2026 ASH®
Annual Meeting & Exposition December 12-15, 2026 ▪ New Orleans, Louisiana

Paper Details

Abstracts, authors, affiliations, and disclosures.

Saturday, December 12, 2026
09:30 a.m. - 11:00 a.m.
MCCNO - 208-210
Technology and Procedures, Molecular Testing, Treatment Considerations, Artificial Intelligence (AI), Natural Language Processing, Biological Processes, Genomics, Diseases, Bleeding and Clotting, Thromboembolism, Thrombotic Disorders, Lymphoid Malignancies, Plasma Cell Disorders, Research, Translational Research, Clinical Research, Clinical Trials, Epidemiology, Health Outcomes Research, Real-World Evidence, Study Population, Human, Adult, Elderly, Young Adult
PAPER

32 Renal infarcts are associated with an increased risk of subsequent stroke in patients with cancer

Divyash Viral Shah, Adriana Wiggins2, Siddhart Kumar3, Kayan Irani4, Andriy Derkach, PhD5, Simon Mantha, MD, MPH6, Jeffrey Zwicker, MD7, Avi Leader, MD8
1 Memorial Sloan Kettering Cancer Center, New York, New York, United States
2 Memorial Sloan Kettering Cancer Center, New York, New York, United States
3 Memorial Sloan Kettering Cancer Center, New York, New York, United States
4 Memorial Sloan Kettering Cancer Center, New York, New York, United States
5 Memorial Sloan Kettering Cancer Center, New York, New York, United States
6 Memorial Sloan Kettering Cancer Center, New York, New York, United States
7 Memorial Sloan Kettering Cancer Center, New York, New York, United States, Weill Cornell Medical College, New York, New York, United States
8 Memorial Sloan Kettering Cancer Center, New York, New York, United States, Weill Cornell Medical College, New York, New York, United States

Introduction

Patients with cancer have an increased risk of venous thromboembolism (VTE) and arterial thromboembolism (ATE). Renal infarcts are thrombotic events often attributed to thromboembolic disease or in situ thrombosis. Data on risk of systemic thrombosis and effect of anticoagulation on outcomes are limited overall and lacking in the context of cancer.  We hypothesized that renal infarcts could indicate embolic risk in the context of cancer and would be associated with increased stroke risk.

 

Aim

Compare the incidence of ischemic stroke at 12 months in matched patients with cancer, with and without renal infarcts.

 

Methods

Using a physician-developed large language model (LLM), we identified patients with active cancer and a radiology report (MRI or CT) indicating renal infarct, as well as potential controls, via negation-aware natural language processing (NLP). Patients with renal infarcts were identified at the first mention of renal infarct in a radiology report and indexed at this date. Potential controls were indexed on the day of abdominal imaging without renal infarcts. Controls were matched hierarchically: primary cancer type; treatment type (systemic, surgery, radiation); number of days between cancer diagnosis and index (less than 180 days difference); age +/- 10 years; index dates within 6 months of each other; sex.

 

The primary and secondary outcome of ischemic stroke or VTE was identified using the CEDARS NLP application that uses predetermined keywords to identify potential clinical events, which were verified by a hematologist. Manual chart review was used to cross-confirm LLM matching criteria and NLP-identified events, and to collect baseline variables and 12-month renal outcomes. Records were followed for 12 months or death. Cumulative incidence of stroke and VTE was assessed with death as a competing risk. Cox proportional hazards models were used to calculate hazard ratios (HR) with 95% confidence intervals (CI) between the exposure groups. Patients with concurrent strokes or VTE (i.e., on index date) and their pair were excluded from the stroke and VTE analyses, respectively.

 

Results

We included 260 patients with active cancer (2011-2025), 130 with renal infarcts, and 130 matched controls without renal infarcts. Median age was 62 years (IQR: 51- 70), and 137 (53%) patients were female. The most common cancer types included pancreatic (63, 24%), lung (40, 15%), and gastric (26, 10%), and 150 (58%) had metastatic disease at index. Median time between cancer diagnosis and index was 92 days (IQR: 28-340.5). Atrial fibrillation was present in 13 patients (6%). Gross hematuria was reported in 10 (4%) patients, and median creatinine was 0.8 mg/dL (IQR 0.6-1). These baseline variables and treatment type were comparable between groups.

More patients with renal infarcts had ATE or VTE prior to index compared to controls (18 [14%] vs. 7 [5%] and 44 [34%] vs. 17 [13%], respectively). Consequently, anticoagulation use differed in the 30 days before index; 42 (33%) patients with renal infarcts and 17 (13%) controls. Renal infarcts were equally distributed between left, right, and bilateral kidneys.  

 

The 12-month cumulative incidence of ischemic stroke was 9.3% (95% CI 5.1-15%) in the renal infarct cohort (n=12) and 0.78% (95% CI 0.07-3.9%) in controls (n=1); HR 17.0 (95% CI 2.20-131.0, p= 0.002). Median time to stroke was 14 days (IQR: 5-25). Two sensitivity analyses excluding patients on anticoagulation pre-index or those with prior ATE yielded similar results. Ten (77%) of 13 strokes were symptomatic, and 11 (85%) were non-lacunar. Among 52 patients with infarcts (41%) prescribed anticoagulation within 5 days after index, the 12-month stroke incidence remained high (12% [95% CI 4.6-22%]).

The 12-month cumulative incidence of VTE was 17% (95% CI 11-25%) in the renal infarct cohort and 12% (95% CI 6.5-18%) in controls (HR 1.93; 95% CI 0.95–3.91, p=0.2). There was no difference between groups in acute kidney injury, new-onset chronic kidney disease, or dialysis during follow-up.

 

Conclusion

In this study of patients with cancer, renal infarcts were associated with a higher 12-month incidence of ischemic stroke, but not VTE, compared to rigorously matched controls. This suggests an embolic nature to renal infarcts in the context of active cancer. Most strokes occurred within one month, indicating a narrow window for intervention. Research is needed on diagnostic work-up and antithrombotic therapy.

Divyash Viral Shah: No relevant conflicts of interest to declare.