included in the study. At week 14, clinical response and remission rates were 87% and 66%, respectively. Endoscopic response rate was 47%. Among week 14 responders, clinical remission rate at week 54 was 69%. No clinical variables were found to predict either clinical or endoscopic outcomes. On the contrary, higher baseline levels of circulating memory CXCR3+CCR6- CD4+ T cells (Th1 cells) were strongly associated with week 14 clinical response (P=0.0001). Reduced baseline levels of lamina propria memory CXCR3-CCR6+ CD4+ T cells (Th17 cells) and CXCR3+CCR6+ CD4+ T cells (Th1/17 cells) were predictive of endoscopic response (P=0.012 and P=0.005 respectively). Circulating levels of Th1 memory T cells predicted clinical remission in IBD patients at week 54. Conclusions.The results of this exploratory study uncovered a panel of circulating and mucosal immunological predictors of response to vedolizumab treatment. These data provide further insights on the mechanism of action of vedolizumab in IBD patients. 837 PILOT STUDY OF EARLY SCREENING IN AFRICAN AMERICANS IN AN ORGANIZED COLORECTAL CANCER SCREENING PROGRAM BASED ON FECAL IMMUNOCHEMICAL TESTING Theodore R. Levin, Christopher Jensen, Neetu Chawla, Jeffrey K. Lee, Wei K. Zhao, Natalia Udaltsova, Molly Landau, Ariel Herm, Eryn Eby, Charles P. Quesenberry, Douglas A. Corley Background: The US Multi-Society Task Force guideline recommends initiating colorectal cancer (CRC) screening for African Americans (AA) at age 45 due to their higher incidence and earlier onset of CRC compared to other racial/ethnic groups. However, screening partici- pation and test yield in AAs 45-50 years of age has not been well studied. Methods: We conducted a pilot evaluation of participation and test yield in an organized, fecal immunochemical test (FIT)-based CRC screening program in an integrated healthcare setting in 2018. We compared 2 groups of AA health plan members in 4 service areas of Kaiser Permanente Northern California: those 45-50 years without previous CRC screening and those 51-55 years who first started receiving CRC screening invitations at age 51. All study participants received a mailed invitation and FIT kit, and robocall and postcard reminders as needed. FIT-positive respondents were contacted to schedule a follow-up colonoscopy. Electronic health records and databases were accessed to evaluate FIT participation and positivity, colonoscopy follow-up, and the detection of any adenoma, adenoma with advanced histology (villous or tubulovillous adenoma), and CRC (adenocarcinoma). Adenomas were identified using Systematized Nomenclature of Medicine codes and CRC was identified by pathology histology codes and/or International Classification of Disease-10 codes. Multivari- able logistic regression models were used to evaluate differences in risk of screening participa- tion and test yield, adjusted for patient sex. Results: Among 10,301 AAs 45-50 years of age, 32.6% completed a FIT up through 11/13/18, 4.0% had a positive FIT result, 77.8% of those who were FIT positive received a follow-up colonoscopy, and of these, 54.4% had any adenoma, 19.4% had an adenoma with advanced histology, and 1.9% had CRC (Table). By comparison, among 7976 AAs 51-55 years of age, nearly half of whom were FIT screened in 2017, 50.3% completed a FIT, 2.8% tested positive, 65.5% received a diagnostic colonos- copy, and of these, 45.9% had any adenoma, 13.5% had an adenoma with advanced histology, and 0% had CRC. After adjusting for sex, AAs 45-50 who had not previously been screened were less likely to participate in screening but more likely to have a positive FIT result and receive a follow-up colonoscopy after a positive test. The detection of neoplasia did not differ significantly between the groups. Conclusions: FIT participation was lower, but FIT positivity and colonoscopy follow-up were higher among AAs 45-50 years of age without previous CRC screening, compared to AAs 51-55 years of age, nearly half of whom were screened the previous year. Colonoscopy findings were similar between the 2 groups. These results suggest that FIT screening of AAs 45-50 years of age is associated with acceptable participation rates and yield for neoplasia. 838 IS 45 THE NEW 50? ADENOMA DETECTION IN PATIENTS BETWEEN THE AGES OF 45-49 YEARS - WORTH THE GUIDELINE CHANGE? Sonia L. Taneja, Sungyoung Auh, Sheila Kumar Introduction: In 2018, the American Cancer Society's Guideline Development Group updated the 2017 U.S. Multi-Society Task Force of Colorectal Cancer recommendation to begin colorectal cancer (CRC) screening in average risk patients 45 years and older. However, there is limited published data on CRC screening outcomes in adults aged 45-49 years or younger. In this retrospective analysis using a large national endoscopic database, we aimed to see if screening colonoscopies, as well as colonoscopies done on symptomatic patients, had significant findings in this younger patient population. Methods: The Clinical Outcomes Research Initiative (CORI) is a large, multi-center database created to study the utilization and outcomes of endoscopies performed in the United States from 1995 to present. We searched all patients between the ages of 45-49 who underwent colonoscopies for CRC screening and then for symptomatic purposes (melena, hematochezia, abdominal pain, S-183 AGA Abstracts constipation, diarrhea, anemia, or weight loss). Colonoscopy quality measures were recorded. Rates of polyp and tumor detection were noted and if found, the underlying pathology was assessed. Results: Within the CORI database, 10181 colonoscopies were performed in patients aged 45-49 years. 