Citations

e18635 Background: With almost 85% of patients treated outside NCI-designated cancer centers, providing comprehensive, state-of-the-art cancer care to millions of patients in the U.S. remains a significant challenge. We aimed to combine cutting edge AI based technology of DLVTB with high value-added services to provide evidence-based care management recommendations and implementation support to the patients and physicians in the community. To further demonstrate the feasibility, reproducibility, scalability, and benefits of DLVTB, we evaluated key outcomes from incorporating DLVTB in a cohort of 35 patients with advanced colorectal adenocarcinoma (CRC). Methods: Our core-enabling technology is a deep learning based Natural Language Understanding Engine that employs (1) natural language processing for medical text digestion and structuring, (2) decision trees and multilayer perception models to produce evidence-based treatment protocols, and (3) natural language based report generation. The technology platform is combined with human support from oncology sub-specialists to deliver a comprehensive Interpretive Report with a prioritized list of recommendations for each patient. These recommendations are operationalized by a case management team to ensure care implementation and monitoring of outcomes. Results: Thirty-five patients with CRC were referred for incorporation of DLVTB into clinical practice. Median age of patients was 57 years with 68.6% males. About 88.6% of the patients were Stage IV and 82.9% were treated by community practice oncologists. Median time since diagnosis was 17 months (1-73 months). Overall, DLVTB-cohort demonstrated an increase in median Overall Survival of 12 months per patient in comparison with historical cohorts. More specifically, DLVTB recommended initial and/or additional biomarker testing for 71% of the patients with further precision oncology-guided treatment recommendation (e.g. an EGFR inhibitor) for 80% of the patients. 63% of the patients were eligible for at least one clinical trial. 58% of the trials identified were within proximity (≤50 miles) of the patient’s primary residence. Thus, DLVTB was able to identify a trial that did not require extensive travel, provided eligibility and actionability closer to the point-of-care. 14% of the DLVTB evaluated patients subsequently enrolled in the recommended clinical trial, far surpassing the national average (3%). Moreover, DLVTB recommended treatments achieved on average savings of $39,194 per patient which is 35% of the average drug cost per patient. Conclusions: Our results demonstrate the feasibility and benefits of incorporating DLVTB into clinical practice. The utilization of DLVTB as a clinical trial enrollment tool and an engine for development of pathways resulting in improved clinical outcomes, in a cost-effective, and innovative care delivery model.

e13573 Background: UpToDate is an evidence based clinical resource designed to provide current clinical information. It is a widely used clinical practice tool providing evidenced based recommendations for diagnosis, management, and therapy. The extent of COI among the UpToDate authors has not been well studied. Our study evaluated the extent of COI of UpToDate authors involved in medical management recommendations for the 10 most common cancers in United States. Methods: We examined the latest November 2020 version of the UpToDate clinical management recommendations for 10 most common cancers. Using disclosure lists, we catalogued COIs for participants in each work group. The categories included: Category I- Grant/Research/Clinical Trial Support; Category II- Consultant/Scientific Advisory Boards/Expert witness; Category III- Speakers Bureau; Category IV- Equity Ownership/Stock Options/Patent holder/Employment/Other Financial Interest; and Category V- Spouse/Domestic Partner/Dependent Potential Conflict. We cataloged the companies/institutions involved in each disclosure. An “episode” describes 1 instance of participation of an individual in 1 company in 1 category for each cancer section. Results: There was a total of 207 authors including section and deputy editors of oncology management section. All authors completed their COI disclosures (100%). 128 (62%) of a total of 207 individuals reported COIs. A total of 1343 episodes were disclosed. We found that each author had an average of 10.49 episodes overall. Authors involved in Category I, II, III, IV & V were 6.3%, 13.5%, 3.2%, 4.7% and 1.6% respectively. 29.36% authors were involved both in Category I and II. A total of 672 company affiliations were associated with COI disclosure. AstraZeneca (6.10%), Merck (4.31%) and Novartis (2.68%) were the companies most frequently reported. The guideline with the maximum episodes (223) was prostate cancer. Conclusions: COIs are prevalent among authors of UpToDate clinical management recommendations. More than ½ of the participants disclosed at least 1 COI, but there appears to be a substantial number of experienced experts without COIs. Further research studies are necessary to determine the impact of these COIs on clinical practice patterns and resource utilization. Distribution of COI and total episodes.[Table: see text]

1503 Background: Patients with cancer and COVID-19 are at risk for poor clinical outcomes. An established multi-site remote patient monitoring (RPM) service was rapidly adapted to support a novel, interdisciplinary COVID-19 program for outpatient management of patients at high-risk for severe illness. The goal of this study was to assess the impact of the RPM program on clinical outcomes and acute care utilization in cancer patients diagnosed with COVID-19. Methods: This is a cross-sectional analysis following a multi-site prospective observational study performed at Mayo Clinic Cancer Center (MCCC). All adult patients with active cancer – defined as currently receiving cancer-directed therapy or in recent remission on active surveillance – and PCR-confirmed SARS-CoV-2 infection between March 18 and July 31, 2020 were included. RPM was comprised of in-home technology to assess symptoms and physiologic data with centralized nurse and physician oversight. Results: During the study timeframe 224 cancer patients were diagnosed with COVID-19 at MCCC. Initial management included urgent hospitalization (within 48 hours of diagnosis) in 34 patients (15%). Of the remaining 190 patients (85%) initially managed in the outpatient setting, those who did not receive RPM were significantly more likely to experience hospitalization than those receiving RPM (OR 3.6, 95% CI 1.036 to 12.01, P = 0.044). Following balancing of patient characteristics by inverse propensity weighting, rates of hospital admission for RPM and non-RPM patients were 3.1% and 11% respectively, implying that RPM was associated with an 8% reduction in hospital admission rate (-0.077; 95% CI: -0.315 to -0.019, P = 0.009). Use of RPM was also associated with lower rates of prolonged hospitalization, ICU admission, and mortality, though these trends did not reach statistical significance. Conclusions: In the midst of a global pandemic associated with inpatient bed, ventilator, and PPE shortages, the RPM program provided an effective strategy for outpatient clinical management and was associated with decreased rates of hospitalization, ICU admission, and mortality in cancer patients with COVID-19. This care model enabled simultaneous opportunity to mitigate the increased risks of exposure, transmission, and resource utilization associated with conventional care.

