OC 36.3 - Epidemiologic Risk Predictors of Venous Thromboembolism (VTE) among Cancer Patients Treated at U.S. Veterans Affairs (VA) Healthcare System

Monday, June 26, 2023

11:15 – 11:30 ET

Room: 510

Session Available As: On-Demand

Presenting Author(s)

Kylee Martens, MD (she/her/hers)

Fellow Physician
Oregon Health & Science University
Portland, Oregon, United States

Co-Author(s)

Ang Li, MD, MS

Assistant Professor
Section of Hematology-Oncology, Baylor College of Medicine
Houston, Texas, United States
JL

Jennifer La

Principal Data Scientist
U.S. Department of Veterans Affairs
Boston, Massachusetts, United States
SM

Sarah May

Statistician
Baylor College of Medicine
Houston, Texas, United States
ND

Nhan V Do

Director of the Informatics Center
Massachusetts Veterans Epidemiology Research and Information Center (MAVERIC)
Boston, Massachusetts, United States
MB

Mary T Brophy

Assistant Professor, Medicine
Boston University; VA Boston Healthcare System
Boston, Massachusetts, United States
JG

J. Michael Gaziano

Adjunct Professor
Boston University Chobanian & Avedisian School of Medicine
Boston, Massachusetts, United States
SL

Saran Lotfollahzadeh

MD, AHA SFRN-Fellow
Renal and Vascular Research Section, Department of Medicine, Boston University Chobanian & Avedisian School of Medicine, Boston, MA 02118, USA
Boston, Massachusetts, United States
VC

Vipul Chitalia

MD, PhD, Associate Professor of Medicine
Renal Section, Department of Medicine, Boston University Chobanian & Avedisian School of Medicine, Boston, MA 02118, USA,Veterans Affairs Boston Healthcare System, Boston MA 02118, USA, Institute of Medical Engineering and Sciences, Massachusetts Institute of Technology, Cambridge, MA, 02139, USA
Boston, Massachusetts, United States
KR

Katya Ravid

DSc, PhD, Barbara E. Corkey Professor of Medicine
Department of Medicine and Whitaker Cardiovascular Institute, Boston University Chobanian & Avedisian School of Medicine, Boston, MA 02118, USA
Boston, Massachusetts, United States
NF

Nathanael Fillmore

Associate Director (Data Science)
VA Boston Healthcare System, Harvard Medical School
Boston, Massachusetts, United States

Background: Several recent retrospective studies reported increasing incidence of cancer-associated thrombosis (CAT) over time; however, reliance on International Classification of Diseases (ICD) codes to capture VTE is challenging due to their evolving nature. This, coupled with the evolution of cancer therapy, highlights the importance of accurately identifying risk predictors of CAT to guide prognostic assessment in the current era.

Aims: Identify epidemiologic patterns of CAT incidence and risk in a racial/ethnically and geographically diverse population.

Methods: We conducted a retrospective cohort study of patients with newly diagnosed cancer at VA healthcare system from 2006-2021. VTE was captured using a rigorously validated algorithm based on ICD codes, anticoagulant medications, and natural language processing radiology reports in inpatient and outpatient encounters (PPV 90%, Sensitivity 96%). Cumulative incidence competing risk functions were used to estimate CAT incidence. Multivariable Cox regression models were built to assess the association between CAT and variables at diagnosis (demographics, NCI comorbidity index, region, rurality, cancer type, staging) or at first-line systemic treatment within 3 months (time-varying covariate).

Results: A total of 434,203 patients met inclusion criteria. The overall incidence of CAT ranged from 4-5% per year over the 15-year study period (Figure 1). In multivariable model, risk of CAT was significantly higher in Non-Hispanic Blacks (HR 1.23 [95% CI 1.19-1.27]) and lower in Asian Pacific Islanders (HR 0.84 [95% CI 0.71-0.93]) compared to Non-Hispanic Whites, with significant heterogeneity across cancer types (Figure 2). Compared with no treatment, first-line chemotherapy (HR 1.44 [95% CI 1.40-1.49]) and immune checkpoint inhibitor (HR 1.49 [95% CI 1.22-1.82]) had higher relative risk than targeted (1.21 [95% CI 1.13-1.30]) or endocrine therapy (1.21 [95% CI 1.12-1.28]).

Conclusion(s): CAT incidence within this large United States cohort remained relatively stable over 15 years. Both novel and known risk predictors were identified, providing valuable insights applicable in the current treatment landscape.

Learning Objectives:

Define changes in the incidence of cancer-associated thrombosis (CAT) over time
Identify pertinent patient-, treatment-, and cancer-specific factors associated with risk of CAT
Characterize differences in patterns of CAT among patients with hematologic malignancies