Age as a surrogate: Mortality and vascular risk in gastrointestinal stromal tumor after frailty and comorbidity adjustment.

M Manaswini Krishnakumar (Saint Vincent Hospital, Worcester, MA) K Kartik Dapke (Mayo Clinic, Jacksonville, Florida, United States) A Arankesh Mahadevan (University of Utah, Salt Lake City, UT) M Madhan Srinivasan Kumar (Saint Vincent Hospital, Worcester, MA) M Masood Pasha Syed (Case Western Reserve University SOM, Cleveland, OH) A Aswanth Reddy (10Mercy Hospital, Fort Smith, United States)

Abstract

e23516 Background: Older adults with gastrointestinal stromal tumor (GIST) have worse survival than younger patients, typically attributed to comorbidity, frailty, or treatment selection rather than age itself. Because prior studies have not isolated the independent effect of age, we conducted an age-stratified cohort study across multiple thresholds using propensity score matching adjusted for validated comorbidity and frailty indices. Methods: We performed a retrospective cohort study using TriNetX (111 healthcare organizations). Incident GIST was identified using ICD-10-CM code C49.A and ICD-O-3 morphology codes 8936/0, 8936/1, and 8936/3 (2018-2022), with a 2-year washout period. Patients with prior malignancies or palliative care were excluded. Four prespecified comparisons were analyzed: ≥65 vs ≤64 years (primary), ≥65 vs 50-64, 50-64 vs ≤49, and ≥65 vs ≤49 years. Cohorts had 1:1 propensity score matching incorporating variables derived from the Charlson Comorbidity Index and Hospital Frailty Risk Score, including demographics, cardiovascular/metabolic comorbidities, renal/pulmonary disease, anemia, substance use, and frailty-related diagnoses. Outcomes assessed from 90 days post-diagnosis included mortality, stroke, acute myocardial infarction (AMI), venous thromboembolism (VTE), healthcare utilization, and tyrosine kinase inhibitor (TKI) exposure. Results: After matching, sample sizes were: ≥65 vs ≤64 (n = 847 pairs), ≥65 vs 50-64 (n = 778), 50-64 vs ≤49 (n = 311), and ≥65 vs ≤49 (n = 234). Despite exhaustive adjustment, patients ≥65 years demonstrated significantly higher mortality versus ≤64 years (OR 1.99 [1.39-2.85], p < 0.001; survival 83.2% vs 89.9%), composite stroke (OR 2.10 [1.23-3.56]), ischemic stroke (OR 2.45 [1.33-4.52]), and AMI (OR 1.85 [1.06-3.23]). Similar patterns were observed for ≥65 vs 50-64 (mortality OR 1.81 [1.23-2.65]; stroke OR 1.96 [1.16-3.31]). Mortality and cardiovascular outcomes could not be analyzed for comparisons involving ≤49 years due to limited events; however, healthcare utilization and TKI exposure showed no significant differences across all four comparisons, including ≥65 vs ≤49 (ED/inpatient OR 0.69 [0.34-1.43]) and 50-64 vs ≤49 (TKI OR 1.13 [0.52-2.46]). Conclusions: Older adults with GIST have higher mortality and cerebrovascular/cardiovascular event rates than younger patients, even after rigorous adjustment for comorbidity and frailty. These differences persist despite similar post-diagnosis healthcare utilization and TKI exposure, arguing against treatment underuse as the primary driver. Risk increased progressively with advancing age, demonstrating a graded age-related effect and suggesting age itself contributes independently to adverse outcomes beyond measured comorbidity and frailty.

Article Details

Volume / Issue Vol. 44, Issue 16_suppl
Published June 01, 2026
ISSN 0732-183X
Publisher Lippincott Williams & Wilkins

Journal Info

Journal of Clinical Oncology

Lippincott Williams & Wilkins

ISSN: 0732-183X Health Sciences

Authors (6)

M

Manaswini Krishnakumar

Saint Vincent Hospital, Worcester, MA

K

Kartik Dapke

Mayo Clinic, Jacksonville, Florida, United States

A

Arankesh Mahadevan

University of Utah, Salt Lake City, UT

M

Madhan Srinivasan Kumar

Saint Vincent Hospital, Worcester, MA

M

Masood Pasha Syed

Case Western Reserve University SOM, Cleveland, OH

A

Aswanth Reddy

10Mercy Hospital, Fort Smith, United States