Background: National Comprehensive Cancer Network (NCCN) guidelines are the most comprehensive and widely used standard for clinical care in malignant hematology by clinicians and payers in the US. The level of scientific evidence in NCCN guidelines for malignant hematological conditions has not been recently investigated. We describe the distribution of categories of evidence and consensus (EC) among the 10 most common hematologic malignancies with regard to recommendations for staging, initial and salvage therapy, and surveillance. Methods: NCCN uses a system of guideline development distinct from other major professional organizations. The NCCN definitions for EC are: category I, high level of evidence such as randomized controlled trials with uniform consensus; category IIA, lower level of evidence with uniform consensus; category IIB, lower level of evidence without a uniform consensus but with no major disagreement; and category III, any level of evidence but with major disagreement. We compared our results with previously published results from 2011 guidelines. Results: Total recommendations increased by 16.6% from 1160 (2011) to 1353 (2020). Of the 1353 recommendations, Category 1, 2A, 2B and 3 EC were 5%, 91%, 4%, 1% while in 2011 they were 3%, 93%, 4% and 0% respectively. Recommendations with category 1 EC were found in all guidelines, except for Burkitt's Lymphoma. 6.3% of therapeutic recommendations were category 1 EC with the majority (56.4%) pertaining to initial therapy. Guidelines with highest proportions of therapeutic recommendations with category 1 EC were Multiple Myeloma (12.4%), CLL/SLL (6.9%) and AML (5.6%). Between 2011 and 2020, the proportion of category I recommendations increased significantly only in Follicular lymphoma and CLL/SLL. No category 1 EC recommendations existed in staging or surveillance. Conclusion: Recommendations issued in the 2020 NCCN guidelines are largely developed from lower levels of evidence but with uniform expert opinion. Despite the major advances in hematology in the past decade, this is largely unchanged. Our study underscores the urgent need and available opportunities to expand the current evidence base in malignant hematological disorders which forms the platform for clinical practice guidelines. Figure Disclosures No relevant conflicts of interest to declare.

Abstract Background Data on efficacy of hepatitis-B vaccine (HBV) in patients with inflammatory bowel disease (IBD) is limited. Our aim was to review the literature and perform meta-analysis of available studies to quantify efficacy of HBV in patients with IBD. Methods We conducted a comprehensive search of several databases (inception to July 2020) to identify studies evaluating efficacy of HBV in patients with IBD. Random effects model was used to calculate the pooled rates and I2 percentage values were used to assess the heterogeneity. Results A total of 14 studies (2375 patients) were included. Four data sets were available from 2 studies that compared HBV response in patients with IBD against healthy controls. The pooled odds ratio of HBV response in IBD patients was 0.13 (95% CI, 0.05–0.33, P = 0.001). The pooled proportion of adequate immune response (AIR) was 64% (95% CI, 55–72.1, P = 0.003) from 13 data sets, and effective immune response (EIR) was 39.7% (95% CI, 30.7–49.5, P = 0.04) from 10 data sets. Conclusion Patients with IBD on immunosuppression demonstrated significantly reduced HBV response as compared with general population.

Background Heterogeneous evidence exists on the effect of coronavirus disease 2019 (COVID‐19) on the clinical outcomes of patients with cancer. Methods A systematic review was performed using the Medline, Embase, and CENTRAL databases and the World Health Organization Novel Coronavirus website to identify studies that reported mortality and characteristics of patients with cancer who were diagnosed with COVID‐19. The primary study outcome was mortality, defined as all‐cause mortality or in‐hospital mortality within 30 days of initial COVID‐19 diagnosis. The pooled proportion of mortality was estimated using a random‐effects model, and study‐level moderators of heterogeneity were assessed through subgroup analysis and metaregression. Results Among 2922 patients from 13 primarily inpatient studies of individuals with COVID‐19 and cancer, the pooled 30‐day mortality rate was 30% (95% CI, 25%‐35%). The overall pooled 30‐day mortality rate among 624 patients from 5 studies that included a mixture of inpatient and outpatient populations was 15% (95% CI, 9%‐22%). Among the hospitalized studies, the heterogeneity (I 2 statistic) of the meta‐analysis remained high (I 2 , 82%). Cancer subtype (hematologic vs solid), older age, male sex, and recent active cancer therapy each partially explained the heterogeneity of mortality reporting. In multivariable metaregression, male sex, along with an interaction between the median patient age and recent active cancer therapy, explained most of the between‐study heterogeneity (R 2 , 96%). Conclusions Pooled mortality estimates for hospitalized patients with cancer and COVID‐19 remain high at 30%, with significant heterogeneity across studies. Dedicated community‐based studies are needed in the future to help assess overall COVID‐19 mortality among the broader population of patients with cancer.