center, Boston, MA

Introduction: National guidelines recommend EUS and MRI for pancreatic cancer screening in high-risk individuals with genetic susceptibility to pancreatic cancer. While studies have assessed the diagnostic yield for pancreatic pathology, there are limited data on prevalence of incidental findings, which can impact cost-effectiveness of screening and patient preferences. Our goal was to determine the prevalence and clinical impact of non-pancreatic incidental findings noted on screening MRI and EUS.

Methods: We searched our institutional database and identified all patients who enrolled in pancreatic cancer screening from 2015 to 2020. We then selected 100 consecutive patients who underwent MRCP and EUS for screening. For patients who underwent more than one test, the first test was used for analysis. Incidental findings were categorized as “potentially serious” or “not serious” as defined by Gibson LM et al (PMID: 30467245). We then reviewed medical records to determine clinical impact of incidental findings by categorizing further assessment as: no follow-up need, referred to specialist, non-invasive testing recommended, or invasive test/procedure recommended.

Results: We identified 90 of 100 consecutive patients who underwent MRI for pancreatic cancer screening for whom complete data was available. We then identified 90 consecutive patients who underwent EUS for screening. Patient demographics and indication for screening are given in Table 1. The average age of study subjects was 62 years, and most were Caucasian and female. Patients who underwent MRI were almost 5-times more likely to have incidental findings when compared to EUS group (RR=4.9; 84% versus 20%, p< 0.001). Most findings in both groups were not serious. In the MRI group, the most common non-serious findings were liver and kidney cysts. Potentially serious findings were noted in 17% of patients undergoing MRI and 7% of patients undergoing EUS (OR=2.8, p=0.03). Clinical follow-up of incidental findings was needed in 32% in MRI group and 4% in EUS group (p< 0.001). Invasive testing for incidental findings was noted in 5.6% in MRI group versus 1% in EUS group (p=0.1).

Discussion: An almost 5-fold difference in prevalence of incidental findings was noted between patients undergoing MRI and EUS for pancreatic cancer screening. These findings should be discussed with patients when counseling them on choosing between these modalities.


Disclosures:


Andy Silva-Santisteban, MD¹, Ryan B. Perumpail, MD¹, Ishani Shah, MD¹, Vaibhav Wadhwa, MD², Katharine Germansky, MD¹, Saad Alrajhi, ¹, Mohammad Bilal, MD³, Jack F. Mlabasati, MD¹, Cinthana Kandasamy, MD¹, Dora Huang, MD¹, Tyler M. Berzin, MD, FACG¹, Douglas Pleskow, MD¹, Abraham F. Bezuidenhout, MD⁴, Leo Tsai, MD¹, Mandeep S. Sawhney, MD¹. P2125 - Prevalence of Non-Pancreatic Incidental Findings Noted on MRI and EUS Performed for Pancreatic Cancer Screening, ACG 2021 Annual Scientific Meeting Abstracts. Las Vegas, Nevada: American College of Gastroenterology.