CERSI Collaborators: Joanne Dorgan, PhD, MPH; Eberechukwu Onukwugha, MS, PhD; Eun-Shim Nahm, PhD, RN, FAAN; Abree Johnson, MS, MBA; Tsung-Ying Lee, BPharm, MClinPharm

FDA Collaborators: Donna Rivera, PharmD, MSc; Bindu Kanapuru, MD; Catherine Lerro, MPH, PhD; Jonathon Vallejo, PhD; Fatima Rizvi, PharmD

Project Start Date: October 1, 2021

Regulatory Science Challenge

Important advances in screening and treatment have improved survival among US cancer patients, however, survival disparities persist among the elderly and racial and ethnic minorities. Based on Surveillance, Epidemiology and End Results (SEER) cancer statistics, in 2012-2018, the five-year relative survival for all cancers (except non-melanoma skin cancer) in the US is poorer for patients aged 75+ (51.4%) compared to patients aged 40-64 years (73.6%).1 The overall age-adjusted cancer mortality rate is higher among African Americans (177.5 per 100,000) compared to Whites (156.3 per 100,000). In addition, because older and minority patients are underrepresented in oncology clinical trials, evidence on the relative efficacy and toxicity of different therapies in these population groups is limited.2 Furthermore, information is needed to: 1) quantify demographic discrepancies between real-world and clinical trial populations; 2) assess, for any given disease, to what extent approved therapies are utilized and what might in practice be considered "available therapy"; and 3) generate hypotheses about the possible reasons certain groups do not participate in clinical trials beyond what is already known. In recognition of the information gap, there is a need for large prospective cohorts of cancer patients with information on demographics, treatment, and tumor characteristics coupled with long-term follow-up to identify multifactorial determinants of disparities in cancer outcomes to advance regulatory science.

Project Description and Goals

The investigators will link data on eligible patients identified in the University of Maryland Greenebaum Comprehensive Cancer Center (UMGCCC) tumor registry with University of Maryland Medical Center (UMMC) Epic electronic health record (EHR) data and Centers for Medicare and Medicaid Services (CMS) Medicare administrative claims data. The objective of this project is to construct the UMGCCC-Medicare database, a prospective cohort of cancer patients aged 65 years and older at diagnosis with data on tumor, clinical, sociodemographic, and behavioral characteristics, as well as availability of stored tumor tissue, healthcare resource utilization, and outcomes. This database will support research in two frequently underrepresented population subgroups: 1) the elderly, and 2) racial and ethnic minorities.

This project has two goals. The first is to create the infrastructure of the UMGCCC-Medicare database to support multidisciplinary translational research to identify determinants of disparities in cancer outcomes among older adults related to the individual, community, treatment, and tumor. Using record linkage to create a prospective cohort of Medicare beneficiaries with linked data on tumor, clinical, sociodemographic and behavioral characteristics as well as access to tumor tissue will significantly advance research capacity at UMGCCC to address multilevel determinants of cancer outcomes. The second goal is to demonstrate proof of concept for the utility of the UMGCCC-Medicare database for cancer health services research. This will be illustrated by evaluating the similarity between a claims-based and a medical record-based comorbidity measure.

The project aligns with the Oncology Center of Excellence (OCE) scientific collaborative interest in enriching oncology real-world data and improving health equity. The UMGCCC catchment area is racially diverse. Approximately 50% of residents in the UMGCCC catchment area are racial/ethnic minorities (33% African Americans and 11% Hispanics) and 16% are ≥ 65 years old while 6% are ≥ 75 years.3 The UMGCCC-Medicare database will address the regulatory science need for increasing high-quality data on underrepresented populations.

Anticipated Research Impacts

The UMGCCC-Medicare linked database will enhance resources and capacity for FDA post-market drug surveillance activities among older Americans (65+ years old) and among racial and ethnic minorities. The linked database will provide FDA and UMB collaborators with an efficient and accessible platform to generate real-world evidence for regulatory science questions on cancer treatment outcomes, including long-term outcomes, in a racially and ethnically diverse population of older Americans. The linked dataset has the potential to serve as proof-of-concept and provide a reproducible approach to expanding surveillance at additional cancer centers, potentially creating a network with linked data. In addition, the database provides opportunities to deliver training on claims data, EHR data, registry data, and biorepository data specially as well as on linked data.

References

1. SEER*Explorer: An interactive website for SEER cancer statistics. Surveillance Research Program, National Cancer Institute. [Cited January 13, 2023]. https://seer.cancer.gov/statistics-network/explorer/.
2. Duma N, Vera Aguilera J, Paludo J, et al. Representation of Minorities and Women in Oncology Clinical Trials: Review of the Past 14 Years. JOP. 2018;14(1):e1-e10. doi:10.1200/JOP.2017.025288
3. United States Census Bureau. American Community Survey S0101: AGE AND SEX: 2021 ACS 1-Year Estimates Subject Tables. Accessed January 17, 2023. https://data.census.gov/