43.5% were male; mean age was 47.3 years. Bowel preparation was rated as excellent/fair/good in 94.3%, and cecum/terminal ileum was reached in 95.2% of procedures. 2685 (26.4%) procedures were done for CRC screening. In this group, 689 procedures (25.7%) had polyps, with a total of 1032 polyps found. 52% of polyps found were at least adenomatous. 3 (0.1%) tumors were found, and at least 1 (33.3%) was adenocarcinoma based on pathology results. 7496 (73.6%) procedures were done for evalua- tion of symptoms. In this group, 1664 (22.2%) had polyps, with a total of 2483 polyps found. 52% of these were at least adenomatous. 56 tumors (0.7%) were found, and at least 37 (66.7%) were adenocarcinoma. Of note, patients with polyps tended to be male (28.0% vs 19.3%, p<0.0001). Incidence of polyps also varied significantly by race (p<0.001, Table 1). Conclusion: In this large, retrospective study, we demonstrate a relatively significant frequency of polyps (specifically adenomas) in patients between the ages of 45-49 years. Although the frequency of cancer was low, the frequency of pre-cancerous polyps detected may suggest a potential benefit to earlier colorectal cancer screening in this age group. Of note, race and gender appear to play a role. Limitations of this study include missing pathology data; however, our results would presumably be strengthened as more adenomas would be identified from detected polyps. Further investigation with randomized control trials will be beneficial to assess the impact this screening will have on overall CRC mortality. 839 DISPARITIES IN COLORECTAL CANCER SCREENING: THE EFFECT OF ACA MEDICAID EXPANSION ON MINORITIES AND LOW-INCOME POPULATIONS Gerald Fletcher, Eric M. Alatevi, Alvaro Genao, Joan A. Culpepper-Morgan, Abraham Ocejo Garcia Background: Colorectal cancer(CRC) is the second leading cause of cancer-related deaths in the United States. Low-income populations and minorities are disproportionately affected and experience a higher incidence of and mortality from CRC. While the U.S. Preventive Services Task Force (USPSTF) recommends screening for all adults aged 50- 75years, screen- ing rates have remained sub-optimal.The data on the effect of the Medicaid Expansion (ME) under the ACA on low income populations and minority groups appears conflicting. The purpose of our study was to investigate the effect of Medicaid expansion under the ACA on CRC screening rates across low-income populations and minorities. Methods: Using a quasi-experimental design, we analyzed data from the Behavioral Risk Factor Surveillance System (BRFSS) from 2011 to 2016 with ME status as the treatment variable. Data from 2011–13 BRFSS constituted the pre-Medicaid expansion while that from 2014-2016 repre- sented post-Medicaid expansion. Self-reported access to at least one of the recommended CRC screening tests (annual high-sensitivity fecal occult blood test (FOBT); sigmoidoscopy every 5 years with high-sensitivity FOBT every 3 years, or colonoscopy every 10 years) were analyzed using a difference-in-differences statistical methodology. The data on change in use of CRC screening tests pre and post ACA by Medicaid eligible respondents were stratified by race and income. Results: 893,004 patients were identified who had undergone at least one kind of CRC screening modality in Medicaid expansion and non-expansion states. Of this number, 62% were in expansion states and 38% in non-expansion states. Of these, 62% had had at least a colonoscopy, 3% had had at least a sigmoidoscopy and 35% had had at least an FOBT. Overall, there was an increase in use of CRC screening tests in Expansion states compared to Non-expansion states (+2.90% CI 95% [2.12, 3.69]). When stratified by income, there was a 4% increase (CI 95% [2.96, 5.07]) in the low-income population of expansion states compared to that of non-expansion. An increase in use of CRC screening tests was found in Non-Hispanic Whites (3.01% CI 95% [2.16, 3.85]) and Hispanics (5.51% CI 95% [2.81, 8.20]). Non-Hispanic Blacks, Multiracial Ethnic groups, and Others in our analysis did not show an uptake in use of CRC screening modalities that was statistically significant (1.72% CI 95% [-0.83,4.26], 4.15% CI 95% [-2.93,11.22] and -1.54% CI 95% [-6.84,3.76] respectively). Conclusion: Our study shows that Medicaid expansion under the ACA has led to an overall increase in self-reported use of colorectal cancer screening tests by adults aged 50-64years in the US. This finding was consistent across low-income populations, but not across all races. Further analysis is needed to investigate other barriers to CRC screening. AGA Abstracts